Acknowledgments
The work from the authors’ own laboratories described in this review was supported by grants from The Wellcome Trust, Medical Research Council, British Heart Foundation, The Deutsche Forschungsgemeinschaft, and The European Union ONE Study, Optistem, TRIAD, AFACTT (European Union COST action – BM1305), and BioDRIM networks.
Introduction
Historical Perspective
In 1951 Billingham and Medawar published a landmark article entitled “The Technique of Free Skin Grafting in Mammals.” These classic experiments provide the foundation for what would become the field of transplantation immunology and the groundwork for many concepts in modern immunology, including immunologic memory. Further work by Medawar and his team, based upon earlier writings of Ray D. Owen, involved skin grafting dizygotic mammalian twin calves. The observation that these grafts were accepted by both hosts led to the hypothesis that a phenomenon of immunologic tolerance to the skin grafts was achieved secondary to “foreign” blood cells that persisted in each twin as a consequence of placental fusion.
These breakthroughs in research translated to the clinic in 1954, when Joseph Murray and colleagues performed the first successful kidney transplant between monozygotic twins at the Peter Bent Brigham Hospital in Boston, Massachusetts. The success of this procedure was in part because of the lack of immunosuppression needed when an organ was transplanted between monozygotic twins. Allografts that were attempted subsequently failed because of uncontrolled acute rejection responses mounted by the immune system. The quest to identify methods of both immunosuppression and tolerance induction in transplantation began.
Definition of “Tolerance”
Generally, the concept of tolerance (operational) refers to the persistent survival of a transplanted allograft in the absence of continuing immunosuppressive therapy and an ongoing destructive immune response targeting the graft. This functional definition is appropriate, because multiple immunologic mechanisms together with donor and recipient factors are involved in both inducing and maintaining tolerance to a defined set of donor antigens in vivo . Achieving functional tolerance in transplant recipients will mandate that specific allograft-destructive responses are “switched off” while the global immune response to pathogens and carcinogens remains intact. The most robust form of transplantation tolerance thus has to be donor-specific, as opposed to mere immuno-incompetence, a requirement that can be tested experimentally by grafting third-party transplants and by challenging tolerant recipients to respond to virus infections and tumors. The concept of graft-specific tolerance is essential both to maintain long-term survival of graft and host, and to eliminate the adverse events associated with lifelong nonspecific immunosuppression.
Need for Tolerance in Clinical Transplantation
The immune response to an allograft is an ongoing dialogue between the innate and adaptive immune system that if left unchecked will lead to the rejection of transplanted cells, tissues, or organs (see Chapter 32 ). Elements of the innate immune system, including macrophages, neutrophils, and complement, are activated as a consequence of tissue injury sustained during cell isolation or organ retrieval and ischemia reperfusion. Activation of the innate immune system inevitably leads to the initiation and amplification the adaptive response that involves T cells, B cells, and antibodies. T cells require a minimum of two signals for activation, antigen recognition (often referred to signal 1) and costimulation (referred to as signal 2). The majority of B cells require help from T cells to initiate antibody production. Antibodies reactive to donor antigens, including major and minor histocompatibility antigens and blood group antigens, can trigger or contribute to rejection early, and late, after transplantation.
Multiple factors determine the decision as to how the immune response to a transplant will be triggered and evolve, including where the antigen is “seen” and the conditions that are present at the time—in particular, the presence or absence of inflammation associated with activation of the innate response. In general, the innate response is neither specific nor is it altered significantly with multiple antigenic challenges. In contrast, the adaptive response is specific for a particular antigen or combination of antigens and “remembers” when it encounters the same antigen again, augmenting its activity and the rapidity of the response at each encounter. When the immune system encounters an antigen, it has to decide which type of response to make. In most cases, even though one component of the immune system may dominate and lead to rejection, the process is usually multifactorial, resulting from the integration of multiple mechanisms.
Understanding the molecular and cellular mechanisms that lead to allograft rejection has provided insights leading to the development of therapeutics that suppress this unwanted immune response after transplantation. A diverse collection of small-molecule and biologic immunosuppressive agents are approved and available for use in the clinic that have the potential to control or inhibit allograft rejection. In the context of solid-organ transplantation, the drugs that currently are available for clinical use include azathioprine, cyclosporine, tacrolimus, mycophenolate mofetil, rapamycin, antithymocyte globulin, anti-CD25 monoclonal antibodies, belatacept, and steroids ( Table 21.1 ). Each immunosuppressive agent acts on a different aspect of the immune response to an allograft and can therefore be used effectively in combination. Unfortunately, all of these agents are globally nonspecific in their suppressive activity, and each has some deleterious side effects.
Class of Agent | Agent |
---|---|
Corticosteroid | Prednisone Methyl prednisolone |
Antiproliferative | Azathioprine Mycophenolate mofetil Mycophenolate sodium |
Calcineurin inhibitor | Cyclosporin Tacrolimus |
mTOR inhibitor | Sirolimus Everolimus |
Polyclonal antilymphocyte antibodies | ALG ATG |
Monoclonal antibodies (with target) | Muromonab (CD3) Basiliximab (IL2α receptor-CD25) Alemtuzumab (CD52) Rituximab (CD20) |
Costimulation blockade | Belatacept (LEA 29Y – CTLA4-Ig) |
These immunosuppressive drugs can be used with good success to prevent or control acute allograft rejection; however, they are less effective at controlling the long-term response to injury and activation of the immune system. They also appear to be unable to induce the development of unresponsiveness or tolerance to the donor alloantigens consistently, at least in the way they are used clinically at present. For nearly all transplant recipients, continued survival of the allograft depends on life-long administration of several immunosuppressive drugs. The exception to this statement is liver transplantation where in a proportion of pediatric and adult recipients it is possible to wean a small proportion of selected patients (<10%–20%) treated with immunosuppressive drugs off their immunosuppression, especially in patients with stable graft function over the first 4 to 5 years.
The inability of current immunosuppressive drug regimens to induce tolerance to donor antigens in the majority of patients may be due in part to the nonspecific nature of the immunosuppression resulting from their inability to distinguish between the potentially harmful immune response mounted against the organ graft and immune responses that could be beneficial, protecting the recipient from infectious pathogens and providing mechanisms to control the development of malignant cells. In general, current immunosuppressive drugs act by interfering with lymphocyte activation and/or proliferation irrespective of the antigen specificity of the responding cells ( Fig. 21.1 ). These mechanisms do not discriminate between effector cells that could be damaging to the transplant and immune regulatory cells that have the potential to control allograft rejection. This lack of immunologic specificity means that the immune system of a patient treated with one or more of these therapeutic agents is compromised not only in its ability to respond to the transplant, but also in its ability to respond to any other antigenic stimuli that may be encountered after transplantation. Therefore patients are more susceptible to infections (see Chapter 15, Chapter 16, Chapter 17, Chapter 18, Chapter 19 ) and are at a higher risk for developing cancer (see Chapters 34 and 35).
The development of immunologic tolerance or specific unresponsiveness to donor alloantigens in the short term or the long term after transplantation appears to offer the best possibility of achieving effectiveness and specificity in the control of the immune system after transplantation in either the absence or at least reduced loads of nonspecific immunosuppressive agents. If tolerance to donor alloantigens could be achieved reliably, it would ensure that only lymphocytes in the patient’s immune repertoire responding to donor antigens were suppressed or controlled, leaving the majority of lymphocytes immune competent and able to perform their normal functions after transplantation, including protecting the body from infection and cancer after transplantation. This chapter is therefore dedicated to discussion of the mechanisms underlying tolerance induction and strategies used to induce unresponsiveness in transplanted allografts.
Understanding the Immunologic Mechanisms Behind Tolerance Induction
Overview of T Cell Activation
Understanding the cellular and molecular mechanisms of activation and immune system regulation is important for the development of novel tolerance induction approaches in the context of transplantation and autoimmunity. The next section of the chapter sets the scene for discussing the different approaches to tolerance induction being explored most actively at present.
Hematopoietic stem cells (HSC) present in the bone marrow give rise to all of the leukocyte populations that participate in innate and adaptive immune responses. The thymus is the key organ that shapes the T cell repertoire. T cell precursors leave the bone marrow and migrate to the thymus where they rearrange the genes that encode the antigen recognition structure, the T cell receptor (TCR). Thymocytes that express TCRs with low affinity for self-antigen presented by self major histocompatibility complexes (MHC) molecules are “neglected” and die as they will be of no use to the host. In contrast, thymocytes expressing TCRs with a high affinity for self-antigen undergo programmed cell death and are “deleted” from the repertoire as they could respond to self-antigens in the periphery and therefore be harmful to the host. This leaves the T cells with receptors that have an intermediate affinity to enter the bloodstream where they recirculate between blood and peripheral lymphoid tissue. A subpopulation of T cells that will be discussed later, so-called thymus-derived or naturally occurring regulatory T cells (Treg), are also selected in the thymus and migrate to the periphery. A mature T cell repertoire is developed through this thymic selection process that is not only diverse, but can also react to foreign antigen while still remaining tolerant to self-antigens.
Naïve T cells encounter antigen in the form of a peptide MHC complex on the surface of antigen-presenting cells (APCs). Antigen presentation to T cells can be performed by a variety of APCs, including dendritic cells (DCs), macrophages, and B cells, although dendritic cells are the most immunostimulatory of all the APCs and the most potent at stimulating naïve T cells to respond.
As a direct consequence of organ retrieval and implantation, the tissue within the transplant is injured and stressed. Cells of the innate immune system express invariant pathogen-associated pattern recognition receptors (PRRs) that enable them to detect not only repeating structural units expressed by pathogens, referred to as pathogen-associated molecular patterns (PAMPs), but also markers of tissue injury or damage-associated molecular patterns (DAMPs). Local tissue damage and ischemia reperfusion injury generates many potential DAMPS, including reactive oxygen species, heat shock proteins, heparin sulfate, and high mobility group box 1 (HMBG1), after capture by the receptor for advanced glycation end products (RAGE) complex and fibrinogen, that can bind to PRRs. There are several families of PRRs, including transmembrane proteins present at the cell surface, such as toll-like receptors (TLRs), and inside the cell, including the NOD-like receptors (NODR).
The sensing of DAMPS by PRRs results in potent activation of the inflammasome, upregulating the transcription of genes and production of microRNAs involved in inflammatory responses setting up amplification and feedback loops that augment the response and trigger adaptive immunity. The end result is production of inflammatory mediators including the proinflammatory cytokines, interleukin (IL)-1, IL-6, and tumor necrosis factor (TNF), type I interferons, chemokines (chemoattractant cytokines), and the rapid expression of P-selectin (CD62P) by endothelial cells. These events identify the transplant as a site of injury and inflammation, modifying the activation status, permeability, and viability of endothelial cells lining the vessels, triggering the release of soluble molecules, including antigens from the graft, inducing the production of acute phase proteins, including complement factors systemically and in some cases by the organ itself, stimulating the maturation and migration of donor-derived DCs from the transplant to recipient lymphoid tissue. This process results in upregulation of donor MHC, costimulatory and adhesion molecules by the donor-derived APC, enabling them to become potent stimulators of naïve T cells. The presentation of donor alloantigens to recipient T cells by donor-derived APCs results in T cell activation via the direct pathway of allorecognition.
Activation of the innate immune system within the allograft also triggers the recruitment of inflammatory leukocytes of recipient origin into the graft. These recipient APCs have the capacity to acquire donor histocompatibility antigens from the graft tissue and process them into peptides that can be presented to T cells via the indirect pathway of allorecognition. A third pathway of presentation of alloantigen to T cells has also been described, the so-called semidirect pathway of allorecognition whereby as a result of close contact between recipient APCs and donor cells, sections of membrane containing histocompatibility molecules are transferred from one cell to the other for presentation to T cells.
The interaction between APCs of donor or recipient origin and T lymphocytes is pivotal to the adaptive arm of the immune response. Immunostimulatory APCs are brought into close proximity to naïve T cells that may have TCRs capable of recognizing either intact donor alloantigens via the direct or semidirect pathways of allorecognition or donor peptides presented by recipient MHC molecules via the indirect pathway. An immunologic synapse (IS) is formed by the close interaction between the APC and the T cell that is dependent on the successful dynamic rearrangement and polarization of the filamentous actin in the DCs cytoskeletal membrane to bring the MHC-peptide complex in close relation to the TCR, thereby initiating an activation response. Specific T cell membrane compartments termed “lipid rafts” serve as recruitment centers for costimulatory molecules to concentrate in the cytoskeleton allowing for closer interactions with molecules on the APC ( Fig. 21.2 ).
For a T cell to become fully activated, a threshold number of TCRs need to be engaged. T cell receptor recognition of a donor MHC-peptide complex present on an APC, often referred to as signal 1, results in signal transduction through the cluster of differentiation (CD3) proteins that associate with the TCR at the T cell surface. This signal transduction initiates a cascade of biochemical signaling pathways that are contributed to by interactions between accessory, costimulatory, and adhesion molecules and ultimately culminates in cytokine production and proliferation of the triggered T cell and its differentiation into an effector cell ( Fig. 21.3 ).
Accessory and costimulatory molecules that have been shown to be important in triggering T cell activation on the T cell side include CD4, CD11b/CD18 (LFA-1), CD28, and CD154 (CD40 ligand). These molecules must engage their ligands on APCs, MHC class II, intracellular adhesion molecule (ICAM), CD86/80 (B7-1/B7-2), and CD40 respectively to ensure that once antigen recognition by TCR (signal 1) has occurred, the threshold for activation of a naïve T cell is overcome by delivering signal 2.
The two-signal model of T cell activation is well accepted, but it is important to note that this is a simplification. The cytokine and chemokine milieu present at the time these molecular engagements occur affects the differentiation pathway a T cell takes and the course of the response. Cytokines and chemokines can modulate expression of the cell surface molecules mentioned previously in addition to the expression of cytokine and chemokine receptors themselves. This modulation can result in differential signaling in the T cell and APC, tipping the balance of the response from full to partial activation or, in some circumstances, inactivation of the cells involved, modifying dramatically the downstream events (i.e., cell migration patterns and the generation of effector cells). Activation signals in the form of cytokines propagate the responses initiated by signals 1 and 2 and are often referred to as the third signal in T cell activation.
Mechanisms of Tolerance to Donor Antigens
The human immune system has evolved naturally to respond to challenges in a precise and controlled way. A constant balance exists to ensure an effective but not excessive response to unwanted stimuli. Many mechanisms of tolerance are, in fact, continuously used by the body to prevent reactions against self-antigens that would ultimately lead to autoimmune pathologies. It is when this balance in the immune system is disrupted that immune pathology leading to disease can occur. Many of these same mechanisms and regulatory cell populations can be harnessed to induce and maintain tolerance to alloantigens, at least in animal models.
The mechanisms identified as responsible for inducing or maintaining tolerance to donor antigens include the following:
- ▪
Deletion of donor-reactive cells centrally in the thymus and in the periphery
- ▪
T cell ignorance, or a state of T cell unresponsiveness that is relevant to grafts placed at “immunologically privileged” sites such as the cornea or brain
- ▪
Exhaustion, in which the ability of donor-reactive cells to harm the allograft is eliminated as a result of overstimulation and cell death
- ▪
Anergy, defined as a state of unresponsiveness that is refractory to further stimulation despite the continuing presence of antigen after transplantation
- ▪
Immunoregulation—an active process whereby the immune response to an allograft is controlled by populations of regulatory immune cells
To exploit regulation of the immune response to an organ graft for therapeutic purposes, a clearer understanding of the mechanisms by which this phenomenon operates is required. Although theoretically regulation could function exclusively through a single mechanism, such as deletion of donor-reactive T cells and B cells from the repertoire (as will be discussed next), at present there is little evidence to support this as the only or even the dominant mechanism for inducing and maintaining unresponsiveness to cell and organ transplants. The more likely scenario is that different mechanisms work in concert and that distinct combinations of mechanisms are brought into play depending on donor and recipient characteristics, immunosuppression, infection, and so on, as the immune response to the transplant evolves.
Mechanisms of Tolerance Induction and Maintenance
Persistence of Donor Antigen
An overriding feature in all of the mechanisms of tolerance mentioned previously is the persistent presence of donor antigen throughout the period of tolerance in vivo . Many experimental models have established that donor antigen must be present continuously to maintain a tolerant state, before or after transplantation, irrespective of the precise mechanisms involved. The source of the antigen can be donor-derived cells introduced before transplantation, as is the case in models of mixed chimerism, or the graft itself after transplantation. In the absence of antigen, tolerance is lost gradually, because the mechanisms responsible for maintaining tolerance are no longer stimulated. During the induction and maintenance phases of tolerance, the presence of alloantigen is the key factor driving the outcome. As is often the case with the immune system, the same element can influence the response both positively and negatively. In the case of donor antigen, presentation in the wrong context, such as in a proinflammatory environment as outlined previously, could lead to activation with the potential of destroying the tolerant state and triggering graft rejection, but once tolerance is established, persistence of antigen is critical for maintaining the tolerant state.
Deletion of Donor-Reactive Leukocytes
T Cells
The death or deletion of lymphocytes capable of recognizing and responding to self-antigens or, after transplantation, donor alloantigens is a very effective mechanism for eliminating lymphocytes from the immune repertoire that have the potential to damage the host or the graft, thereby creating unresponsiveness or tolerance to self or donor alloantigens. Both T cells and B cells can be deleted from the repertoire in this manner. Importantly, if this is the only mechanism in operation to either induce or maintain tolerance, deletion needs to be sustained indefinitely.
Central tolerance by clonal deletion of T cells in the thymus is the major mechanism by which tolerance to self-antigens is induced. This mechanism can be exploited for inducing tolerance to donor antigens.
Central deletion of donor alloantigen-reactive T cells has been particularly successful in the context of therapeutic strategies using donor bone marrow in combination with nonmyeloablative therapy, such as T cell depletion or costimulation blockade, for the induction of tolerance. The clinical applicability of this strategy can be demonstrated by kidney transplant recipients who have previously undergone bone marrow transplantation from the same donor because of hematologic indications. The hematopoietic macrochimerism developing in these patients is in itself evidence for tolerance, and leads to long-term graft acceptance without the need for life-long immunosuppression. In mixed allogeneic chimeras in the mouse, donor-derived dendritic cells were shown to reside and persist in the recipient thymus, resulting in continuous deletion of donor-reactive thymocytes leading to the absence of donor-reactive T cells in the periphery and hence tolerance to donor alloantigens.
The challenge of these approaches is to achieve a sufficient level of chimerism reliably without using a treatment regimen that is excessively toxic or leads to potentially devastating graft-versus-host disease (GVHD). Moreover, data regarding the necessity for durable chimerism are conflicting. Data from some clinical studies suggest that transient rather than persistent chimerism may be sufficient in the presence of other immunosuppressive agents to achieve tolerance in some individuals, whereas others suggest that only the achievement of full donor chimerism is consistent with the ability to withdraw immunosuppression and maintain graft survival.
Antigen-reactive T cells may also be deleted from the T cell repertoire in the periphery either in the presence of high doses of antigen or because of the persistent presence of antigen such as would occur after transplantation. The introduction of high doses of defined antigens intravenously or orally has been shown to result in deletion of mature T cells in the peripheral lymphoid organs. Both CD4 + and CD8 + T cells can be eliminated by this mechanism, but in many cases deletion is incomplete even when high doses of antigen are used.
The mechanisms by which T cells are deleted in the thymus and the periphery is an area of active investigation. Two distinct modes of apoptosis have been implicated as the mechanism essential for T cell death in these settings. Activation induced cell death (AICD) is a process through which T cells undergo cell death in the periphery. After restimulation through TCR, a number of different molecules including CD95 (Fas), tumor necrosis factor receptor 1 (TNFR1), and tumor necrosis factor-related apoptosis inducing ligand receptor (TRAILR) can play a role in AICD depending on the circumstances, triggering a complex series of signaling events, which ultimately lead to caspase activation, DNA fragmentation, cytoskeletal degradation, and cell death. Either high doses of antigen or repetitive stimulation is necessary for AICD in the periphery.
The Fas pathway may also play a role, in combination with other mechanisms, in deletion of T cells at particular sites in the periphery, so-called immune privileged sites that include the testis and the eye. At these sites, transplantation of allogeneic tissues results in prolonged survival relative to that obtained after transplantation of the same tissue at other sites. Fas-ligand expression has been shown to be an important contributor enabling these sites to maintain their immune privilege status. The Fas pathway also has been implicated in deletional tolerance after administration of allogeneic bone marrow. However, attempts to harness the immunologic potential of these immune privileged sites have been met with varying degrees of success, suggesting that multiple mechanisms are involved. These include upregulation of expression of CD152 (CTLA4), a negative regulator of T cell activation on T cells, and other costimulatory molecules, such as programmed cell death 1 receptors (PD-1) that can inhibit lymphocyte activation when it binds to its ligands PD-L1 and PD-L2.
B Cells
The elimination of B cells from the repertoire is also a mechanism that has evolved to ensure that polyreactive B cells capable of binding self-antigens are eliminated in the bone marrow before they enter the periphery. As rearrangement of immunoglobulin genes occurs at random to enable as many foreign protein and carbohydrate antigens to be recognized as possible, it inevitably leads to the generation of B cells that express B cell receptors (BCRs) that can recognize self-antigens. Estimates vary, but suggest that up to 70% of the immature B cells produced are self-reactive. Of the order of one-third of these immature B cells are eliminated by receptor editing whereby new gene rearrangements result in the production of an alternate light chain that can pair with the existing heavy chain, altering the antigen recognition properties of the expressed BCR. When an immature B cell recognizes self-antigen with high avidity, it rapidly internalizes the antigen and undergoes a period of developmental arrest. As a result, lymph node homing receptors such as CD62 ligand (CD62L) are not expressed, and the receptors for B-cell activating factor (BAFF), a cytokine required for B cell survival, are not induced. In addition, recombinase-activating genes remain switched on, which allows the BCR to be replaced by editing the light chain. Any B cell undergoing this process will die after 1 to 2 days if it fails to express a nonautoreactive receptor. Death through this pathway does not require Fas and is in part because of antigen-induced expression of the Bcl-2 interacting mediator of cell death (Bim), which inhibits B cell survival proteins from the Bcl-2 family.
Receptor editing is a mechanism that could be used to delete immature B cells capable of recognizing donor alloantigens from the B cell repertoire and particularly in situations where the repertoire is reshaped after leukocyte depletion or the induction of mixed chimerism and when organs are transplanted into young infants.
If receptor editing fails to eliminate all of the self-reactive B cells generated, residual immature B cells expressing highly self-reactive receptors are triggered to die by interaction with self-antigen. This mechanism of control or regulation has been studied using immunoglobulin transgenic mice and the data obtained suggest that B cells are deleted efficiently when the antigen they recognize is membrane-bound. The efficiency of this process was found to be dependent on the probability that the immature B cells encountered the relevant self-antigen and therefore its efficiency is clearly related to antigen density/frequency.
Deletion of B cells capable of recognizing donor alloantigens from the immune repertoire of a transplant recipient is a mechanism that can be harnessed in transplantation. This mechanism of regulation is most efficient when the antigens recognized are present at high doses. Infant recipients of ABO-incompatible heart allografts have been shown to delete B cells capable of making antibodies to blood group antigens presented on the heart transplant. Mixed chimerism strategies that result in the coexistence of donor and recipient bone marrow should also enable donor-reactive B cells to be deleted if the level of chimerism achieved is sufficient to ensure that B cells encounter donor cells.
Regulation of Immune Responses
Although the concept of antigen-specific suppression is not new, there has been a resurgence of interest in the characterization and functional dissection of T cell mediated suppression, now more often called immunoregulation, since it became possible to identify the cells responsible both phenotypically and functionally.
Importantly, transplantation provided some of the earliest evidence for suppression or immune regulation by T cells in vivo . Data from neonatal tolerance studies suggested that additional mechanisms beyond deletion of donor-reactive cells were involved. In the 1970s it was demonstrated that antigen-specific unresponsiveness could be transferred from one recipient to another.
It is now clear that different populations of immune regulatory cells can play a role in controlling the immune response after transplantation, including Treg, regulatory B cells (Breg), myeloid-derived suppressor cells (MDSC), dendritic cells, and regulatory macrophages (Mreg).
Regulatory T Cells
Many different types of T cells with regulatory activity have been described including: CD4 + T cells ; CD8 + T cells ; CD4 − CD8 − double-negative T cells ; natural killer (NK) T cells ; and γδ T cells. Because CD4 + T cells with regulatory functions have been studied in greater depth to date, this section of the chapter will focus on this population of regulatory cells.
CD25 + forkhead box P3 (Foxp3) + Treg can arise via two distinct developmental pathways. First, as mentioned previously, thymus-derived or naturally occurring (tTreg) differentiate in the thymus and are thought to function primarily to suppress responses to self-antigen and hence prevent autoimmune disease. Evidence for this comes from studies in patients with rare genetic defects and in mice with either naturally occurring or genetically engineered defects. For example, profound immune dysregulation leading to autoimmunity is observed in patients with the immune dysregulation, polyendocrinopathy, enteropathy, and X-linked (IPEX) syndrome. IPEX patients have been found to have a point mutation in the gene encoding factor Foxp3, the master gene for T cell regulation, resulting in functional impairment of Treg activity in vitro . Scurfy mice also have mutations in foxp3 and a similar immune profile. Second, when CD4 + T cells encounter antigen in a tolerogenic microenvironment in the periphery, such as when antigen is presented by immature DCs or in the presence of immunosuppressive cytokines, the CD4 + T cells differentiate into “adaptive” or antigen-induced Treg, also known as peripheral Treg (pTreg). In the context of transplantation it can be argued that this pathway may be an important route to generating donor alloantigen-reactive Treg after transplantation and be used to sustain the unresponsive state through a process often referred to as infectious tolerance. Interestingly, despite their distinct developmental origins, both tTreg and pTreg rely on sustained expression of high levels of foxp3 transcription for their suppressive function.
A clear indication that long-term allograft survival in the absence of long-term immunosuppression, a status referred to as operational transplantation tolerance, involved the presence of T cells with the ability to regulate the function of naïve alloreactive T cells, thereby preventing the rejection of a fresh graft was reported. Subsequently, this form of cellular regulation was found to be associated with CD4 + T cells, and Hall and colleagues were the first to suggest that CD25 might be a useful marker for identifying CD4 + T cells with regulatory activity. Similar data were obtained in a rat renal allograft model where operational tolerance was induced by donor-specific blood transfusion (DST). As a direct demonstration that CD4 + T cells expressing high levels of CD25 could regulate rejection, Hara and colleagues showed that cotransfer of CD4 + CD25 + T cells from tolerant animals led to indefinite skin graft survival in 80% of immunodeficient mice reconstituted with naïve CBA effector T cells.
In the absence of any previous exposure to alloantigen there are usually insufficient numbers of tTreg to prevent rejection of a fully allogeneic (MHC + minor histocompatibility antigen mismatched) graft, because the frequency of T cells present in the repertoire capable of making a destructive response to the graft far outnumbers the relatively small number of tTreg present and rejection occurs. The fact that tTreg cannot prevent destruction of an allograft in the absence of immunosuppression does not mean that the cells do not function. However, under these circumstances, the balance between rejection and regulation is against regulation as the suppressive activity of any Treg present is overwhelmed by the destructive response mediated by effector T cells. The presence of preexisting donor alloantigen-reactive memory T cells in the recipient can also overwhelm immune regulation as the kinetics of activation of the memory cells is very rapid and unless very high numbers of Treg are present at the outset of the response, the balance between rejection and regulation is pushed markedly in favor of rejection. Importantly, this critical balance between graft destruction and acceptance can be shifted in a number of ways, notably by employing strategies that increase the relative frequency and/or the activation status and consequently the functional activity of tTreg that can then respond to donor alloantigens before or in the early period after transplantation, including by the infusion of ex vivo expanded Treg, and/or by inhibiting or reducing the activity of the effector cells, for example by using immunosuppressive agents.
Although the observation that mice with long-term surviving allografts contain populations of alloantigen-reactive CD25 + Treg was important, these experiments were unable to distinguish between Treg that were generated by the induction strategy itself and those that arose simply by the presence of the accepted allograft. In terms of developing potential clinical approaches, it is important to clarify whether induction strategies that ultimately lead to long-term operational tolerance can drive Treg development independently of the graft itself. Data demonstrating that exposure to alloantigen in the absence of a transplant can lead to the induction of CD4 + CD25 + Treg were obtained in a number of studies. For example, when CD4 + CD25 + were isolated from mice 28 days after pretreatment with donor alloantigen in combination with nondepleting anti-CD4 therapy, these cells prevented rejection of a test skin graft in a sensitive adoptive transfer model. Critically, protection of the test graft was not observed with similar populations isolated from naïve, or mice treated with anti-CD4-only or DST-only, demonstrating that tolerance mediated by CD25 + Treg can indeed be induced in vivo before transplantation if recipients are exposed to donor alloantigen under permissive conditions. Moreover, data demonstrating that the presence of the allograft alone can lead to the development of T cells with regulatory properties that can protect a challenge graft from rejection, even when the allograft itself has been rejected, have been reported. These examples demonstrate that T cells with regulatory activity can be induced in the presence of alloantigen in the form of the allograft or an infusion of alloantigen or indeed both, and that these cells can contribute to controlling subsequent responses to alloantigen in vivo .
Another important issue is understanding where Treg that can control allograft rejection function most effectively and where they can be found or detected in vivo . There is evidence that the location in which Treg function may change with time after transplantation. Early in the response after transplantation, Treg have been shown to be present and functionally active in the draining lymph nodes, whereas later in the posttransplant course Treg have been shown to function within the allograft itself. Indeed there is increasing evidence that an important site of immune regulation is within the allograft itself, where Treg function to create an environment that is permissive of immune control. Moreover, reexposing Treg to antigen in a tissue may enable them to become more potent suppressors and therefore more effective at controlling rejection.
Although a significant body of work has demonstrated that Treg can control responses to alloantigen, most studies have used either in vitro assays or experimental models whereby cells are adoptively transferred into immunodeficient recipients where allograft rejection is driven by relatively small numbers of effector T cells compared with the number that would be found in a full immune repertoire in vivo . Arguably, a much more relevant question for clinical application of this approach is what role do Treg play in an intact immune system? In transplantation, Treg-specific inactivation was used to show that in the anti-CD4/DST tolerance induction model described previously, the survival of primary heart allografts in normal, lymphoreplete recipients is also unequivocally dependent on iTreg driven by the tolerance induction protocol. These data suggest that it should indeed be possible to boost the function of Tregs in nonlymphopenic transplant patients.
The mechanisms used by Treg to control the activity of other cell populations include cell-cell contact; CD152 has been shown to play a critical role in this respect and the creation of a microenvironment through molecules such as IL-10 78 and transforming growth factor (TGF)-β.
The microenvironment created by the presence of Treg in the lymphoid tissues and the allograft contribute to the phenomenon of linked unresponsiveness, a powerful mechanism that allows tolerance to “spread” from the initiating antigen to those present on cells in the immediate vicinity. For example, if a recipient’s immune system is exposed to a defined alloantigen before transplantation either alone or in combination with immunomodulating agents, the immune response to that antigen will be modulated and subsequently the unresponsiveness achieved will be linked to other molecules present on the transplanted tissue.
Regulatory B Cells
B cells with the capacity to suppress the development of autoimmune diseases in mice were first described in the 1980s. Breg express high levels of CD1d, CD21, CD24, and IgM and moderate levels of CD19, although some heterogeneity has been described, suggesting that different subsets of Breg may exist. One of the characteristics of B cells with regulatory activity is their ability to secrete IL-10. CD40 stimulation appears to be required to stimulate IL-10 production and has been reported to be necessary for activation of Breg, enabling them to manifest their functional activity and suppress Th1 differentiation. It has been suggested that there is a link between regulatory B cells and T cells, with Breg acting as potent generators of Treg. Breg have been described in both mouse and human. Mouse Breg express T cell Ig domain and mucin domain protein 1 (Tim-1). Although human regulatory B cells comprise a small subset of the total B cell pool, they share some properties with their mouse counterparts, including an immature phenotype.
In transplantation, there is evidence that Breg may play a role in controlling immune responses to alloantigens. In experimental models, a shift in both peripheral and intragraft gene expression from IgG to IgM was observed, and IgM + , but not IgG + B cell clusters within rat kidney allografts. In mice, Tim-1 ligation on B cells induced Tim-1 + B cells with regulatory activity, suggesting a potential therapeutic strategy for increasing the number of Breg in vivo .
Interestingly, in renal transplant recipients with a functioning graft in the absence of immunosuppression, a B cell signature was found to be associated with this state of operational tolerance to donor alloantigens. A distinct but also B cell dominated signature of tolerance was identified in a separate cohort of long-term immunosuppression-free renal transplant recipients. The potential role that immunosuppressive drug therapy plays in enabling a B cell signature to emerge in these immunosuppression-free renal transplant recipients has also been investigated. Not surprisingly, data are emerging demonstrating that different immunosuppressive drugs have a significant effect.
Regulatory Macrophages
Macrophages have both protective and pathogenic functions and can be divided into subgroups on the basis of their tissue location and their functional properties. Mature macrophages are present in tissues where they contribute to immune surveillance by sensing tissue damage or infection. Macrophages activated via TLRs differentiate into “classically activated” macrophages (also referred to as M1 macrophages) and produce inflammatory cytokines, including TNF and IL-1. In transplantation, macrophage activation occurs initially as a result of the tissue injury that is associated with ischemia and reperfusion and can contribute to early graft damage. Alternatively activated macrophages (also referred to as M2 macrophages) are induced by exposure to the cytokine IL-4, are less proinflammatory than classically activated macrophages, and in some settings can create a microenvironment that downregulates inflammatory immune responses. Indeed, alternatively activated macrophages can inhibit the production of proinflammatory cytokines by classically activated macrophages. Alternatively activated macrophages also play an important role in wound healing and tissue repair by producing growth factors that can stimulate epithelial cells and fibroblasts. This function is important in organ transplantation in the early posttransplant period where wound healing will allow tissue homeostasis to be reestablished. However, later in the response, tissue repair responses may be less desirable because they have been shown to contribute to delayed allograft failure by causing occlusion of blood vessels within the allograft, a process referred to as transplant arteriosclerosis or transplant-associated vasculopathy.
Macrophages are positioned within the lymphoid architecture to interact with lymphocytes and therefore have the potential to influence each other’s functional activity. The interaction of Treg cells with macrophages can result in the macrophages acquiring the properties of alternatively activated or regulatory macrophages (see Fig. 21.2 ). Although the interaction of macrophages with B1 B cells results in the formation of a regulatory macrophage population that produces reduced levels of inflammatory cytokines and increased levels of IL-10. Thus macrophages can influence the character of an adaptive immune response, and they also can change their physiology as a result of interactions with other populations of regulatory immune cells during an immune response.
It has been proposed that “regulatory macrophages” may represent an additional, distinct macrophage population whose main physiologic role is to dampen inflammatory immune responses and prevent the immunopathology associated with prolonged classical macrophage activation. In vitro, regulatory macrophages are efficient APCs that induce highly polarized antigen-specific T cell responses dominated by the production of Th2-type cytokines. A specific subpopulation of human regulatory macrophages, referred to as Mreg, can be isolated from the peripheral blood and are characterized by their morphology, expression of HLA-DR, and high levels of CD86 with low or absent cell-surface expression of CD14, CD16, CD80, CD163, and CD282 (TLR2); Mreg strongly suppress mitogen-driven T cell proliferation in vitro. Human Mreg express high levels of indoleamine-pyrrole 2,3-dioxygenase (IDO), have the capacity to regulate immune responses to alloantigens, can delete activated T cells, and in a pilot clinical study donor-derived Mreg were shown to reduce the need for immunosuppressive drugs when administered intravenously to kidney transplant recipients. Recently a marker for specific for Mreg has been identified.
Groups have also shown that host macrophages can also have a protective effect after transplantation. Reducing the pool of the host macrophages in recipient mice increased donor T cell expansion and aggravated GVHD mortality after allogeneic stem cell transplantation, suggesting that host macrophages may have an important protective effect by both engulfing and inhibiting the proliferation of donor allogeneic T cells. As with the advancement of Treg research, further studies to specifically identify regulatory macrophage populations will help further dissect their role in vivo.
Tolerogenic Dendritic Cells
DCs are crucial for priming antigen-specific T cell responses, including T cell responses to alloantigens, but they can also promote the development of immunologic unresponsiveness, either in their own right or by interacting with other leukocyte populations.
Initially, immature conventional myeloid DCs that express low levels of MHC class II and costimulatory molecules at the cell surface were identified as the dominant form of DC that had the capacity to induce T cell tolerance. Indeed, immature DCs can promote tolerance to solid-organ allografts and bone marrow grafts, and their tolerogenic effects can be enhanced by other immunomodulatory agents, such as drugs that block the CD40–CD40L costimulatory axis. However, as the characterization of DC has progressed it is clear that the state of maturity of the DC is not the only factor that enables them to induce unresponsiveness.
Plasmacytoid DC (pDCs) produce type-I IFNs in response to single-stranded nucleic acids, which activate pDCs via both TLR-dependent and TLR-independent pathways, thus promoting antiviral innate and adaptive immune responses. Compared with conventional DCs, pDCs express lower levels of the costimulatory molecules CD80 and CD86 and higher levels of the inhibitory molecule PD-L1, and consequently, exhibit poor immunostimulatory activity. Indeed, pDC interaction with naïve T cells has been shown to promote the generation of Treg cells in the thymus and in the periphery. The molecular mechanisms used by pDCs to promote tolerance are complex. In experimental models, pDCs can acquire alloantigen in the allograft and then migrate to the draining lymphoid tissue where they interact with T cells and induce the generation of CCR4 + CD4 + CD25 + Foxp3 + Treg cells. In mice, prepDC appear to be the principal cell type that facilitates hematopoietic stem cell engraftment and induction of donor-specific skin graft tolerance in allogeneic recipients.
In humans, pDCs produce significant amounts of IL-10, low levels of IFN-γ, and no detectable IL-4, IL-5, or TGF-β, creating a microenvironment that promotes immune regulation. Tolerogenic human DCs that secrete high levels of IL-10 in the absence of IL-12 induce adaptive IL-10-producing regulatory type-1 (Tr1) T cells. Data demonstrating that DC regulate the suppressive ability, expansion, and/or differentiation of Treg cells, with the loss of DC resulting in reduced numbers of Treg cells, has been reported in mice. Thus tolerogenic DCs and T cells with regulatory activity may be linked and act in concert with one another in some situations.
The tolerogenic phenotype of human dendritic cells has been characterized by high cell surface expression of the inhibitory receptor ILT3. In DCs, immunoglobulin-like transcript 3 (ILT3) signaling on binding cognate ligands such as MHC class I, HLA-G, and CD1d inhibits tyrosine phosphorylation, NF-κB, and mitogen-activated protein kinase (MAPK) p38 activity, transcription of certain costimulatory molecules, and the secretion of proinflammatory cytokines and chemokines. In addition, ILT3 can interact with its ligands on T cells to promote inhibitory signaling in T cells. Stimulation through ILT3 appears to be a prerequisite for DC tolerization. Both ILT3 high tolerogenic dendritic cells and soluble ILT3 have been shown to induce CD4 + T cell anergy and differentiation of antigen-specific CD8 + suppressor T cells.
Higher numbers of pDCs than myeloid DCs have been found in the peripheral blood of pediatric liver transplant recipients who were operationally tolerant to their allograft and in those receiving low-dose immunosuppressive therapy during prospective immunosuppressive drug weaning, compared with patients on maintenance immunosuppression. In addition, higher expression levels of PD-L1 and CD86 by pDC were found to correlate with elevated numbers of CD4 + CD25 high Foxp3 + Treg cells in liver transplant recipients who were free from immunosuppressive drug regimens. These data suggest that pDCs and Treg cells may contribute to immune regulation in liver transplant recipients.
The ability of different populations of DC to induce tolerance combined with clinical observations suggest that both myeloid DCs and pDCs can promote tolerance to alloantigens, and that DC maturation in itself is not the distinguishing feature that separates immunogenic DC functions from tolerogenic ones. Indeed, maturation is more of a continuum than an “on-off” switch, and a “semimature” state, in which DCs are phenotypically mature but remain poor producers of proinflammatory cytokines, and appears to be better linked with tolerogenic function. Maturation-resistant regulatory DC do show some promise in nonhuman primates at prolonging kidney allograft survival. However, the use of DCs to facilitate the induction of operational tolerance is not without risk. DCs are arguably better known for the ability to prime the immune system. Indeed, DCs pulsed with antigen are being used clinically as vaccines to stimulate immune responses to tumor antigens. Using DCs as a cellular therapy in transplantation may carry the risk of sensitizing the recipient. One possible approach to reducing this risk is to combine DC administration and costimulatory blockade, with the objective of presenting donor alloantigens to induce T cell unresponsiveness. Thus far, therapeutic application of donor antigen pulsed DC has recently shown some modest positive effect on kidney allograft survival in nonhuman primates.
Myeloid-Derived Suppressor Cells
MDSC have been associated with many antigen specific and nonspecific suppressive functions, including regulation of innate immunity, T cell activation, and tumor immunity. MDSCs are a heterogeneous population of progenitor cells that can accumulate in tissues during an inflammatory immune response, where they differentiate into macrophages, DCs, and granulocytes. The expansion and activation of MDSCs are regulated by factors produced by other cells that are present in the same microenvironment, including stromal cells, activated T cells, and in tumors, the tumor cells themselves.
Several MDSC subsets have been described in both mice and humans. Despite their heterogeneity, common phenotypical markers are expressed by most MDSCs, including Gr1 and CD11b in mice, and CD33, CD11b, CD34, and low levels of MHC class II in humans. Activated MDSCs suppress proliferation and cytokine production by effector T cells, B cells, and natural killer (NK) cells in vitro through mechanisms that include their expression of inducible nitric oxide synthase (iNOS) 1, which induces nitric oxide production, and expression of arginase 1, which depletes arginine. MDSCs also can inhibit T cell proliferation and modify T cell differentiation pathways. For example, they can promote Treg cell differentiation in a process requiring interferonγ (IFNγ) and IL-10. Interestingly, interactions between MDSCs and macrophages result in a shift of macrophages toward an alternatively activated phenotype and increased production of IL-10 by MDSCs.
In experimental transplant models, MDSCs have been shown to promote tolerance to alloantigens. Direct evidence of a tolerogenic role for MDSCs has been obtained for heart and islet allografts in mice and by iNOS-expressing MDSCs in a rat kidney allograft model. In bone marrow transplantation, indirect evidence for a role of MDSCs has come from the observation that transplantation tolerance could not be induced in mice that did not express MHC class II on circulating leukocytes, although given that many other leukocytes’ populations may also be altered in this setting. The mechanisms used by MDSCs to promote tolerance to alloantigens require further clarification. Some evidence suggests that they may act partly through the induction or sparing of Treg. Alternatively, MDSCs have been found to upregulate heme oxygenase 1, an enzyme that has immunoregulatory activity through the inhibition of DC maturation and preservation of IL-10 function in addition to cytoprotective properties. It has been shown recently that granulocyte macrophage-colony stimulating factor (GM-CSF) signaling pathways through IFN-γ receptor/interferon regulatory factor-1 and AKT/mechanistic target of rapamycin (mTOR) provide monocyte licensing for suppressor function.
Mesenchymal Stromal Cells
Mesenchymal stromal cells (MSCs) are a subpopulation of multipotent cells within the bone marrow that support hematopoiesis and possess immunomodulatory and reparative properties. Bone marrow derived MSCs have the ability to migrate to sites of inflammation, including to an allograft, but when injected intravenously are trapped in the lung and difficult to find alive thereafter. One mechanism by which MSCs may exert their immunosuppressive effects is by undergoing apoptosis in vivo; moreover, already apoptotic MSCs can be effective immunomodulators by delivering cellular material that is phagocytized. Regarding live cells, MSCs exposed to an inflammatory microenvironment have been found to be capable of regulating many immune effector functions through specific interactions with leukocytes that participate in both innate and adaptive responses. The proinflammatory cytokines IFNγ, TNF-α, and/or IL-1β have been shown to activate bone marrow derived MSCs ; studies indicate that their suppressive functions depend on activation signals associated with inflammation, which makes these cells potentially important in the situation of organ transplantation, where retrieval and transplantation of the allograft inevitably results in ischemia, reperfusion injury, and an inflammatory microenvironment within the graft. Indeed, in vivo, MSC activation by IFN-γ has been shown to be essential for preventing GVHD. Once activated MSCs can control the activity of T cells, B cells, DCs, NK cells, and macrophages. Research indicates that mediation of their immunomodulatory properties may depend on the presence of macrophages, which convert to an antiinflammatory phenotype that expresses suppressive cytokines such as IL-10 152,154 and take on the function of suppressive macrophages. MSC themselves have also been shown to possess immunosuppressive activity via production of IDO, prostaglandin E 2 (PGE 2 ), and PD-L1. Moreover, among the cocktail of other factors secreted by MSCs include matrix metalloproteinases (MMPs), with the production of MMP2 being linked directly to a decrease in CD25 expression on CD4 + T cells and the inhibition of alloantigen-driven proliferation resulting in long-term survival of allogeneic islets of Langerhans. Furthermore, regarding the potential enhancement of Treg, MSC are known to produce factors such as TGF-β, and through their activation by the inflammatory cytokine IL-17A, promote Treg numbers via the cyclo-oxygenase-2/PGE2 pathway. The immunomodulatory properties of MSCs on B cell function could also contribute to suppressing graft rejection by inhibiting alloantibody production. Clearly, the mechanisms by which MSC suppress the immune response are growing in complexity and are actively being studied, especially in the context of the inflammatory conditions associated with organ transplantation.
Information From Analyzing Tolerant Recipients
Operational tolerance, whereby an allograft remains functional and rejection-free for more than 1 year without immunosuppression, is often described as the holy grail of transplantation. Clearly, to be able to achieve operational tolerance would have a major benefit for the patient, reducing the morbidity and mortality associated with lifelong immunosuppression. However, operational tolerance remains a relatively rare event in the clinical setting and thus far not reliably predictable through biomarkers.
A small number of bone marrow transplant recipients who subsequently required a renal transplant were transplanted with a kidney from their bone marrow donor. In these cases long-term immunosuppression was unnecessary because the recipient already was unresponsive to the donor alloantigens as a result of the allogeneic chimerism that developed after the successful bone marrow transplant. Clinical reports of patients exhibiting spontaneous tolerance to an allograft after withdrawal of immunosuppression in the absence of a bone marrow transplant are still infrequent but cases are accruing. Based on these case reports, a number of groups proposed that by studying these patients in depth it would be possible to define a molecular signature of tolerance in immunosuppression-free kidney and liver transplant recipients (see later).
The majority of patients who are able to stop immunosuppression without rejection of their allograft arrive at that point as a result of either (1) discontinuation of immunosuppression by the clinical team as a consequence of the side effects of immunosuppression, nonadherence/compliance; or (2) as a result of weaning of immunosuppression based on a clinical protocol (particularly in liver transplant recipients) designed to minimize immunosuppressive drugs without compromising the function of the allograft. There is also a smaller group of patients who have been treated with protocols designed with the deliberate aim of inducing tolerance to a kidney transplant that will be discussed in detail later.
It is generally accepted that there is a hierarchy with respect to the ease of inhibition or suppression of immune responses directed against different organs. Liver allografts and skin grafts (when they are not part of a vascularized composite allograft) are at the two opposite ends of this spectrum. In experimental transplantation, liver allografts are sometimes accepted spontaneously in the absence of any immunosuppression or immunomodulation. Furthermore, in clinical transplantation, it is often noted that the liver “protects” other organs that are transplanted alongside it from the full force of the rejection response—the so-called liver effect . Based upon these observations it has been suggested that liver allografts have the capacity to promote the development of immunologic unresponsiveness. A number of mechanisms have been proposed to account for the liver effect, including many of those discussed previously, for example, the large antigen load delivered by the liver itself, the presence of a large numbers of passenger leukocytes that could result in deletion of donor-reactive cells, and the establishment of long-lasting microchimerism in some recipients in addition to the production of soluble MHC class I by the liver. As a result, liver transplant recipients have proved to be an interesting population of immunosuppression-free patients for investigations designed to determine whether a molecular signature of tolerance exists.
Estimates vary, but on the order of 10% to 20% select adult and pediatric liver transplant recipients appear to have the potential to be weaned from immunosuppression without the risk of rejection. Early investigations focused on phenotyping the populations of leukocytes that were present in immunosuppression-free liver transplant recipients; the majority were weaned from immunosuppression as a result of participation in well controlled clinical weaning protocols. Interestingly, Foxp3 + cells were found to be a component of lymphocytes present in the peripheral blood and infiltrating the liver allografts. Further analysis using transcriptional profiling demonstrated that there was a molecular signature of tolerance associated with tolerance in immunosuppression-free liver transplant recipients and that this signature can be used to guide weaning of immunosuppression. In other words, the molecular signature and the transplant biopsy can be used to provide a risk assessment of which patients are more likely to wean from immunosuppression without the risk of rejection, which is an exciting development for the future of clinical liver transplantation. This biomarker signature for liver transplant tolerance, however, is best derived through sampling of the liver tissue itself, rather than blood markers.
In kidney transplantation, there are far fewer patients able to continue to have a functioning allograft in the absence of immunosuppression. Nevertheless, a small number of patients who are immunosuppression-free can be identified and have been studied in depth to determine whether there is a molecular signature of tolerance after kidney transplantation. Two independent studies, one in Europe and the other in the US, found that there was a signature of tolerance in immunosuppression-free kidney transplant recipients that could be cross-validated between the two cohorts of patients studied. Interestingly, the signature of tolerance in kidney transplant recipients was distinct from that identified in immunosuppression-free liver transplant recipients. The predominance of B cells in the peripheral blood in the absence of donor-specific antibody was found in the “kidney” signature as mentioned previously and was supported by gene expression data. Further work examining the prevalence of this signature in kidney transplant recipients still being treated with immunosuppression has demonstrated that different drugs in clinical use have a distinct influence on the molecular signature in a particular patient. More work is therefore required to determine whether a molecular signature can be used to identify patients currently on immunosuppression who could potentially benefit from withdrawal or minimization of immunosuppression without risk to the graft.