Brenner and Rector's The Kidney, 8th ed.

CHAPTER 63. Transplantation Immunobiology

Anil Chandraker   John J. Iacomini   Mohamed H. Sayegh



Characteristics of the Allogeneic Immune Response, 2103



Tolerance and Immunity: Self-Nonself Discrimination, 2104



Antigen Recognition, 2104



Immune Tolerance, 2104



Transplantation Antigens, 2104



The Major Histocompatibility Complex, 2105



HLA Molecules: Class I, 2106



HLA Molecules: Class II, 2107



Inheritance of HLA, 2108



HLA Typing, 2109



Relative Strengths of HLA Loci, 2110



Minor Histocompatibility Antigens, 2110



ABO Blood Group Antigens, 2111



Monocyte and Endothelial Cell Antigens, 2111



Effects of Blood Transfusions, 2111



The Immune Response to Allografts, 2111



T Cell-APC Interactions, 2111



T Cell Receptor Complex, 2113



CD4 and CD8 Coreceptors, 2113



Adhesion Molecules, 2113



Costimulatory Molecules, 2113



Cytokines/Chemokines, 2115



Effector Mechanisms of Allograft Rejection, 2116



Acute Cellular and Humoral Rejection, 2117



Chronic Rejection, 2117



Mechanisms of Immunosuppression, 2118



Corticosteroids, 2118



Azathioprine, 2118



Mycophenolate Mofetil, 2119



Cyclosporine, 2119



Tacrolimus (FK-506), 2120



Rapamycin (Sirolimus), 2120



Polyclonal Immune Globulins, 2120



Monoclonal Antibodies, 2121



Experimental Immunosuppressive Drugs, 2121



Genomics, 2121



Xenotransplantation, 2122


The term allogeneic refers to the genetic relationship between two individuals of the same species, in contrast with xenogeneic for different species, and syngeneic for human monozygous twins, or completely inbred animals of the same genetic background or strain. The immune system evolved to protect us from invading microorganisms. It appears also to play a protective role in surveillance for altered normal cells, such as ones that have undergone malignant transformation. The components of the immune system that are recruited when there is a threat to body integrity are multiple, and they vary according to the nature of the threat (e.g., viral versus bacterial) and the anatomic location of the insult (e.g., skin or gastrointestinal tract). Many important components of immunity that are ancient in evolutionary time are described as components of natural immunity. These include phagocytic cells, natural killer (NK) cells, complement, and some cytokines and chemokines. Tissue injury from any cause can activate the natural immune system. In contrast, immune reactants that specifically recognize foreign molecules or antigens recognized from the microbial world are generated as a result of an adaptive immune response. The main components of such antigen-specific adaptive immunity are immunoglobulins, which are made by B cells, and (thymus-dependent) T cells. Some authors refer to an inflammatory response that involves only components of natural immunity as being nonimmune, although it is better referred to as an innate immune response. Indeed, the natural or innate immune response is very important to the success of the overall immune response. In transplantation immunity, recognition of donor antigens is crucial in developing antigen-specific clones of T and B cells, which then direct an even larger array of cellular and humoral responses, many of which do not use specific antigen receptors. Fundamentally, the “firestorm” of allograft rejection does not occur in the absence of cell-mediated immune responses initiated by specific T lymphocytes, but the full force of the rejection uses components of the natural/innate immune system. Furthermore, the innate immune response against tissue injury (such as ischemia) augments antigen-specific immune responses, thus contributing ultimately to graft destruction.

The first successful renal transplantation was performed in 1954 at The Peter Bent Brigham Hospital in Boston, Massachusetts between identical twins. In the 1960s, recognition of the immunosuppressive properties of azathioprine in combination with corticosteroids made it possible to successfully perform transplant surgeries between nonidentical donors. Improvement in surgical techniques and the development of newer immunosuppressive agents, including cyclosporine, tacrolimus, mycophenolate mofetil (MMF), as well as polyclonal and specific monoclonal antibodies, have reduced the incidence and intensity of acute rejection, but the problem of chronic graft dysfunction remains a persistent obstacle. Because transplantation was a very rare event until modern times, the immune system clearly did not evolve to mediate graft rejection. It was the investigation of the alloimmune response to transplanted tissues that led to the identification of the major histocompatibility complex (MHC) molecules, also called HLA antigens. This investigation, in turn, provided crucial insights into the process of immune recognition, leading to the discovery that MHC molecules play a central role in presenting foreign antigenic peptides molecules, to T cells in a way that they can be recognized by the antigen-specific T cell receptors (TCRs). Although there are many similarities between immune recognition of conventional antigen and the recognition of allogeneic transplantation antigens, a major difference exists between the response to allogeneic and conventional antigens that has major implications for transplantation biology. The most striking difference is the markedly increased frequency of responding T cells in the allogeneic response. In addition, because recipient T cells recognize intact allogeneic MHC molecules directly, they are stimulated maximally by the high density of MHC on the surfaces of transplanted cells. T cells do not contact allo- or foreign MHC molecules during development in the thymus and thus escape the deletion (negative selection) imposed by interaction with self-MHC. The molecular basis for the high frequency of T cells responding to an allogeneic stimulus remains incompletely understood; however, the high frequency of alloantigen-specific T cells contributes to the vigorous nature of the immune response that causes early acute graft rejection, and it is the major initial obstacle to successful organ transplantation.

Tolerance and Immunity: Self-Nonself Discrimination

The principal tenet of immune recognition is that the immune system must discriminate self from nonself. This process has been viewed as being dependent on two components. First, the immune system must not respond to any self-antigens, which it does principally through the mechanisms of self-tolerance. Second, nonself (foreign)-antigens derived from numerous sources, including pathogens and tumor cells, must be effectively recognized to prevent infections and tumors. In this paradigm, the immune system simply recognizes transplanted tissues as nonself, causing graft rejection. In reality, it now seems that autorecognition of self-antigens is not uncommon, but only rarely results in autoimmune disease. The key resides in an effective immunoregulatory system that is more complex than was previously imagined. The ultimate goal of clinical transplantation is to develop protocols to induce specific tolerance to the graft, so that the immune system is regulated to accept the graft as self without maintenance immunosuppression.

Antigen Recognition

Clearly, the immune system did not evolve to reject transplanted organs. It is commonly accepted that the major function of the immune response is to recognize foreign antigens derived from pathogens and tumors. However, we now know that the processes by which the immune system recognizes conventional and transplantation antigens have many similarities and involve the same antigen recognition molecules. Specifically, the antigen-specific receptors include the TCR on T lymphocytes and antibody molecules produced by B lymphocytes, that when expressed on the B-cell surface comprise the major component of the B-cell receptor. As we discuss in more detail later, the TCR recognizes peptide fragments of processed antigen only when they are bound to MHC molecules expressed on the surface of antigen-presenting cells (APCs), whereas immunoglobulins or the B-cell receptor can bind directly to peptide fragments or to the same peptide sequences in the intact native molecule. In the early days of study of MHC antigens, investigators wondered why hematopoietic cells, especially B cells and monocyte-macrophages, have high expression of MHC molecules. It is now quite clear that the true role of the MHC is to present potentially immunogenic peptides of foreign antigens to T lymphocytes. Whereas the T cell response to a conventional antigen can be experimentally detected only after previous immunization, an allogeneic response, as assayed in a mixed lymphocyte culture (MLC) in vitro, can be readily detected in previously unimmunized (naïve) lymphocytes. At least part of the basis for the greater magnitude of the allogeneic response is the increased frequency of responding cells. For example, the frequency of specific T cells to conventional antigens is approximately 1 in 104 to 105, whereas the frequency responding during allogeneic stimulation can be as high as 1 in 101 to 102. The graft, which includes donor bone marrow-derived APCs, usually expresses several class I and class II MHC molecules that differ from the recipient's MHC molecules, and which can directly stimulate recipient T cells (direct allorecognition). Alternatively, donor antigens can be processed and peptide fragments presented by the host MHC molecules on self-APCs, indirectly stimulating the recipient T cells (indirect allorecognition) by way of the same pathway used for responses to microbial antigens (see later in this chapter). Relevant to clinical transplantation, the greater intensity of an allogeneic response can produce vigorous episodes of acute allograft rejection that can be difficult to control and may require large doses of immunosuppressive agents (see Chapter 65 ).

Immune Tolerance

Immune tolerance is a state of unresponsiveness to specific antigens derived from either self- or nonself-proteins. From numerous studies it is clear that maintenance of immune tolerance to self-antigens involves multiple mechanisms. First, tolerance can occur at the level of either T or B lymphocytes. [1] [2] Second, tolerance can be induced in either immature lymphocytes during the early steps of differentiation or in mature lymphocytes after they have migrated to the peripheral lymphoid tissues, including lymph nodes and spleen. [2] [3] [4] Third, tolerance, or immune unresponsiveness, can be mediated by several mechanisms, including clonal deletion of antigen-specific lymphocytes, anergy (inactivation of lymphocytes that nevertheless remain viable), and suppression involving regulatory processes between different subsets of lymphocytes.[5] Evidence from both human studies and animal models indicates that the maintenance of self-tolerance involves many, if not all, of these mechanisms.

The major mechanism of self-tolerance is the elimination of potentially autoreactive T cells at an immature stage of development during maturation in the thymus by a process called negative selection.[6] It is estimated that more than 95% of thymocytes die because of a high-affinity encounter with self-MHC before they migrate to the peripheral lymphoid organs.[7] Thymocyte clones that are positively selected for differentiation into mature peripheral T lymphocytes have receptors that are less avid for self-MHC[8] and also possess the necessary repertoire to recognize a large number of peptide configurations when bound to self-MHC, and by cross-reaction (molecular mimicry), to the intact surfaces of allo-MHC. Although negative selection is a major mechanism for maintaining self-tolerance, the process of negative selection is not complete and potentially self-reactive T cells are able to escape and emigrate to the periphery. The immune system has therefore evolved mechanisms to maintain tolerance in the periphery. [9] [10] [11] [12] [13] [14] In clinical transplantation, the main focus has been to prevent T cells that have already been selected in the thymus from rejecting an allograft. Thus, efforts to achieve donor-specific allograft tolerance have focused largely on inducing tolerance in mature peripheral T cells (see later in this chapter and Chapter 66 ).


Normal persons maintain a state of self-tolerance to self-tissues. Allografts, however, express non-self antigens to which the recipient is not tolerant, and they cause an antigraft immune response that initiates rejection. Several types of transplantation antigens have been characterized, including the MHC molecules, minor histocompatibility antigens, ABO blood group antigens, and monocyte/endothelial cell antigens. The antigenic stimulus for initiation and progression of the rejection response to grafted tissue is provoked by cell surface molecules that are polymorphic (i.e., that vary in structure from individual to individual and that are treated as foreign intruders that the body must recognize and destroy). The most important transplantation antigens are those of the MHC. In both animals and humans, this series of linked genes provides the strongest incompatibilities for any sort of tissue and organ transplant. The MHC antigens were originally discovered during tumor transplants between different inbred strains of mice, but their evolutionary conservation is based on defense against the microbial world. The MHC antigens are the strongest transplantation antigens and can stimulate a primary immune response without priming. The minor histocompatibility antigens are processed peptides derived from cellular antigens that are presented by MHC molecules but are not derived from the MHC. T cells recognize a combination of the antigen and MHC molecule through a trimolecular interaction involving the TCR, MHC molecule, and processed antigen in the form of a short peptide. The minor histocompatibility antigens require priming and can be detected only in a secondary immune response.

The Major Histocompatibility Complex

Transplantation antigens are classified as either major or minor, according to their relative potencies in eliciting rejection. The major antigens in all mammalian species studied are encoded by a closely linked series of genes in a chromosomal region called the MHC. The MHC was first defined in the mouse by Gorer[15] and Snell[16] as the genetic locus responsible for rapid rejection of tumor transplants between inbred strains of mice. This antigen system was called H-2, and was found to function in rejection of normal tissues as well. Rejection elicits serum antibodies that are used for typing of H-2 antigens. It was subsequently shown that cytotoxic T cells also arise in response to H-2 differences and that the H-2 genes are all clustered in a single region on chromosome 17. [17] [18] [19] Except for some details of the ordering of genes, the human lymphocyte antigen (HLA) and rat (RT1) MHC regions are quite homologous to H-2 of the mouse. HLA is located on the short arm of chromosome 6 ( Fig. 63-1 ). [20] [21] The species' chromosome numbers are different only because they have not been numbered in a manner that reflects the locations of actual genes. Transplants compatible for the MHC antigens can still be rejected because of minor antigen (e.g., H-1, H-3, H-4) incompatibilities but not with the same intensity as with MHC-incompatible grafts. Modification of rejection by drugs or other means is more readily accomplished when the donor and recipient MHC antigens are matched. Extensive work in the mouse skin graft model with a large number of different major (H-2) and minor incompatibilities has shown, that, in general, the sum total of multiple non-H-2 (minor) incompatibilities, once the recipient has become immunized to such antigens, can be equal to the strength of the H-2 barrier alone in the unimmunized, or first-set, rejection response.[22] For both MHC or non-MHC barriers, placement of a second graft from the same donor is rejected at an accelerated rate (second-set rejection). The distinction of first- and second-set rejection phenomena in humans was first made by Holman[23] in 1924 with skin grafts in burn patients. During World War II, the problem of extensive burn injuries prompted fundamental studies of skin grafting by Medawar.[24]

FIGURE 63-1  Schematic map of the HLA region on the short arm of chromosome 6. Distances are shown in kilobases derived from sequencing of DNA nucleotides. The centromere is to the left (5′). The boxes along the central line represent coding regions for the HLA polypeptides expressed on cell surfaces. GLO is a polymorphic red blood cell enzyme more than 4000 kilobases from HLA. On the right are the HLA-A, -B, and -C loci for the three sets of class I heavy chains, and on the left are the loci for the three sets of class II molecules, HLA-DP, -DQ, and -DR. The latter are composed of two chains, a and b, each the product of linked genes of the DP, DQ, or DR subregions. The DRA gene, for example, encodes the α-chain, and several DRB genes encode the β-chains, of the heterodimeric HLA-DR molecules. In the case of DR, DRA is not polymorphic, so all the antigenicity lies with DRB. As shown in the expansion at the bottom, each subregion contains tandem sets of exons for one or more α- and β-chains. The gene for DR private specificities is B1; the more public DR51, DR52, and DR53 are encoded by the other B genes (see text). Pseudogenes, which are not expressed on the cell surface, are shown as white boxes. The HLA-DP and -DQ molecules are also heterodimers, and both the α- and β-chains are polymorphic. Genes involved in proteolysis of proteins and the intracellular transport of peptides to meet up with class I molecules are in the class II region near DQ (shown as P/T). Between the class I and II loci are genes for the complement components C2, BF (factor B), and C4. Genes for 21OH (steroid 21-hydroxylase) are not shown but are between C4 and BF. An expansion of the class III region would show that both C4 and 21OH are reduplicated, so the order is this: (C4A-21OHA-C4B-21OHB-BF-C2). The genes of this region are sometimes referred to as class III, although they bear little homology to class I and II, which do show homology for each other. The genes for TNF-α and -b, and heat shock protein 70 (hsp70), have been mapped between the complement region and HLA-B.  (From Carpenter CB: Histocompatibility systems in man. In Ginns LC, Cosimi AB, Morris PJ [eds]: Transplantation. Boston, Blackwell Science, 1999, p 61.)



Together with the emerging concept of a major transplantation antigen system, [15] [16] these studies laid the groundwork for the development of clinical transplantation in the second half of the 20th century. Although the initial discovery and definition of the MHC came from studies in transplantation, its central role in the initiation and expression of the immune response in general has become increasingly evident. The ability to produce an efficient immune response to many antigens is inherited in a mendelian autosomal dominant fashion, and the controlling genes, called “Ir” (for immune response) are of the MHC. In fact, the failure to mount a response to a peptide antigen may now be attributed to a genetically determined inability to bind properly the antigenic fragment to MHC molecules.[25] The complex of antigen and MHC provides an efficient mode of presentation to clones of T lymphocytes that bear the appropriate antigen receptors. Indeed, it is clear that the TCR recognizes the total configuration of self-MHC plus the antigen fragment (peptide), and not antigen alone. Many experimental systems have demonstrated this MHC restriction phenomenon; that is, T cells, in order to respond, must share MHC antigens with the APCs.[26] The combined recognition of self-MHC plus antigen (self plus X hypothesis) is now being visualized at the molecular level (see later in this chapter).

The HLA region on the short arm of chromosome 6 (see Fig. 63-1 ) contains more than 3 million nucleotide base pairs. It encodes two structurally distinct classes of cell surface molecules, termed class I and II ( Fig. 63-2 ). The term MHC antigen has been traditionally applied to the product of a given locus that displays polymorphism in a population of individuals. Now that the sequence and structure of molecules bearing MHC antigens have been extensively elucidated, it is known that the polymorphic, or antigenic, portions of MHC molecules are, indeed, quite small, often involving only one to four amino acid substitutions in regions of sequence hypervariability. In an MHC molecule, the specific substituted area that causes a change in antigenicity is called an epitope. Normal pregnancy induces antibodies against the HLA antigens of the fetus derived from the father's genes. Although blood from normal human pregnancies is the main source of antisera used for HLA typing, the first appreciation of the HLA system came from the studies of Dausset[20] on blood transfusion reactions due to antileukocyte antibodies. One such antibody from patient MAC proved to be a pregnancy-induced anti-HLA-2 response. Subsequent studies by Payne and coworkers[27] showed that such antibodies marked a codominantly expressed antileukocyte system that segregated in a mendelian distribution in families. International workshops on the HLA system began in 1962, and today they continue to accelerate progress in the definition and technical aspects of typing for the polymorphic antigens of this chromosome region.[28] The World Health Organization Nomenclature Committee has defined HLA as the “logo” for the human MHC. Although the individual letters of the logo may have different colloquial meanings, such as human or histocompatibility, leukocyte, lymphocyte, or locus, antigen, for example, HLA is to be used as the prefix to a locus or subregion designation that marks all that follows as a product of the human MHC chromosomal region (e.g., HLA-A2, HLA-DR4).[29] The term human leukocyte antigen is to be avoided, because it has a broader meaning that includes cluster differentiation (CD) antigens, which are not histocompatibility antigens.

FIGURE 63-2  Class I and class II HLA molecules on a cell membrane as seen from the side of the molecules. The 44-kd α-chain is inserted through the lipid bilayer of the membrane and has three domains (a1, α2, α3) formed in part by disulfide bonding. β2-Microglobulin, encoded by a gene on chromosome 15, is noncovalently bound and is not membrane-inserted. The α2 and α1 domains from the β-strands and α-helices, which form the base and sides, respectively, of the groove that binds a potentially immunogenic peptide. The variable amino acids that confer the antigenic differences between individual HLA types are arrayed along this groove. Class II molecules consist of one α-chain (34 kD) and one β-chain (28 kD), each having two domains, and each is membrane inserted. A similar peptide-binding groove is formed by the α1 and b1 domains. (From Carpenter CB: Histocompatibility systems. In Ginns LC, Cosimi AB, Morris PJ [eds]: Transplantation. Boston, Blackwell Science, 1999, p 62.)



Class I and II molecules show some structural homology to immunoglobulins, the T cell antigen receptor, and to molecules bearing the T cell differentiation antigens CD4 and CD8 (see later in this chapter). The latter are part of the system whereby T cells interact preferentially with class II or class I molecules, respectively, on APCs or on cells that are targets for immune destruction. This family of cell surface and extracellular recognition and interaction structures may have evolved from the same progenitor gene, diversifying by duplication and mutation, and in the process the new genes have moved to multiple chromosome sites. All mammalian species studied thus far have structural and functional representations of this immunoglobulin supergene family.[30]

HLA Molecules: Class I

Class I HLA molecules consist of two polypeptide chains in noncovalent association on cell surfaces. The heavy chain (44 kD) is inserted into the plasma membrane and contains the antigenic portions. The light chain (12 kD) is β2-microglobulin, encoded by a gene on chromosome 15.[21] There are three domains of the class I heavy chain, formed in part by disulfide bonding to make loops (see Fig. 63-2 ). The amino acid sequence variable regions are on the first (a1) and second (a1) domains. Class I molecules are expressed on almost all nucleated body cells, including the endothelium of blood vessels. Tissue typing is performed on peripheral blood, lymph node, or spleen lymphocytes, all of which strongly express HLA class I. Platelets are not commonly used for typing, but they are useful for absorbing anticlass I antibodies from serum, because they do not express class II antigens. Some organ-specific anatomic variations occur in endothelial expression, and in states of active inflammation the density of class I can be locally increased. There are three class I heavy chain loci, HLA-A, -B, and -C (see Fig. 63-1 ). Each locus product in a given individual bears a unique, so-called “private” antigenic epitope plus additional “public” epitopes that are shared more widely among the population. There are more than 80 serologically defined A and B locus private antigens and 10 C locus ones, but more than 800 class I alleles have been defined by DNA techniques.[31] The independence of the HLA-A, -B, and -C molecules on the cell surface can be demonstrated by observing with fluorescent markers the separate aggregation and capping of each set of molecules when antibodies specific to each locus are added. When antibodies to β2-microglobulin are added, all class I molecules are capped.

The first MHC molecule to be crystallized was HLA-A2, and the structure of this class I allele was determined by x-ray diffraction studies to a resolution of 3.5Å ( Fig. 63-3 ).[32] This accomplishment afforded visualization of how the amino acid sequence relates to the folding of the chains into a three-dimensional structure. The two membrane distal domains,α1 andα2, form a groove along the top surface of the molecule facing away from the cell membrane. The margins of the groove are formed by α-helices, and the base is floored by a series of eight parallel β-strands, with theα1 andα2 domains contributing more or less equally to each side of the structure. In the crystallographic study, the groove, approximately 25Å long and 10Å wide, contained an unidentified molecule that in subsequent studies was shown to represent the bound peptide fragment, eight to nine amino acids long. These peptides have an extended linear core structure, and binding is to a large measure determined by side chain interactions. When the locations of amino acid variations, already known from study of sequences and interactions with antibodies or cytotoxic T cells, are related to the crystal structure, it is remarkable that the HLA variable sites lie along the α-helical and β-strand surfaces that form the margins of the groove (see Fig. 63-3 ).[33] In other words, the polymorphisms serve to define the shape of the binding groove on MHC molecules and thus determine which peptides are bound and recognized by T cells. The sites that determine whether or not a given peptide binds may also confer a conformational change on the fragment. The result is that the TCR binds to the unique topography of the MHC surface formed by a given MHC and peptide combination.[34] The TCR has two chains, α and β, which form a heterodimer ( Fig. 63-4 ). The membrane distal surface of the assembled TCR has six variable loops, called CDRa1, -2, and -3, and CDRb1, -2, and -3, which provide the specificity of binding to the MHC α-helices and bound peptide antigen.[35] Additional human and mouse class I molecules have been crystallized, and the hypothesis that allelic polymorphisms determine binding of different peptide sequences has been confirmed. Peptides found in eluates from class I crystals or purified molecules are usually eight or nine residues in length.[36] Their origin is in the intracellular pool of polypeptides derived from metabolic turnover of housekeeping proteins or intracellular infections such as viruses. There is selective proteolysis and transmembrane transport from lysosomal compartments into the Golgi, where 8 to 9 mer peptides are placed in class I-binding sites before transport to the cell surface. Some of the genes that control this process are in the class II region of the MHC (see Fig. 63-1 ). [37] [38]

FIGURE 63-3  The structure of the HLA-A2 molecule derived from x-ray crystallographic study. A, The flat ribbons represent the β-sheets or -strands, and the spiral areas at the top (membrane distal) are the α-helices, which form the sides of a groove approximately 10Å wide and 25Å long. The floor of the groove is formed by eight parallel β-sheets. The COOH-terminal end of the α3 domain is inserted in the membrane. B, This view looks down on the top of the molecule, so that the groove goes from left to right. Diagrammed is the core structure of the amino acid sequence, consecutively numbered from 1 (bottom left) to 180 (center left). The symbols show the variable amino acid substitutions that have been identified indifferent human or mouse haplotypes as relating to sites of reactivity to alloreactive T cells (gray squares), or to monoclonal antibodies (black circles). Sites that relate to both are shown as dual symbols. Whereas the antibody sites are on the external surface of the helices, many of the T cell sites are at the base of the groove. The variable sites determine which peptide sequences can be bound by a given MHC allele (see text).  (A from Bjorkman P, Saper M, Samroui B, et al: Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329:506–512, 1987; B modified from Bjorkman P, Saper M, Samroui B, et al: The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329:512–518, 1987.)



FIGURE 63-4  Illustration of the area of surface contact between the T cell receptor (TCR) and major histocompatibility peptide-binding groove of a class I molecule. The α-helices are shown from the same perspective as in Figure 63-3B . The footprint of the TCR, which consists of an α- and β-chain, is shown to contact the α-helices and the peptide in the groove (not shown). Each chain has three hypervariable regions (CDRs), and they are aligned in such a way that the CDR3 regions of the TCR, the sites of greatest sequence variability, make contact with the peptide.  (From Carpenter CB: Histocompatibility antigens and immune response genes. In Dale DC, Federman DD [eds]: Scientific American Medicine. New York, Scientific American, 1998.)



HLA Molecules: Class II

Class II HLA molecules consist of two membrane-inserted and noncovalently associated glycosylated polypeptides, called a (34 kD) and b (28 kD) (see Fig. 63-2 ). Each of these chains has two domains, and, again, the polymorphic regions are mostly on the outer, NH2-terminal, domains. The region of HLA encompassing class II genes is generally referred to as HLA-D. There are 21 HLA-DR alleles that have been defined serologically, but more than 320 alleles can be identified using DNA-typing techniques.[31] Although three class II molecules, HLA-DP, -DQ, and -DR, are generally recognized on cell surfaces, the situation is not entirely analogous to class I, because the α- and β- chains of each class II molecule are encoded by separate, closely linked genes (see Fig. 63-1 ). Although α- and β-chains of different parental haplotypes can associate, such hybrid associations are restricted to products of the same DP, DQ, or DR subregion. The naming of HLA-D-region genes is based on knowledge of the biochemistry of expressed antigens and on the growing database of DNA nucleotide sequencing. The gene that encodes the HLA-DR α-chain, for example, is called DRA. It has no sequence variation (i.e., it is monomorphic). All the variability of different alleles lies in the DRB chains: DRB1 encodes the private DR antigens 1 to 21, and DRB3, DRB4, and DRB5 encode β-chains for DR52, DR53, and DR51. The HLA-DQ subregion contains the genes DQA1, DQB1, DQA2, DQB2. The latter two are nonexpressed pseudogenes; the products of the first two, DQa and DQb, are both polymorphic. HLA-DP is similarly organized, and has polymorphisms on both chains. Study by the Southern blot technique to determine restriction fragment length polymorphisms (RFLPs) of DNA digested with various nucleases and hybridized to cDNA probes specific for the HLA genes has proved to be an alternative detection technique that has been particularly informative for class II genes. [39] [40] More rapid and precise detection of actual DNA sequences can now be accomplished by selective polymerase chain reaction amplification of polymorphic gene regions, followed by hybridization with short oligonucleotide probes specific for a given HLA sequence or by RFLP analysis of the amplified product. [41] [42] Comparative studies in ongoing workshops have demonstrated a strong correlation between serologically defined polymorphisms and those identified by T cell clones reactive to class II molecules.

Analysis of crystals of class II HLA-DR1 shows a remarkable similarity to class I molecules in the peptide-binding region (see Fig. 63-3 ). The α-helical and β-stranded core structures of class I and II are virtually superimposable. The main difference is at the ends of the groove, which in class II are more open allowing for binding of longer peptides. Typically, the length of eluted peptides from class II molecules are 13 to 26 residues, and there is protrusion of the linearly arrayed peptide at both ends of the groove. [43] [44] Class II antigens are limited in expression to B lymphocytes, monocytes/macrophages, dendritic cells, and activated T lymphocytes. Human endothelium generally does not express class II, but this situation can change rapidly when there is inflammation in the vicinity. In addition, epithelial cells of skin, intestine, and renal proximal tubule can synthesize and express class II molecules in response to injury and inflammation.[45]

Peptides bound to class II molecules are derived from proteolysis in acidic endosomal compartments and represent endocytosed proteins or microorganisms coming from outside the APC. Thus, the extracellular compartment, in contrast to the intracellular compartment for class I, is the responsibility of the typical class II-positive APC. When foreign polypeptide antigens are added to cultured APCs, peptide fragments appear on class II (but not on class I) molecules in a matter of minutes. Peptide fragments of self-MHC classes I and II are found in the eluates from class II molecules,[44] indicating that there is representation of intracellularly synthesized products on class II at the cell surface or that secreted molecules reenter the cell by the endocytic pathway.

Inheritance of HLA

Because chromosomes are paired, each person has two sets of HLA antigens, one set from each parent. The genetically linked antigens of the entire HLA region inherited from one parent collectively are called a haplotype, and, according to the rules of simple mendelian-dominant inheritance, any sibling pair has a 25% chance of inheriting the same two parental haplotypes, a 50% chance of sharing one haplotype, and a 25% chance of having two completely different haplotypes ( Fig. 63-5 ). The main evidence that HLA is the major transplantation barrier in humans originally came from the observation that, in renal transplantation, HLA-identical sibling donor grafts provided the best long-term graft survival and required the least aggressive immunosuppression. In the absence of a recombination (crossover), the entire HLA-A to HLA-DP region, including the specific electrophoretic polymorphisms of C4, BF, and C2, is expressed with each inherited haplotype. Recombination rates within the region are in the vicinity of less than 1%; thus, generally, it is not necessary to “type” for the expressed products of all the loci to identify haplotypes within a family. Rarely, when recombination has occurred or when several common antigens are present on both sides of a family, complete molecular typing of HLA may be necessary.

FIGURE 63-5  Inheritance of HLA haplotypes. The HLA-A, Cw, -B, and -DR genes are shown to represent the entire region. Children inherit one of each of the parental HLA haplotypes and they are inherited as a block unless a recombination has occurred during meiosis of an ovum or sperm, shown here as an arrow between B and DR in the mother. Such events occur less than 1% of the time for HLA. The chances of an HLA-identical sibling or an entirely nonmatching pair in a family is 1:4. Haploidentical siblings occur with odds of 1:2, and all children are haploidentical with each parent.


The distribution of HLA antigens in the general population is not random. Some are more common than others, and racial and ethnic patterns are well known. Furthermore, within a given racial or ethnic group, certain HLA haplotypes or portions thereof are likely to be found with higher frequency than one would predict by random distribution. For example, HLA-A1, -B8, and -DR3 very often occur on the same haplotype in northern Europeans. These alleles are not in equilibrium, and thus, are said to be in linkage disequilibrium. When entire, so-called extended haplotypes[46] are found in apparently unrelated persons, it is likely that they were inherited from a common and relatively recent ancestor. When alleles of adjacent loci of the HLA region (e.g., B and DR) are in linkage disequilibrium, the possibility exists that selective pressures were exerted over many generations to sustain the coexpression of a combination that favors defense against infectious diseases. Review of the associations of HLA alleles with a number of diseases is beyond the purpose of this chapter,[47] but the existence of linkage disequilibrium is relevant to considerations of HLA antigen distribution throughout the general population, a matter of direct concern when matching donors for transplantation.

HLA Typing

The main sources of antibodies for typing come from large-scale screenings of thousands of serum samples from multiparous women. Immunizations among humans yield the most highly specific antibodies to private HLA determinants. Anticlass I (HLA-A, -B, -C) antibodies react with both B and T lymphocytes, whereas anticlass II antibodies react with B, but not T, cells. Generally, a positive reaction is marked by cell lysis in the presence of rabbit complement. In addition, more broadly reactive antisera, originally thought to contain several antibodies in mixture, may have reactivity to the public determinants. For example, HLA-B molecules contain at least three immunogenic regions, one for private and two for public polymorphisms.[48] Monoclonal antibodies derived from the immunization of mice or rats with human lymphocytes only occasionally bind to the same antigenic sites defined by human antibodies. Although there are several examples of monclonal antibodies that may substitute for human antisera, a large number of monoclonals react in a “public” fashion, but not necessarily in the same patterns that human antisera do.

The mixed lymphocyte response (MLR) occurs when lymphocytes of one individual are cultured with those of another. Proliferation occurs over 5 to 7 days and is measured by the incorporation rate of 3H-thymidine into newly replicated DNA.[49] Usually, one population of cells is irradiated to prevent proliferation; then, the readout represents the response of nonirradiated helper T cells to the class II antigens present on stimulator B cells or macrophages or dendritic cells. Genetic identity for HLA yields a negative, nonproliferating MLR; similarity is revealed by weak proliferation. Before the HLA-DP, -DQ, and -DR subregions were clearly defined, incompatibility for the MLR was attributed entirely to HLA-D. The MLR is rarely used now as a clinical test, because the complexity of HLA-D can be fully assessed by typing. HLA-DR determinants provide the strongest MLR stimulus, whereas DQ plays a lesser role. HLA-DP is recognized only by primed (i.e., previously immunized) cells. The MLR itself is a complex series of cellular responses. Helper cell clones are first activated to proliferate but then induce the proliferative burst of CD8+ cytotoxic T cells. The latter are generally directed to class I incompatibilities and injure appropriate target cells after direct cell-to-cell contact is initiated by T cell-antigen receptors. When cytotoxic T cells are tested against a large number of individuals typed by classic serologic techniques, a good, but not perfect, correlation is observed overall. Some antigenic sites, or epitopes, recognized by T cells are actually different from those on the same class I molecule that are recognized by antibodies. This is explained by the fact that immunoglobulins can recognize small epitopes on intact whole molecules with tertiary structures whereas T cells “see” only the complex surface made up of a peptide fragment bound in the MHC-binding groove. Furthermore, there is more than one diversity site per HLA molecule that provides targets for rejection. The private specificities are, by definition, most immunogenic in the human antihuman alloresponse, but a given specificity usually involves a composite of small amino acid differences at more than one site in an HLA molecule.

The marked superiority of HLA haplotype-identical sibling donors for organ and bone marrow transplantation has been demonstrable since the early 1970s.[50] When azathioprine and prednisone were used as standard therapy, initial graft loss rates were in the 5% to 10% range during the first year, and graft survival half-lives in the 25- to 30-year range.[51] Now that HLA serologic typing is more reliable, a major question is whether similar results can be obtained if unrelated donors are matched for all, or most, of the HLA loci. The answer is evolving, because many other factors in clinical transplantation must be considered, but it is indisputable that both improved immunosuppression and reductions in histocompatibility barriers have contributed to better graft survival. In addition, management of infections and judicious attention to treatment of rejection activity contribute much to clinical success. Overall, patient and graft survival rates have been improving over the past decade, and had been improving even before use of cyclosporine became widespread.[52] One of the major problems in interpretation of any data relates to the fact that 1-year graft survival rates have been rising at the rate of 1.5% per year for more than 2 decades, and no single factor can explain this ( Fig. 63-6 ). The technology of HLA typing need not be absolutely accurate for selection of HLA-identical sibling donors, but on the other hand, true definition of a zero-mismatched cadaver donor for HLA-A, -B, and -DR may require molecular DNA typing. It is fair to say that widespread competency in both class I and class II serologic typing has been achieved only in the past decade, but it is still imperfect, and there is direct evidence that technical difficulties with HLA serologic methods can account for poor correlations with grafting results.[53] Pooled information from a large number of collaborating centers and that include tens of thousands of cases provides strong evidence of the role of HLA phenotypic matching in the success of cadaver renal transplantation. [54] [55]

FIGURE 63-6  Long-term graft survival as a function of donor source and HLA matching. Data are from the Scientific Registry of Transplant Recipients. Recipients of living kidney transplants have improved short and long term outcomes compared to recipients of deceased donor grafts indicating the influence of the ischemic insult that deceased donor kidneys are exposed to during harvesting and transportation. The survival for randomly matched living donor kidneys (Living all) is still superior to zero HLA mismatched deceased donor kidneys (Deceased 0 mm). (Survival curves drawn from Scientific Registry of Transplant Recipients data,


There is rather strong evidence that long-term graft survival is improved by avoiding HLA mismatches. Log-linear plots of graft survival always show a straight-line decline after the first year, which makes it possible to calculate a half-life and to project survival rates over time. Recent data show that recipients of first deceased donor kidneys having no HLA-A, -B, or -DR antigen mismatches have graft half-lives of 14.5 years and a projected 73% chance of 5-year survival.[54] With one or two mismatches, the value is a 12.5-year half-life and 70% 5-year survival for recipients 50 years of age or younger. These long-term results are to be compared with HLA-identical living related donors (32 years half-life), and all other living donors (17 years half-life). Avoiding ischemic and storage injury of organs from living donors certainly enhances outcomes as compared with those predicted from the cadaver data (see Fig. 63-6 ). There is considerable opportunity both for matching more cases and for avoiding completely mismatched grafts. It is worth noting that current projections give the same, or better, 10-year graft survival rates to cadaver donor organs that are not mismatched with recipients for HLA-A, -B, and -DR, as with HLA haplotype-identical family donors. It is for this reason that United Network of Organ Sharing (UNOS) has a mandatory share policy for 0 mismatched deceased donor kidneys. Because of the size of the national pool available, about 15% to 20% of cadaver kidneys are now shared on this basis.

Relative Strengths of HLA Loci

Clinical data do not support the simple assumption that each mismatch for antigens of various loci has equal weight in causing graft loss. The major impact comes from the effects of B and DR antigens; little additional effect comes from the A locus.[56] Transplant data from the United Kingdom show that DR matching has a much greater effect than matching of A or B.[57] As compared with the results when no mismatches are present for A, B, or DR, the addition of a single mismatch for A, or B, or DR increases the chances of graft loss twofold for A, threefold for B, and fivefold for DR. The UNOS system uses as the mandatory sharing criterion of a zero mismatch for HLA-A, -B, and -DR (i.e., “six-antigen match”).[58] There also appears to be a temporal effect HLA-A, -B, and -DR mismatching: HLA-DR matching being the most important in the first 6 months after transplantation, the HLA-B effect emerges during the first 2 years, and HLA-A matching does not show an effect before 3 years. [55] [57] [58] [59] [60] These findings are similar to those of Opelz and colleagues,[59] who found that during the first post-transplant year, the class II HLA-DR locus had a stronger impact than the class I HLA-A and HLA-B loci. In subsequent years, however, the influence on graft survival of the three loci was found to be equivalent and additive. This would indicate that, in the absence of prior sensitization, HLA-A mismatches have a deleterious effect only on long-term graft survival.

The more effective current immunosuppressive therapy produces 1-year cadaver graft survival rates as high as those associated with one-haplotype matched living, related donors. The short-term results are around 95% graft survival for first grafts at 1 year (see Fig. 63-6 ). This is in keeping with the steady increase in early graft survival and late graft survival with newer and more powerful immunosuppressive agents. Although 1-year graft survival rate has steadily improved to more than 90% over the past few decades with better immunosuppressant strategies, this has not been successfully translated to long term kidney allograft survival. [61] [62] Five-year overall adjusted graft survival of expanded criteria donor (ECD), non-ECD, living donor kidney recipients is only 54%, 69% and 80% respectively.[63] The effect of HLA matching on long-term allograft survival is less apparent. Patients experiencing an acute rejection episode have shorter long-term allograft survival. Acute rejections were of less importance when donor and recipient were HLA matched. It should be noted that the degree of mismatch was not in itself directly predictive of early rejection, but it did predict the prognosis once a rejection episode had occurred. Acute vascular rejection has poor longer term outcome than less severe tubulointerstitial rejection. [64] [65] Multiple acute rejection episodes are associated with a greater risk of chronic allograft nephropathy. [66] [67] A poor long-term outcome is significantly more common in patients who had more than one acute rejection episode compared with those with only one episode (34.8% versus 8.9% respectively). Likewise, a single late acute rejection episode (occurring >3 months after transplantation) carries a much higher risk of graft failure than early (<3 months) acute rejections (relative risk of graft failure 5.27 versus 3.07).[68]

Meier-Kriesche and colleagues[69] have shown that among patients with acute rejection episodes who recovered their baseline renal function to greater than 95% at 1-year post-transplantation had significantly higher 6-year graft survival compared to their counterpart who had less than 75% baseline recovery (72.7% versus 38% respectively).

Intrinsic responsiveness of individual patients may to a large extent determine the rejection rate, but in those patients who do reject, the HLA barrier becomes important to short-term survival rates. Alternatively, the rejection tempo may be intrinsically more powerful with more mismatched antigens. The question of defining intrinsic responsiveness remains open. Failure to reject may reflect the patient's susceptibi-lity to immunosuppressive agents, or it may be related to different donor-recipient histocompatibility combinations. One example would be a patient who does well with a second transplant after vigorously rejecting a first graft, with different HLA antigens.

The use of CREGS (for serologically Cross-REacting GroupS), has been suggested as an alternative to exact HLA matching to increase the potential pool of donors for any given recipient. CREGs classify HLA antigens more broadly into families of structurally related HLA antigens that share public antigenic sites.[70]

Minor Histocompatibility Antigens

Minor histocompatibility antigens are fundamentally different from MHC antigens. [22] [71] They are recognized as peptides bound to self-MHC. Because most minor incompatibilities elicit class I restricted CD8+ T cell responses, it seems likely that they are peptides derived from intracellular proteins of the same cell. [72] [73] [74] In mice, Y chromosome-encoded as well as mitochondrial gene-encoded minor antigens have been found. Many autosomal minor systems have also been identified. Minor antigenic peptides do not have sequence homologies with other sequences whose functions are known. It should be noted that these systems require that the recipient have the restricting MHC class I allele if a given antigen is to be operative as a transplantation antigen. This means that recipient T cells are capable of binding the minor donor peptide, and that the donor cells express a shared MHC antigen that presents the same peptide. Although the targets for cytotoxic CD8+ T cells are class I molecules that bear the peptide, it is clear that a CD4+-mediated helper response is needed during immunization. Another feature of minor antigens is that they do not provoke an antibody response. Why such CD4 response does not characteristically help B cells to make immunoglobulins is not known.

It has been appreciated for several years that minor histocompatibility antigens can be immunogenic in humans, from observations based on organ graft rejections and bone marrow graft-versus-host reactions in cases of genetically matched HLA antigens. A few cases of donor- directed, HLA class I-restricted, cytotoxic T cell responses have been demonstrated in such situations. [75] [76] [77]

ABO Blood Group Antigens

The ABO blood group antigens were initially identified as the cause of transfusion reactions during red blood cell transfusions. The A and B groups are glycosylated differentially, whereas group O lacks the enzymes necessary for glycosylation. The antigens are readily recognized by natural antibodies, termed hemagglutinins because they cause red cell agglutination. They are relevant to transplantation because they are also expressed on other cell types, including the endothelium. Thus, they cause hyperacute rejection of vascular allografts owing to preformed natural antibodies. Specifically, individuals with group A or B types produce natural antibodies to the opposite type, and those with group O produce antibodies to both A and B. However, because group O does not express the glycosylated moiety, both A and B fail to produce antibodies to group O. Thus, group O grafts can be transplanted into recipients with blood types A, B, and AB. This is not commonly done, for ethical reasons, so as to prevent depletion of available grafts for type O recipients, who must receive an O organ.

Allograft rejection due to red blood cell type mismatching can be readily prevented by routine blood typing before transplantation. The rhesus (Rh) factor and other red cell antigens are of little concern, because they are not expressed on endothelial cells. In recent years, a variety of desensitization protocols using plasmapheresis, pooled intravenous immunoglobulin, antilymphocytic antibody, monocloclonal anti CD20 Ab and sometimes splenectomy have enabled patients with preexisting anti-HLA (and sometimes anti-ABO) antibodies to receive transplants they would have previously rejected (see Chapter 65 ).[78]

Monocyte and Endothelial Cell Antigens

Occasionally, allografts undergo hyperacute rejection, despite appropriate ABO matching. Some rejection episodes have been attributed to additional non-ABO antigens expressed on endothelial cells and monocytes. [79] [80] The characterization of these antigens has been an area of active research; however, these antigen systems remain poorly understood. Pretransplant tissue typing does not currently evaluate the endothelial/monocyte antigens, owing to the apparent rarity of such antibodies and the lack of accurate reagents for typing.

Effects of Blood Transfusions

The historical data on the beneficial effects of blood transfusions are well known.[81] In recent years, the effect has been diminishing until now it appears to have little impact on graft survival. Indeed, only in the nontransfused group is there some evidence of a detriment to graft survival. With the introduction of erythropoetin the need for blood transfusions pretransplant has dramatically diminished and transplant survival has increased enough to make the beneficial effects of transfusion less important. More important clinically is the need to minimize blood exposure to avoid transmission of viruses, and this can be accomplished with proper erythropoietin therapy.


Acute allograft rejection in an unsensitized host is characterized by a lymphocytic infiltration into the allograft. The vast majority of these cells are T cells that are not alloantigen specific. T cells as part of their immunosurveillance function pass from the circulatory system, through the endothelium and into tissues before returning by means of the lymphatic system into the circulation. The inevitable tissue damage that occurs during transplant surgery results in a nonantigen specific inflammatory response, which greatly increases leukocyte recruitment (including T cells). This is mediated through interactions between surface molecules (known as adhesion molecules) and their receptors on endothelial cells and leukocytes, and also through the secretion and binding of small soluble proteins known as chemokines. The site of alloantigen recognition had until recently been believed to the allograft itself, recent data seems to indicate that antigen recognition may occur in lymphoid tissues. [82] [83] Regardless of the site, in the context of allograft rejection T cells can recognize alloantigen through the usual pathway of recognition of foreign antigen, confusingly known as the indirect pathway of allorecognition (see later in this chapter). Unique to the transplant setting, T cells may also recognize foreign MHC and antigen through a process of molecular mimicry, in which the foreign MHC and antigen complex resemble the intended receptor of the TCR and activate the T cell. After T cell interaction with APCs through the TCR, a number of other cellular interactions between the cells stabilize the interaction and lead to full T cell activation. The context of antigen presentation, most strongly defined by the type of cell that presents antigen is also increasingly recognized as important in whether the T cell mounts an aggressive or passive response to the antigen.


In the MLR, allo-MHC molecules induce very strong primary immune responses in vitro.[84] It has long been recognized that the normal T cell repertoire contains a large cohort (1% to 10%) of total T cells, which are capable of responding to allo-MHC molecules.[85] This translates to a precursor frequency at least 100 times that of antigen-specific self-restricted T cells, which recognize conventional antigens as peptide fragments bound to self-MHC. The vigor of the primary alloimmune response has puzzled transplant immunologists for almost 3 decades. [86] [87] [88] The original hypothesis on the high frequency of alloreactive T cells[89] proposed that the repertoire of lymphocytes was preselected by evolutionary pressure to have specificity for the MHC molecules of the species. It was also proposed that the lymphocyte repertoire that recognizes conventional antigens was derived by somatic mutation of the TCR specific for self-MHC.[8] Today we know that TCRs do not undergo somatic mutations, and there is evidence from studies with T cell clones that the antigen-specific and alloreactive T cell repertoires may be contained in the same clones. [90] [91] [92] [93]

The two fundamental questions in allorecognition are these: First, why is the frequency of alloreactive T cells so high? Second, how can positively selected (in the thymus) self-MHC-restricted T cells recognize foreign antigens as well as allo-MHC? For one thing, it is apparent that there are at least two distinct, but not necessarily mutually exclusive, pathways of allorecognition. [82] [94] [95] In the so-called direct pathway, T cells recognize intact allo-MHC molecules on the surface of donor or stimulator cells. In the indirect pathway, T cells recognize processed alloantigen as peptides in the context of self-APCs, which is the normal route of T cell recognition of foreign antigens. Both pathways have been shown in experimental animals to contribute independently to allograft rejection. [96] [97] The relative contributions of the direct and indirect pathways of allorecognition to allograft rejection in humans are under investigation, although recent data in cardiac, renal, and lung transplant recipients suggest that indirect allorecognition of donor HLA peptides may play a key role in chronic rejection ( Fig. 63-7 ). [98] [99] [100]

FIGURE 63-7  A, Mechanisms of allorecognition and T cell response. T cells recognize antigen through direct and indirect pathways. In the direct pathway, T cells recognize intact major histocompatibility molecules on donor APCs. In the indirect pathway, T cells recognize processed alloantigen in the form of peptides presented by recipient antigen-presenting cells. Recipient monocytes are recruited by endothelial cells to the graft tissue. They are also transformed to become highly efficient antigen-presenting dendritic cells that may need to recirculate to peripheral lymphoid organs for maturation. The dendritic cells and intragraft macrophages present donor peptides by way of the indirect pathway to recruited CD4+ T cells. CD8+ T cells, on the other hand, are activated by donor endothelial cells and can either directly kill endothelial cells or traverse the endothelium and kill parenchymal graft cells. B, The alloreactive T cell can undergo a number of different fates. They may provide help for macrophages, B cells, and monocytes by secreting cytokines and by cell-cell contact-dependent mechanisms or kill graft cells in an antigen-specific manner through the release of toxic agents and by FAS-mediated apoptosis. As a process of activation, some T cells undergo cell death or become unresponsive (anergic) to antigenic stimulation through either regulation soluble factors or direct cell contact. Yet others may become memory T cells and await antigenic restimulation to provide a recall response.  A (From Briscoe DM, Sayegh MH: Rendezvous before rejection: Where do T cells meet transplant antigens? Nature Medicine 8:220–222, 2002.); B (From Salama AD, Remuzzi G, Harmon WE, Sayegh MH: Challenges to achieving clinical transplantation tolerance. J Clin Invest 108:943–948, 2001.)



Direct recognition of intact MHC molecules, although focused on the polymorphic MHC epitopes, is strongly influenced by the presence of peptide in the MHC groove.[101] MHC molecules “empty” of peptide generally are not recognized unless the missing self-peptides are reconstituted. [102] [103] [104] It has also been shown that changing the bound peptide can alter the allorecognition of a given MHC molecule.[105] These observations have provided the rationale for studying the immunomodulatory functions of synthetic peptides, particularly MHC peptides in vitro and in vivo.[106]

The basic premise for indirect allorecognition as a mechanism for initiation or amplification of allograft rejection is that donor alloantigens are shed from the graft, taken up by recipient APCs, and presented to CD4+ T cells.[95]Indeed, it has been demonstrated that intact HLA molecules are present in the circulation of renal transplant recipients. Therefore, during transplantation,[107] shed fragments of allo-MHC could be processed by host APCs and presented as allopeptides to T cells on self-MHC. This indirect pathway of allorecognition may lead to activation of helper T cells, which secrete cytokines and provide the necessary signals for the growth and maturation of effector cytotoxic T lymphocytes, B cells, and monocytes/macrophages, leading to allograft rejection (see Fig. 63-7 ). [82] [94] Other important yet poorly understood mechanisms that may contribute to graft destruction include nonspecific tissue injury and repair and graft cell apoptosis. Ischemia plus reperfusion injury of the grafted organ leads to up-regulation of MHC class II and costimulatory molecules, which in turn increase the immunogenicity of the graft and amplify the immune response to it. The clinical syndromes of rejection are described in detail in Chapter 65 .

T Cell Receptor Complex

T cell recognition of alloantigens on APCs is the primary and central event that initiates allograft rejection.[95] The interaction between T lymphocytes and APCs involves multiple T cell surface molecules and their counter-receptors expressed by APCs.

Antigen specificity is determined by the TCR, which recognizes processed antigen in the form of short peptides bound to an MHC molecule. The specificity of antigen recognition is exquisitely precise; the alteration of a single amino acid in the peptide antigen or MHC molecule can alter recognition by the TCR. Thus, T cell recognition of antigen involves a trimolecular interaction involving the TCR on the surface of the T cell, the MHC molecule on the surface of the APC, and the antigenic peptide bound to the MHC molecule. Although the TCR is responsible for antigen recognition, the invariant proteins CD3 and ζ, (which are noncovalently bound to the TCR) are responsible for signal transduction through the activation of protein tyrosine kinases associated with the cytoplasmic tails of CD3 and other TCR associated receptors. This interaction provides signal 1 of T cell activation (see later in this chapter). Binding of the TCR to a peptide/MHC complex is of itself usually insufficient to cause T cell activation. Provision of signal 1 alone leads instead to a state of T cell unresponsiveness or “anergy.”[108]

The OKT3 monoclonal antibody binds to the CD3 complex. OKT3 also induces signal transduction, including activation of specific kinases, activation of phospholipase C, and the increase in intracellular calcium and phosphoinositol metabolites.[109] The immunosuppressive effects of OKT3 are also dependent on signal transduction. The major immunosuppressive effect of OKT3 is the down-modulation of cell surface expression of the TCR.[110] In addition, OKT3 causes the clearance of T cells from the circulation and their sequestration in the peripheral lymphoid organs, such as lymph nodes and spleen.[110] OKT3 is not cytotoxic and does not lyse peripheral T cells. Also, there is no convincing evidence that OKT3 directly interferes with antigen recognition by the TCR molecule. The side effects of OKT3 are also due to partial T cell activation causing the release of cytokines, principally tumor necrosis factor a (TNF-α), from large numbers of T cells.[111] For efficient T cell activation, a number of other interactions between the APC and the T cell need to occur.[109]

CD4 and CD8 Coreceptors

The two major subsets of T cells, cytotoxic CD8+ T cells and helper CD4+ T cells, recognize processed antigen on MHC class I and II molecules, respectively. Although not directly involved in antigen recognition, the CD4 and CD8 coreceptors bind to nonpolymorphic regions of the MHC molecules. Thus, the specificity of class I versus class II recognition is determined by whether a T cell expresses CD4 or CD8 in conjunction with the specificity of the TCR. In addition, CD4 and CD8 promote and stabilize the immunologic synapse between the T cell and the APC. The cytoplasmic tails of CD4 and CD8 molecules are associated with the tyrosine kinase p56lck, which plays an important role in T cell activation.[112] Binding of the CD4 or CD8 co-receptor approximates the cytoplasmic tail to immunoreceptor tyrosine-based activation motifs on the CD3 complex. This, in turn, leads to the phosphorylation of a series of intracellular proteins, resulting in the activation of a variety of enzymes including calineurin (the target of the immunosuppressive drugs cyclosporine and tacrolimus), and the activation of transcription factors, such as nuclear factor of activated T cells (NFAT) and NF-kB. These transcription factors are responsible for the increased and, in some cases, decreased expression of a variety of gene products associated with T cell activation. This signaling process is also dependent on the binding of T cell costimulatory receptors with their specific ligands present on antigen presenting cells.

Adhesion Molecules

Cells of the immune system infiltrate the graft from nearby lymphoid organs and the bloodstream by a three-step process. First, they roll along the vessel wall by way of interactions between selectins on the endothelium and receptors on immune cells. Second, they adhere to vessel endothelium. Third, chemoattractant cytokines (chemokines) are released. Adhesion molecules and chemokines are important regulators of rejection and appear to be targets for immunotherapy. Adhesion molecules on T cells include lymphocyte function-associated antigen (LFA)-1, which interacts with intercellular adhesion molecules (ICAM)-1 and -2; CD2, which interacts with CD58 (LFA-3); and VLA-4 (a4, b1 integrin, CDw49d, CD29), which interacts with vascular cell adhesion molecules (VCAM-1, CD106). These receptors are of two large structural families. The integrins, including LFA-1 and VLA-4, are made up of a, b heterodimers, whereas members of the immunoglobulin superfamily, including CD2, LFA-3, VCAM-1, and the ICAMs, are made up of disulfide-linked receptor domains. Some of these receptors have also been shown to transduce signals and, thus, are more appropriately called accessory molecules. The inhibition of adhesion/accessory cell function has been shown to be immunosuppressive.[113] Previous studies in a primate model showed increased graft survival with anti-ICAM-1 monoclonal antibodies, although clinical trials have failed to show this in human renal transplant recipients.[114] Blockade of LFA-1, on the other hand, is being studied in the context of preventing delayed graft function after transplantation.[115]

Costimulatory Molecules

A classic understanding of T cell function is based on Bretscher and Cohn's[116] two-signal hypothesis of T cell activation. One signal is transduced by the antigen-specific TCR when it recognizes processed antigen bound to an MHC molecule on the surface of an APC (see previous material). The second signal is mediated by costimulatory molecules and is independent of antigen. The TCR and costimulatory signal transduction pathways are distinct and use different second messengers. At the level of the regulation of transcription, as shown for the interleukin-2 (IL-2) gene, the two pathways interact by poorly understood mechanisms to control gene expression. By controlling the level of expression of IL-2 and other genes, the TCR and costimulatory pathways can regulate T cell activation and function. The best-characterized costimulatory molecule is CD28, which is constitutively expressed on the surface of essentially all CD4+ and approximately 50% of CD8+ peripheral T lymphocytes.[117] CD28 binds a family of counter-receptors termed B7 (B7-1 and B7-2 or CD80 and CD86, respectively), which are expressed by APCs. Costimulatory pathways may also down-regulate T cell responses. Following activation, CTLA4, which also binds B7 but with greater affinity than CD28, is expressed by the T cell. CTLA4 interaction with B7 transduces a negative signal to the T cell, resulting in physiologic termination of the immune response ( Fig. 63-8 ). [95] [118] [119] [120] CTLA4 knockout mice die within a few weeks of birth from a disease process characterized by massive multiorgan lymphocytic infiltration, emphasizing the importance of this pathway in turning off a lymphoproliferative response.[121] Interestingly, in humans CTLA4 polymorphisms are associated with diabetes and other autoimmune diseases.[122]

FIGURE 63-8  CD28-B7 and TNF-TNF-R T-cell costimulatory superfamilies. Following ligation of the receptor-ligand pair there is a net stimulatory (+) or inhibitory (-) signal transduced. In some cases, there is bidirectional signaling and for certain pairs both the ligand and receptor are expressed on the same cell constitutively or more commonly following activation. The expression of ligands is also not limited to professional antigen-presenting cells and may also be observed on endothelial and parenchymal cells.  (From Clarkson MR, Sayegh MH: T cell costimulatory pathways in allograft rejection and tolerance. Transplantation 80:555–563, 2005.)



Signaling by means of the TCR plus the CD28 costimulatory molecule is sufficient to activate T cells. In contrast, signaling by means of the TCR alone without costimulation induces long-term T cell unresponsiveness (i.e., anergy) or ignorance. [108] [123] Experimental analysis of the minimal signal transduction event necessary to induce anergy showed that an increase in intracellular calcium concentration was sufficient.[124] Because increased intracellular calcium is produced by signaling by means of the TCR, these results are consistent with the observation that anergy can be induced by monoclonal antibodies to the TCR or by APCs expressing the appropriate MHC molecule but lacking the costimulatory counter-receptor. [108] [123] In vivo, anergy is defined as failure of T cell clonal expansion after immunization with antigen. [108] [123] Anergic T cells remain viable but are unresponsive for at least several weeks in experimental murine models in vitro and in vivo. [108] [123] [125] In vitro, some anergic states can be reversed by cytokines such as IL-2.[124] The fate and function of anergic T cells in vivo remain undetermined; however, evidence from experimental models suggests that anergic T cells can be reactivated by some processes (e.g., viral infections).[126] These observations suggest that anergy is a reversible state, and that therapeutic use of anergy during clinical transplantation, although potentially very useful, requires thorough evaluation and careful monitoring. Certain anergic states may not be reversible and may be associated with T cell death by apoptosis.[127] Strategies targeted at inducing such states in vivo are clinically desirable. Inhibition of costimulation with soluble receptors to (e.g., CTLA4-Ig) or monoclonal antibodies against the B7 molecules has been shown to dramatically prolong graft survival and induce tolerance in some animal models. [95] [119] However, the fact that blockade of the CD28-B7 pathway alone does not appear to be effective at inducing tolerance in the more stringent models of transplantation, including primate models, [128] [129] and CD28 deficient mice[130] are capable of rejecting allografts albeit in a delayed fashion, has led to speculation that other pathways are capable of providing T cell costimulation.

Additional members of the CD28 family with positive or negative costimulatory function have now been recognized. ICOS is a homolog of CD28 and binds to B7h.[120] It does not interact with B7-1 and B7-2 and in contrast to CD28 it is not constitutively expressed, but is up-regulated on activated T cells.[131] Blocking this pathway leads to increased allograft survival in experimental models.[132] PD-1, which binds to its ligands, PD-L1 and PD-L2, is the latest member of the CD28 family to be recognized.[119] Like CTLA4, it appears to be a negative regulator of T cell activation, however, knockout animals do not produce the same dramatic phenotype as the CTLA4 knockouts but do result in spontaneous autoimmune disease.[133] The negative T cell costimulatory pathways appear to regulate alloimmune response, and the challenge is to develop new therapies that can target these pathways and harness the physiologic mechanisms of regulation of immune responses in vivo. [134] [135]

A separate family of costimulatory molecules that belong to the TNF:TNF-R family of molecules also play a role in T and B cell activation. [119] [136] CD40 and its ligand CD40L (CD154) were the first members of this family to have demonstrable costimulatory function. CD40 is expressed on B cells and other APCs, including dendritic cells and endothelial cells and belongs to the TNF superfamily of molecules. Its ligand, CD40L (CD154), is expressed early on activated T cells. CD40 is critical in providing cognate T cell help for B cell immunoglobulin production and class switching. A defect in CD154 expression is responsible for the hyper-IgM syndrome, in which there is failure to switch from IgM to IgG, seen in humans.[95] Blockade with anti-CD154 antibodies has been shown to be effective at prolonging allograft survival in rodent and primate models of transplantation. There is evidence that CD154 acts directly to transduce a costimulatory signal to the T cell, or indirectly, by ligation of CD40 on APCs and induction of B7 expression, thus enhancing CD28-B7 costimulation. Other members of group include the CD134-CD134L, CD30-CD30L, CD27-CD70, and the 4-1BB-4-1BBL pathways (see Fig. 63-8 ).[119] Their function in transplantation is under investigation. In particular, CD134-CD134L appears to be an important pathway in costimulation of effector/memory CD4+ T cells, [137] [138] whereas the CD27-CD70 appears to be important for costimulation of CD8+ effector/memory T cells.[139]


In addition to cell-cell interactions, cell function can be directed through proteins produced by a variety of cell types. These factors can function as growth, activation, and differentiation factors (cytokines) or as chemotracttants (chemokines) of inflammatory cells to the site of immune responses. They can act locally or systemically through signaling cell surface, receptors resulting in changes in gene expression of the cell. Cytokines are produced by cells that participate in the immune response, including T cells, B cells, and APCs. Nonimmune cells, such as endothelial cells, also produce cytokines that can modulate an immune response. The complex regulatory networks of cytokines is incompletely understood but appears to modulate the antigraft immune response. In addition, it has been shown that different subsets of T cells can produce different patterns of cytokine secretion.[140] The two subsets of CD4+ T cells defined in terms of cytokine production are the Th1 and Th2 subsets. Both subsets produce some cytokines such as IL-3, TNF-b, and GM-CSF, whereas each subset produces a predominant set of cytokines. The Th1 subset preferentially produces IL-2, gamma interferon (IFN-γ), and TNF-α. In contrast, the Th2 subset preferentially produces IL-4, IL-5, IL-10, and IL-13. The net effect of producing each cytokine profile is differential regulation of the immune response. The Th1 subset is considered proinflammatory, promoting delayed-type hypersensitivity (DTH) reactions and cytotoxic T lymphocyte expansion. The Th2 subset is considered a helper of B cells; however, the functional segregation between the Th1 and Th2 subsets remains incompletely understood. For example, IFN-γ secreted by Th1 cells regulates B cells by promoting antibody isotype switching to the IgG2a isotype. Also, secretion of IL-5 and IL-6 by Th2 cells increases eosinophil recruitment and expansion. In addition, some T cell clones analyzed in vitro express overlapping profiles of cytokines.

Chemokines are chemoattractant cytokines[141] that are structurally related by amino acid homologies, in particular the placement of cysteines. The nomenclature of chemokines is becoming increasingly complex; four chemokine families are now recognized, of which the majority of members belong to the C-C chemokine family, represented by RANTES, or the C-X-C chemokine family, typified by IL-8. In general, C-C chemokines attract monocytes and T lymphocytes, and C-X-C chemokines attract granulocytes.[141] Receptors on the surface of immune cells are named after the family of chemokines they interact with and can bind with a variety of chemokines in that family. For example CCR1 and CCR5 can respectively bind with the chemokines MCP-3 and MIP-1β and both can bind RANTES and MIP-1α. They act to create a chemoattractant gradient across tissues to move cells into sites of inflammation. Detection of altered chemokine mRNA in experimental models of rejection suggests that they play an important role in this process, however, because of redundancy and differences in between the function of chemokines in rodents and humans for the most part the exact role individual chemokines play in an alloimune response remains unclear,[141] although recent data may suggest that expression of CCR5 is associated with acute allograft rejection. [142] [143] [144]

In experimental models of transplantation, altering the T cell response from Th1 to Th2 has been shown to prevent acute graft rejection.[145] Except for blockade of the IL-2R, [146] [147] therapeutic strategies that modulate cytokines have not proved highly effective. This may be due to the pleiotropic effect of each cytokine and to the complex interactions exhibited by regulatory networks of multiple cytokines. It has also has been suggested that a Th1 profile is associated with rejection, whereas a Th2 phenotype promotes tolerance. However, this paradigm has been challenged by several investigators, although it may hold true in cases of lower mismatched allografts.[148] For example, IL-2 has been shown by several criteria to promote graft rejection, and blockade of the IL-2R has been shown to promote graft survival. [146] [147] However, “knockout mice,” which do not express any IL-2 owing to inactivation of the IL-2 gene by homologous recombination, reject grafts as easily as do normal mice.[149] Rejection in animals that lack IL-2 is likely mediated by other chemokines that compensate for the IL-2 deficit by having overlapping functions. Furthermore, recent studies with specific Th1 or Th2 cytokine gene knockout animals indicate the complex-ity of the Th1-Th2 paradigm in graft rejection and tolerance. [150] [151] [152] Interesting data from animal and human studies has shown that Th2 clones propagated from patients with stable renal transplant function, or animals tolerant to kidney transplants, can regulate a proliferative response from Th1 clones isolated from patients or animals undergoing active rejection. [153] [154] Therefore, although manipulation of cytokine functions may hold promise as a therapeutic modality, we will have to better understand the role of cytokines in graft rejection and tolerance under physiologic conditions if we are to develop effective treatments.

In summary, cell-cell interactions between T cells and APCs can be mediated by five classes of receptors: the antigen-specific TCR, the CD4 or CD8 coreceptor, costimulatory molecules, accessory or adhesion molecules, and cytokine receptors. Therapeutic or experimental manipulation of members of each class of receptors has been shown to prolong graft survival. The most effective in clinical studies to date have been the anti-TCR monoclonal antibody OKT3 and antibodies targeted at the IL-2R complex,[113] and more recently CTLA4-Ig targeting B7 costimulation.[155] Interaction of alloreactive T cells and alloantigen does not uniformly lead to an aggressive T cell response. Activated T cells may undergo a number of different fates after recognition of alloantigen (see Fig. 63-7B ). Some take on the role of effector cells and mediate an alloimmune B cell response, delayed type hypersensitivity or direct cell killing through a cytolytic T cell response. Others become memory cells awaiting antigenic restimulation at a latter time point to provide a rapid recall response. Alternatively, the T cell response may be terminated through anergy, activation induced cell death or active regulation by other cells or soluble factors.[156] The challenge in the transplant setting is to drive the immune response away from activation and development of effector/memory cells towards regulation and tolerance (see Chapter 66 ).


There are both cellular (DTH responses, cell-mediated cytotoxicity) and humoral components to transplant rejection. Once fully activated, CD4+ helper T cells produce cytokines that orchestrate various effector arms of the alloimmune response (see Fig. 63-7 ). Th1 cytokines activate macrophages, and both CD4+ Th1 cells and activated macrophages effect DTH responses. Although initiated by a specific immune response, DTH results in nonspecific tissue injury and repair. Although the exact mechanisms by which DTH leads to graft destruction remain unclear, it is hypothesized that some of the cytokines produced by T cells and macrophages (TNF-α) mediate apoptosis of graft cells. It has been suggested that DTH responses, presumably initiated by indirect allorecognition, are particularly important in the pathogenesis of chronic rejection.

Several effector mechanisms that participate in graft destruction have been identified. Adaptive immune responses mediated by B and T cells play a significant role in graft destruction as evidenced by the fact that inhibiting these cell types can often prevent graft destruction. T cells mediate destruction of allografts by direct lysis of donor tissues by CD8+ cytotoxic T lymphocytes, and by the production of proinflamatory cytokines by CD4+ T cells such as occurs in delayed-type hypersensitivity reactions. Alloantigen-specific antibodies produced by B cell are able to mediate endothelial cell activation leading to graft loss, as well as mediate lysis of allogenic cells within the graft by activating the complement cascade. Innate immune responses also play a role, although it is less clear whether they are sufficient to mediate allograft rejection in the absence of an adaptive response. CD8+ precytolytic T lymphocytes recognize specific HLA class I antigens on the surface of donor cells, and in the presence of helper T cytokines (IL-2, IL-4, and IL-5), differentiate and divide. Mature cytolytic T lymphocytes (CTL) damage target cells displaying foreign HLA class I by at least two mechanisms ( Fig. 63-9 ).[25] A secretory pathway involves granule-mediated exocytosis of soluble factors, including granzymes (serine esterases) and a complement-like molecule, perforin. These proteins induce cell death by means of both DNA degradation and osmotic lysis secondary to pore formation in the target cell membrane. Detection of both the perforin and granzyme mRNA from cells isolated from the graft, peripheral blood lymphocytes of urinary lymphocytes have been shown to be able to predict whether acute rejection is ongoing within the allograft. [157] [158] A second cytolytic pathway involves interaction between Fas (CD95), a TNF-like protein, and Fas ligand. Fas ligand is induced on CTL through triggering of the TCR. Fas-expressing target cells undergo apoptosis when Fas-ligand is engaged. The role in the rejection process of CD8+ T lymphocyte-mediated killing of graft cells remains controversial. It is possible to demonstrate ex vivo specific cytotoxicity against donor cells in rejecting animals and humans. In addition, passive transfer of specific “killer” T cell clones to naive animals induces allograft rejection, particularly skin allografts. However, recent data using CD8 gene knockout mice indicate that these animals are capable of rejecting skin and vascularized allografts as easily as normal controls, whereas CD4 gene-knockout mice were incapable of rejecting an allograft, indicating that CD4+ helper T cells, but not CD8+ T cells, are essential to allograft rejection.[159] It is now thought that CD8+ T cells, although they may contribute to acute allograft rejection through killer mechanisms, are not essential for allograft rejection to proceed. In addition, CD8+ killer lymphocytes may not play an important role in the process of chronic rejection.

FIGURE 63-9  Mechanisms of CD8+ T cell-mediated killing. Cytotoxic T cells (top) activated through the class I major histocompatibility peptide complex can cause target cell killing by two pathways. One involves granule exocytosis, releasing perforin, granzymes, and granulysin, which, in turn, cause membrane damage and cell lysis by osmotic dysregulation and activation of caspase cascade, giving rise to programmed cell death (apoptosis). The second involves expression of Fas ligand on activated T cells. When Fas on target cells interacts with its ligand on activated T cells, the target cell undergoes apoptosis by activation of caspase cascade.  (From Germain R: MHC-dependent antigen processing and peptide presentation: Providing ligands for T lymphocyte activation. Cell 76:287–299, 1994.)



B lymphocytes express clonally restricted antigen-specific cell surface receptors, immunoglobulins. When cell surface immunoglobulin binds specific antigen in the context of soluble helper factors such as IL-4, IL-6, and IL-8, B cells are activated. They differentiate, divide, and become plasma cells, which secrete soluble forms of antigen-specific antibodies that are displayed on their cell surfaces. These antibodies, in turn, can bind allogeneic target antigens and induce graft damage by binding complement or directing antibody-dependent cellular cytotoxicity. Both IgM and IgG alloantibodies can be detected in the serum and in the allografts (of animals and humans) that are being rejected. Preformed anti-HLA class I antibodies, and occasionally, antiendothelial antibodies, play an important role in hyperacute rejection and accelerated vascular rejection observed in previously sensitized transplant recipients. In xenotransplantation, naturally occurring xenoreactive antibodies play a critical role in hyperacute rejection of grafts.[160] Finally, alloantibodies, particularly IgG, play important pathogenic roles in the development of chronic rejection and graft arteriosclerosis. CD4 T cells are also able to mediate allograft rejection through a mechanisms that is most similar to deplayed-type hypersensitivity (DTH). CD4 T cells activated following recognition of alloantigen produce a variety of cytokines that are able to recruit other cell types to the graft site including CD8+ lymphocytes, B cells and macrophages. Resulting inflammation leads to graft destruction by production of cytokines such as TNF-alpha. CD4 T cells are also able to secrete cytokines that help other effect subsets differentiate into effector cells.

Other effector mechanisms include cell death through NK cells. NK cells express cell surface receptors called “killer inhibitory receptors (KIR)” that recognize HLA class I molecules.[161] Although the role of NK cell-mediated cytotoxicity in allograft rejection remains controversial, NK cells appear to play a key role in mediating delayed xenograft rejection.[162]


Acute rejection is the clinical syndrome that occurs as the result of an alloimmune response against a transplanted organ and can be due to either a cellular or humoral response. An acute cellular rejection normally occurs 5 to 7 days after transplant surgery in an unsensitized recipient but can occur in an accelerated fashion if this is the result of a secondary immune response and previously primed T cells are present. Clinically acute cellular rejection is characterized by a mononuclear cellular interstitial infiltrate, edema, and tubulitis. A humoral response, by definition, requires CD4+ T cell help and preformed anti-HLA antibodies that have long been recognized as a cause of accelerated rejection. The ability to detect evidence of antibody mediated injury through peritubular C4d staining within the allograft along with the characteristic histological changes of neutrophilic infiltration of the peritubular capillaries and the detection of circulating anti-HLA antibodies, have lead to increasing awareness of acute antibody mediated rejection. The clinical aspects of acute rejection are covered in more detail in Chapter 65 .


The slow progressive deterioration in renal dysfunction is characterized clinically by a rise in serum creatinine, increasing proteinuria, and progressive hypertension; and histologically by tubular atrophy, interstitial fibrosis, and fibrous neointimal thickening of arterial walls (an almost universal finding in renal transplant recipients). The terminology most commonly used to describe these changes is chronic allograft nephropathy. The term chronic rejection is more usually used to denote an immunologic cause of injury. The Banff classification system for chronic rejection grades the degree of allograft injury according to the severity of interstitial fibrosis and tubular injury, from grade I-III, correlating with mild to severe injury. In addition to these pathologic findings, allografts may demonstrate peritubular C4d staining, suggesting a role for antibody mediated injury, whereas a subset of transplanted kidneys have changes of transplant glomerulopathy characterized by swollen glomeruli, infiltration of the glomeruli with mononuclear cells, mesangial matrix expansion, mesangiolysis, and splitting of the glomerular basement membrane (GBM) with a subendothelial deposition of electron lucent material. Chronic allograft nephropathy (CAN) is an almost universal finding in transplanted kidneys and it can be detected in the majority of allografts within the first couple of years of transplantation. Serial protocol biopsy studies have shown that CAN is a progressive disorder, with two thirds of the fibrosis present by 10 years had already appeared within the first year.[163]

A number of different antigen-independent or so called nonimmunologic factors have been associated with progression of CAN ( Fig. 63-10 ). [164] [165] These include a number of factors before transplantation such as ischemia-reperfusion injury suffered by the kidney at the time of transplant, brain death in the donor, and nonspecific injury due to donor age, hypertension and diabetes, and so on. Post-transplant factors have also been shown to accelerate the course of chronic rejection including reduced renal mass, calcineurin inhibitor toxicity, hypertension in the recipient, as well as hyperlipidemia and cytomegalovirus (CMV) infection. However, in nearly every transplant, there is a degree of tissue incompatibility, whether this is a mismatch of the minor histocompatibility antigens alone or in combination with the MHC. Although the immune response may be the prime driver (at least initially) of chronic rejection, the actual immune mechanisms responsible for this are still not clear. The same effector mechanisms that are responsible for acute rejection are thought to be active chronically, although the relative importance of each may be different. Indirect as opposed to direct allorecognition is thought to play an important role in the process. The evidence for this comes from a number of studies that have shown that donor MHC derived peptides are capable of eliciting an immune response long after donor derived APCs (the main target of the direct alloimmune response) have disappeared from the allograft.[166] Alloantibodies have always been thought to be important in the development of chronic rejection, and recent studies have supported this hypothesis. [167] [168]

FIGURE 63-10  Mechanisms of chronic allograft dysfunction. Many different forms of injury can contribute to the development of chronic allograft dysfunction. Immunologically mediated injury, in the form of acute cellular and humoral rejection, is associated with poorer graft function, as is ongoing subclinical rejection. This ongoing alloimmune response can take the form of indirect alloreactivity in which recipient T cells recognize allopeptides present on recipient antigen-presenting cells or a chronic antibody response to the allograft. Antigen independent factors such as ischemic injury at the time of transplantation are thought to be important in enhancing the immune response by upregulating cell surface molecules involved in antigen presentation.  (From Sayegh MH, Carpenter CB: Transplantation 50 years later—Progress, challenges and promises. N Engl J Med 351:2761–2766, 2004.)



The overall pathways involved in development of fibrosis and tissue remodeling in transplanted kidneys have not been elucidated as yet. However, a number of recent studies have shed some light on individual components that influence this process. TGF-β has been shown to play an important role in the development of fibrosis in native kidney disease. However, it is well known that TGF-β has both immunosuppressive and profibrotic properties,[169] and although the tolerance and immunosuppressive aspects of TGF-β production are desirable in transplantation, the remodeling and fibrosing are damaging. Recent data have shown that anti-TGF-β antibody in high doses can abrogate long-term allograft survival induced by cyclosporine administration in a cardiac rat transplant model indicating the importance of TGF-β in mediating the immunosuppressive properties of cyclosporine.[170] In the same model, both high and low doses of anti-TGF-β antibodies prevent cyclosporine-related fibrotic renal injury. TGF-β has also been shown to be crucial in the regulation of transplantation and other forms of tolerance. [171] [172] In rodent models of transplantation, tolerance induction does not prevent the development of chronic rejection, and an animal that has undergone transplant surgery can exhibit tolerance to a second donor specific allograft and have evidence of chronic allograft dysfunction within the original allograft simultaneously. [173] [174] Regulation of TGF-β associated fibrosis has been achieved by a variety of inhibitors that inhibit TGF-β expression or signaling, including decorin, pifenidone, relaxin, and bone morphogenic protein (BMP)-7, or by agents that interfere with TGF-b-associated profibrotic pathways including angiotensin II, endothelin-1, and CTGF.[175] Of particular interest is epithelial to mesenchymal transformation (EMT). This describes the process of phenotypic change leading to fibrosis that affects cells of a variety of origins, including mesenchymal cells, resident fibroblasts, and epithelial cells. Both BMP-7 and hepatic growth factor (HGF) have been used to reduce TGF-b-associated EMT in experimental models of renal fibrosis.[176]

Clinically, prevention of chronic allograft nephropathy is focused on reduction or elimination of calcineurin inhibitor-based immunosuppression protocols. Numerous approaches have been studied in clinical trials, including use of IL2-R Abs with mycophenolate and steroids, substitution of calcineurin inhibitors with the mTORs everolimus and sirolimus, as well as most recently the use of belatacept, a modified CTLA4-Ig fusion protein that blocks T cell CD28 costimulation. However, because multiple factors have been shown to contribute to the development of CAN, it is unlikely that the modification of one cause of CAN, such as calcineurin inhibitor associated toxicity will result in a complete solution to this problem. Recently there has been growing evidence that the humoral response plays an important role in the development of CAN. Studies have shown that development of anti HLA antibodies after transplantation is associated with a substantial decrease in transplant survival. [177] [178] A multicenter study sponsored by the National Institutes of Health is currently examining the use of monoclonal anti-CD20 antibody to treat renal transplant recipients who develop de novo anti-HLA antibodies post-transplant (rituximab in kidney transplantation;



Most immunosuppressive drug regimens use an adrenocorticosteroid such as prednisone in combination with other immunosuppressive agents. Corticosteroids provide the most proximal block in the T cell-activation cascade: They indirectly interfere with T cell proliferation, at least partially through their ability to block expression of the IL-1 and IL-6 genes. [113] [179] Macrophages treated with corticosteroids do not produce IL-1 or IL-6 mRNA, even after incubation with powerful macrophage stimulants. Because IL-2 release depends in part on IL-1–stimulated IL-6 release, corticosteroids also indirectly block IL-2. Corticosteroids modulate the immune response by regulating gene expression. The steroid molecule enters the cytosol, where it binds the steroid receptor, inducing a conformational change in the receptor. The complex then migrates to the nucleus and binds regulatory regions of DNA called “glucocorticoid response elements,” which regulate the transcription of many genes. Thus, corticosteroids' many effects on multiple cell types account for both their efficacy and the diverse array of complications.

Conventional treatments for acute renal allograft rejec-tion include high-dose pulses of glucocorticoids, drugs that have broad, nonspecific immunosuppressive and anti-inflammatory effects. In addition to their effects, oncytokines, glucocorticoids reduce the migration of monocytes to sites of inflammation. A major drawback to the use of glucocorticoids in the treatment of acute rejection is that they inhibit the entire immune and inflammatory systems and alter many other steroid-responsive systems. Large doses of glucocorticoids can thus produce undesirable side effects, including decreased inflammatory and phagocytic capacity, resulting in increased susceptibility to infection, hyperglycemia, hyperkalemia, osteoporosis, increased capillary fragility, and growth suppression in children.


Azathioprine is a purine analogue that is enzymatically converted to 6-mercaptopurine and other derivatives in vivo, molecules that function as antimetabolites.[180] After metabolic conversion, it has multiple activities, including incorporation into DNA, inhibition of purine nucleotide synthesis, and alteration of RNA synthesis. The major immunosuppressive effect is thought to be due to blocking of DNA replication, which prevents lymphocyte proliferation after antigenic stimulation. Although it is useful for inhibiting primary immune responses, azathioprine has little effect on secondary responses or the reversal of acute allograft rejections, which are not dependent on lymphocyte proliferation. Azathioprine also decreases the number of migratory mononuclear and granulocytic cells while inhibiting proliferation of promyelocytes in bone marrow. As a result, the number of circulating monocytes capable of differentiating into macrophages is decreased. Among the possible deleterious effects of azathioprine administration are severe leukopenia, and occasionally, thrombocytopenia, gastrointestinal disturbances, hepatoxicity, and increased risk of neoplasia.

Mycophenolate Mofetil

In many centers, MMF is now replacing azathioprine in standard immunosuppression protocols for new kidney and pancreas-kidney transplants.[181] The rationale for this switch is that MMF is a selective inhibitor of the pathway of de novo purine synthesis. In the normal mammalian cell, guanine and adenine nucleotides are manufactured to form smaller precursors through two mechanisms, a de novo pathway or a salvage pathway (where purine bases are recycled). MMF is a reversible inhibitor of inosine monophosphate dehydrogenase (IMPDH), which is the rate-limiting enzyme in the de novo synthesis of guanosine nucleotide and nucleosides.[113] Inhibition of this enzyme results in the selective inhibition of T and B lymphocyte proliferation. MMF has minimal effects on other cell populations because proliferation of T and B cells is dependent on the de novo pathway for purine synthesis, whereas other cells are capable of using a salvage pathway. The main active metabolite of MMF is glucuronidated mycophenolic acid (MPAG), which is excreted into bile and undergoes enterohepatic recirculation. Cyclosporine but not FK506 inhibits MPAG excretion into bile and thereby decreases circulating levels of mycophenolic acid,[182] which would suggest that lower doses of MMF are required when used in combination with FK506. Three large multicenter trials have compared MMF to azathioprine or placebo in renal transplant patients receiving cyclosporin A and corticosteroids.[45] These studies show that MMF therapy is associated with a 50% reduction in the incidence of acute rejection in the first year after transplantation and less need for high-dose steroids and antilymphocyte antibody therapy to treat rejection.[183] In experimental animal models, MMF prevents development of chronic rejection,[184] and most but not all studies have shown that there is some reduction in the rate of progression of chronic rejection if mycophenolate is substituted for azathioprine in renal transplant recipients Side effects of MMF include gastrointestinal upset and diarrhea (which may be dose limiting) and some increased risk of tissue-invasive CMV infection. MMF may also cause bone marrow suppression but is associated with less bone marrow suppression than azathioprine. In an attempt to reduce the gastrointestinal side effects of MMF, enteric-coated mycophenolate sodium is also now marketed. This drug is metabolized to MPA, and therefore, its efficacy is the same as MMF. Its ability to reduce gastrointestinal side effects is currently the subject of ongoing clinical studies.


Cyclosporine, a small cyclic peptide of fungal origin, has played a major role in preventing graft rejection and has improved graft survival rates.[185] Although it is highly effective in blocking the initiation of an immune response, cyclosporine, like azathioprine, is of limited value in treating acute allograft rejection. [113] [186] Its primary action is to block the expression of cytokine genes produced by T cells, including IL-2, IL-3, IL-4, IFN-γ, and TNF-α, but it does not interfere with IL-1, TNF-α, or TGF-β produced by APCs, including macrophages. There is likewise no evidence that NK cells are affected by cyclosporine. In the presence of cyclosporine, T cell proliferation is indirectly inhibited owing to the absence of cytokines; however, the addition of exogenous IL-2 has been shown to restore T cell proliferation.[187]

It is now understood that cyclosporine blocks the calcium-dependent component of the TCR signal transduction pathway ( Fig. 63-11 ). Previous work showed that cyclosporine binds to a family of cytoplasmic molecules termed cyclophilins, which have peptidyl-prolyl isomerase activity. After binding cyclophilin, cyclosporine inhibits the isomerase activity; however, it is now apparent that the immunosuppressive effects are not a result of this inhibition. This was shown by analogs of cyclosporine that bind cyclophilin and inhibit isomerase activity but do not cause immunosuppression.[188] More recent work has shown that the cyclophilin-cyclosporine complex binds to and inhibits calcineurin, which is a cytoplasmic serine threonine phosphatase.[189] After T cell activation in the absence of cyclosporine, calcineurin dephosphorylates the cytosolic component of nuclear factor of activated T cells termed NFAT. After dephosphorylation, NFAT is translocated from the cytosol to the nucleus, where it forms a complex with other DNA-binding proteins, including fos and jun.[190] The complex of DNA-binding proteins regulate gene transcription, including the IL-2 gene. During treatment with cyclosporine, the inhibition of the formation of the nuclear NFAT complex has been shown to prevent transcription of the IL-2 gene,[191] and a similar complex has been shown to regulate TNF-α gene transcription. The net result is a reduction of IL-2 and resultant T cell activation. Clinically, a reduction in IL-2 concentrations minimizes the immune response associated with allograft rejection. Patients maintained on therapeutic levels of cyclosporine experience approximately a 50% reduction in calcineurin activity, allowing the patient to retain a degree of immune responsiveness sufficient enough to maintain host defenses. It is likely that identical or similar DNA-binding complexes regulate the transcription of multiple cytokine genes. The mechanism of action of cyclosporine in nonlymphocytes remains poorly understood; however, it has been shown that analogs of cyclosporine that lack toxicity also lack immunosuppressive effects.[192] These results suggest that the toxic and immunosuppressive effects are mediated by similar signal transduction mechanisms. The differential susceptibility to cyclosporine of lymphocytes, as compared with other cell types, may be due to different levels of expression of calcineurin and the cyclophilins.

FIGURE 63-11  Mechanisms of action of immunosuppressive agents viewed as a function of inhibiting T cell activation. Signal 1: The calcium-dependent signal induced by T cell receptor stimulation results in calcineurin activation, a process inhibited by cyclosporine and tacrolimus. Calcineurin dephosphorylates nuclear factor of activated T cells (NFAT), enabling it to enter the nucleus and bind to the IL-2 promoter. Corticosteroids bind to cytoplasmic receptors, enter the nucleus, and inhibit cytokine gene transcription in both T cells and the APCs. Corticosteroids also inhibit NFkB activation (not shown). Signal 2: Costimulatory signals are necessary to optimize T cell IL-2 gene transcription, prevent T cell anergy, and inhibit T cell apoptosis. Experimental agents, but not current immunosuppressive agents, interrupt these intracellular signals. Signal 3: IL-2 receptor stimulation induces the cell to enter the cell cycle and proliferate. Signal 3 may be blocked by IL-2 receptor antibodies or by sirolimus, which inhibits S6 kinase activation. Following progression into the cell cycle, azathioprine and mycophenolate mofetil interrupt DNA replication by inhibiting purine synthesis.


Tacrolimus (FK-506)

Tacrolimus, another calcineurin inhibitor, is a potent immunosuppressive agent that inhibits T cell activation in vitro. [189] [190] [191] [192] [193] Tacrolimus, in contrast to cyclosporine, is a macrolide antibiotic produced by fungi, yet the immunosuppressive effects on T cells are similar. [189] [190] [191] [192] [193] Because of structural differences, tacrolimus binds a family of cytosolic proteins termed FK-506-binding proteins (FKBP), which are different from the cyclosporine-binding cyclophilins. Interestingly, both FKBP and cyclophilin have peptidyl-prolyl isomerase activity and collectively are called immunophilins; however, the similar immunosuppressive properties of the two drugs are attributable to the fact that both agents inhibit the phosphatase activity of calcineurin.[194] Thus, although tacrolimus and cyclosporine have different structures, their immunosuppressive effects are mediated through a common final pathway. Thus, it is not surprising that both drugs block the induction of cytokine mRNA, including IL-2; inhibit cytokine production; and indirectly inhibit T cell proliferation. [189] [190] [191] [192] [193] Owing to the similar mechanisms of action, the side effect profiles of tacrolimus and cyclosporine are similar with certain exceptions such as hirsutism, and gum hyperplasia caused by cyclosporine and alopecia and neurotoxicity caused by tacrolimus. In experimental studies, the combination of tacrolimus and cyclosporine show additive increases in toxicity and in efficacy. Therefore, in clinical transplantation, immunosuppressive protocols involving multiple drugs use either cyclosporine or tacrolimus.

Rapamycin (Sirolimus)

Rapamycin was first used as an immunosuppressive through a search for drugs with a similar structure to FK506. Like FK506, it is a macrolide antibiotic and binds to the same family of FKBP isomerase proteins. However, whereas cyclosporine and tacrolimus inhibit calcineurin in the calcium-dependent component of the TCR signal transduction pathway, rapamycin binds to mTOR (mammalian target of rapamycin) and prevents phosphorylation of p70 S6 kinase in the CD28 costimulatory and IL-2R signal-transduction pathways (see Fig. 63-11 ). In functional studies, activation of T cells by monoclonal antibodies to the TCR was inhibited by either cyclosporine or tacrolimus but not rapamycin.[195] Conversely, the activation of T cells by exogenous IL-2 plus protein kinase C stimulation with a phorbol ester was inhibited by rapamycin but not by cyclosporine or tacrolimus. Similarly, T cell activation by monoclonal antibodies to the CD28 costimulatory molecule plus protein kinase costimulation with a phorbol ester was also inhibited by rapamycin but not by cyclosporine or tacrolimus. Cell cycle analysis shows that rapamycin blocks T cell proliferation during late G1 of the cell cycle and before S phase. [196] [197] [198] Thus, rapamycin inhibits late signals in T cell activation that are transduced by either the IL-2R or CD28 costimulatory signal-transduction pathways. [196] [197] [198] In contrast, cyclosporine and FK-506 inhibit an early signal in T cell activation, which is transduced by the TCR signal transduction pathway. Despite interacting with same binding proteins as FK506, there is no competitive inhibition of these drugs, because the binding proteins are present in great excess compared with the FK506 and rapamycin. In experimental models, the combination of rapamycin with a calcineurin inhibitor results in a synergistic effect.[113] This synergistic action has now been used with good effect in clinical studies.[199] Everolimus (RAD) is a rapamycin derivative and has a similar mode of action to sirolimus. The mTor inhibitors have recently been shown to be potent inhibitors of VEGF, which may explain their role in preventing progression of many forms of cancer.[200] Because the mTOR inhibitors have powerful immunosuppressive properties, they are also being used as calcineurin inhibitor-sparing agents in transplant recipients (see Chapter 65 ).

Polyclonal Immune Globulins

Polyclonal antilymphocyte globulin (ALG) or antithymocyte globulin (ATG) preparations have been available for approximately 2 decades and have proven more effective than steroids alone for reversing acute renal allograft rejection.[201] Polyclonal immune globulins are produced by injecting animals such as horses or rabbits with human lymphoid cells to obtain the purified γ-globulin fractions of the resulting immune sera. Cultured lymphoblasts (to produce ALG) and human thymocytes (to produce ATG) typically have been used. Polyclonal immune globulin may exert its immunosuppressive effect by several mechanisms, including classic complement-mediated lysis of lymphocytes, clearance of lymphocytes through reticuloendothelial uptake, masking of T cell antigens, and the possible expansion of regulatory cells. [113] [202] Administration of immune globulins produces prompt and profound lymphopenia. It soon abates, however, and the number of circulating T cells gradually increases, even while treatment continues, but the proliferative response continues to be impaired. It has been suggested that regulatory cells may be responsible for the prolonged immunosuppressive effect that persists after the resolution of lymphopenia. Thus, the resolution of cell-mediated graft rejection results from elimination of circulating T cells, and the subsequent inhibition of proliferative responses sustains the immunosuppressive effect.

Each polyclonal immune globulin preparation has different constituent antibodies. Because of the unpredictable nature of the antibody mixtures, treatment is associated with variable efficacy and risks of adverse reactions. Batch standardization and assessments of immunosuppressive potency are, therefore, difficult. Unwanted antibodies could cause thrombocytopenia, granulocytopenia, serum sickness, or glomerulonephritis. Although polyclonal immune globulins are potent immunosuppressive agents, the major concern is the potential for excessive immunosuppression, which frequently results in opportunistic infections. Therefore, caution is the rule when combining immune globulins with other immunosuppressive agents. There has been an increased interest in the use of ALG and ATG as induction therapy in renal transplantation partially because of the suggested benefit in protecting against delayed graft function and also as a means to reduce or eliminate CNIs or steroids from maintenance immunosuppressive regimens.[203]

Monoclonal Antibodies

The development of monoclonal antibodies to T cell surface molecules offers the advantage of homogeneous preparations and more predictable therapeutic agents. A number of monoclonal antibodies have been shown in clinical trials to produce immunosuppression,[113] but the most effective monoclonal antibody tested to date is OKT3, which binds the CD3 of the TCR complex. [109] [204] The TCR is a complex of six or seven polypeptides, including the polymorphic α- and β-chains, which provide the antigen recognition component, and the γ-, Δ-, and ε-chains, plus a ζζ-homodimer or ζη-heterodimer, which provide the signal transduction function of the receptor. OKT3 binds to the TCR ε-chain, which is a pan-T cell nonpolymorphic component of the antigen receptor.[109] Immunosuppression with OKT3 blocks both CD4+ and CD8+ T cell function.

Treatment with OKT3 produces multiple effects on T lymphocytes. The acute effect, which commences within minutes, is partial activation of T cells due to signal transduction induced by the antibody. [113] [204] The activated T cells produce large amounts of cytokines, among them TNF-α, which causes many of the acute side effects of treatment, including fever, chills, nausea, vomiting, diarrhea, headache, anorexia, and a capillary leak syndrome. In severe cases, patients can become hypotensive secondary to vasodilation and diarrhea. Alternatively, volume-overloaded patients are susceptible to pulmonary edema from the capillary leak syndrome. Therefore, patients should be within 3% of their estimated dry weight before commencing therapy. Later side effects include aseptic meningitis and serum sickness due to the development of antimouse antibodies. The development of antimouse antibodies results in rapid clearance of OKT3 from the serum and renders it ineffective. The presence of antimouse antibodies can be monitored by indirect immunofluorescence by flow cytometry. Owing to the profound immunosuppression, OKT3 treatment can be complicated by opportunistic infections, including an increased incidence of CMV infection. Multiple courses of therapy have been associated with an in-creased incidence of malignancy, principally non-Hodgkins lymphoma. There at least two components to the mechanism of immunosuppression by OKT3. [109] [111] [113] [204] [205] Within hours after administration, OKT3 causes profound depletion of peripheral T cells. Because OKT3 is not cytotoxic, the depletion is attributed to sequestration of the T cells. In addition, T cells down-modulate the expression of the TCR complex. Thus, after a few days of treatment, by flow cytometry circulating CD4+ and CD8+ cells can be shown to lack detectable TCR. It is thought that the TCR is internalized by endocytosis or shedding. The modulation of TCR expression is reversible, and after elimination of OKT3, the TCR-CD3 complex is again expressed on the cell surface. A new humanized, less antigenic, form of an anti-CD3 antibody has recently been used for the treatment of type I diabetes as well as in renal transplant patients. [205] [206]

Chimeric and humanized antibodies targeted against the IL2 receptor have also been developed. They are created by splicing together the DNA encoding the antigen recognition portion of the murine antibody and with human immunoglobulin DNA. Humanized antibodies contain 90% human and 10% murine antibody sequences. Chimeric antibodies are composed of mouse light and heavy chains and the human Fc portion. AntiIL2 R antibodies bind to the alpha subunit of the high affinity IL-2 receptor, which is only expressed on activated lymphocytes (15). This agent competitively inhibits the IL-2 activation of lymphocytes. Therapy with IL-2 receptor antibodies results in a highly specific inhibition of the lymphocytic immune response of activated T cells, [147] [207] therefore suppressing T cell activity against the allograft. However, in patients, 10% to 15% of T cells are IL2Ra positive; therefore, use of IL2R antibodies may also have more of a generalized anti-inflammatory effect. These antibodies are indicated for prophylaxis rather than treatment of acute allograft rejection.

Experimental Immunosuppressive Drugs

A number of newer biologic agents that have not been approved for clinical transplantation are currently being used clinically, and some are undergoing evaluation in phase III studies. Campath1H is a lymphocyte-depleting antibody (anti-CD52) that has been approved for use in leukemias. It has been used for induction therapy in renal transplantation with low-dose cyclosporine,[208] sirolimus,[209] or alone[210] with variable results. [208] [209] [210] Interestingly, a high rate of acute rejection rates including humoral rejection was reported when the antibody was used alone or with sirolimus monotherapy. [209] [210] T cell costimulatory blockade, in the form of CTLA4Ig (LEA29Y),[155] is in phase III trials in kidney transplantation as a calcineurin inhibitor-sparing agent (see Chapter 65 ).


Recent completion of a draft of the sequence of the human genome[161] in addition to concurrent technologic advances have made the simultaneous analysis of tens of thousands of genes feasible. If the gene profile associated with rejection and overimmunosuppression can be determined from biopsy material or even a blood or urine sample, it would have great clinical applicability. The application of DNA microarrays to the analysis of transplantation is in its infancy, but investigations in animal models and clinical studies are emerging. A study of rejection in human recipients of kidney allografts analyzed gene expression in biopsies of renal allografts from patients on triple immunosuppression with a confirmed diagnosis of acute rejection.[211] Expression of gene transcripts from seven renal biopsies with histopathologic evidence of acute cellular rejection were compared with tissue from three normal renal biopsies. Using a minimum threshold of fourfold induction, four genes were identified in each acute rejection sample: human monokine induced by interferon-γ (a CXC chemokine crucial for murine acute allograft rejection), TCR-active β-chain protein, IL-2–stimulated phosphoprotein (expressed predominantly in T cells) and RING4 (important in intracellular antigen processing). Murine cardiac allografts analyzsed for the expression of approximately 6500 genes and expressed sequence tags (ests) following transplantation identified 181 genes with significantly modulated expression in at least one of the experimental groups of ischemia, stress, innate, or adaptive immunity. In an additional study, differentially expressed genes were identified using self-organizing maps. Twenty-nine genes were up-regulated as early as 6 hours after transplantation.[212] These included IL-1 receptor, IL-6, and haptoglobin, which have all been associated with the acute phase response. In current clinical practice, acute rejection is typically diagnosed by histologic analysis of graft biopsies. The application of these newer technologies could produce a molecular profile of each biopsy specimen that could provide more precise diagnostic criteria. Studies analyzing urine samples from kidney transplant recipients have found a correlation between acute rejection and the cytotoxic T cell products. [213] [214] Another study has found that urinary levels of T cell regulatory marker FOXP3 mRNA were inversely correlated with serum creatinine during acute rejection episodes.[215]


One of the most important problems currently encountered in kidney transplantation is the insufficiency of donor organs to satisfy the increasing recipient demand. In that regard, strategies targeted at expanding the donor pool without jeopardizing graft function and outcome are extremely important. The need to address this problem has stimulated research into xenotransplantation, organogenesis and cellular transplantation to augment failing organs.

An area of obvious importance is xenotransplantation (i.e., transplantation across species). More than 25 years ago, baboon and chimpanzee organs were transplanted into humans, with short-term success. Currently, xenotransplantation studies are targeted at pig-to-primate xenografting, as a preclinical model for developing pig-to-human transplantation. Several hurdles have yet to be overcome, however. They can be classified as immunologic and nonimmunologic.[162] The immunologic response to a xenotransplant can be envisaged as a series of waves consisting of hyperacute, acute vascular, cellular, and chronic rejection.[216] Hyperacute rejection is mediated by preformed complement binding xenoreactive antibodies, which are present in primates and humans and are reactive to a carbohydrate epitope (Gal1a-3Gal) on endothelial cells of lower mammals and occurs within hours of transplantation. Xenografts also express many other antigens, which are unfamiliar to the host. These antigens can elicit a vigorous immune response resulting in acute vascular rejection within days to weeks.[217] Experimental approaches to deal with these problems include development of transgenic pigs that express complement-regulatory proteins such as decay-accelerating factor (CD55),[218] and the generation of knockout animals that express lower levels of antigenic proteins.[219] Recently, pigs have been developed using cloning by nuclear transfer that do not express the Gala1 Gal epitope. [220] [221] [222] Unfortunately, the data presented seem to suggest that anti-Gal antibodies play only a relatively small role in mediating pig xenograft rejection,[223] highlighting the enormous difficulty of host immunity currently facing the field. It now appears that hyperacute rejection is only the beginning of the problem and raises the concern that immune regulatory responses that normally curb the immune response would work as efficiently across species. Nonimmunologic hurdles to xenotransplantation include risk of transmitting infections, physiologic compatibility, and ethical and religious issues related to using organs from animals for humans. These issues have also slowed down the development of xenotransplantation.[224] Organogenesis, or the growing of organs, is an exciting area. Research in this field remains rudimentary; however, the demonstration that rat metanephroi transplanted into the omentum of another rat are capable of growing, differentiating, becoming vascularized, and functioning has generated interest in this field.[225] Because fetal tissue expresses lower levels MHC and other antigens, this approach may have the advantage of stimulating less of a xenoimmune response when transplanted across species.[226]

Recently, it has been demonstrated that cellular transplantation of skeletal myoblasts appear to be able to augment cardiac function in a damaged heart.[227] The complexity of the kidney very likely precludes such straightforward replacement strategies. Although it has clearly been demonstrated that recipient cells are capable of incorporating into donor kidneys and presumably function at some level.[228] Similarly, a number of investigators are looking at the potential of both adult and embryonic stem cells to repair and regenerate damaged tissue before complete organ failure has occurred, or cloning of stomatic cells through nuclear reprogramming. [229] [230]


1. Goodnow CC, Adelstein S, Basten A: The need for central and peripheral tolerance in the B cell repertoire.  Science  1990; 248:1373-1379.

2. Shevach EM: CD4+ CD25+ suppressor T cells: More questions than answers.  Nat Rev Immunol  2002; 2:389-400.

3. Kappler JW, Roehm N, Marrack P: T cell tolerance by clonal elimination in the thymus.  Cell  1987; 49:273-280.

4. Murphy KM, Weaver CT, Elish M, et al: Peripheral tolerance to allogeneic class II histocompatibility antigens expressed in transgenic mice: Evidence against a clonal-deletion mechanism.  Proc Natl Acad Sci U S A  1989; 86:10034-10038.

5. Schwartz RH: Acquisition of immunologic self-tolerance.  Cell  1989; 57:1073-1081.

6. Nossal GJ: Negative selection of lymphocytes.  Cell  1994; 76:229-239.

7. Fry AM, Jones LA, Kruisbeek AM, Matis LA: Thymic requirement for clonal deletion during T cell development.  Science  1989; 246:1044-1046.

8. von Boehmer H: Positive selection of lymphocytes.  Cell  1994; 76:219-228.

9. Gallegos AM, Bevan MJ: Central tolerance: Good but imperfect.  Immunol Rev  2006; 209:290-296.

10. Okazaki T, Honjo T: The PD-1-PD-L pathway in immunological tolerance.  Trends Immunol  2006; 27:195-201.

11. Powell JD: The induction and maintenance of T cell anergy.  Clin Immunol  2006; 120:239-246.

12. Sakaguchi S, Sakaguchi N: Regulatory T cells in immunologic self-tolerance and autoimmune disease.  Int Rev Immunol  2005; 24:211-226.

13. Sakaguchi S, Setoguchi R, Yagi H, Nomura T: Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in self-tolerance and autoimmune disease.  Curr Top Microbiol Immunol  2006; 305:51-66.

14. Suciu-Foca N, Cortesini R: Reviewing the mechanism of peripheral tolerance in clinical transplantation.  Contrib Nephrol  2005; 146:132-142.

15. Gorer PA: The genetic and antigenic basis of tumor transplantation.  J Pathol Bacteriol  1937; 44:691-699.

16. Snell GD: Methods for the study of histocompatibility genes.  J Genetics  1948; 49:87-91.

17. Klein J: An attempt at an interpretation of the mouse H-2 complex.  Contemp Top Immunobiol  1976; 5:297-336.

18. Klein J: Evolution and function of the major histocompatibility system: Facts and speculations.   In: Goetze D, ed. The Major Histocompatibility System in Man and Animals,  Berlin: Springer-Verlag; 1977:339.

19. Shreffler DC, David CS: The H-2 major histocompatibility complex and the I immune response region: Genetic variation, function, and organization.  Adv Immunol  1975; 20:125-195.

20. Dausset J: Iso-leuco anticorps.  Acta Haematol  1958; 20:156-166.

21. Lamm LU, Friedrich U, Petersen CB, et al: Assignment of the major histocompatibility complex to chromosome No. 6 in a family with a pericentric inversion.  Hum Hered  1974; 24:273-284.

22. Graff RJ, Silvers WK, Billingham RE, et al: The cumulative effect of histocompatibility antigens.  Transplantation  1966; 4:605-617.

23. Holman E: Protein sensitization in iso-skin grafting. Is the latter of practical value?.  Surg Gynecol Obstet  1924; 38:100-111.

24. Medawar PB: The behaviour and fate of skin autografts and skin homografts in rabbits.  J Anat  1944; 78:176-179.

25. Germain RN: MHC-dependent antigen processing and peptide presentation: Providing ligands for T lymphocyte activation.  Cell  1994; 76:287-299.

26. Zinkernagel RM, Doherty PC: Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system.  Nature  1974; 248:701-702.

27. Payne R, Tripp M, Weigle J, et al: A new leucocyte isoantigen system in man.  Cold Spring Harb Symp Quant Biol  1964; 29:285-295.

28.   Charron D: Genetic diversity of HLA. Functional and medical implication. Twelfth International Histocompatibility Workshop and Conference. Paris, 1997.

29. Bodmer JG, Marsh SG, Albert ED, et al: Nomenclature for factors of the HLA system, 1995.  Tissue Antigens  1995; 46:1-18.

30. Williams AF: A year in the life of the immunoglobulin superfamily.  Immunol Today  1987; 8:298-303.

31. Immunogenetics Data Base for HLA, December 2002.

32. Bjorkman PJ, Saper MA, Samraoui B, et al: Structure of the human class I histocompatibility antigen, HLA-A2.  Nature  1987; 329:506-512.

33. Bjorkman PJ, Saper MA, Samraoui B, et al: The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens.  Nature  1987; 329:512-518.

34. Davis MM, Bjorkman PJ: T-cell antigen receptor genes and T-cell recognition.  Nature  1988; 334:395-402.

35. Garboczi DN, Ghosh P, Utz U, et al: Structure of the complex between human T-cell receptor, viral peptide and HLA-A2.  Nature  1996; 384:134-141.

36. Rotzschke O, Falk K: Naturally-occurring peptide antigens derived from the MHC class-I- restricted processing pathway.  Immunol Today  1991; 12:447-455.

37. Patel SD, Monaco JJ, McDevitt HO: Delineation of the subunit composition of human proteasomes using antisera against the major histocompatibility complex-encoded LMP2 and LMP7 subunits.  Proc Natl Acad Sci U S A  1994; 91:296-300.

38. Powis SH, Rosenberg WM, Hall M, et al: TAP1 and TAP2 polymorphism in coeliac disease.  Immunogenetics  1993; 38:345-350.

39. Bidwell J: DNA-RFLP analysis and genotyping of HLA-DR and DQ antigens.  Immunol Today  1988; 9:18-23.

40. Marcadet A, Cohen D, Dausset J, et al: Genotyping with DNA probes in combined immunodeficiency syndrome with defective expression of HLA.  N Engl J Med  1985; 312:1287-1292.

41. Higuchi R, von Beroldingen CH, Sensabaugh GF, Erlich HA: DNA typing from single hairs.  Nature  1988; 332:543-546.

42. Milford EL: HLA molecular typing.  Curr Opin Nephrol Hypertens  1993; 2:892-897.

43. Brown JH, Jardetsky TS, Gorga JC, et al: Three dimensional structure of the human class II histocompatibility antigen HLA-DR1.  Nature  1993; 364:33-39.

44. Chicz RM, Urban RG, Gorga JC, et al: Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles.  J Exp Med  1993; 178:27-47.

45. Halloran P, Mathew T, Tomlanovich S, et al: Mycophenolate mofetil in renal allograft recipients: A pooled efficacy analysis of three randomized, double-blind, clinical studies in prevention of rejection. The International Mycophenolate Mofetil Renal Transplant Study Groups.  Transplantation  1997; 63:39-47.

46. Alper CA, Larsen CE, Dubey DP, et al: The haplotype structure of the human major histocompatibility complex.  Hum Immunol  2006; 67:73-84.

47. Jones EY, Fugger L, Strominger JL, Siebold C: MHC class II proteins and disease: a structural perspective.  Nat Rev Immunol  2006; 6:271-282.

48. Schwartz BD, Luehrman LK, Rodey GE: Public antigenic determinant on a family of HLA-B molecules.  J Clin Invest  1979; 64:938-947.

49. Bach FH, van Rood JJ: The major histocompatibility complex—-Genetics and biology. (First of three parts).  N Engl J Med  1976; 295:806-813.

50. Human Renal Transplant Registry: Report No. 12.  JAMA  1995; 233:787-796.

51. Opelz G: Correlation of HLA matching with kidney graft survival in patients with or without cyclosporine treatment.  Transplantation  1985; 40:240-243.

52. Terasaki PI: Overview.   In: Terasaki PI, Cecka M, ed. Clinical Transplantation,  Los Angeles: UCLA Tissue Typing Laboratory; 1987:467.

53. Schreuder GM, Hendriks GF, D'Amaro J, Persijn GG: An eight-year study of HLA typing proficiency in Eurotransplant.  Tissue Antigens  1986; 27:131-141.

54. Cecka JM, Terasaki PI: Clinical Transplantation 2001,  Los Angeles, UCLA Tissue Typing Laboratory, 2002.

55. Opelz G, Wujciak T, Dohler B, et al: HLA compatibility and organ transplant survival. Collaborative Transplant Study.  Rev Immunogenet  1999; 1:334-342.

56. Opelz G, Mytilineos J, Scherer S, et al: Survival of DNA HLA-DR typed and matched cadaver kidney transplants. The Collaborative Transplant Study.  Lancet  1991; 338:461-463.

57. Morris PJ, Johnson RJ, Fuggle SV, et al: Analysis of factors that affect outcome of primary cadaveric renal transplantation in the UK. HLA Task Force of the Kidney Advisory Group of the United Kingdom Transplant Support Service Authority (UKTSSA).  Lancet  1999; 354:1147-1152.

58. Takemoto SK, Terasaki PI, Gjertson DW, Cecka JM: Twelve years' experience with national sharing of HLA-matched cadaveric kidneys for transplantation.  N Engl J Med  2000; 343:1078-1084.

59. Opelz G, Mytilineos J, Wujciak T, et al: Current status of HLA matching in renal transplantation. The Collaborative Transplant Study.  Clin Invest  1993; 70:767-772.

60. Zantvoort FA, D'Amaro J, Persijn GG, et al: The impact of HLA-A matching on long-term survival of renal allografts.  Transplantation  1996; 61:841-844.

61. Meier-Kriesche HU, Schold JD, Kaplan B: Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.  Am J Transplant  2004; 4:1289-1295.

62. Meier-Kriesche HU, Schold JD, Srinivas TR, Kaplan B: Lack of improvement in renal allograft survival despite a marked decrease in acute rejection rates over the most recent era.  Am J Transplant  2004; 4:378-383.

63. Cohen DJ, St Martin L, Christensen LL, et al: Kidney and pancreas transplantation in the United States 1995-2004.  Am J Transplant  2006; 6:1153-1169.

64. Sijpkens YW, Doxiadis II, De Frijter JW, et al: Sharing cross-reactive groups of MHC class I improves long-term graft survival.  Kidney Int  1999; 56:1920-1927.

65. van Saase JL, van der Woude FJ, Thorogood J, et al: The relationship between acute vascular and interstitial renal allograft rejection and subsequent chronic rejection.  Transplantation  1995; 59:1280-1285.

66. Humar A, Kerr S, Gillingham KJ, Matas AJ: Features of acute rejection that increase risk for chronic rejection.  Transplantation  1999; 68:1200-1203.

67. Humar A, Payne WD, Sutherland DE, Matas AJ: Clinical determinants of multiple acute rejction episodes in kidney transplant recipients.  Transplantation  2000; 69:2357-2360.

68. Joseph JT, Kingsmore DB, Junor BJ, et al: The impact of late acute rejection after cadaveric kidney transplantation.  Clin Transplant  2001; 15:221-227.

69. Meier-Kriesche HU, Schold JD, Srinivas TR, Kaplan B: Lack of improvement in renal allograft survival despite a marked decrease in acute rejection rates over the most recent era.  Am J Transplant  2004; 4:378-383.

70. Takemoto S: HLA amino acid residue matching.   In: Cecka JM, Terasaki PI, ed. Clinical Transplants 1997,  Los Angeles: UCLA Tissue Typing Laboratory; 1997:397.

71. Takiff H, Cook DJ, Himaya NS, et al: Dominant effect of histocompatibility on ten-year kidney transplant survival.  Transplantation  1988; 45:410-415.

72. Roopenian DC, Christianson GJ, Davis AP, et al: The genetic origin of minor histocompatibility antigens.  Immunogenetics  1993; 38:131-140.

73. Yard BA, Claas FH, Paape ME, et al: Recognition of a tissue-specific polymorphism by graft infiltrating T-cell clones isolated from a renal allograft with acute rejection.  Nephrol Dial Transplant  1994; 9:805-810.

74. Yard BA, Kooymans-Couthino M, Reterink T, et al: Analysis of T cell lines from rejecting renal allografts.  Kidney Int Suppl  1993; 39:S133-S138.

75. Garovoy MR, Franco V, Zschaeck D, et al: Direct lymphocyte-mediated cytotoxicity as an assay of presensitisation.  Lancet  1973; 1:573-576.

76. Goulmy E, Termijtelen A, Bradley BA, van Rood JJ: Alloimmunity to human H-Y.  Lancet  1976; 2:1206.

77. Pfeffer PF, Thorsby E: HLA-restricted cytotoxicity against male-specific (H-Y) antigen after acute rejection of an HLA-identical sibling kidney: clonal distribution of the cytotoxic cells.  Transplantation  1982; 33:52-56.

78. Crew RJ, Ratner LE: Overcoming immunologic incompatibility: Transplanting the difficult to transplant patient.  Semin Dial  2005; 18:474-481.

79. Lucchiari N, Panajotopoulos N, Xu C, et al: Antibodies eluted from acutely rejected renal allografts bind to and activate human endothelial cells.  Hum Immunol  2000; 61:518-527.

80. Perrey C, Brenchley PE, Johnson RW, Martin S: An association between antibodies specific for endothelial cells and renal transplant failure.  Transpl Immunol  1998; 6:101-106.

81. Opelz G, Terasaki PI: Improvement of kidney-graft survival with increased numbers of blood transfusions.  N Engl J Med  1978; 299:799-803.

82. Briscoe DM, Sayegh MH: A rendezvous before rejection: Where do T cells meet transplant antigens?.  Nat Med  2002; 8:220-222.

83. Lakkis FG, Arakelov A, Konieczny BT, Inoue Y: Immunologic “ignorance” of vascularized organ transplants in the absence of secondary lymphoid tissue.  Nat Med  2000; 6:686-688.

84. Bach FH, Graupner K, Dan E, Klostermann H: Cell kinetic studies in mixed leukocyte cultures: An in vitro model of homograft reactivity.  Proc Natl Acad Sci U S A  1969; 62:374-384.

85. Sherman LA, Chattopadhyay S: The molecular basis of allorecognition.  Annu Rev Immunol  1993; 11:385-402.

86. Eckels DD: Alloreactivity: Allogeneic presentation of endogenous peptide or direct recognition of MHC polymorphism? A review.  Tissue Antigens  1990; 35:49-55.

87. Lechler R, Batchelor R, Lombardi G: The relationship between MHC restricted and allospecific T cell recognition.  Immunol Lett  1991; 29:41-50.

88. Lechler RI, Lombardi G, Batchelor JR, et al: The molecular basis of alloreactivity.  Immunol Today  1990; 11:83-88.

89. Jerne NK: The somatic generation of immune recognition.  Eur J Immunol  1971; 1:1-9.

90. Braciale TJ, Andrew ME, Braciale VL: Simultaneous expression of H-2-restricted and alloreactive recognition by a cloned line of influenza virus-specific cytotoxic T lymphocytes.  J Exp Med  1981; 153:1371-1376.

91. Finberg R, Burakoff SJ, Cantor H, Benacerraf B: Biological significance of alloreactivity: T cells stimulated by Sendai virus-coated syngeneic cells specifically lyse allogeneic target cells.  Proc Natl Acad Sci U S A  1978; 75:5145-5149.

92. Matis LA, Sorger SB, McElligott DL, et al: The molecular basis of alloreactivity in antigen-specific, major histocompatibility complex-restricted T cell clones.  Cell  1987; 51:59-69.

93. Srendni B, Schwartz RH: Alloreactivity of an antigen-specific T cell clone.  Nature  1980; 287:855-857.

94. Sayegh MH: Why do we reject a graft? Role of indirect allorecognition in graft rejection.  Kidney Int  1999; 56:1967-1979.

95. Sayegh MH, Turka LA: The role of T-cell costimulatory activation pathways in transplant rejection.  N Engl J Med  1998; 338:1813-1821.

96. Auchincloss Jr H, Lee R, Shea S, et al: The role of “indirect” recognition in initiating rejection of skin grafts from major histocompatibility complex class II-deficient mice.  Proc Natl Acad Sci U S A  1993; 90:3373-3377.

97. Rogers NJ, Lechler RI: Allorecognition.  Am J Transplant  2001; 1:97-102.

98. Ciubotariu R, Liu Z, Colovai AI, et al: Persistent allopeptide reactivity and epitope spreading in chronic rejection of organ allografts.  J Clin Invest  1998; 101:398-405.

99. SivaSai KS, Smith MA, Poindexter NJ, et al: Indirect recognition of donor HLA class I peptides in lung transplant recipients with bronchiolitis obliterans syndrome.  Transplantation  1999; 67:1094-1098.

100. Vella JP, Spadafora-Ferreira M, Murphy B, et al: Indirect allorecognition of major histocompatibility complex allopeptides in human renal transplant recipients with chronic graft dysfunction.  Transplantation  1997; 64:795-800.

101. Lechler RI, Heaton T, Barber L, et al: Molecular mimicry by major histocompatibility complex molecules and peptides accounts for some alloresponses.  Immunol Lett  1992; 34:63-69.

102. Cotner T, Mellins E, Johnson AH, Pious D: Mutations affecting antigen processing impair class II-restricted allorecognition.  J Immunol  1991; 146:414-417.

103. Elliott TJ, Eisen HN: Cytotoxic T lymphocytes recognize a reconstituted class I histocompatibility antigen (HLA-A2) as an allogeneic target molecule.  Proc Natl Acad Sci U S A  1990; 87:5213-5217.

104. Heath WR, Hurd ME, Carbone FR, Sherman LA: Peptide-dependent recognition of H-2Kb by alloreactive cytotoxic T lymphocytes.  Nature  1989; 341:749-752.

105. Bluestone JA, Kaliyaperumal A, Jameson S, et al: Peptide-induced changes in class I heavy chains alter allorecognition.  J Immunol  1993; 151:3943-3953.

106. Magee CC, Sayegh MH: Peptide-mediated immunosuppression.  Curr Opin Immunol  1997; 9:669-675.

107. Suciu-Foca N, Reed E, D'Agati VD, et al: Soluble HLA antigens, anti-HLA antibodies, and antiidiotypic antibodies in the circulation of renal transplant recipients.  Transplantation  1991; 51:593-601.

108. Kearney ER, Walunas TL, Karr RW, et al: Antigen-dependent clonal expansion of a trace population of antigen-specific CD4+ T cells in vivo is dependent on CD28 costimulation and inhibited by CTLA-4.  J Immunol  1995; 155:1032-1036.

109. Clevers H, Alarcon B, Wileman T, Terhorst C: The T cell receptor/CD3 complex: A dynamic protein ensemble.  Annu Rev Immunol  1988; 6:629-662.

110. Caillat-Zucman S, Blumenfeld N, Legendre C, et al: The OKT3 immunosuppressive effect. In situ antigenic modulation of human graft-infiltrating T cells.  Transplantation  1990; 49:156-160.

111. Gaston RS, Deierhoi MH, Patterson T, et al: OKT3 first-dose reaction: Association with T cell subsets and cytokine release.  Kidney Int  1991; 39:141-148.

112. Shaw AS, Amrein KE, Hammond C, et al: The lck tyrosine protein kinase interacts with the cytoplasmic tail of the CD4 glycoprotein through its unique amino-terminal domain.  Cell  1989; 59:627-636.

113. Denton MD, Magee CC, Sayegh MH: Immunosuppressive strategies in transplantation.  Lancet  1999; 353:1083-1091.

114. Haug CE, Colvin RB, Delmonico FL, et al: A phase I trial of immunosuppression with anti-ICAM-1 (CD54) mAb in renal allograft recipients.  Transplantation  1993; 55:766-772.

115. Hourmant M, Bedrossian J, Durand D, et al: A randomized multicenter trial comparing leukocyte function-associated antigen-1 monoclonal antibody with rabbit antithymocyte globulin as induction treatment in first kidney transplantations.  Transplantation  1996; 62:1565-1570.

116. Bretscher P, Cohn M: A theory of self-nonself discrimination.  Science  1970; 169:1042-1049.

117. Linsley PS, Greene JL, Tan P, et al: Coexpression and functional cooperation of CTLA-4 and CD28 on activated T lymphocytes.  J Exp Med  1992; 176:1595-1604.

118. Bluestone JA: Is CTLA-4 a master switch for peripheral T cell tolerance?.  J Immunol  1997; 158:1989-1993.

119. Rothstein DM, Sayegh MH: T-cell costimulatory pathways in allograft rejection and tolerance.  Immunol Rev  2003; 196:85-108.

120. Sharpe AH, Freeman GJ: The B7-CD28 superfamily.  Nat Rev Immunol  2002; 2:116-126.

121. Tivol EA, Borriello F, Schweitzer AN, et al: Loss of CTLA-4 leads to massive lymphoproliferation and fatal multiorgan tissue destruction, revealing a critical negative regulatory role of CTLA-4.  Immunity  1995; 3:541-547.

122. Gough SC, Walker LS, Sansom DM: CTLA4 gene polymorphism and autoimmunity.  Immunol Rev  2005; 204:102-115.

123. Judge TA, Tang A, Spain LM, et al: The in vivo mechanism of action of CTLA4Ig.  J Immunol  1996; 156:2294-2299.

124. Schwartz RH: A cell culture model for T lymphocyte clonal anergy.  Science  1990; 248:1349-1356.

125. Perkins DL, Wang Y, Ho SS, et al: Superantigen-induced peripheral tolerance inhibits T cell responses to immunogenic peptides in TCR (beta-chain) transgenic mice.  J Immunol  1993; 150:4284-4291.

126. Rocken M, Urban JF, Shevach EM: Infection breaks T-cell tolerance.  Nature  1992; 359:79-82.

127. Chen W, Sayegh MH, Khoury SJ: Mechanisms of acquired thymic tolerance in vivo: Intrathymic injection of antigen induces apoptosis of thymocytes and peripheral T cell anergy.  J Immunol  1998; 160:1504-1508.

128. Kirk AD, Harlan DM, Armstrong NN, et al: CTLA4-Ig and anti-CD40 ligand prevent renal allograft rejection in primates.  Proc Natl Acad Sci U S A  1997; 94:8789-8794.

129. Levisetti MG, Padrid PA, Szot GL, et al: Immunosuppressive effects of human CTLA4Ig in a non-human primate model of allogeneic pancreatic islet transplantation.  J Immunol  1997; 159:5187-5191.

130. Yamada A, Kishimoto K, Dong VM, et al: CD28-independent costimulation of T cells in alloimmune responses.  J Immunol  2001; 167:140-146.

131. Yoshinaga SK, Whoriskey JS, Khare SD, et al: T-cell co-stimulation through B7RP-1 and ICOS.  Nature  1999; 402:827-832.

132. Harada H, Salama AD, Sho M, et al: The role of the ICOS-B7h T cell costimulatory pathway in transplantation immunity.  J Clin Invest  2003; 112:234-243.

133. McAdam AJ, Greenwald RJ, Levin MA, et al: ICOS is critical for CD40-mediated antibody class switching.  Nature  2001; 409:102-105.

134. Ito T, Ueno T, Clarkson MR, et al: Analysis of the role of negative T cell costimulatory pathways in CD4 and CD8 T cell-mediated alloimmune responses in vivo.  J Immunol  2005; 174:6648-6656.

135. Sanchez-Fueyo A, Sandner S, Habicht A, et al: Specificity of CD4+CD25+ regulatory T cell function in alloimmunity.  J Immunol  2006; 176:329-334.

136. Clarkson MR, Sayegh MH: T-cell costimulatory pathways in allograft rejection and tolerance.  Transplantation  2005; 80:555-563.

137. Vu MD, Clarkson MR, Yagita H, et al: Critical, but conditional, role of OX40 in memory T cell-mediated rejection.  J Immunol  2006; 176:1394-1401.

138. Yuan X, Salama AD, Dong V, et al: The role of the CD134-CD134 ligand costimulatory pathway in alloimmune responses in vivo.  J Immunol  2003; 170:2949-2955.

139. Yamada A, Salama AD, Sho M, et al: CD70 signaling is critical for CD28-independent CD8+ T cell-mediated alloimmune responses in vivo.  J Immunol  2005; 174:1357-1364.

140. Mosmann TR, Cherwinski H, Bond MW, et al: Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins.  J Immunol  1986; 136:2348-2357.

141. Hancock WW, Gao W, Faia KL, Csizmadia V: Chemokines and their receptors in allograft rejection.  Curr Opin Immunol  2000; 12:511-516.

142. Fischereder M, Luckow B, Hocher B, et al: CC chemokine receptor 5 and renal-transplant survival.  Lancet  2001; 357:1758-1761.

143. Abdi R, Means TK, Ito T, et al: Differential role of CCR2 in islet and heart allograft rejection: Tissue specificity of chemokine/chemokine receptor function in vivo.  J Immunol  2004; 172:767-775.

144. Charo IF, Ransohoff RM: The many roles of chemokines and chemokine receptors in inflammation.  N Engl J Med  2006; 354:610-621.

145. Sayegh MH, Akalin E, Hancock WW, et al: CD28-B7 blockade after alloantigenic challenge in vivo inhibits Th1 cytokines but spares Th2.  J Exp Med  1995; 181:1869-1874.

146. Nashan B, Moore R, Amlot P, et al: Randomised trial of basiliximab versus placebo for control of acute cellular rejection in renal allograft recipients. CHIB 201 International Study Group.  Lancet  1997; 350:1193-1198.

147. Vincenti F, Kirkman R, Light S, et al: Interleukin-2-receptor blockade with daclizumab to prevent acute rejection in renal transplantation. Daclizumab Triple Therapy Study Group.  N Engl J Med  1998; 338:161-165.

148. Strom TB, Roy-Chaudhury P, Manfro R, et al: The Th1/Th2 paradigm and the allograft response.  Curr Opin Immunol  1996; 8:688-693.

149. Steiger J, Nickerson PW, Steurer W, et al: IL-2 knockout recipient mice reject islet cell allografts.  J Immunol  1995; 155:489-498.

150. Dai Z, Konieczny BT, Baddoura FK, Lakkis FG: Impaired alloantigen-mediated T cell apoptosis and failure to induce long-term allograft survival in IL-2-deficient mice.  J Immunol  1998; 161:1659-1663.

151. Konieczny BT, Dai Z, Elwood ET, et al: IFN-gamma is critical for long-term allograft survival induced by blocking the CD28 and CD40 ligand T cell costimulation pathways.  J Immunol  1998; 160:2059.

152. Lakkis FG, Konieczny BT, Saleem S, et al: Blocking the CD28-B7 T cell costimulation pathway induces long term cardiac allograft acceptance in the absence of IL-4.  J Immunol  1997; 158:2443-2448.

153. Kist-van Holthe JE, Gasser M, Womer K, et al: Regulatory functions of alloreactive Th2 clones in human renal transplant recipients.  Kidney Int  2002; 62:627-631.

154. Waaga AM, Gasser M, Kist-van Holthe JE, et al: Regulatory functions of self-restricted MHC class II allopeptide-specific Th2 clones in vivo.  J Clin Invest  2001; 107:909-916.

155. Vincenti F, Larsen C, Durrbach A, et al: Costimulation blockade with belatacept in renal transplantation.  N Engl J Med  2005; 353:770-781.

156. Salama AD, Sayegh MH: Alternative T-cell costimulatory pathways in transplant rejection and tolerance induction: hierarchy or redundancy?.  Am J Transplant  2003; 3:509-511.

157. Strehlau J, Pavlakis M, Lipman M, et al: Quantitative detection of immune activation transcripts as a diagnostic tool in kidney transplantation.  Proc Natl Acad Sci U S A  1997; 94(2):695.

158. Suthanthiran M: Transplantation tolerance: Fooling mother nature.  Proc Natl Acad Sci U S A  1996; 93:12072-12075.

159. Krieger NR, Yin DP, Fathman CG: CD4+ but not CD8+ cells are essential for allorejection.  J Exp Med  1996; 184:2013-2018.

160. Bracy JL, Cretin N, Cooper DK, Iacomini J: Xenoreactive natural antibodies.  Cell Mol Life Sci  1999; 56:1001-1007.

161. Lander ES, Linton LM, Birren B, et al: Initial sequencing and analysis of the human genome.  Nature  2001; 409:860-921.

162. Auchincloss Jr H, Sachs DH: Xenogeneic transplantation.  Annu Rev Immunol  1998; 16:433-470.

163. Nankivell BJ, Borrows RJ, Fung CL, et al: The natural history of chronic allograft nephropathy.  N Engl J Med  2003; 349:2326-2333.

164. Halloran PF: Call for revolution: A new approach to describing allograft deterioration.  Am J Transplant  2002; 2:195-200.

165. Sayegh MH, Carpenter CB: Transplantation 50 years later—-Progress, challenges, and promises.  N Engl J Med  2004; 351:2761-2766.

166. Womer KL, Vella JP, Sayegh MH: Chronic allograft dysfunction: Mechanisms and new approaches to therapy.  Semin Nephrol  2000; 20:126-147.

167. Lee PC, Terasaki PI, Takemoto SK, et al: All chronic rejection failures of kidney transplants were preceded by the development of HLA antibodies.  Transplantation  2002; 74:1192-1194.

168. Mauiyyedi S, Pelle PD, Saidman S, et al: Chronic humoral rejection: Identification of antibody-mediated chronic renal allograft rejection by C4d deposits in peritubular capillaries.  J Am Soc Nephrol  2001; 12:574-582.

169. Schmidt-Weber CB, Blaser K: Regulation and role of transforming growth factor-beta in immune tolerance induction and inflammation.  Curr Opin Immunol  2004; 16:709-716.

170. Khanna AK, Plummer MS, Hilton G, et al: Anti-transforming growth factor antibody at low but not high doses limits cyclosporine-mediated nephrotoxicity without altering rat cardiac allograft survival: Potential of therapeutic applications.  Circulation  2004; 110:3822-3829.

171. Knechtle SJ, Burlingham WJ: Metastable tolerance in nonhuman primates and humans.  Transplantation  2004; 77:936-939.

172. von Boehmer H: Mechanisms of suppression by suppressor T cells.  Nat Immunol  2005; 6:338-344.

173. Dong VM, Yuan X, Coito AJ, et al: Mechanisms of targeting CD28 by a signaling monoclonal antibody in acute and chronic allograft rejection.  Transplantation  2002; 73:1310-1317.

174. Spriewald BM, Ensminger SM, Billing JS, et al: Increased expression of transforming growth factor-beta and eosinophil infiltration is associated with the development of transplant arteriosclerosis in long-term surviving cardiac allografts.  Transplantation  2003; 76:1105-1111.

175. Mannon RB: Therapeutic targets in the treatment of allograft fibrosis.  Am J Transplant  2006; 6:867-875.

176. Zeisberg M, Hanai J, Sugimoto H, et al: BMP-7 counteracts TGF-beta1-induced epithelial-to-mesenchymal transition and reverses chronic renal injury.  Nat Med  2003; 9:964-968.

177. Lee PC, Terasaki PI, Takemoto SK, et al: All chronic rejection failures of kidney transplants were preceded by the development of HLA antibodies.  Transplantation  2002; 74:1192-1194.

178. Terasaki PI, Ozawa M: Predicting kidney graft failure by HLA antibodies: A prospective trial.  Am J Transplant  2004; 4:438-443.

179. Inaba K, Granelli-Piperno A, Steinman RM: Dendritic cells induce T lymphocytes to release B cell-stimulating factors by an interleukin 2-dependent mechanism.  J Exp Med  1983; 158:2040-2057.

180. Bach JF: The mode of action of immunosuppressive agents: Thiopurines,  Amsterdam, North-Holland Publishing Company, 1985.

181. Sievers TM, Rossi SJ, Ghobrial RM, et al: Mycophenolate mofetil.  Pharmacotherapy  1997; 17:1178-1197.

182. van Gelder T, Klupp J, Barten MJ, et al: Comparison of the effects of tacrolimus and cyclosporine on the pharmacokinetics of mycophenolic acid.  Ther Drug Monit  2001; 23:119-128.

183. Anonymous : A blinded, randomized clinical trial of mycophenolate mofetil for the prevention of acute rejection in cadaveric renal transplantation. The Tricontinental Mycophenolate Mofetil Renal Transplantation Study Group.  Transplantation  1996; 61:1029-1037.

184. Azuma H, Binder J, Heemann U, et al: Effects of RS61443 on functional and morphological changes in chronically rejecting rat kidney allografts.  Transplantation  1995; 59:460-466.

185. Kahan BD: Cyclosporine.  N Engl J Med  1989; 321:1725-1738.

186. Kronke M, Leonard WJ, Depper JM, et al: Cyclosporin A inhibits T-cell growth factor gene expression at the level of mRNA transcription.  Proc Natl Acad Sci U S A  1984; 81:5214-5218.

187. Emmel EA, Verweij CL, Durand DB, et al: Cyclosporin A specifically inhibits function of nuclear proteins involved in T cell activation.  Science  1989; 246:1617-1620.

188. Dumont FJ, Staruch MJ, Koprak SL, et al: Distinct mechanisms of suppression of murine T cell activation by the related macrolides FK-506 and rapamycin.  J Immunol  1990; 144:251-258.

189. Schreiber SL, Crabtree GR: The mechanism of action of cyclosporin A and FK506.  Immunol Today  1992; 13:136-142.

190. Liu J, Farmer Jr JD, Lane WS, et al: Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes.  Cell  1991; 66:807-815.

191. Jain J, McCaffrey PG, Valge-Archer VE, Rao A: Nuclear factor of activated T cells contains Fos and Jun.  Nature  1992; 356:801-804.

192. Bierer BE, Mattila PS, Standaert RF, et al: Two distinct signal transmission pathways in T lymphocytes are inhibited by complexes formed between an immunophilin and either FK506 or rapamycin.  Proc Natl Acad Sci U S A  1990; 87:9231-9235.

193. Tocci MJ, Matkovich DA, Collier KA, et al: The immunosuppressant FK506 selectively inhibits expression of early T cell activation genes.  J Immunol  1989; 143:718-726.

194. Clipstone NA, Crabtree GR: Identification of calcineurin as a key signalling enzyme in T-lymphocyte activation.  Nature  1992; 357:695-697.

195. Bierer BE, Schreiber SL, Burakoff SJ: The effect of the immunosuppressant FK-506 on alternate pathways of T cell activation.  Eur J Immunol  1991; 21:439-445.

196. Calvo V, Crews CM, Vik TA, Bierer BE: Interleukin 2 stimulation of p70 S6 kinase activity is inhibited by the immunosuppressant rapamycin.  Proc Natl Acad Sci U S A  1992; 89:7571-7575.

197. Kuo CJ, Chung J, Fiorentino DF, et al: Rapamycin selectively inhibits interleukin-2 activation of p70 S6 kinase.  Nature  1992; 358:70-73.

198. Morice WG, Wiederrecht G, Brunn GJ, et al: Rapamycin inhibition of interleukin-2-dependent p33cdk2 and p34cdc2 kinase activation in T lymphocytes.  J Biol Chem  1993; 268:22737-22745.

199. MacDonald AS: A worldwide, phase III, randomized, controlled, safety and efficacy study of a sirolimus/cyclosporine regimen for prevention of acute rejection in recipients of primary mismatched renal allografts.  Transplantation  2001; 71:271-280.

200. Guba M, von Breitenbuch P, Steinbauer M, et al: Rapamycin inhibits primary and metastatic tumor growth by antiangiogenesis: involvement of vascular endothelial growth factor.  Nat Med  2002; 8:128-135.

201. Webster AC, Pankhurst T, Rinaldi F, et al: Monoclonal and polyclonal antibody therapy for treating acute rejection in kidney transplant recipients: a systematic review of randomized trial data.  Transplantation  2006; 81:953-965.

202. Taylor AL, Watson CJ, Bradley JA: Immunosuppressive agents in solid organ transplantation: Mechanisms of action and therapeutic efficacy.  Crit Rev Oncol Hematol  2005; 56:23-46.

203. Meier-Kriesche HU, Li S, Gruessner RW, et al: Immunosuppression: Evolution in practice and trends, 1994-2004.  Am J Transplant  2006; 6:1111-1131.

204. Norman DJ, Kahana L, Stuart Jr FP, et al: A randomized clinical trial of induction therapy with OKT3 in kidney transplantation.  Transplantation  1993; 55:44-50.

205. Chatenoud L, Baudrihaye MF, Kreis H, et al: Human in vivo antigenic modulation induced by the anti-T cell OKT3 monoclonal antibody.  Eur J Immunol  1982; 12:979-982.

206. Norman DJ, Vincenti F, de Mattos AM, et al: Phase I trial of HuM291, a humanized anti-CD3 antibody, in patients receiving renal allografts from living donors.  Transplantation  2000; 70:1707-1712.

207. Hong JC, Kahan BD: A calcineurin antagonist-free induction strategy for immunosuppression in cadaveric kidney transplant recipients at risk for delayed graft function.  Transplantation  2001; 71:1320-1328.

208. Calne R, Moffatt SD, Friend PJ, et al: Campath IH allows low-dose cyclosporine monotherapy in 31 cadaveric renal allograft recipients.  Transplantation  1999; 68:1613-1616.

209. Knechtle SJ, Pirsch JD, H Fechner Jr J, et al: Campath-1H induction plus rapamycin monotherapy for renal transplantation: results of a pilot study.  Am J Transplant  2003; 3:722-730.

210. Kirk AD, Hale DA, Mannon RB, et al: Results from a human renal allograft tolerance trial evaluating the humanized CD52-specific monoclonal antibody alemtuzumab (CAMPATH-1H).  Transplantation  2003; 76:120-129.

211. Akalin E, Hendrix RC, Polavarapu RG, et al: Gene expression analysis in human renal allograft biopsy samples using high-density oligoarray technology.  Transplantation  2001; 72:948-953.

212. Christopher K, Mueller TF, Ma C, Liang Y, Perkins DL: Analysis of the innate and adaptive phases of allograft rejection by cluster analysis of transcriptional profiles.  J Immunol  2002; 169:522-530.

213. Kotsch K, Mashreghi MF, Bold G, et al: Enhanced granulysin mRNA expression in urinary sediment in early and delayed acute renal allograft rejection.  Transplantation  2004; 77:1866-1875.

214. Li B, Hartono C, Ding R, et al: Noninvasive diagnosis of renal-allograft rejection by measurement of messenger RNA for perforin and granzyme B in urine.  N Engl J Med  2001; 344:947-954.

215. Muthukumar T, Dadhania D, Ding R, et al: Messenger RNA for FOXP3 in the urine of renal-allograft recipients.  N Engl J Med  2005; 353:2342-2351.

216. Cascalho M, Platt JL: Xenotransplantation and other means of organ replacement.  Nat Rev Immunol  2001; 1:154-160.

217. Leventhal JR, Matas AJ, Sun LH, et al: The immunopathology of cardiac xenograft rejection in the guinea pig-to-rat model.  Transplantation  1993; 56:1-8.

218. Zaidi A, Schmoeckel M, Bhatti F, et al: Life-supporting pig-to-primate renal xenotransplantation using genetically modified donors.  Transplantation  1998; 65:1584-1590.

219. Lin SS, Hanaway MJ, Gonzalez-Stawinski GV, et al: The role of anti-Galalpha1-3Gal antibodies in acute vascular rejection and accommodation of xenografts.  Transplantation  2000; 70:1667-1674.

220. Dai Y, Vaught TD, Boone J, et al: Targeted disruption of the alpha1,3-galactosyltransferase gene in cloned pigs.  Nat Biotechnol  2002; 20:251-255.

221. Lai L, Kolber-Simonds D, Park KW, et al: Production of alpha-1,3-galactosyltransferase knockout pigs by nuclear transfer cloning.  Science  2002; 295:1089-1092.

222. Phelps CJ, Koike C, Vaught TD, et al: Production of alpha 1,3-galactosyltransferase-deficient pigs.  Science  2003; 299:411-414.

223. Chen G, Qian H, Starzl T, et al: Acute rejection is associated with antibodies to non-Gal antigens in baboons using Gal-knockout pig kidneys.  Nat Med  2005; 11:1295-1298.

224. Cascalho M, Ogle BM, Platt JL: Xenotransplantation and the future of renal replacement.  J Am Soc Nephrol  2004; 15:1106-1112.

225. Hammerman MR: Growing kidneys.  Curr Opin Nephrol Hypertens  2001; 10:13-17.

226. Hammerman MR: Xenotransplantation of developing kidneys.  Am J Physiol Renal Physiol  2002; 283:F601-F606.

227. Menasche P, Hagege AA, Scorsin M, et al: Myoblast transplantation for heart failure.  Lancet  2001; 357:279-280.

228. Grimm PC, Nickerson P, Jeffery J, et al: Neointimal and tubulointerstitial infiltration by recipient mesenchymal cells in chronic renal-allograft rejection.  N Engl J Med  2001; 345:93-97.

229. Piedrahita JA, Mir B, Dindot S, Walker S: Somatic cell cloning: the ultimate form of nuclear reprogramming?.  J Am Soc Nephrol  2004; 15:1140-1144.

230. Strom TB, Field LJ, Ruediger M: Allogeneic stem cell-derived “repair unit” therapy and the barriers to clinical deployment.  J Am Soc Nephrol  2004; 15:1133-1139.

If you find an error or have any questions, please email us at Thank you!