The President's Council on Bioethics click here to skip navigation


This paper was discussed at the Council's June 2003 meeting. It was prepared solely to aid discussion, and does not represent the official views of the Council or of the United States Government.


Silviu Itescu, MD
Director of Transplantation Immunology
Columbia University, New York




The Human Leukocyte Antigens (HLA)
T Cell Recognition Of Antigen Presented By HLA Molecules
Thymic Education Of T Cells
T Cell Recognition Of Alloantigens
Tolerance Induction



Materno-Fetal Tolerance
Immunogenic Characteristics Of Embryonic And Adult Stem Cells
Tolerogenic Effects Of Stem Cell Transplantation


The Major Histocompatibility Complex (MHC) is located on the short arm of chromosome 6 in humans and encodes the alloantigens known as Human Leukocyte Antigens (HLA), polymorphic cell surface molecules which enable the immune system to recognize both self and foreign antigens. The class II HLA molecules (HLA-DR, HLA-DP, and HLA-DQ) are usually found only on antigen-presenting cells such as B lymphocytes, macrophages, and dendritic cells of lymphoid organs, and initiate the immune response to foreign proteins, including viruses, bacteria, and foreign HLA antigens on transplanted organs.

Following binding of foreign proteins, class II HLA on antigen-presenting cells activate CD4+ T cells, which in turn activate cytotoxic CD8+ T cells to recognize the same foreign antigen bound to HLA class I (HLA-A, HLA-B, and HLA-C, molecules found on the surface of all cells) and destroy the target. The actual recognition of foreign HLA transplantation antigens by T cells is referred to as allorecognition. Two distinct pathways of allorecognition have been described, direct and indirect. The direct pathway involves receptors on the host T cells that directly recognize intact HLA antigens on the cells of the transplanted organ. The indirect pathway requires an antigen-presenting cell that internalizes the foreign antigen and presents it via its own HLA class II molecule on the surface of an antigen-presenting cell to the CD4+ helper T cells.

Once recognition has taken place, an important cascade of events is initiated at the cellular level, culminating in intracellular release of ionized calcium from intracellular stores. The calcium binds with a regulatory protein called calmodulin, forming a complex that activates various phosphatases, particularly calcineurin. Calcineurin dephosphorylates an important cytoplasmic protein called nuclear factor of activated T cells (NFAT), resulting in its migration to the nucleus and induction of the production of various cytokines such as IL-2. These cytokines recruit other T cells to destroy the transplanted organ, ultimately resulting in rejection and loss of the graft.

Immunosuppressive regimens used to prevent allograft rejection are aimed at inhibiting the various arms of the immune response, typically require multiagent combinations, and need to be maintained for the duration of life. The currently used armamentarium confers significant side-effect risks, including infectious and neoplastic complications. Moreover, despite success at preventing early allograft rejection, long-term survival of transplanted organs remains difficult to achieve and novel methods to achieve long-term tolerance are being actively sought.

Stem cells obtained from embryonic or adult sources differ from other somatic cells in that they express very low levels of HLA molecules on their cell surfaces. This endows these cell types with the theoretical potential to escape the standard mechanisms of immune rejection discussed above. However, under conditions that enable cellular differentiation in vitro and in vivo each of these stem cell populations acquires high level expression of HLA molecules, suggesting that their long-term survival following transplantation in vivo may be limited by typical immune rejection phenomena. Recent experimental data, however, provide striking counterintuitive examples that stem cells from both embryonic and adult sources may evade the recipient's immune system and result in long-term engraftment in the absence of immunosuppression despite acquisition of surface HLA molecule expression. These observations may have significant impact on the emerging field of regenerative medicine.


The Human Leukocyte Antigens (HLA)

Differences between individuals which enable immune recognition of non-self from self are principally due to the extreme polymorphism of genes in the Major Histocompatibility Complex (MHC) on chromosome 6 in man which encode the cell surface HLA molecules. These molecules are cell surface glycoproteins whose biologic function is to bind antigenic peptides (epitopes) derived from viruses, bacteria, or cancer cells, and present them to T cells for subsequent immune recognition. Each HLA gene includes a large number of alleles and the peptide binding specificity varies for each different HLA allele. The 1996 WHO HLA Nomenclature Committee report lists more than 500 different HLA class I and class II alleles.

Crystallographic x-ray studies have demonstrated that the hypervariable regions encoded by polymorphic regions in the alleles correspond to HLA binding pockets which engage specific “anchor” residues of peptide ligands. One HLA molecule will recognize a range of possible peptides, whereas another HLA molecule will recognize a different range of peptides. Consequently, no two individuals will have the same capability of stimulating an immune response, since they do not bind the same range of immunogenic peptides. It is estimated that >99% of all possible peptides derived from foreign antigens are ignored by any given HLA molecule. Since in the absence of HLA polymorphism a large number of immunogenic peptides would not be recognized, the extensive HLA polymorphism in the population reduces the chance that a given virus or bacterium would not be recognized by a sizable proportion of the population, reducing the likelihood of major epidemics or pandemics.

T Cell Recognition Of Antigen Presented By HLA Molecules

Since HLA molecules regulate peptide display to and activation of the immune system, considerable effort has been devoted to understanding the molecular basis of peptide-HLA interactions. These issues are important for defining the biology of T cell antigen recognition and the properties of a protein that make it immunogenic or non-immunogenic. Specific antigen recognition by T cells is dependent on recognition by the T cell receptor of a three-dimensional complex on the surface of antigen-presenting cells (APC) comprised of the HLA molecule and its bound peptide. The peptides are produced by complex antigen processing machineries within the APC (i.e. proteolytic enzymes, peptide transporters and molecular chaperones) which generate a pre-selected peptide pool for association with the HLA molecules. The different types of T cells require different HLA molecules for antigen presentation, so-called “HLA restriction” phenomena. T cell receptors on CD8+ cytotoxic T cells (CTLs) bind peptides presented by HLA class I molecules, whereas CD4+ T helper cells (Th) recognize peptides bound to HLA class II molecules. Of the 8-13 amino acid residues of a bound peptide within a class I or II HLA molecule, only three to four amino acid side chains are accessible to the T cell receptor, and a similar number of amino acids are involved in binding to the HLA molecule.

Thymic Education Of T Cells

T cells mature in the thymus to appropriately respond to foreign pathogens without inadvertently attacking the host. Under the influence of various thymic resident cells and factors they elaborate, maturing T cells fall into two categories: those that are able to discriminate between self and non-self and can appropriately respond to foreign pathogens without inadvertently attacking the host, and those which are unable to appropriately discriminate between self and non-self. Dendritic cells have been implicated in the deletion, or inhibition, of T cells reactive to self-antigens, particularly in the thymus during T cell development or in peripheral lymphoid organs. The process of self/non-self discrimination by the maturing T cells is dependent on thymic dendritic cell (DC) presentation of self-antigens in the context of self-HLA molecules. When maturing thymic T cells are highly reactive with self-antigen/HLA complexes, they are deleted so that potentially autoreactive T cells will not be released into the periphery. If a particular foreign antigen can be presented in such a way in the thymus as to fool the maturing T cells into believing that the antigen is part of self tissue, then T cells capable of reacting with this antigen will also be eliminated. Indeed, it has been demonstrated that when mouse thymic DC present transgenically introduced foreign antigens to developing T cells, the mature peripheral T cell repertoire of the mouse lacks T cells capable of reacting with the specific foreign antigen, i.e. it is tolerant to the foreign antigen. This has raised the possibility that injection of dendritic cells into an allogeneic recipient might induce tolerance to a subsequent allograft by causing deletion or inhibition of alloreactive T cells.

T Cell Recognition Of Alloantigens

Recognition of foreign, or allogeneic, HLA antigens by the recipient immune system is the major limitation to the survival of solid organ grafts. The central role of HLA molecules in allograft rejection is due to their role as restriction elements for T cell recognition of donor antigens and the extensive polymorphism displayed by the HLA molecules, which elicit host immune responses. Although progress has been made in the short-term survival of transplants, chronic immunologic rejection remains an impediment to long-term survival.

The primary cause of acute rejection of transplanted organs is so-called "direct" recognition of whole allogeneic HLA antigens by receptors on the surface of recipient T cells. The direct recognition pathway involves recognition by recipient T cells of donor HLA class I and class II molecules, resulting in the generation of cytotoxic and helper T lymphocytes which play a pivotal role in the rejection process. In contrast, chronic rejection of transplanted organs results from so-called "indirect recognition" of donor HLA peptides derived from the allogeneic HLA molecules shed by the donor tissue. These foreign HLA molecules are taken up and processed by recipient antigen presenting cells (APC), and peptide fragments of the allogeneic HLA molecules containing polymorphic amino acid residues are bound and presented by recipient’s (self) HLA molecules to recipient (self) T cells. Although direct and indirect recognition of alloantigen generally leads to adverse graft outcome, tolerance induction may occur following exposure of the recipient to donor alloantigens prior to transplantation. Since this strategy is based on the nature and dose of the antigen as well as the route of administration, understanding how to control the balance between activation and unresponsiveness mediated by the direct and/or indirect recognition of alloantigen is a an area of active research which could lead to development of new therapies to prolong graft survival.

Indirect allorecognition has been implicated in recurrent rejection episodes in various transplantation models of cardiac, kidney and skin grafts. Determinants on donor HLA molecules can be divided into two main categories: (a) the dominant allodeterminants that are efficiently processed and presented to alloreactive T cells during allograft rejection; and (b) the cryptic allodeterminants that are potentially immunogenic but do not normally induce alloreactive responses, presumably due to incomplete processing and/or presentation. Indirect recognition of allo-HLA peptides is important for the initiation and spreading of the immune response to other epitopes within the allograft. So-called "spreading" of indirect T cell responses to other allo-HLA epitopes expressed by graft tissue is strongly predictive of recurring episodes of rejection. Tolerance induction to the dominant donor determinants represents potential effective strategy for blocking indirect alloresponses and ensuring long-term graft survival in animal models.

Tolerance Induction

Advances in surgical methods and current immunosuppressive therapies have led to significant improvement in short-term graft survival, however long-term survival rates remain poor. For example, whereas both kidney and heart allografts have one-year graft survival rates of 85 to 95 percent, only about 50% of transplanted hearts survive five years and only about 50% of kidney grafts survive ten years. Thus, despite being able to achieve short-term success, these relatively poor long-term graft survival rates demonstrate the limitations of the current clinical immunosuppressive regimens to enable long-term immune evasion by the graft. Consequently, a major goal of transplantation immunobiologists is to induce donor-specific tolerance, allowing the long-term survival of human allografts without the need of HLA-compatibility and without the continuous recipient immunosupression leading to the concomitant risks of infection, malignancy, and/or other specific drug side effects. This would theoretically improve long-term graft survival, reduce or eliminate the continuing need for expensive, toxic and non-specific immunosuppressive therapy and enhance the quality of life.

Insight into some of the mechanisms involved in tolerance induction has been gained from pre-clinical and clinical studies in numerous animal models and in patients, particularly those with liver allografts which typically do not induce a prominent immune response leading to rejection. One possible mechanism by which liver transplantation results in allograft tolerance tolerance may be that the donor or "passenger" lymphoid cells in the transplanted liver emigrate and take up residence in the recipient's immune organs, such as the thymus or lymph nodes. Donor lymphocytes at these sites might "re-educate" the recipient immune system so that the donor organ is not recognized as foreign. In an attempt to initiate a similar process in other organ recipients, transfusions of donor blood or bone marrow have been used to enhance solid organ graft survival in animal models and in clinical trials. These studies are currently ongoing in various organ systems.

Molecular understanding of the cellular immune response has led to new strategies to induce a state of permanent tolerance after transplantation. Several approaches have shown promise, including the use of tolerizing doses of class I HLA-molecules in various forms for the induction of specific unresponsiveness to alloantigens, and the use of synthetic peptides corresponding to HLA class II sequences. Other approaches include alteration in the balance of cytokines that direct the immune response away from the TH1 type of inflammatory response and graft rejection to the TH2 type of response that might lead to improved graft survival, and the use of agents to induce "co-stimulatory blockade" of T cell activation. This latter approach is based on the concept that blockade of a "second signal" to the T cell enables the signal provided to the T cell receptor by the HLA-peptide complex to induce antigen specific tolerance.

The experimental use of human dendritic cells as tolerogenic agents has been limited due to the low frequency of circulating dendritic cells in peripheral human blood, the limited accessibility to human lymphoid organs, and the terminal state of differentiation of circulating human dendritic cells making their further expansion ex vivo difficult. Dendritic cells are migratory cells of sparse, but widespread, distribution in both lymphoid and non-lymphoid tissues. Although the earliest precursors are ultimately of bone marrow origin, the precise lineage of dendritic cells is controversial and includes both myeloid-derived and lymphoid-derived populations. Recent work has revealed that an expanded population of mature human dendritic cells can be derived from non-proliferating precursors in vitro is by culturing bone-marrow derived cells with a combination of cytokines. This method of enrichment for human dendritic cells from a precursor population can result in the production of dendritic cells that are tolerogenic to foreign antigens. Whether such cells could be useful when co-administered with an allograft transplant remains to be determined. Nevertheless, it is clear that considerable progress has been made in the past few years using approaches to manipulate the immune response to enable routine donor-specific tolerance, and there is reason to be optimistic that with better understanding of molecular and cellular mechanisms this goal could be attained.



Cyclosporine has been the single most important factor associated with improved outcomes after organ transplantation over the past two decades. CyA binds to a cytosolic cell protein, cyclophilin (CyP). The CyA-CyP complex then binds to calcineurin and subsequently blocks interleukin-2 (IL-2) transcription. The binding of IL-2 to the IL-2 receptors on the surface of T lymphocytes is a key stimulant in promoting lymphocyte proliferation, activation, and ultimately allograft rejection. A review of the first decade of experience with heart transplantation revealed a total of 379 cardiac allograft recipients worldwide; actuarial survival rates in this cohort of patients at 1 year and 5 years were 56% and 31% respectively; the main causes of death being acute rejection and the side effects of immunosuppression. With the introduction and widespread use of CyA over the next decade, survival rates dramatically improved to 85% and 75% at 1 and 5 years respectively. Similar results were obtained with other organ transplants, including kidney and lung.

The major adverse effects of CyA are nephrotoxicity, hypertension, neurotoxocity and hyperlipidemia; less common side effects include hirsuitism, gingival hyperplasia and liver dysfunction. CyA nephrotoxicity can manifest as either acute or chronic renal dysfunction. It is important to note that a number of drugs commonly used in transplant patients, such as aminoglycosides, amphotericin B and ketoconazole can potentiate the nephrotoxicity induced by CyA. More than half the patients receiving CyA will require treatment for hypertension within the first year following transplantation. Corticosteroids also potentiate the side effects of CyA such as hypertension, hyperlipidemia and hirsuitism. Frequent monitoring of the serum level is essential to minimize the adverse effects. One of the major limitations of the original oil-based CyA formulation (Sandimmune) is its variable and unpredictable bioavailability. In the mid-90s Neoral was introduced, a new microemulsion formula of CyA, which has greater bioavailability and more predictable pharmacokinetics than Sandimmune.


Tacrolimus (FK506) is a macrolide antibiotic that inhibits T-cell activation and proliferation and inhibits production of other cytokines. The product of Streptomyces tsurubaensis fermentation, FK 506 was first discovered in 1984 and first used in clinical studies in 1988 at the University of Pittsburgh. While the mechanism of action of tacrolimus is similar to that of CyA, and comparative clinical trials have suggested similar efficacy, it has been suggested that some groups of patients may benefit from tacrolimus rather than CyA as primary immunosuppressive therapy. Unlike CyA, hirsuitism and gingival hyperplasia occur infrequently with tacrolimus; thus, tacrolimus-based therapy may improve compliance and quality of life in female and pediatric transplant recipients. It should be noted that alopecia has been documented with tacrolimus, but is known to improve with dose reductions. The decreased incidence of hypertension and hyperlipidemia with tacrolimus makes it preferable to CyA in patients with difficult to treat hypertension or hyperlipidemia. A final indication for tacrolimus has been as a rescue immunosuppressant in cardiac transplant recipients on CyA with refractory rejection or intolerance to immunosuppression (severe side effects). Since tacrolimus is metabolized using the same cytochrome P450 enzyme system as CyA, drug interactions are essentially the same. Thus, drugs that induce this system may increase the metabolism of tacrolimus, thereby decreasing its blood levels. Conversely, drugs that inhibit the P450 system decrease the metabolism of tacrolimus, thereby increasing its blood levels. It is important to note that some studies have indicated a higher incidence of nephrotoxicity with tacrolimus as compared to CyA.

Azathioprine and Mycophenolate Mofetil (MMF)

Despite being available for more than 35 years, azathioprine is still a useful agent as an immunosuppressive agent. Following administration, azathioprine is converted into 6-mercaptopurine, with subsequent transformation to a series of intracellularly active metabolites. These inhibit both an early step in de novo purine synthesis and several steps in the purine salvage pathway. The net effect is depletion of cellular purine stores, thus inhibiting DNA and RNA synthesis, the impact of which is most marked on actively dividing lymphocytes responding to antigenic stimulation. In currently used immunosuppressive protocols, azathioprine is used as part of a triple therapy regimen along with CyA or tacrolimus and prednisone. Mycophenolate mofetil (MMF), which is rapidly hydrolyzed after ingestion to mycophenolic acid, is a selective, noncompetitive, reversible inhibitor of onosine monophosphate dehydrogenase, a key enzyme in the de novo synthesis of guanine nucleotides. Unlike other marrow-derived cells and parenchymal cells that use the hypoxanthine-guanine phosphoribosyl transferase (salvage) pathway, activated lymphocytes rely predominantly on the de novo pathway for purine synthesis. This functional selectivity allows lymphocyte proliferation to be specifically targeted with less anticipated effect on erythropoiesis and neutrophil production than is seen with azathioprine.

Early studies in human kidney and heart transplant recipients showed that MMF, when substituted for azathioprine in standard triple-therapy regimens, is well tolerated and more efficacious than azathioprine. In a large, double-blind, randomized multicenter study comparing MMF versus azathioprine (with CyA and prednisone) involving 650 patients, the MMF group was associated with significant reduction in mortality as well as a reduction in the requirement for rejection treatment. However, there was noted to be an increase in the incidence of opportunistic viral infections in the MMF group. The overall greater efficacy of MMF compared to azathioprine has resulted in MMF generally replacing azathioprine in triple immunosuppressive protocols together with steroids and cyclosporine in most solid organ recipients.


Steroids are routinely used in almost all immunosuppressive protocols after organ transplantation. The metabolic side effects of steroids are well known and lead to significant morbidity and mortality in the post-transplant period. Almost 90% of organ recipients continue to receive prednisone at 1-year post-transplant and 70% at three-years post-transplant. A recent review of over 1800 patients from a combined registry outlined the morbid complications that patients suffer within the first year after transplantation. Many of these complications are known side effects of prednisone, including hypertension (16%), diabetes mellitus (16%), hyperlipidemia (26%), bone disease (5%) and cataracts (2%). It is thereby obvious that avoidance of steroids may decrease morbidity and mortality after organ transplantation. Two general approaches are used to institute prednisone-free immunosuppression: early and late withdrawal.

Withdrawal of prednisone during the first month post-transplant has resulted in long-term success of steroid withdrawal in 50–80% of patients. In these studies, the use of antilymphocyte antibody induction therapy appears to increase the likelihood of steroid withdrawal. Several centers have reported their results with immunosuppressive regimens that did not include steroids in the early post-transplant period. Studies reporting high success rates of 80% have used specific enrolment criteria, such as excluding patients with recurrent acute rejections or those with female gender. Review of numerous studies demonstrate that steroid free maintenance immunosuppression is possible in atleast 50% of patients, is as safe as triple drug therapy and may reduce some of the long-term complications of steroids. Owing to the fact that the majority of acute rejection episodes occur in the first three months post-transplant, steroid withdrawal is made after this time period, resulting in long-term success in about 80% of patients. Generally, there is no need for conventional induction agents when late withdrawal of steroids is done.

Anti-Lymphocyte Antibody Therapy

Despite the extensive use of induction therapy using anti-lymphocyte antibody in solid organ transplantation, their exact role is unclear. There is no doubt that routine use of these agents is unwarranted as the generalized immunosuppression induced by then increased the risk of infections and malignancy. Despite the lack of consistent data supporting the routine use of induction therapy with anti-lymphocyte antibody agents, there is a role in certain select situations. Specifically, patients with early post-operative renal or hepatic dysfunction may benefit especially by the avoidance of cyclosporine therapy while using these induction agents. Anti-lymphocyte antibody therapy can provide effective immunosuppression for atleast 10 to 14 days without CyA or tacrolimus therapy. It has also been suggested that patients with overwhelming postoperative bacterial infections or diabetics with severe postoperative hyperglycemia may benefit from the comparatively low doses of corticosteroids required during anti-lymphocyte induction therapy.

The two main types of induction agents have been either the polyclonal antilymphocyte or antithymocyte globulins and more recently the murine monoclonal antibody OKT3. While these agents have been shown to be effective in terminating acute allograft rejection and in treating refractory rejection, the results of comparative studies of outcomes with and without monoclonal induction therapy have varied, with most studies demonstrating an effect on rejection that is maintained only while antibody therapy is ongoing. Without repeated administration, these agents only delay the time to a first rejection episode without decreasing the overall frequency or severity of rejection. More importantly, their use has been associated with an increased risk of short-term (infections) and long-term (lympho-proliferative disorders) complications. A complication specific to OKT3 is the development of a “flu-like syndrome” characterized by fever, chills and mild hypotension, typically seen with the first dose.

Since antilymphocyte antibodies are produced in nonhuman species, their use is associated with the phenomenon of sensitization, leading to decreased effectiveness with repeated use as well as the possibilty of serum sickness. The development of sensitization has been linked with an increased risk of acute vascular rejection. While this association has not been reported by other centers using OKT3 prophylaxis, it is believed that the development of immune-complex disease, inadequate immunosuppression due to decreased OKT3 levels or that OKT3 sensitization may be a marker for patients at higher risk for humoral rejection may be responsible for this phenomenon.

Interleukin-2 Receptor Inhibition

A new class of drugs has been developed which targets the high affinity IL-2 receptor. This receptor is present on nearly all activated T cells but not on resting T cells. In vivo activation of the high-affinity IL-2 receptor by IL-2 promoted the clonal expansion of the activated T cell population. A variety of rodent monoclonal antibodies directed against the a chain of the receptor have been used in animals and humans to achieve selective immunosuppression by targeting only T-cell clones responding to the allograft. Chimerisation or humanisation of these monoclonal antibodies resulted in antibodies with a predominantly human framework that retained the antigen specificity of the original rodent monoclonal antibodies. A fully humanized anti-IL2R monoclonal antibody, daclizumab, and a chimeric anti-IL-2R monoclonal antibody, basiliximab, have undergone successful phase III trials demonstrating their efficacy in the immunoprophylaxis of patients undergoing renal and cardiac transplantation.

Both agents have immunomodulatory effects that are similar to those of other monoclonal antibody-based therapies (i.e., induction of clonal anergy rather than clonal deletion). The advantages of these agents include their lack of immunogenicity, long half-lives, ability to repeat dosing, and short-term safety profile. Daclizumab appears to be an effective adjuvant immunomodulating agent in cardiac allograft recipients. It has advantages over conventional induction therapy as it is more selective and can be used for prolonged and potentially repeated periods. Studies with larger cohorts are needed to further study the short-term and long-term survival benefits for patients following organ transplantation and should determine the optimal dosing schedules of these new agents.


Materno-Fetal Tolerance

As outlined above, when tissues from an HLA-disparate donor are transplanted into a recipient they are always recognized as foreign, and immunosuppression is required to prevent rejection. An important exception to this is observed in pregnant women who tolerate their unborn fetus despite the fact that it expresses a full set of non-maternal HLA antigens inherited from the father. The mechanisms by which embryonic tissue demonstrates immune privilege during prenatal development have not yet been fully elucidated, however it is evident that interactions between fetus and mother differ substantially from the events triggered by a classical allograft. Consequently, much work is being dedicated to the emerging field of materno-fetal immunobiology in order to enable the development of innovative strategies to induce tolerance and prevent allogeneic graft rejection.

When maternal T cells encounter the fetus they demonstrate adaptive tolerance. In part this may be due to the absence of expression of MHC class II antigens and low levels of expression of MHC class I antigens on fetal cells. However, this can only partly explain the state of prolonged maternal tolerance since induction of HLA class I and II molecules inevitably occurs as the fetus matures and differentiates, yet rejection still does not occur. Consequently, non-fetal aspects of the placental barrier must be of critical importance in maintaining prolonged tolerance to the fetus. An important mechanism may relate to upregulation of the human non-classical HLA class Ib antigen, designated HLA-G, by the syncytiotrophoblast. HLA-G molecules bind to inhibitory receptors on natural killer cells and subsequently protect against maternal rejection responses. The placenta produces high levels of the anti-inflammatory cytokine interleukin 10 which stimulates HLA-G synthesis while concomitantly downregulating MHC class I antigen production, thus contributing to the tolerance-inducing local environment. The trophoblast also produces high levels of the enzyme indoleamine 2,3-dioxygenase, which catabolizes tryptophan, an essential amino acid necessary for rapid T cell proliferation. Annexin II, found in isolated placental membranes in vitro is present in placental serum, exerts immunosuppressive properties, and additionally contributes to fetal allograft survival. Together, these features indicate that materno-fetal tolerance results from a combination of transiently reduced antigenicity of the fetus in combination with a complex tolerance-inducing milieu at the placental barrier.

Immunogenic Characteristics Of Embryonic
And Adult Stem Cells

Murine and human embryonic stem (ES) cells do not express HLA class I and II antigens, and demonstrate reduced surface expression of co-stimulatory molecules important for T cell activation. Transplantation of murine ES cells demonstrates long-term graft survival despite the fact that these cells do acquire HLA class II antigen expression after in vivo differentiation. Since they are able to accomplish long-term engraftment without the need for immunosuppression, their inability to induce an immune response is not likely to be the result of escaping immune surveillance, but rather due to their ability to colonize the recipient thymus and induce intrathymic deletion of alloreactive recipient T cells.

Recently, a population of cells has been described in human adult bone marrow that has similar functional characteristics to embryonic stem cells in that they have high self-regenerating capability and capacity for differentiation into multiple cell types, including muscle, cartilage, fat, bone, and heart tissue. While such cells, termed adult mesenchymal stem cells (MSC), appear to have a more restricted self-renewal capacity and differentiation potential than ES cells, their functional characteristics may be sufficient for clinically meaningful tissue regeneration. A striking recent observation is that MSC can broadly inhibit T-cell proliferation and activation by various types of antigenic stimulation, including allogeneic stimuli. MSCs have been shown to inhibit both naive and memory T cell responses in a dose-dependent fashion and affect cell proliferation, cytotoxicity, and the number of interferon gamma (IFN-gamma)-producing T cells. MSCs appear to inhibit T cell activation through direct contact, and do not require other regulatory cellular populations. Similarly to ES cells, adult bone marrow-derived mesenchymal stem cells (MSCs) do not express HLA class II molecules, and only low levels of HLA class I molecules. Despite the fact that MSC can be induced to express surface HLA class II molecules by in vitro culture with cytokines such as interferon-gamma, their ability to inhibit T cell activation results in induction of T cell non-responsiveness to the MSC themselves, endowing them with potential survival advantages in the setting of transplantation.

Tolerogenic Effects Of Stem Cell Transplantation

Extending the approaches discussed above using donor-derived blood transfusions to induce a tolerogenic state to the subsequent organ, the most promising clinical strategy for tolerance induction at present is the use of donor-derived hematopoietic stem cells in conjunction with reduced myeloablative conditioning. The objective of this therapy is to achieve a state of so-called mixed chimerism, or the permanent co-existence of donor- and recipient-derived blood cells comprising all the different hematopoietic lineages in the same host. This approach has been tested in a variety of small and large animal settings and currently available data suggest that stable engraftment of donor bone marrow reliably renders the host tolerant to donor antigens and subsequently to any cellular or solid organ graft of the same donor.

The two underlying mechanisms by which creation of a mixed-chimeric host results in tolerance induction are (1) thymic deletion of potentially donor-specific alloreactive T cells, and (2) nonthymic peripheral mechanisms, such as blocking costimulatory T cell activation, which facilitate the process of donor bone-marrow or stem cell engraftment. However, despite the efficacy of an approach using fully HLA-mismatched stem cells in an allogeneic host to induce tolerance to a subsequent organ allograft, the host is placed at a high risk of substantial morbidity and mortality due to toxicity of the myeloablative conditioning regimen and potential for graft-versus-host disease, or immune-mediated attack of the host by the implanted allogeneic stem cells.

In an attempt to overcome these potential limiting toxicities, investigators have suggested the use of either adult bone marrow-derived mesenchymal stem cells or preimplantation-derived embryonic stem (ES) cells for induction of mixed chimerism. The theoretical advantages of these cell types is their low level of surface expression of HLA class I and II antigens, and reduced surface expression of co-stimulatory molecules important for T cell activation. Rat preimplantation stage derived embryonic-like stem cells have been shown to successfully engraft in the recipient bone marrow without the need for pre-conditioning therapies such as irradiation, cytotoxic drug regimens or T cell depletion. Long-term partial mixed chimerism by use of rat preimplantation stage derived embryonic-like stem cells did not trigger graft-versus-host reactions, in contrast to the high frequency of this complication in the clinical setting of allogeneic hematopoietic stem cell transplantation Of most interest, the induced partial chimerism enabled the recipient animals to be tolerant to a subsequent heart allograft. Allograft acceptance required the presence of an intact thymus, and rat ES cells were present in the recipient thymus.

Similar results have been reported following transplantation of human adult bone marrow-derived mesenchymal stem cells (MSC) into fetal sheep early in gestation, before and after the expected development of immunologic competence. In this xenogeneic system, human MSC engrafted, differentiated in a site-specific manner, and persisted in multiple tissues for as long as 13 months after transplantation, including the thymus. Since MSCs do not present alloantigen and do not require MHC expression to exert their inhibitory effect on alloimmune reactivity, the possibility exists that they could theoretically be derived from a donor irrespective of their HLA type and used to inhibit T-cell responses to transplantation antigens of an unrelated third party. In initial human clinical studies, the use human adult bone marrow-derived mesenchymal stem cells has been shown to successfully enable engraftment of subsequently infused allogeneic bone marrow in transplant recipients, reduce the risk of graft-versus-host disease, and reduce the need for concomitantly administered immunosuppression. Whether similar results will be obtained when combining adult bone marrow-derived mesenchymal stem cells with solid organ allografts remains to be determined, and this is an area of active research for clinical transplant immunobiologists. Of broader relevance, if the results relating to long-term engraftment and survival of adult bone marrow-derived MSC are confirmed and extended in human clinical studies, they will have broad implications for the field of tissue and organ regeneration.


1. Billingham RE, Brent L, Medawar PB (1953) Actively acquired tolerance of foreign cells. Nature 172:603-606
2. Auchincloss H Jr (2001) In search of the elusive holy grail: the mechanisms and prospects for achieving clinical transplantation tolerance. Am J Transplant 1:6-12
3. Wekerle T, Kurtz J, Ito H, Ronquillo JV, Dong V, Zhao G, Shaffer J, Sayegh MH, Sykes M (2000) Allogeneic bone marrow transplantation with co-stimulatory blockade induces macrochimerism and tolerance without cytoreductive host treatment. Nat Med 6:464-469
4. Watkins DI (1995) The evolution of major histocompatibility class I genes in primates. Crit Rev Immunol 15:1-29
5. Salter-Cid L, Flajnik MF (1995) Evolution and developmental regulation of the major histocompatibility complex. Crit Rev Immunol 15:31-75
6. Fowlkes BJ, Ramsdell F (1993) T-cell tolerance. Curr Opin Immunol 5:873-879
7. Marrack P, Kappler J (1990) T cell tolerance. Semin Immunol 2:45-49
8. Van Parijs L, Abbas AK (1998) Homeostasis and self-tolerance in the immune system: turning lymphocytes off. Science 280:243-249
9. Davies JD, Martin G, Phillips J, Marshall SE, Cobbold SP, Waldmann H (1996) T cell regulation in adult transplantation tolerance. J Immunol 157:529-533
10. Thomson AW, Lu L (1999) Dendritic cells are regulators of immune reactivity: implications for transplantation. Transplantation 68:1-8
11. Brent L (1997) Fetally and neonatally induced immunologic tolerance. In: Brent L (ed) A history of transplantation immunology. Academic, San Diego.
12. Matzinger P (1994) Tolerance, danger, and the extended family. Annu Rev Immunol 12:991-1045
13. Janeway CAJ (1992) The immune system evolved to discriminate infectious nonself from noninfectious self. Immunol Today 13:11-16.
14. Mellor AL, Munn DH (2001) Extinguishing maternal immune responses during pregnancy: implications for immunosuppression. Semin Immunol 13:213-218.
15. Fändrich F, Lin X, Chai GX, Schulze M, Ganten D, Bader M, Holle J, Huang DS, Parwaresch R, Zavazava N, Binas B (2002) Preimplantation-stage stem cells induce allogeneic graft tolerance without supplementary host conditioning. Nat Med 8:171-178
16. Liechty KW, et al (2000). Human mesenchymal stem cells engraft and demonstrate site-specific differentiation after in utero transplantation in sheep. Nat Med 6(11):1282-6.
17. Krampera M, et al (2002). Bone marrow mesenchymal stem cells inhibit the response of naive and memory antigen-specific T cells to their cognate peptide. Blood, 101(9):3722-3729.
18. Le Blanc K, Tammik L, Sundberg B, Haynesworth SE, Ringden O (2003). Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex.Scand J Immunol. 57(1):11-20.


  - The President's Council on Bioethics -  
Home Site Map Disclaimers Privacy Notice Accessibility NBAC HHS