Minor Transplantation Antigens

Evidence for there being transplantation alloantigens in addition to those encoded within the major histocompatibility complex (MHC*)(H2 in mice, HLA in humans) came originally from the work of Snell in mice(1). He defined transplantation (histocompatibility, H) antigens as those that elicited rejection of skin or tumor grafts from genetically nonidentical donors and devised methods for analyzing the genetic loci that encoded them. Following the identification of H2 as the strongest of these H antigens (hence the term “major”), the others were termed“minor,” as individually they elicit slow graft rejection in comparison with the rapid rejection of H2 disparate grafts. The chromosomal regions encoding each of the H antigens were genetically isolated by backcrossing to make a series of congenic mouse strains, each differing from the parental strain by expression of the selected minor H alloantigen(2, 3). The human counterparts of mouse histocompatibility alloantigens were evident from the rejection of tissues and organs transplanted between genetically dissimilar individuals. That the human MHC, HLA, is the homolog of the mouse H2 was initially established at the functional level by showing that skin grafts exchanged between HLA-matched siblings (HLA identical by descent) survived significantly longer than those between HLA-mismatched siblings(4). Subsequently, this homology was shown by the MHC class I and II proteins and the close similarity of the genetic organization. This was apparent from the mapping and cloning of MHC class I and II and the very large number of linked genes. These included those encoding tumor necrosis factor-α, complement components, as well as proteins important for the biosynthesis of MHC class I and II molecules. The focus of transplantation immunology on MHC genes and antigens paved the way for the discovery of the central physiological role of cell surface-expressed class I and II molecules. This derives from their incorporation during biosynthesis of peptides from other proteins, either endogenous or from invading pathogens, for display at the cell surface. The recognition by T cells of self-MHC molecules incorporating foreign peptides is the basis of protective immune responses to pathogens, and this has probably driven the evolution of the extraordinary polymorphism of MHC molecules, which in turn accounts for their strong immunogenicity in the transplant situation. The central function of MHC molecules in presenting peptides also underlies their involvement in the recognition by T cells of the other, non-MHC transplantation (minor H) antigens. The peptidic nature of one of the essential components of minor H antigens, and the involvement of an MHC molecule as the other, is implied from a number of lines of evidence. First, minor H antigens, like viruses, are recognized by cytotoxic T cells (CTL) in an MHC-restricted fashion(5-7); second, virus-specific CTL recognize short peptides (8) bound within the peptide binding grooves of MHC class I molecules (9, 10); and third, peptides eluted from cell surface MHC molecules can sensitize target cells for lysis by minor H-specific CTL clones (11, 12). The existence of human counterparts of mouse minor H antigens was shown by the finding that HLA-matched donor-recipient pairs, even those with HLA identity by descent (i.e., siblings), still reject grafts (host-versus-graft, HVG) and, in the case of bone marrow transplantation, are susceptible to graft-versus-host (GVH) reactions. By analogy with mice, it would be expected that humans would have numerous loci encoding minor H antigens, and these would include those on the Y chromosome controlling expression of male-specific transplantation antigens (HY). The HY antigens were identified by grafting experiments in mice in the mid-1950s (13) and twenty years later by cytotoxic T cell responses generated in vitro from female mice (6) and humans (14) previously immunized with male tissue. The molecular identification of several different peptide epitopes of male-specific antigens in mice and humans, and the genes encoding them, has allowed investigation of the extent to which these should be regarded as true homologs. The molecular nature of any minor H epitope is that of a complex between an MHC class I or II molecule and a peptide derived by breakdown from a polymorphic protein in one of the cellular compartments: nuclear, mitochrondrial, cytoplasmic, or one of the cytoplasmic vesicles. In principle, one would expect MHC class I molecules to be loaded with endogenous peptide in the endoplasmic reticulum (ER) during biosynthesis and MHC class II molecules with exogenously derived peptide in an endosomal compartment. However, because traffic between cellular compartments is “leaky,” and very small numbers of cell surface MHC molecules loaded with a particular peptide suffice to trigger T cell recognition, it is likely that for some endogenous proteins that comprise components of minor H antigens, the distinction between MHC class I and II peptide loading pathways may be blurred, i.e., not all class I-presented minor H peptides may be endogenously derived, and not all those presented by class II may be exogenous to that cell and brought into it by endocytosis. Both pathways could be involved in loading minor H peptides into either class of MHC molecule. This is important for understanding the relative roles of direct and indirect T cell recognition of minor H antigens by T cells in vivo. It is also relevant for understanding immune responses to autoantigens and tumor antigens, whose T cell-recognized components are also peptides derived from endogenous proteins and are displayed on the cell surface in the peptide binding grooves of self-MHC class I or II molecules. WHAT IS A MINOR HISTOCOMPATIBILITY (TRANSPLANTATION) ANTIGEN“LOCUS”? The generation of congenic mouse strains established that there were certain chromosomal regions on autosomes in which genes encoding minor H antigens were located (2, 3). The male-specific transplantation antigens mapped one such region onto the Y chromosome(13). However, the isolation of both CD4 and CD8 T cell clones in vitro from mice immunized across these “single” minor H locus barriers hinted at a greater complexity: the use of such clones to test for the expression of minor H epitopes on tissues from recombinant mice provided evidence that within each “locus” there existed separate genes encoding the class I- and II-restricted minor H epitopes recognized by CD8 and CD4 T cells, respectively(15-18). The existence of both genes, isolated together in a congenic strain, allows rejection of grafts exchanged between the congenic and the parental strain, graft rejection being a function of the synergistic interaction of helper (CD4) and cytotoxic (CD8) cells. However, when grafts are exchanged between mouse strains matched at the MHC, H2, but differing at multiple minor H “loci,” a situation which parallels the exchange of grafts between HLA-matched humans, rejection is much more rapid than those differing at a single minor H “locus”(1-3). The effect of multiple minor H differences is synergistic, and the “help” provided by a CD4 T cell responding to one epitope is not limited to boosting the activity of a CD8 T cell responding to a class I-restricted epitope encoded by an adjacent gene. It is enough that CD4 and CD8 epitopes are present on the same cell to evoke synergy between responding T cells. That single minor H epitopes are not in themselves transplantation antigens is shown by experiments in which genes encoding single peptide epitopes have been introduced as transgenes into an innbred mouse strain. Skin grafts from transgenic mice are not rejected when grafted onto recipients of the inbred strain and indeed can induce tolerance to subsequent grafts expressing the transgene plus an additional helper epitope (19, 20). These results also illustrate how readily the immune system can enter a state of nonresponsiveness when exposed to “weak” transplantation antigens, a situation also exemplified by the failure of some individual mice in a genetically identical group grafted with skin from a congenic strain, from which grafts are usually rejected, to do so. Knowledge of the nature of the immunogenicity of endogenously processed epitopes is required for a better understanding of how to manipulate immune responses to cells expressing them, to clinical advantage. THE MALE-SPECIFIC ANTIGEN HY: A MODEL TO EXAMINE HOMOLOGS Because any polymorphic protein, especially one expressed ubiquitously, could give rise to a peptide whose presentation by an appropriate MHC molecule could elicit a minor H response, there is no reason to suppose, a priori, that homologous genes in different species will contribute the peptide component of a particular minor H antigen. It is, however, a difficult question to address at the moment, as so few are both cloned and identified at the peptide level. One example of minor transplantation antigen genes that can be examined for homology are those whose peptide component is encoded by a Y chromosome gene, as some HY peptide genes have been identified in mice and humans. Two different, complementary approaches have been taken to identify genes and peptides of minor H antigens. Both depend on the use of minor H-specific T cell clones to determine whether a target or antigen-presenting cell expresses the minor H epitope. One approach, pioneered for the identification of the genes encoding tumor antigens (21-24), is expression cloning, in which DNA or cDNA is transfected into recipient cells expressing the relevant MHC restriction element, and transfectants expressing the epitope are identified by peptide-specific T cells. The minimum DNA fragment conferring expression is sequenced, and from that information candidate peptides are synthesized and tested by addition over a wide dose range to target cells of the appropriate MHC. The peptide sensitizing the target cells for recognition in the nanomolar or picomolar range is the likely cognate peptide for the T cell-recognized epitope. Initial selection of appropriate DNA or cDNA for transfection can assist in this approach and was used to identify two mouse Y chromosome genes that encode HY peptides. One was in a cosmid, which mapped to a region on the Y known to be deleted in a mutation that extinguished expression of HY (25); the other was a cDNA that was mapped within that deletion interval(18). The cosmid contained the ubiquitously expressed gene Smcy and was found to encode the first HY peptide to be identified, HY/Kk, of which the peptide TENSGKDI is presented by the H2-Kk allele (26). The X homolog, Smcx, has a 3-base pair deletion within this stretch of the sequence, resulting in a Y-specific peptide differing at six amino acid residues from that of the X. The cDNA was part of the ubiquitously expressed Uty gene, which encodes another HY peptide, HY/Db, in which the peptide WMHHNMDLI is presented by the H2-Db allele. The X homolog, Utx, differs at three amino acid residues from Uty in this part of the sequence(27). An expression cloning approach has also recently identified a novel gene on chromosome 2 encoding the H13 minor H peptide antigen in mice, thus clarifying the molecular bases of this autosomally encoded minor H antigen (28). A conservative single amino acid difference in this peptide accounts for each of the alleles. Remarkably, the peptides (SSVVGVWYL and SSVIGVWYL) do not display the conventional motif even though they both bind the H2-Db molecule. The second approach to identifying peptide components of minor H antigens was that pioneered by Rammensee and his colleagues. It is based on the elution of peptide from MHC molecules expressing the relevant MHC allele and the minor H epitope in question and testing of separated peptide peaks for their ability to sensitize targets for recognition by minor H-specific T cells. The positive peptide peaks are then sequenced, and testing of synthetic peptides of the deduced sequence is the final step in identification of the cognate peptide. Examination of the amino acid sequence can provide clues about the likely DNA sequence from which it is derived, although redundancy of the genetic code makes this step in gene identification dependent on genes already identified and in the DNA database. This method was originally used to establish direct evidence for the peptidic nature of the associated H2-Db components of the mouse minor H antigens H4 and HY (12). A modification of the method, using mass spectrometric analysis for peptide identification, was used to determine the first human minor H peptide, HA-2, as YIGEVLVSV, which seems to be the product of a myosin-related gene on an unknown autosome, recognized in association with HLA-A*0201 (29). The method was subsequently used to identify two human HY peptides, components of HY/B7 (30) and HY/A2 (31), as SPSVDKARAEL and FIDSYICQV, respectively, both products of different regions of the human SMCY gene, which had been mapped to a position on the long arm of the Y chromosome known to contain gene(s) controlling HY expression in humans (32-34). From these data, it seems that for some peptide components of HY at least, a homologous gene, Smcy/SMCY in mice and humans, is the source of the endogenous protein from which they are derived. However, the peptide components themselves cannot be the same, as the peptide sequence that allows it to be loaded into the different alleles of the human and mouse MHC molecules is different. It is also clear that Smcy is not the only Y chromosome gene to encode HY peptide epitopes, as in mice at least, a second gene, Uty(27), encodes at least one(HY/Db), and there are additional linked candidate genes, some of which may encode MHC class II-restricted epitopes that are essential for immunogenicity (18). However, class II peptide epitopes may also be discovered within genes already identified; Smcy/SMCY for example is a large gene (with a 6.5-kilobase cDNA) and has already been found to encode two HY peptide epitopes in mice and a further two in humans(26, 30, 31, and Scott et al., unpublished results). From sequence analysis, both Smcy and Uty seem to encode regulatory proteins that would be expected to be expressed in the nucleus; Smcy shares considerable sequence homology with the retinoblastoma binding protein, RBP2 (26), and that ofUty is homologous to a family of tetratricopeptide repeat proteins(27). These gene products may therefore interact with DNA or other proteins and hence play a role in transcription regulation, although their function is quite unknown, and it is unclear whether the Y-specific sequence, differing so little from that of the X homolog, has a male-specific role, for example by involvement in spermatogenesis. It is also of interest that the one crucial male-specific gene on the Y, Sry/SRY, which controls testis differentiation during embryonic development(35), lies outside the region where all the HY genes are located in mice. That it is not a minor H peptide-encoding gene may, however, be due to the fact that it does not have ubiquitous tissue expression. MITOCHONDRIAL PEPTIDES ARE TRANSPLANTATION ANTIGENS A mitochondrial peptide was the first minor H peptide to be identified(36), some time before those of HY described above and also taking a genomic approach. The tracking of a minor transplantation antigen to the small mitochrondrial genome came from studies of a maternally transmitted transplantation antigen, which could also be detected by cytotoxic T cells in vitro (37), and these were then used to test candidate peptides derived from examination of sequence differences between alleles of the enzyme gene, mt-ND1. A single amino acid difference in the peptide f-MFFINILTL was found to account for immunogenicity. Subsequently, additional mitochondrial genes in mouse and rat have been found to encode minor H peptides, and several are presented to T cells by nonclassical, MHC class Ib molecules (38, 39). WHY ARE ENDOGENOUS PEPTIDES IMMUNOGENIC? There are a number of different answers to this, depending on the epitope in question. The first requirement for immunogenicity is that the peptide should be generated in, or transported to, an appropriate cellular compartment for incorporation during the biosynthesis of MHC molecules. The second is that the peptide sequence should be capable of being bound in the MHC peptide binding groove of that allele with an appropriate affinity, and the complex be displayed on the surface of an appropriate antigen-presenting cell in sufficient quantity for T cell recognition. Finally, there needs to be in the T cell repertoire sufficient cells able to recognize the complex and be activated by it in a milieu with the right cytokines for clonal expansion and differentiation to effector function. Negative selection in the thymus will remove from the repertoire those T cells that generate antigen-specific receptors able to interact with high affinity to endogenous peptides expressed by cells in the thymic environment. This will remove most T cells with receptors of high affinity for peptides from ubiquitously expressed genes. This process, however, may leave T cells with lower affinity receptors for such peptides, which could subsequently be triggered by expression of high levels of the peptide/MHC complex in the periphery or in the presence of certain cytokines. This may be the basis of some autoimmune reactions and may account for some tumor antigens(21). Genes expressed in a can generate this accounts for some peptides exemplified by those of the genes It may contribute to the generation of some although in both immunogenicity on the cell expressing the gene also expressing molecules of antigen-presenting cells. There are also of transplantation although of these is identified at the level of a cloned gene or a peptide. in peptide sequence due to or differences can in strong even when the difference is limited to a single amino acid the sequence can be loaded into an MHC restriction molecule. This some tumor and several of the minor H peptide 30, 31, However, the two alleles of the H13 peptide are an of although the T cell clones used to are activated by picomolar of the cognate peptide, that of the other allele will at a two This the of in although whether such high levels would be by is thymic selection on peptides may, however, contribute to the presence in the periphery of T cells with receptors of affinity for the peptide. There have been a number of of the number of minor H antigens in mice. From the number of congenic strains a minimum of several would An genetic approach of between and and from of of endogenous proteins, that be an However, of of all proteins does not into account how of these peptides could be loaded into MHC molecules and presented at the cell surface in a that would elicit the of T cells, how other the T cell In in the presence of multiple minor H differences between and as is the case between mouse strains and a very small number of seem to elicit a T cell These few are although the basis of is very and the and minor H antigens to the This could be a of ability of the peptide binding the MHC restriction molecule more This in turn could the number of at the cell surface the interaction the identification of more of the relevant peptides, these can be TO MINOR TRANSPLANTATION ANTIGENS One for of the nature and identity of minor H antigens is the to use it to manipulate immune responses to such epitopes in clinical These include not only in autoimmune and any novel antigens introduced by gene but also of immune responses to antigens expressed on tumor cells. is a to and and from the experiments of and on of this an The of cells to can in a state of tolerance as the recipient the for this in are not for this to be or at least the presence of the seems to of but when it does the is seems clear is that tolerance can in in the of the existence of T cells, which therefore seem to have been However, their presence makes this state a It is much to induce to minor H antigens than their MHC this is apparent from the lower of to grafts from HLA-matched than siblings and the difference in of when and recipient are HLA-matched by descent than being when HLA are tolerance in both and mice is to induce when the genetic differences are limited to minor H antigens, and it when differences are limited to a “single” for example The discovery of which linked recognition of minor H antigens to which tolerance has been an of to with additional minor H antigens on a test the way to use of the defined minor H peptides to tumor antigens is an but may especially during the of the each one as a than an The incorporation of appropriate helper in any may be to The in use of T cells generated in vitro is an It has been for the of the that in using T cells before and may be used to the effect of minor H-specific T cells generated from the bone marrow The use of an source of effector T cells, and with for a tumor peptide presented by an MHC molecule of the has been in both mice and and could provide high affinity T cells, the of clonal deletion and of the T cell The molecular identity of minor H antigens is the is in The information about the peptide and MHC components of these endogenous products is to the and immunology of not only the minor alloantigens in transplantation but also autoantigens and tumor antigens. only a of are identified at the level of peptide and gene, but already there is the of and clinical The physiological roles of the genes encoding the peptides to be their MHC they are likely to be The is to for about the genes, molecules, and immune responses to minor transplantation antigens, and to for in of the