• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Xenograft Transplantation: Ethical Issues and Public Policy. Xenotransplantation: Science, Ethics, and Public Policy. Washington (DC): National Academies Press (US); 1996.

Cover of Xenotransplantation

Xenotransplantation: Science, Ethics, and Public Policy.

Show details

2Assessing the Science Base


Xenotransplants, or xenografts, are viewed as one solution to the growing problem of the inadequate supply of human organs and as a source of cells, tissues, and organs that show promise for treating a variety of human diseases. This chapter was drawn largely from the workshop Session I: Assessing the Science Base. Thus, the majority of the chapter summarizes workshop presentations. Where useful for background, some sections have been supplemented with additional information. The chapter, however, is not intended as an in-depth analysis and summary of the field of transplant immunology. The committee is aware that there are views of this field other than those presented here and that there are facets of transplant immunology not discussed.

The immunological response to a xenograft depends, in part, on the phylogenetic distance between the source animal and the host. Transplants between closely related species (e.g., example, humans and nonhuman primates or rats and guinea pigs) are called concordant transplants and are less likely to elicit an immediate reaction. However, transplants between more distantly related species, such as those from swine to humans, are called discordant transplants and elicit an immediate immunological reaction that destroys the endothelial lining of the blood vessel of the graft. Transplants of cells and tissues are not subject to immediate rejection since they lack blood vessels. Thus, discordant transplants of whole organs are rejected immediately, whereas such a reaction does not occur with concordant transplants.

For the most part, rejection of a xenograft is more vigorous than rejection of an allograft. Explanations for this are the stronger nature of the immunological responses against xenografts than against allografts and the existence of immunological reactions by the host that are unique to xenotransplantation. Although immunosuppressive drugs are highly effective in allotransplantation, rejection, especially late after transplantation, still occurs. In contrast, immunosuppressive agents cannot overcome certain aspects of xenograft rejection. It is largely for this reason that investigators worldwide are seeking novel therapies that can supplement the use of immunosuppression to achieve survival of xenografts. New immunosuppressive drugs will also be needed to suppress certain aspects of xenograft rejection. Xenografts of cells and tissues are the first to reach small-scale human trials because means of preventing or blunting the immune response to those transplants are being developed rapidly.

The promise of success of xenografts is derived from research on newer therapies and strategies in animals. Heart transplants from monkeys to baboons (closely related species) have survived for months—some for more than a year. These encouraging results have prompted researchers to propose human trials in which baboon hearts are envisioned as "bridges" for patients awaiting human organs (i.e., as a means of sustaining life until a human donor becomes available). Success has also been achieved with long-term survival of transplanted organs between various rodent combinations (e.g., hamster and rat), but questions remain about the applicability of results of experiments conducted in rodents to human xenografting.

Although the major problem in xenotransplantation is rejection, biochemical and physiological aspects of xenograft function are also unanswered concerns. For example, transplanted bone marrow can react with the heart to cause complications. As new agents are developed and tested, the adverse side effects of those agents will require close monitoring (see Appendix C)

Strategies to counteract human immune rejection of xenografts vary with the tissue or organ being transplanted, the disease being treated (e.g., diabetes, AIDS, or Parkinson's disease), the nature of the transplant (tissues or organs), and the phylogenetic dissimilarity between the patient and the source animal. Both tissues and organs are being modified by either immunological or genetic engineering approaches. Cells or tissues such as pancreatic islets have been encapsulated to deny access to the recipient's immune system. Whole organ xenografts between phylogenetically distant species are the most difficult because a series of new rejection mechanisms, not previously encountered in allograft rejection, must be addressed.

Stages Of The Immune Response

The immunological response to an organ xenograft can be divided into different phases, although the division is to some extent arbitrary, both because the manifestations of one can continue into, or be present in, another and because they can occur as a continuum. The first phase, hyperacute rejection, does not ordinarily occur in allotransplants, although in certain situations there can be hyperacute rejection of an allograft if the recipient has certain types of antibodies against the donor cells (Platt, 1995). Current practices detect the presence of such antibodies, and hyperacute rejection is avoided. Hyperacute rejection is a serious problem because conventional immunosuppressive drugs cannot control it. Hyperacute rejection occurs in discordant but not in concordant transplantation.

Each phase of rejection is characterized by the time of its appearance after transplantation and by the rejection mechanisms that seem to be involved (Table 2-1). Hyperacute rejection begins in minutes and results in rejection of the organ within one to two hours in the great majority of cases. If hyperacute rejection is avoided, then the graft is rejected after a period of some days by a process known as delayed xenograft rejection (also referred to as acute vascular rejection), which begins within hours and results in rejection after several days (Table 2-1). Although not yet proven, since delayed xenograft rejection has not been completely overcome in a discordant organ transplant model, it is exceedingly likely that the next phase of rejection would involve the xenograft counterpart of a T-cell (cell-mediated) immune response that is responsible for rejection of an allograft. Only if all of these phases are overcome does one expect that a discordant organ xenograft may undergo chronic rejection , after a period of weeks or months.

TABLE 2-1 Different Types of Xenotransplant Rejection .


TABLE 2-1 Different Types of Xenotransplant Rejection .

Apart from the pace and time of onset, the phases are also distinguishable on the basis of whether the immune response is humoral, cellular, or both. Humoral immunity refers to host responses to an infectious agent or foreign protein by molecules such as antibodies and complement that are found in body fluids. Cellular immunity refers to reactions mediated by whole cells such as T-lymphocytes. Both of these reactions can secondarily involve cells other than T-cells including neutrophils, macrophages, and natural killer cells, which are found in blood and tissues. Hyperacute and delayed xenograft rejection are due primarily to humoral immunity in which preexisting antibodies of the recipient act together with the recipient's complement to initiate the rejection response, although it is hypothesized that other stimuli can initiate delayed xenograft rejection (Bach et al., 1995). Cell-mediated (T-cell) rejection is initiated by T-cells and can involve other cells. The causes of chronic rejection are poorly understood but likely involve both cellular and humoral immune responses.

Hyperacute rejection is a formidable obstacle to successful whole organ xenografting. It is easily recognized clinically because the graft very rapidly becomes swollen and turns black. Hyperacute rejection results from preformed xenoreactive natural antibodies binding to endothelial cells that line the blood vessels of the graft. Endothelial cells stimulated by xenoreactive antibodies plus complement become activated (Bevilacqua et al., 1984; Pober and Gimbrone, 1982). This activation very rapidly results in a profound disruption of endothelial cell integrity and function. Endothelial cells are normally linked tightly to one another to create a barrier that prevents leakage of cells and proteins from the blood into the extravascular space of the organ. Also, endothelial cells normally express, on their surfaces, molecules that prevent clotting and platelet aggregation. When the endothelium is activated, as in hyperacute rejection and delayed xenograft rejection, its barrier and anticoagulant functions are lost, blood cells and fluid leak out into tissues resulting in hemorrhage and edema, and thrombosis occurs in the graft. The sequence of molecular events that culminates in hyperacute graft rejection is described in the section dealing with mechanisms of hyperacute rejection.

Hyperacute rejection is not a factor in cellular xenografts, such as transplants of pancreatic islet cells, because such grafts are revascularized by the recipient; thus, the endothelial cells are self (i.e., not "foreign").

If hyperacute rejection is prevented by depleting xenoreactive antibodies or blocking complement in the recipient, then the graft survives to experience delayed xenograft rejection. This form of rejection also involves endothelial cell activation, but in this instance it appears to be due primarily to upregulation of new sets of genes on the activated endothelial cells. It is also possible that some of the features of endothelial cell activation associated with hyperacute rejection, which precipitate the earlier thrombosis and inflammation, actually cause delayed xenograft rejection.

Cell (T-lymphocyte) mediated rejection is postulated to be the next phase of the immune response against a discordant organ xenograft. By analogy with allograft rejection, a T-lymphocyte-mediated response in xenograft rejection is initiated when the recipient's T-lymphocytes recognize foreign antigens on the cells of the graft. The antigens recognized are those associated with the major histocompatibility complex (MHC).

Immunosuppressive drugs are used to suppress the T-cell response. For the most part, suppression of this response is quite successful for allotransplantation, although episodes of acute rejection can occur despite immunosuppression. In the nonimmunosuppressed patient, cell-mediated immunity leads to rejection of an organ allotransplant in 7-10 days. If immunosuppression is unsuccessful, rejection of the organ will occur at this time or a little later. A deterioration or change in graft function typically prompts clinicians to suspect acute rejection. Finding immune cell infiltrates in biopsies of the graft is diagnostic. Immune cells can react either with graft endothelium passenger white blood cells (leukocytes) in the graft or with graft parenchymal cells (cells essential for organ or tissue function).

T-cell-mediated rejection, especially that associated with the action of helper (CD4+) T-lymphocytes, involves recruitment and activation of macrophages, plus cytokine-mediated effects on graft endothelial cells. The result is development of graft thrombosis and ischemia due to loss of the anticoagulant molecule, thrombomodulin, and induction of a procoagulant tissue factor that is present on both infiltrating macrophages and graft endothelial cells (Hancock, 1984; Platt, 1994).

Finally, chronic rejection is marked by a slow but progressive loss of graft function that begins months to years after transplantation. Pathologically, graft tissue architecture deteriorates during chronic rejection. Both cellular and humoral immune responses have been implicated but have not been well characterized. Chronic rejection is difficult to manage with conventional immunosuppressive medication and may progress to the point where another transplant is needed. What is known about chronic rejection comes from the study of allografts, primarily because discordant xenografts have not remained viable long enough to meet with chronic rejection.

Progress in Molecular and Cellular Biology

Much scientific progress has been made in defining the molecular basis of hyperacute and delayed xenograft rejection, the first and second phases of the host immune response to a discordant organ xenograft. This understanding allows design of potential therapeutic strategies for overcoming immune rejection.

Mechanisms of Hyperacute Rejection

Hyperacute rejection occurs when vascularized whole organs are transplanted between species combinations that are phylogenetically distant, such as pigs and humans. Hyperacute rejection begins when preformed xenoreactive natural antibodies (which are present at the time of transplantation) in host blood combine with cell surface molecules on graft endothelial cells. The binding of antibodies to these cell surface molecules induces a conformational change in the antibody that exposes a binding site for complement, another key component of humoral immunity. These events on the surface of the endothelium cause the endothelium to lose two critical properties, its barrier and antithrombotic functions. Disruption of the endothelium leads to edema and hemorrhage as blood migrates from the vessel into host tissue. To seal the gaps in the endothelium, thereby stopping blood loss, platelets aggregate and blood clots. These clots obstruct the flow of blood that feeds the graft. Without blood flow, the graft dies.

Although the initial step in this sequence of events is antibody binding to graft endothelium, the subsequent binding of complement is an important event in the immune response (termed an effector function). Although complement is best recognized for its ability to destroy its target by lysing the cell membrane, the role of complement in hyperacute rejection does not appear to be through lysis of the graft endothelial cell. Rather, as discussed below, the role of complement is to induce endothelial cells to change their shape, and thus lose their barrier function, to release several molecules from their surface, including thrombomodulin, heparan, sulfate, and ecto-ADPase (adenosine disphosphatase). These molecules maintain an antithrombotic environment around the endothelial cells. Later, if hyperacute rejection is avoided, complement acts to regulate a large number of genes that contribute to thrombosis and inflammation.

Complement is activated when it binds to an antibody molecule. Host antibodies that bind to the xenograft, xenoreactive antibodies, are mostly of the IgM isotype, one of the five classes of antibodies (the other four are IgA, IgD, IgE, and IgG). The more distantly related the species combination, the higher is the concentration of these xenoreactive IgM antibodies in host blood (Hammer, 1989). Pathology studies using markers that bind to IgM have identified IgM antibodies along endothelial cell surfaces of xenografts; depletion of serum IgM antibodies before transplantation prevents hyperacute rejection.

Recent studies have identified the specific target (epitope) for xenoreactive antibodies on the surfaces of endothelial cells (Cooper, 1995; Squinto and Fodor, 1995). It is a complex molecule, ending in a simple sugar called α-galactose, which is attached to a cell surface protein to form a glycoprotein.

These glycoproteins (those having the α-gal epitope) are found in most mammals, with the exception of apes, Old World monkeys, and humans. Because most nonhuman primates lack these glycoprotein antigens, they form antibodies against the α-gal and thus have the antibodies present before the transplant. The removal of the galactose portion of surface glycoproteins on endothelial cells of pigs reduces the binding of xenoreactive antibodies of primate hosts by 80–90 percent and, thus, also reduces the binding of complement.

These findings are important because binding of xenoreactive antibody to surface glycoproteins activates the complement system and initiates a cascade of enzymatic reactions that can be likened to a domino effect. To understand this cascade, it is essential to point out that complement actually refers to one of several classes of proteins that circulate in the blood as proenzymes (i.e., enzymes whose active site is masked). When one of the complement proteins is activated through antibody binding, it becomes active as an enzyme and causes a change in another complement protein. This reaction generates yet another complement enzyme, which affects yet another complement protein. The multistep process ultimately forms a complex called the membrane attack complex and other intermediary complement products that take part in the immune reaction. Both membrane attack complexes and intermediary products participate in hyperacute rejection.

One complement product (C5a), together with the preformed antibodies, induces endothelial cells to lose heparan sulfate, a substance that prevents blood clotting. This promotes the formation of platelet clots, leading to ischemia. Another complement protein causes changes in the structural integrity of endothelial cells, which leads to gaps in the endothelial surface and subsequent edema and hemorrhage (Platt et al., 1990).

The critical role of antibody binding with subsequent complement activation in hyperacute rejection is supported by experiments in which either xenoreactive antibody or complement is depleted prior to xenotransplantation. These procedures have been shown to prevent hyperacute rejection in that the graft is no longer rejected within hours. In a number of experiments, antibodies were successfully removed by plasmapheresis, immunoabsorption, and other procedures. In another series of animal studies, complement was reduced to nearly undetectable levels before transplant by administration to the host of purified cobra venom factor, which depletes complement. When the whole organ was then transplanted from guinea pigs to rats or pigs to baboons, graft survival was extended from approximately 90 minutes to several days. However, the graft was destroyed after a few days in what is referred to as delayed xenograft rejection (Bach et al., 1995). Cobra venom factor is not a realistic long-term therapeutic strategy for human recipients because of its toxicity and the requirement for continuous infusion. Nonetheless, these studies indicate that techniques that avoid hyperacute rejection may allow the use of xenografts as temporary or bridge measures for patients awaiting a suitable allograft.

Because complement is such a basic part of the host immune system, cells must be protected when complement molecules are activated on their surface. (Complement is usually activated by antibody, as described above, but can also be activated without the participation of antibody.) To protect cells from activated complement, mammals have evolved a number of mechanisms. The most significant mechanism, from the standpoint of potential therapies for hyperacute rejection, involves a group of cell surface proteins called complement regulatory proteins, or regulators of complement activation (RCA). These proteins are species-specific and serve to inhibit complement activation at various points in the complement cascade. The surge in knowledge about complement regulatory proteins and the genes that encode them has spawned a variety of approaches to counteract hyperacute rejection, some of which are discussed in a later section of this chapter.

Mechanisms of T-Cell-Mediated Rejection

Most of our knowledge about T-cell-mediated reactions comes from studies of allografts. Such studies are the basis of the discussion that follows (Faustman, 1995).

T-cell-mediated rejection of organ transplants occurs within days to weeks of transplantation and has been characterized only in allografts. T-cell-mediated rejection is characterized by gradual loss of graft function brought about by a cellular immune response. Host immune cells react with a variety of graft cells, including vascular endothelial cells and parenchymal cells. Humoral immunity may also play a role, but the presence in graft biopsies of cellular infiltrate suggests that cellular immunity predominates.

It has long been known that the response of host T-cells to histocompatibility antigens on the surfaces of cells in the graft is responsible for acute allograft rejection. The detailed molecular mechanisms of recognition and response, however, were poorly understood before the great progress made over the past years in the fields of immunology and molecular biology. Much of this progress has been to identify antigenic determinants, identify how antigen is presented to the host immune system, and identify how the immune system reacts.

Cell-mediated rejection is initiated by activation of host T-lymphocytes (T-cells). T-cells play a pivotal role in controlling the immune response that destroys graft cells. T-cells participate in the activation of macrophages, natural killer cells, B-lymphocytes, and other immune cells. Some of the most effective immunosuppressive drugs in clinical use today work by disabling critical T-cell function (see Appendix C for discussion of desired characteristics of immunosuppressive drugs; based on Kahan, 1995, and Kahan and Ghobrial, 1994).

The graft antigens that most commonly activate host T-cells are MHC Class I and II proteins. These are cell surface markers that not only distinguish one species from another but also distinguish among members of a given species. The amino acid variation in MHC proteins is greater between species than within species, but within-species variability is enormous and elicits an immune response. Put simply, MHC molecules dictate whether the graft is accepted as self or is rejected as nonself. An allograft recognized as self has MHC Class I and II surface markers identical to the host and has a number of other histocompatibility antigens, referred to as minor histocompatibility antigens, that are identical to the host. A mismatch in MHC antigens between graft and host leads to acute T-cell rejection in both allografts and xenografts. The response by the host to foreign MHC antigens is very strong, even stronger than the host's reactions to other foreign antigens, such as viruses and bacteria. Subsequent in vitro studies have shown that the response to foreign MHC molecules is greater than that to other antigens by a factor of up to 100 (Kaufman et al., 1995).

MHC molecules are expressed on all cells in the body, although the class of MHC molecule varies. MHC Class I molecules are found on all mammalian nucleated cells, but Class II molecules are expressed on only a few cell types: endothelial cells and immune cells such as B-cells, dendritic cells, monocytes, and macrophages. Accordingly, a graft possesses MHC Class I molecules on all of its cells, but MHC Class II molecules are present only on endothelial cells in its blood vessels and on its passenger leukocytes, which are immune cells residing in any tissue. It is important to understand the distribution of Class I and Class II molecules on the cells of the graft, because their presence or absence determines which type of T-cell is activated. For example, porcine graft endothelial cells have both Class I and Class II molecules on their surface and thereby activate two types of host T-cells, whereas endothelial cells of some other species do not constitutively express Class II antigens.

Xenograft cells having Class I antigens activate host CD8+ T-cells (cytotoxic T-lymphocytes), while those having Class II antigens activate host CD4+ T-cells (helper T-cells). The designations CD4+ and CD8+ refer to molecular markers on the surface of the T-cell. When a CD4+ T-cell recognizes a foreign Class II molecule, it does not directly destroy the target cell, but rather stimulates other immune cells to do so. First, an activated CD4+ T-cell proliferates; then, its daughter cells release intercellular messengers termed cytokines or lymphokines. Cytokines signal B-cells (B-lymphocytes) to differentiate and to secrete antibodies that react with antigens. Cytokines secreted by CD4+ cells also stimulate CD8+ T-cells, macrophages, and other immune cells to destroy a given tissue (Table 2-1). In short, foreign Class II antigens on the surface of appropriate cells activate host CD4+ T-cells that orchestrate an immune response by a variety of immune cells.

MHC Class I molecules are recognized by CD8+ T-cells. When this type of T-cell is activated, it proliferates into a population of effector cells called cytotoxic T-lymphocytes (CTLs), which have the capability to bind to cells carrying the appropriate antigen. It should be noted that xenografts possess other antigens besides Class I and II molecules that are recognized and destroyed by the host immune system, but they do not appear to play as important a role, at least in the systems thus far studied.

How host T-cells recognize foreign antigens has been the focus of much research that has led to the development of immunosuppressive medications. T-cell recognition has been found to be a more complex process than antibody recognition of antigen. Host T-cells can recognize an antigen through either direct or indirect presentation of the antigen. With direct presentation, the antigens on the surface of the graft cells (e.g., Class I and II molecules) are recognized directly by the T-cell. With indirect presentation, in contrast, the antigen must first be processed by another kind of host immune cell: an antigen-presenting cell. After the antigen-presenting cell internalizes the antigen, it displays a fragment of that antigen on its cell membrane in the exterior groove of its Class I or Class II molecule. Thus, the host MHC molecule essentially acts as a guidance system or recognition structure for the T-cell. What is intuitively difficult to grasp is that the antigen is a fragment of the graft's MHC molecule, but the ability to recognize this foreign antigen depends on the fragment being displayed to the host T-cell by the host's MHC molecule. Both indirect and direct antigen presentation take place with xenografts, depending on the species combination and on whether the graft is a whole organ or a tissue.

Novel Therapeutic Approaches

Limitations of Immunosuppression in Xenotransplantation

Rejection reactions are an inevitable result of all organ and cell transplants, except those between identical twins. Many new immunosuppressive medications have been developed in the past two decades for the treatment of rejection in allotransplantation (see Appendix C). These medications attenuate or slow the immune reactions responsible for rejection, although all too frequently they do not completely abolish them since acute rejection episodes and chronic rejection still occur. Although not without serious side effects, the new generation of immunosuppressive drugs has significantly prolonged graft survival of kidneys, livers, hearts, and combined heart–lung transplants, among others.

Medications successful at treating allograft rejection are being investigated in xenograft research, with mixed results. Some are effective for organ transplants between concordant species, particularly in hamster-to-rat transplants. For discordant organ transplants, there are some data showing that immunosuppressive drugs can maintain the organ in the presence of lowered levels of xenoreactive natural antibodies and can prolong graft survival by a few days through mechanisms that have not been elucidated yet. The major obstacle to solid organ discordant xenografts is hyperacute rejection, as discussed above, which cannot be overcome with immunosuppression.

Clinical experience with xenografts is so limited that it is difficult to forecast the effectiveness of existing immunosuppressives in combating immune rejection. Patients receiving xenografts three decades ago did not have the benefit of today's immunosuppressives. In two patients receiving baboon liver transplants, Starzl and colleagues (1993) administered tacrolimus, corticosteroids, and prostaglandin, a regimen routinely used for liver allografts. To prevent humoral rejection, they added cyclophosphamide, which has broad-ranging effects on many cells. The regimen appeared effective in preventing graft rejection, but left the patients vulnerable to infection, from which they died. It should be pointed out, however, that one of the patients was infected with the human immunodeficiency virus, which attacks the immune system. A compromised immune function prior to transplant does not allow conclusions to be drawn about the efficacy of immunosuppression to prevent graft rejection.

Advances in molecular biology and immunology have suggested new therapeutic strategies beyond immunosuppression to overcome xenograft rejection. The range of approaches can be grouped into three major categories: modification of the source animal, development of chimeras by bone marrow transplantation, and encapsulation of the cells or tissues to be transplanted. The three approaches, their advantages and disadvantages, and their applications to xenografts are discussed below and summarized in Table 2-2.

TABLE 2-2 Novel Antirejection Strategies in Xenotransplants.


TABLE 2-2 Novel Antirejection Strategies in Xenotransplants.

Modification of the Source Animal

Through modification of animal tissue, investigators can create designer tissues to render them less immunogenic or susceptible to rejection. Tissue is designed or modified by one of several methods that use immunological or genetic engineering techniques. The first method is through masking of antigen with antibodies. The second is through genetic manipulation with transgenic technology, which includes the use of blocking strategies at the RNA (ribonucleic acid) level to stop translation of the antigen or another factor. The third is through gene knockout technology, which selectively inactivates the gene coding for the antigen or another factor that is involved in rejection. Each of these approaches is described below. The idea of modifying the source animal is to avoid immunosuppression or any other treatment of the host that might be toxic.

Antibody Masking

Antibody masking strives to conceal antigens derived from the source animal from the host's immune system by covering them with antibody fragments. Antigens on the surface of the cell derived from the animal become hidden from cytotoxic T-lymphocytes of the host. This approach has demonstrated success in animals and represents an example of successful xenotransplantation of tissues without the need for immunosuppression. The cardinal studies introducing the concept of designer tissues by donor antigen masking were performed in rodents transplanted with human cadaveric islet cells and liver cells (Faustman and Coe, 1991). Before transplantation, the human cells were treated in vitro with antibody fragments, which shielded the MHC Class I antigens responsible for acute rejection by host cytotoxic T-lymphocytes. Class I antigens were selected for antibody masking because of their expression on islet cells and because of the sparse expression of other epitopes such as ICAM-1 and LFA-3, which on other cell types promote adhesion between host cells and the graft. It was not necessary to mask epitopes responsible for hyperacute rejection, because the islet and liver cells were nonvascularized and, as such, eventually carried endothelial cells of the host.

In these experiments, human islets implanted into rodents survived more than 200 days, without the need for host immunosuppression. Their proper functioning was demonstrated by physiological and histological assays. Liver cells and neuronal cells similarly treated have been shown to survive in xenogeneic experiments in animals. The importance of masking Class I antigens was substantiated by experiments that eliminated (by depleting passenger leukocytes) or masked other immune determinants, such as CD29, which did not result in similar survival in this model system.

Success with antibody masking of Class I antigens in xenotransplantation experiments in animals has led to human clinical trials. Antibody-treated fetal pig neurons rich in dopamine are being transplanted into the brains of patients with Parkinson's disease.

Source animal modification through antigen masking has proved effective with cells and homogeneous tissue such as islets in a few selected models tested to date. Source animal modification is also feasible, although far more difficult, for whole organs (Faustman, 1995). Not only do whole organs contain a multiplicity of cell types with a variety of antigenic determinants, but organs also contain vascular endothelial cells foreign to the recipient and, thus, are complicated by hyperacute immune rejection. Therefore, alternative strategies have been developed for modification of organs, including the production of transgenic animals (i.e., animals that express an additional gene introduced by genetic methodology).

Transgenic Modification

The production of transgenic organs also has potential as a strategy to shield animal organs (and tissues) from rejection by humans without the need for immunosuppression and represents another form of source animal modification (Squinto and Fodor, 1995). The source organ can be modified at the genetic level before implantation. Genes that are important to prevent rejection can be added through transgenic technology. In transgenic modification, either all cells of the animal contain the foreign gene (transgene) which is incorporated stably into their genome expressing the protein, or only selected cells contain it due to the use of promoters (genetic elements that control expression of a gene) that are specific for a single cell type. The transgenes most commonly used to date in xenograft research molecules that inhibit the human complement cascade, and are thus, intended to block hyperacute rejection.

The idea of creating transgenic donors that express human inhibitors of complement was proposed by Dalmasso and Bach (Bach et al., 1991; Dalmasso et al., 1991). These investigators showed that incorporating human decay-accelerating factor (DAF) (which blocks the complement cascade), at the stage of C3, into porcine endothelial cells in vitro blocked the ability of the human complement to lyse the porcine endothelial cells. They based their suggestion for creating such transgenic pigs on the fact that complement inhibitors such as DAF are species-specific; therefore, the porcine DAF present in the endothelial cells of a porcine organ that is transplanted to a human would not effectively inhibit human complement and would not prevent hyperacute rejection. Thus, the human gene DAF should be expressed in source animals by microinjection of the gene into the nucleus of a fertilized porcine egg by conventional transgenic technology.

Such manipulation gives rise to a mature source animal that incorporates the transgene into the genome of all cells. The mere presence of the human gene for DAF in every cell of the source animal does not, however, ensure its expression at levels high enough to inhibit complement. High level are required, particularly on the surface of source animal endothelial cells, where complement is activated. Therefore, a promoter DNA sequence that boosts the level of expression of the transgene is attached to the human DAF gene. The hope in these experiments is that high enough levels of expression of human DAF will be achieved in porcine endothelial cells to block complement action. It appears that transgenic source organs that express human DAF are not rejected hyperacutely, at least in some cases.

Another gene encoding an inhibitor of complement that has been used to make transgenic source animals is CD59, which inhibits the final reaction in the complement cascade (i.e., formation of the membrane attack complex). Studies of the efficacy of transgenic organs expressing human CD59 have yielded mixed results. Porcine hearts, kidneys, and lungs that express human CD59 have been transplanted into Old World monkeys. These recipients are similar to humans insofar as they carry preformed antibodies causing hyperacute rejection of porcine organs. After transplantation, transgenic organs survived up to 48 hours, a significant improvement over nontransgenic organs, which survived only about an hour. The use of transgenic animals expressing inhibitors of complement may succeed in inactivating complement proteins, but it does not prevent binding of preformed antibodies. Although it was originally thought that the binding of preformed antibodies was benign as long as the ensuing complement attack was arrested, further research suggests that the binding of preformed antibodies does more than just activate complement. Antibody binding to the vascular endothelium of the xenotransplant, by itself, can cause deleterious changes in the endothelial cell (Platt, 1994). An alternative approach to blocking hyperacute rejection is by abolishing expression of the gal α (1–3)gal epitope that is recognized by preformed antibodies in the host on the graft endothelial cells, an idea proposed by Sandrin McKenzie and colleagues (McKenzie et al., 1995), and others. This idea is being tested by inserting a gene that causes fucose instead of galactose to be added at the end of the carbohydrate chains, yielding the carbohydrate surface marker found on universal donor, or O blood group, cells. This fucose-containing surface glycoprotein does bind the xenoreactive antibodies. One research team has developed a transgenic mouse cell line that over expresses the enzyme, human transferase (H-transferase). This enzyme competes with the one that catalyzes placement of the galactose at the end of the carbohydrate chain, thereby reducing the formation of the α-galactose antigenic epitope on the surface glycoprotein and blocking hyperacute rejection. Herds of transgenic pigs and mice expressing H-transferase are currently being produced to determine the utility of this approach for transplantation to humans.

Gene Knockout

Another technique from molecular biology that can be used to modify a donor organ is referred to as gene knockout technology (Koller, 1995). It is used to inactivate a given gene(s). In theory, gene knockout can be used to delete permanently from animal organs and tissues genes for any antigens or other factors that elicit rejection. To date, gene knockout technology has been limited to use in mice.

Thus far, investigators have concentrated on inactivating the complex of genes associated with the expression of MHC Class I antigens involved in rejection. Knockout animals have been used to study both tissue and organ allografts, but only tissue xenografts. Knockout mice have also been created that do not make β2-microglobulin, a protein that forms an integral part of Class I antigens. Kidneys and liver cells from these knockout mice have prolonged survival when implanted into other mice; liver cells implanted into frogs also fared well, without immunosuppression.

Development of technology that could allow gene knockouts in pigs or other species suitable for organ donation to humans would represent a major step forward in xenotransplantation. Embryonic stem cell lines, needed for all current approaches to create knockouts (see below), have not yet been established for these higher species. When gene knockouts are developed for these other species, it would permit, for example, inactivation of the gene encoding the enzyme that is responsible for forming the antigenic epitope on surface glycoproteins, thereby potentially blocking hyperacute rejection. T-lymphocyte-mediated rejection could perhaps be avoided by disabling the set of genes encoding Class I or II MHC antigens. In short, advances in knockout technology could lead to the creation of xenogeneic organs that could bypass rejection by human recipients.

An even newer version of gene knockout technology is capable of inactivating the gene in selected cell types, such as T-lymphocytes or endothelial cells. In standard knockouts, the gene is inactived in all cells, which can be lethal. With the new technology, gene inactivation can be restricted to certain types of cells, allowing, for example, selective elimination of MHC Class I and/or Class II antigens on specific cells, such as endothelial cells.

Antisense RNA

Antisense RNA is a therapeutic strategy aimed at preventing expression of a given gene in the graft. It is designed to block the translation of functional gene products into protein antigens by the incorporation of antisense oligonucleotides—synthesized messenger RNA (mRNA) sequences that are complementary to, and thus hybridize with, specific mRNA sequences responsible for protein synthesis. Once hybridized to the proper native mRNA, antisense RNA interrupts the synthesis of selected proteins in the graft. However, antisense RNA is a technique that appears to be more complicated than originally thought and is far from realization for xenotransplantation.

Modification of the Host

Bone Marrow Chimerism

Bone marrow chimerism represents another strategy to circumvent the immune response in either allotransplantation or xenotransplantation (Ildstad, 1995). The strategy calls for bone marrow to be transplanted into a host along with a solid organ from the same donor or source animal, or from a second donor carrying the same antigens as the first. Bone marrow engraftment may confer permanent acceptance of organs and tissues from the same source or donor without the need for immunosuppression, a state called tolerance. The host is described as a chimera because its bone marrow contains cells from both source or donor and recipient. Bone marrow chimerism has been shown to induce tolerance in studies of allotransplants and xenotransplants in rodents, and clinical trials with allotransplants in humans are under way.

Bone marrow is the source of many types of immune cells and, hence, is a major part of the immune system. In bone marrow transplantation, source or donor bone marrow can essentially replace the host's immune system with that of the source or donor, and the immune cells produced by this reconstituted bone marrow do not reject tissue or organs from the donor, since those tissues or organs are seen as self. For this strategy to work, the host's immune system must first be destroyed with radiation and/or chemotherapy, an extremely high-risk procedure. In contrast to modification of the source animal, bone marrow transplantation represents modification of the host. The host undergoes more than just immunosuppression: the host's immune system is either entirely or partially destroyed and replaced with transplanted cells. The justification for giving radiation and/or immunosuppression is to make room for the graft, although recent research suggests this may not be needed. At least for allotransplants, if the donor marrow engrafts permanently, long-term immunosuppressive medication is very likely no longer necessary.

Bone marrow chimerism has been studied in both humans and animals for decades, but recent progress has laid the foundation for human clinical trials of bone marrow implantation in patients requiring allogeneic or xenogeneic whole organ transplants (described later). Experiments in animals in the 1950s revealed that allotransplants of skin would not be rejected if they were accompanied by bone marrow or spleen transplants from the same source animal. However, a skin transplant from a genetically different source animal was rejected. The problem was that recipient animals later died, mostly likely from graft-versus-host disease (GvHD).

GvHD, a serious complication of bone marrow transplantation, is caused by immune cells from the graft recognizing the host as foreign or nonself. It can occur either in bone marrow transplants alone or in combination with organ transplants using either allogeneic or xenogeneic tissue. Graft leukocytes initiate an immune response that rejects host tissues, most commonly skin, gastrointestinal tract, and liver. The host immune system does not reject the immune cells of the graft because host cells have been destroyed by pretreatment with radiation and/or cytotoxic drugs. Even today, GvHD is lethal in 15 percent of the patients who develop it after allotransplants of bone marrow, and approximately 50-70 percent of all bone marrow transplant patients show some symptoms of GvHD.

A major research goal of the past decade has been to understand GvHD. After determining that GvHD was caused by mature donor T-cells, investigators used donor marrow depleted of mature T-cells in several human clinical trials. However, the remaining bone marrow cells did not engraft in up to 70 percent of these patients, a lethal complication for those patients whose own bone marrow had been totally destroyed. Because of these results, GvHD was thought to be an unavoidable complication of bone marrow allotransplants, rendering the prospect for bone marrow xenotransplants even more remote.

Recent research in rodents, however, has yielded evidence of a new cell type in bone marrow that acts to facilitate engraftment in allotransplants and xenotransplants of bone marrow. The insight came from experiments showing that syngeneic bone marrow grafts (grafts from the same or a genetically identical individual) were successful, even if T-cells were depleted. Yet the same graft of purified bone marrow cells did not engraft in genetically different individuals of the same species (allografts). Suzanne Ildstad reported at the workshop that she and her colleagues (Kaufman et al., 1994) reasoned that the purification process had inadvertently removed a population of cells that helped or facilitated engraftment across histocompatibility barriers. They referred to these as facilitating cells. Facilitating cells comprise less than a half a percent of total bone marrow cells and must be matched genetically to the bone marrow donor. As few as 30,000 of these cells ensure engraftment of allogeneic and xenogeneic marrow that is devoid of mature T-cells. With facilitating cells, donor marrow reconstitutes an immune system in the host without causing GvHD. Facilitating cells administered without other marrow cells do not engraft. These cells share some surface markers with T-cells, which explains why they were removed when the marrow was purged of T-cells. However, they do not perform some of the conventional functions of T-cells, at least in part because they do not have T-cell receptors on their surface. How facilitating cells aid in engraftment is not yet known.

Facilitating cells have been shown to be effective in promoting bone marrow chimerism in a variety of animal species with allografts and xenografts. Researchers also have discovered that complete ablation of the host's immune system may not be necessary for bone marrow chimerism to succeed. Partial ablation carries the distinct advantage of a lesser risk.

To produce a functioning immune system, transplanted bone marrow cells must mature, and some of the mechanisms of this process have been discovered. It appears that some bone marrow cells from the source or donor, known as stem cells, migrate to the host's thymus, the normal site of development of certain immune cells. In the thymus, the source or donor stem cells destined to become T-cells undergo a process of maturation and selection. It is at this site that they are thought to become tolerant to the host.

Clinical trials are under way at several transplant centers to determine the effectiveness of allogeneic bone marrow transplants along with transplants of kidneys or livers. In at least one trial, bone marrow is depleted of mature T-cells, leaving behind stem cells, progenitor cells, and a putative facilitating cell population. In another trial, two unmodified bone marrow infusions are given separately to increase the dose of marrow (Ricordi, 1995). The degree of destruction of the hosts' immune systems varies between trials. In some, patients receive partial instead of complete ablation of their immune system; in others, no ablation is undertaken. If any of these clinical trials prove successful, allogeneic bone marrow transplants may have applications to immune and autoimmune disorders. Indeed, this evidence led to proposals for human trials with xenogeneic bone marrow, including the controversial clinical trial discussed at the workshop and performed later that year, which involved transplant of baboon bone marrow into an AIDS patient.

Over the last decade David Sachs, Suzanne Ildstad, and their colleagues have tested a mixed chimerism approach in which bone marrow from the donor is given together with host bone marrow (Ildstad, 1995). This procedure allows a lower dose of irradiation to be used and has the advantage of more fully reconstituting immune function in the recipient than does total replacement of the bone marrow with only allogeneic bone marrow. This procedure shows promise for permitting survival of xenografts with less loss of immune competence. The application of bone marrow chimerism to solid organ discordant xenografts would require additional steps to circumvent hyperacute rejection.

In summary, establishment of bone marrow chimerism represents a high-risk, high-reward strategy for whole organ allografts and, potentially, xenografts. The high risks are incurred when the patient's immune system is destroyed, or severely impaired, to make way for the donor marrow. The procedure also carries the risk of GvHD, but this risk may be reduced by depletion of mature T-cells from the donor. Bone marrow offers the tantalizing prospect of permanent acceptance of allogeneic whole organs without the need for long-term immunosuppression, and likely would contribute to the survival of a discordant xenograft.


Another strategy that could be employed to increase host acceptance of whole organs is microchimerism, which is caused by the migration of leukocytes out of the graft and dispersion throughout the host. The patient becomes chimeric, as defined by the coexistence of foreign and host cells, but only at a microlevel because of the small number of donor cells present in the host.

Microchimerism occurs naturally after transplantation. It also occurs naturally after pregnancy, as was discovered by obstetricians who found fetal cells in maternal serum, sometimes years after pregnancy. The existence of microchimerism after transplantation was discovered only recently by Thomas Starzl and his group at Pittsburgh with the advent of better detection technologies, such as use of the polymerase chain reaction (PCR). Hypotheses about its role in graft acceptance came from studies of long-term survivors of kidney transplants. Patients who survived transplants from the 1960s were examined in the early 1990s. Biopsies revealed donor cells throughout host tissues, although few in number. The presence of donor cells in a host years after transplantation meant that cells of donor origin were not destroyed by the host immune system. The identification of coexisting immune cell populations in healthy transplant recipients led Starzl and his colleagues to propose that microchimerism is a desirable, rather than an undesirable, consequence of transplantation that ought to be fostered. It also is seen as a predictor of graft success. These investigators hypothesized that the two-way communication between graft and host leads to a state of mutual tolerance, which can occur only under what they term the protective umbrella of immunosuppression. If the balance is tilted too far in either direction, GvHD or rejection of the graft will occur (Starzl et al., 1993). Once mutual tolerance occurs, immunosuppression should in theory no longer be necessary. To promote microchimerism, these and other investigators are infusing bone marrow cells at the time of whole organ transplant in ongoing clinical trials.

Modification of the Graft: Encapsulation

Encapsulants are semipermeable barriers designed to surround transplanted cells to protect them from the host's immune response. The foreign cells or tissues, either allogeneic or xenogeneic, are placed in a polymer sheath having pores that are selectively permeable to molecules of low molecular weight. The pores are impermeable to immune cells and large molecules, including antibodies and complement if desired. The pores are sufficiently large, however, to permit passage of such molecules as insulin and other hormones, growth factors, and other molecules of small molecular weight. The encapsulated cells survive and function because the pores allow uptake of oxygen and nutrients from blood and interstitial fluid, and they allow exit of waste products. Encapsulants are feasible only for cells and tissues, not for organs. Because of the size of organs, interior cells would die from lack of nutrients if transplanted in an encapsulated form.

Encapsulants are versatile enough to shield a variety of secretory cells, but much of the research by industry and academic investigators has focused on encapsulating islet cells, including those from the pig, for the treatment of human diabetes. Insulin secreted by pig islets is nearly as effective as human insulin and has enjoyed widespread clinical use for decades. Islet cells are superior to insulin for regulation of blood glucose levels because injections do not achieve the fine control over glucose levels needed physiologically. This lack of fine control leads to serious and debilitating complications such as atherosclerosis of larger vessels and painful neuropathies. Besides this therapeutic advantage, encapsulated cells offer other advantages: they are only minimally invasive; the biocompatible encapsulant sheath is durable and, if large enough, can be retrieved along with its contents in the event of rejection or dysfunction; and because the encapsulated cells are shielded from cellular rejection, the recipient should require little or no immunosuppression. Unencapsulated islet allografts encounter immune cellular rejection and therefore require immunosuppression of the patient.

What remains to be demonstrated is whether encapsulants can protect against the cytokine-mediated autoimmune damage that caused the disease in the first place, since cytokines can enter freely through the membrane.

Cytokines are believed to play a key role in diabetes because they are directly toxic to islet cells. Type I diabetes involves a poorly understood autoimmune process in which the immune system reacts to autoantigens from islet cells, and destroys the cells. Encapsulated cells likely release some (porcine) islet antigens, which are seen as foreign by the host and lead to T-cell activation with release of cytokines that are of small enough molecular weight to diffuse into the encapsulant.

Problems that must be considered even if encapsulants are used include the following. The pores in the encapsulants are large enough to permit the exit of toxins from infectious agents in the donor tissue (either of allogeneic or xenogeneic origin), if such agents are present in the tissue. Without access to the encapsulated tissue, the host's immune system cannot destroy the tissue harboring the infection. Prescreening potential donor tissue for infections would decrease this possibility, but as noted in Chapter 3, previously unexpected infections may still be present. Another possible disadvantage of encapsulation may be eventual immune reaction to the encapsulant itself. Years of research have been devoted to identifying the best mechanical design for encapsulating cells. Early animal studies placed islets in a relatively large plastic chamber surrounding a shunt that was connected to an artery. This approach failed because the islets died when coagulated blood blocked the pores or the host suffered life-threatening clotting complications such as stroke. Newer approaches with smaller devices have employed hollow fibers and microspheres. To prevent cell clumping inside the capsule, which deprives internal cells of nutrients and oxygen and thereby leads to cell death, the cells can be suspended in alginate or other types of matrix that immobilize them. The matrix does not block the passage of small molecules.

Success with glucose regulation in allograft animal models has led to human clinical trials, in which islets from human cadavers were microencapsulated in hollow fiber devices and implanted into nonimmunosuppressed diabetic patients (Scharp et al., 1994). Over the course of the two-week-long safety trial, no complications were recorded and 90 percent of the encapsulated cells survived. Because there was no detectable short-term immune response, immunosuppression was unnecessary. Insufficient numbers of islet cells were transplanted to draw any conclusions about the efficacy of the procedure, but further trials are planned. If allogeneic tissue proves successful, xenogeneic islet cells are the obvious next step since they are easier to harvest and in plentiful supply. Human pancreases, like other human organs, remain in short supply.

Copyright © 1996, National Academy of Sciences.
Bookshelf ID: NBK45531
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...