The driving forces that control nearly all immune responses are Ir (immune response) genes encoded within the MHC supergene of the host (1). The peptide specificity of class I and II MHC molecules is important for the selection of relevant peptides. Moreover, the Ir genes regulate both T- and B-cell response to the same antigen, although their specificities are independent and are most frequently directed against structurally different determinants on the same antigen.
The function of MHC molecules is the transfer of information that defines the current stock of proteins within a cell to the cell surface. This enables the immune system to react if necessary by inducing cytotoxic T-lymphocytes to kill transplanted organs and tissues or by activating B cells via a helper T-lymphocyte. Two distinct routes of T-cell allorecognition are considered to be responsible for the host immune response against an allogeneic graft (2). The direct pathway involves T cells that recognize foreign HLA antigens on the surface of donor cells. It has been suggested that direct recognition plays an important role in early acute rejection. However, the indirect pathway, by which shed donor MHC antigens are processed into allopeptides and presented to the host’s CD4+ helper T-lymphocytes by the class II HLA antigens expressed on autologous antigen-presenting cells (APCs), may dominate during late rejections and during the chronic phase of the cellular response (3–5).
The induction of humoral-mediated responses should exclusively use the indirect pathway of allopeptide presentation. The CD4+ helper T-lymphocytes are required to provide the cognate secondary signals that are essential for switching the B-cell response to producing immunoglobulin (Ig) G HLA donor-specific antibodies (6). Moreover, a recent study strongly supports the concept that humoral responsiveness correlates with the HLA-DR phenotype of the responder (7).
In this study, we evaluated distinct HLA-DR alleles to determine possible class II restriction during the production of HLA-specific antibodies against the HLA-A2 group of epitope(s) in renal transplant patients. We can attribute certain positions as unique to HLA-A2, and we can, therefore, associate them with the ability to form specific antibodies. What we cannot prove is whether they operate at the level of induction only (and, hence, the antibody when formed reacts through a conformational specificity) or whether the antibody includes this position as a critical part of its footprint. Because the true answer is likely to lie somewhere between these extremes and may vary according to the particular position, we are using the term “epitope” while fully realizing that in some cases this term may not be strictly correct.
MATERIALS AND METHODS
The study group comprised 217 patients (123 males and 94 females) with a mean age 46±8 years. All the patients had received an HLA-A2–mismatched renal graft (72 live-related donors and 145 cadaveric donors), and none had developed HLA-specific antibodies before the transplantation.
The diagnosis of acute rejection was defined using clinical, biochemical, and histological criteria including, in all cases, a rise in creatinine >15% above baseline and characteristic transplant biopsy features. If the creatinine failed to fall or return to baseline, patients underwent another biopsy to confirm the diagnosis of continuing acute rejection. These individuals were then treated with antilymphocyte agents. Individuals who failed to meet these criteria for the presence or absence of rejection were excluded from the study. Transplant failure was defined as transplant nephrectomy or return to dialysis.
To define the HLA-specific antibodies during the follow-up time (posttransplantation), sera were screened every 3 months. Moreover, sera from each patient were screened 3 weeks after an acute rejection episode, transfusion, and nephrectomy.
All the recipients were typed for class II HLA-DRB1 alleles with high-resolution DNA polymerase chain reaction–sequence-specific primer (8). Only the recipients who had one of the following common eight HLA-DRB1 alleles were included in the study: -*0101, -*0301, -*0401, -*0701, -*1101, -*1301, -*1401, and -*1501. Furthermore, all the recipients were divided into eight subgroups according their examined HLA-DRB1 allele.
To define the HLA class I–specific antibodies posttransplantation, test sera were separated from the clotted blood of the patients and screened with the complement-dependent cytotoxicity (CDC). One microliter of each patient’s serum was aliquoted in duplicate into 60-well Terasaki trays and frozen at −30°C until required. Upon thawing the tray, 1μl of 0.005 M dithiothreitol (8) (Sigma, St. Louis, Mo) was added to one duplicate well for each serum and incubated for 30 min at −37°C. One microliter of fresh or freeze/thawed peripheral blood mononuclear cell/lymphoblastoid cell line at a concentration of 2×106 ml−1 was added to each well and incubated at −22°C for 30 min, and a 5-μl well−1 rabbit complement was added and the trays were incubated at −22°C for 60 min. One microliter of staining cocktail was added to each well, and the cells were assessed for percentage viability within 24 hr with a Leitz Diavert microscope.
In this assay, biotinylated recombinant HLA-A2 monomers were conjugated to streptavidin-coated microtiter plates and used individually in an ELISA technique (monomer-ELISA, “monoLISA”) (9). Serum was added to the bound HLA-A2 monomers, and any anti-HLA IgG would bind to the target. Peroxidase-conjugated anti-human IgG was added and allowed to interact with bound anti-HLA. A peroxidase substrate was added, and this underwent a color change in the presence of the enzyme. This change in color was detected with a microplate spectrophotometer, and the resulting absorbance (OD) values gave a measure of alloantibody binding. The absorbance (OD) was evaluated at 490 nm with a reference wavelength of 630 nm using a Dynatech MRX plate reader. Results were corrected (δA) for nonspecific binding (background) by subtracting the absorbance of the blank well (A no antigen), which contained serum but no antigen, from that of the test well (A test).
Examined HLA-A2 Group of Epitopes
The epitope identification was based on class I HLA-antigen sequencing available at http://www.anthonynolan.com. All the patients of the study were examined for the detection of HLA-specific antibodies against a unique (private) epitope configuration or shared (public) epitope(s) of the HLA-A2 group of antigens. The following epitope(s) were examined: 74H (HLA-A2)1, 65–66GK (HLA-A23 and HLA-24), 62G (HLA-A2-B17), 114H (HLA-A2-A9), 142–145TTKH (HLA-A2-A28), and 107W-127K (HLA-A2-A9-A28).
Analysis: Class II Restriction
Table 1 shows the methodological approach followed for the definition of the HLA-DR permittors based on a class II restriction table in one patient of the study. The definition of the recipient HLA-DR alleles that may respond to an HLA class I–specific epitope was based on a table designed for this purpose. In brief, the analysis included the following steps.
A table was constructed for each HLA class I–specific epitope examined. One column corresponded to each main HLA-DR allele, and one row corresponded to each second HLA-DR allele. The number of patients (who responded to the relevant class I epitope) with each combination of HLA-DR alleles was tallied in the appropriate intersection of main and second alleles. The total number of responders who possessed each main allele was listed at the end of each main column.
The HLA-DR permittors to the examined HLA class I–specific epitope were defined as those main HLA-DR alleles that were more frequent when compared to the other main alleles.
The response was not considered to be due to the second or other alleles if, when tested as main alleles, they were associated with little or no response.
The correlation (P <0.05) between the examined HLA-DR alleles and the development, or nondevelopment, of HLA-specific antibodies against the HLA-A2 group of epitopes was studied. Positive correlation was considered for the cases that had the examined main HLA-DR allele and developed HLA antibodies against the HLA-A2 group. The cases that had the examined main HLA-DR allele and did not develop HLA-specific antibodies were considered to have negative correlation with the examined main HLA-DR allele.
The epitope prediction was based on the MHC database “SYFPEITHI” (the first MHC-eluted peptide that was directly sequenced) available at http://188.8.131.52 (10). The scoring system evaluates every amino acid within a given peptide, taking into consideration the amino acids in the anchor and auxiliary anchor positions as well as the other frequent amino acids. Ideal anchors are given 10 points, unusual anchors are given 6 to 8 points, auxiliary anchors are given 4 to 6 points, and slightly preferred residues are given 1 to 4 points. Amino acids that are regarded as having a negative effect on the binding ability are given values between −1 and −3. All predicted class II MHC ligands are 15 mers and consist of three N-terminal– flanking residues, the nonamer core sequence located within the binding groove, and three C-terminal flanking residues (11). Thus, anchor residue P1 appears in position 4 of the peptides predicted with “SYFPEITHI.”
The chi-square test with Yates correction was used for assessment of the significance in association analysis. P <0.05 was considered statistically significant.
In 217 recipients who received an HLA-A2–mismatched renal graft, the correlation of the examined HLA-DRB1 alleles with the production of HLA-specific antibodies against the HLA-A2 group of epitopes was studied.
The patients were divided into three groups according the examined HLA-DRB1 alleles (Table 2). All the responders to the HLA-A2 group of epitopes and to the other class I HLA mismatches or II experienced transplant failure. In the first group (“responders to the HLA-A2 group of epitopes”), the following patients were included: (1) the patients who produced only HLA-A2 donor–specific antibodies and (2) the patients who developed, after the interruption of the immunosuppression, alloreactivity spreading and a broader reactivity directed to other HLA-A2 group of epitopes. The second group (“responders to the other HLA mismatches”) included the patients who produced HLA donor-specific antibodies against the other class I HLA mismatches or II. Finally, the third group included patients who did not develop HLA donor–specific antibodies.
The CDC defined specificity and the δA values for each of the 60 sera after testing against the HLA-A2 monomers is shown in Table 3. Briefly, 4 of 60 CDC-A2–negative sera were positive with the A2 monomer, while no CDC-A2–positive sera were negative with A2 monomer. The remaining 56 sera were concordant between the two methods (Table 3). Table 4 shows the examined and the new defined shared (public) epitopes between HLA-A2 and other class I HLA antigens, both in HLA-A and -B locus.
In each one of the examined HLA-A2 group of epitopes, the “HLA-DR permittors,” the possible positive and negative correlation with the examined epitope was defined (Table 5). Because almost none of the correlations between the presence of HLA-A2–specific antibodies and the presence of a particular HLA-DR allele were lower than P <0.05, there are no statistically significant correlations in the data; therefore, there is the possibility of the trend toward a correlation. Thus, three HLA-DRB1 alleles (DRB1-*0101, -*1401, and -*1501) had a trend toward a correlation with the development of HLA-specific antibodies against an HLA-A2 group of epitopes. Only the DRB1-*1501 allele had a higher trend toward a positive correlation (P <0.07) with the production of antibodies against the 74H epitope (unique epitope configuration of the HLA-A2). Five of the examined HLA-DRB1 alleles (DRB1-*0101, -*0701, -*1101, -1301, and -*1401) had a trend toward a negative correlation with the development of HLA class I–specific antibodies against the HLA-A2 group of epitope(s) (Table 5).
The HLA-DRB1-*0101 allele, which had a trend toward a positive correlation with the production of antibodies against the epitope 66–65GK, appeared to have a statistically significant (P <0.05) negative correlation with the production of antibodies against the epitope 74H (Table 5). Moreover, the HLA-DRB1-*1401 allele, which had a trend toward a positive correlation with the production of antibodies against the epitope 62G, seemed to have a trend toward a negative correlation with the production of antibodies against the epitope 65–66GK and a statistically significant (P <0.05) negative correlation with the epitope 114H (Table 5).
In 42 patients with the HLA-DRB1-*1501 allele, the non–HLA-A2 mismatches and the produced HLA-specific antibodies are shown in Table 6. Eleven patients (26%) produced HLA class I–specific antibodies, and two patients produced only HLA-A2 donor–specific antibodies. In nine patients, spreading of the alloreactivity against other antigens, which did not belong to the HLA-A2 group of epitope(s) but had new defined shared epitopes with the HLA-A2 group, was detected (Table 6).
The immunogenic and nonimmunogenic HLA-A2 groups of epitopes were predicted according the examined HLA-DRB1 alleles (Table 7). All the selected class II HLA antigens of the patients were included in the HLA panel of the MHC database. The epitope prediction showed that the peptide with a specific amino acid sequence (e.g., position 104–118), which include the examined epitope (e.g., 114H), had a high score (score 31) when it was bound to a specific HLA-DR allele (e.g., DRB1-*0101) and a low score (score 1) when it was bound to another HLA-DR allele (e.g., DRB1-*0301) (Table 7). Moreover, the epitope prediction MHC database confirmed the lack of correlation between the examined HLA-DRB1-*1101 allele and the epitope 142–145TTKH (Table 7).
In clinical transplantation, it appears that not all the patients have the innate capacity to respond to a vascularized allograft, even in a situation of complete HLA incompatibility. For example, rarely do patients who reject an HLA incompatible renal transplant produce HLA-specific antibodies against all incompatibilities (12). Moreover, in 1988, Welsh et al. reported the successful transplantation of kidneys bearing previously mismatched HLA-A and -B locus antigens (13,14). In these patients, satisfactory graft survival rates were achieved provided the patient had not produced antibody to the mismatched antigen or to a cross-reactive group. Furthermore, a recent study shows that it is not always necessary to exclude repeat mismatches for a subsequent transplantation (15). The obvious question is whether the immunogenicity of mismatched donor HLA antigens is affected by the recipients’ HLA type. Such an influence would mean that the same donor mismatches would have different effects in recipients with different HLA types (16). In addition, a recent study shows that a small but significant proportion of people who receive the hepatitis B vaccine do not produce anti-hepatitis B antibodies, a phenomenon associated with certain HLA class II haplotypes (17).
The above evidence strongly supports the fact that there is a genetic basis for control of allograft responses. Furthermore, from the study of Fuller and Fuller, it seems that the humoral response to a specific foreign class I HLA allopeptide (Bw4) could be possibly predicted according to the recipient’s class II HLA phenotype (7). The results of our study support, in part, this concept, because a trend toward a correlation of the class II HLA phenotype of the patients with the production of HLA-specific antibodies against the HLA-A2 group of epitopes has been detected. Thus, the patients, who produced HLA-specific antibodies against the epitopes 74H (HLA-A2), 62G (HLA-A2-B17), and 65–66G (HLA-A9) tended to express HLA-DRB1*1501, -*1401, and -*0101, respectively. The role of the HLA-DQ and -DP alleles was not examined, but, because of their linkage disequilibrium with the HLA-DR alleles, their contribution to observed correlation could not be ruled out.
In this study, the HLA-DRB1-*1501 allele had a trend toward a positive correlation with the production of alloantibodies against the HLA-A2 group. To the best of our knowledge, there is no data on the circumstances in which the presentation of HLA-A2 allopeptides is linked to specific class II HLA alleles of the host. Recently, a case of an HLA-A2–negative patient whose HLA-A2–positive kidney transplant failed as a result of chronic rejection was described. Both recipient and donor had HLA-DRB1-*1502 and DRB1-*1501, respectively (18). It seems that the HLA phenotype of the recipient and the type of bound allopeptide (immunogenic/activating or nonimmunogenic/suppressor peptides) might influence the humoral and cellular response. Thus, some class II (or class I) HLA alleles could interact particularly well with immunogenic (activating) HLA-A2 peptides, which elicit strong cellular and humoral immune response.
In nine patients with the HLA-DRB1-*1501 allele, we observed expansion of the allorecognition to a number of other HLA antigens that did not belong to the HLA-A2 group according the classification of the HLA antigens into cross-reactive epitope groups. Detailed analysis of their linear amino acid sequence revealed that these HLA antigens, against which HLA-specific antibodies were detected, had shared epitopes with the HLA-A2 group and a significant number of them belonged to the HLA-B locus. For example, until now, the only B-locus product known to share an epitope with HLA-A2 was HLA-B17. The above observation could be of great importance because it links the HLA-A and -B locus and could lead to development of new strategies for organ allocation using epitope analysis to define acceptable mismatches in highly sensitized patients (19). Moreover, this observation could lead to a revision of the cross-reactive epitope group clusters, giving a more detailed list of the HLA antigens with shared epitopes.
Several possibilities could explain the hyporesponsiveness and the induction of T-cell anergy and/or tolerance in 31 recipients with HLA-DRB1-*1501 who did not develop HLA-specific antibodies against the HLA-A2 group of epitopes. (1) Donor-derived nonimmunogenic (suppressor) HLA-A2 peptides might be structurally closely related to “self” (recipients) peptides (“mimicry of the “self” peptides phenomenon) and competing for T-cell receptor recognition (20). (2) The competition among immunogenic (activating) and nonimmunogenic HLA-A2 peptides for binding to class II molecules might be a significant factor for the kind of peptides that were bound to them (21–23). Moreover, in vivo competition between different peptides derived by processing of the same protein has been shown to profoundly influence the immunodominance of T-cell determinants (24). (3) The response or nonresponse against the HLA-A2 group of epitopes in the patients with the HLA-DRB1-*1501 allele could depend on the ligand density displayed by the activated T cell. Thus, distinct anergic phenotypes can be induced in the responding T cells upon subsequent interaction with professional APCs that present the same peptide. These can range from the absence of T-cell anergy (i.e., T-cell activation) to an anergic phenotype to a suppressive anergic phenotype that can be persistently present (25). (4) The class II HLA molecules are capable of signal transduction that leads to the generation of second messengers, homotypic aggregation, and either cellular activation or programmed cell death (26–28). Moreover, a role for class II HLA signaling in increasing sensitivity to apoptosis of the activated B-lymphocytes, via the Fas molecule, has also been recently described (29). (5) Finally, several lines of evidence suggest that antigen presentation by tissue parenchymal cells, such as renal tubular epithelial cells, induce T-cell nonresponsiveness (18).
In the cases that had HLA-DRB1*0101, -*0701, -*1101, -1301, or -*1401 alleles, HLA-specific antibodies against some of the HLA-A2 group of epitopes were not detected. It seems that these HLA molecules could not bind effectively and present peptides with epitopes 62G, 65–66GK, 107W, 114H, and 127K. Furthermore, the HLA-DRB1-*0101 and -*1401 alleles, which had a trend toward a positive correlation with the production of antibodies against the epitopes 65–66GK and –62G, respectively, also had statistical significant negative correlation with the development of antibodies against the epitopes 74H and 114H, respectively. This could be due to determinant selection in which HLA molecules of the recipient bind some of the produced allopeptides while it ignores others (30). On the other hand, the type of determinants on donor HLA antigens (allodeterminants) could influence the “determinant selection.” Thus, some allodeterminants (dominant allodeterminants) are efficiently processed, bound, and presented to alloreactive T cells. Other allodeterminants (cryptic allodeterminants) do not normally reach the threshold of presentation required to induce alloreactive response, presumably because of incomplete processing, binding, and/or presentation (31). Moreover, the set of peptides for display also depends on other factors, including the enzymatic machinery of the APC, the route of antigen processing, and the presence of protein cofactors (31). Furthermore, the type of the cells, which present the peptide, seem to be important for the activation of the immune response. Thus, peptide presentation by resting B cells is reported to induce unresponsiveness or tolerance (32,33), whereas peptide presentation by activating B cells induces T-cell activation with preferential stimulation of TH2 T-cell response (34,35). Finally, the ability of different epitope HLA-DR complexes to trigger T-cell receptor for antigen (TCR) responses via receptor antagonism could explain the fact that HLA-DRB1-*0101 and -*1401 alleles, which had a trend toward a positive correlation with the production of antibodies against the epitopes 65–66GK and –62G, respectively, also had statistically significant negative correlation with the development of antibodies against the epitopes 74H and 114H, respectively. Agonists are capable of binding to a receptor and triggering a response, whereas antagonists bind to a receptor but are incapable of triggering a response. Peptides that bind class II HLA molecules can be used as agonists or antagonists. Once a peptide-MHC complex is formed, this complex then acts as agonist or antagonist for the TCR (36).
It would be of great importance to predict the immunogenic and nonimmunogenic HLA-A2 epitopes and confirm them by cell culture experiments with synthetic peptides analogues, thus gaining support for their clinical use. For example, we predicted that peptides corresponding to the residues 58–69 (HLA-A2-B17) are nonimmunogenic (suppressor) peptides. This prediction is in accordance with a previous study that showed, in cell cultures, that synthetic peptides corresponding to the same linear sequence of the HLA-A2 could specifically inhibit cytotoxic T-lymphocyte recognition of the HLA-A2 alloantigen (37). Instead of synthesizing and testing dozens or even hundreds of peptides, which will include shared epitopes between HLA-A2 and other HLA antigens, a preselection of a small set of peptides could be made. Thereafter, the confirmed nonimmunogenic synthetic peptide analogues could be used for the induction of anergy and/or tolerance in transplanted patients (38).
We believe, based on these results, that the class II HLA phenotype of the recipient defines, at least in part, the immunogenicity of mismatched donor HLA antigens and that this is further influenced by the class I HLA phenotype of the patients. The ultimate goal of our approach would be the accurate ability to predict peptide binding to MHC molecules to develop the tools to perform a rational identification of immunogenic epitopes. This would eventually require that other specific mechanisms involved in antigen processing and presentation (e.g., proteasomes, the transporter associated with antigen presentation, etc.) be considered in addition to the specificity of the MHC complex (30). Additional sections are already available from serological studies going back decades. For example, the ability of HLA-A68 individuals to respond specifically to HLA-A2, while the reverse is never observed, is easily explained in terms of shared or nonshared inducer positions.
In conclusion, these studies, which are in progress, if confirmed and found to be generally applicable to the formation of HLA-specific antibodies against the diverse array of class I HLA polymorphism, could have far reaching clinical significance in the field of transplantation. Identification of the immunodominant class I allopeptides recognized by CD4+ T-lymphocytes may offer the possibility, together with factors already defined, to predict risks for humoral and cellular responses to HLA incompatibility. Production of synthetic peptide analogues, which do not stimulate the T-cell response and are able to block the recognition of the dominant determinant, may provide a tool for selective immune intervention against CD4+ T-lymphocytes that participate in indirect allorecognition (39,40). These peptide analogues may be useful in the future for induction of T-cell anergy and/or tolerance. In addition, new strategies for organ allocation with epitope matching and acceptable mismatches would be applied.
We wish to thank Dr. Sara Marshall for kind comments and helpful and stimulating discussions during the study.
1. Butcher G, Corvalán JR, Licence DR, Howard JC. Immune response genes controlling responsiveness to major transplantation antigens: specific major histocompatibility complex-linked defect for antibody responses to class I alloantigens. J Exp Med 1982; 155: 303.
2. Auchincloss HJ, Sultan H. Antigen processing and presentation in transplantation. Curr Opin Immunol 1996; 8: 681.
3. Benichou G, Takizawa PA, Olson CA, McMillan M, Sercarz EE. Donor major histocompatibility complex (MHC) peptides are presented by recipient MHC molecules during graft rejection. J Exp Med 1992; 175: 305.
4. Watschinger B, Gallon L, Carpenter CB, Sayegh MH. Mechanisms of allo-recognition: recognition of in vivo-primed T cells of specific major histocompatibility complex polymorphisms presented as peptides by responder APCs. Transplantation 1994; 57: 572.
5. Liu Z, Colovai AI, Tugulea S, et al. Indirect recognition of donor HLA-DR peptides in organ allograft rejection. J Clin Invest 1996; 98: 1150.
6. Steele DJR, Laufer TM, Smiley ST, et al. Two levels of help for B cell alloantibody production. J Exp Med 1996; 183: 699.
7. Fuller TC, Fuller A. The humoral immune response against an HLA class I allodeterminant correlates HLA-DR phenotype of the responder. Transplantation 1999; 68: 173.
8. Bunce M, O’Neill CM, Barnardo NCNM, et al. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 and DQB1 by PCR with 144 primer mixes utilising sequence-specific primers (PCR-SSP). Tissue Antigens 1995; 46: 355.
9. Barnardo MCNM, Harmer AW, Shaw OJ, et al. Detection of HLA-specific IgG antibodies using single recombinant HLA alleles: the monoLISA assay. Transplantation 2000; 70: 531.
10. Rammensee HG, Bachmann J, Emmerich NPN, Bachor OA, Stevanovic S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 1999; 50: 213.
11. Chicz RM, Urban RG, Gorga JC, Vignali AAD, Lane WS, Strominger JL. Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J Exp Med 1990; 178: 27.
12. Sanfilippo F, Vaughn WK, Bollinger RR, Spees EK. Comparative effects of pregnancy, transfusion, prior graft rejection on sensitisation and renal transplant results. Transplantation 1982; 34: 360.
13. Welsh KI, Van Dam M, Bewick ME, et al. Successful transplantation of kidneys bearing previously mismatched HLA-A and B locus antigens. Transpl Int 1988; 1: 190.
14. Doxiadis IIN, De Lange P, D’Amaro J, De Meester J, Schreuder GMT, Claas FHJ. Repeated HLA mismatches in cadaveric renal transplantation is safe to transplant? Transplant Proc 1997; 29: 1408.
15. Van Kampen CA, Versteeg-van der Voort Maarschalk MFJ, Roelen DL, Ten Berge IJM, Claas FHJ. Rejection of a kidney transplant does not always lead to priming of cytotoxic T cells against mismatched donor HLA class I antigens. Transplantation 2001; 71: 869.
16. Doxiadis IIN, Smits JMA, Schreuder GMT et al. Association between specific HLA combinations and probability of kidney allograft loss: the taboo concept. The Lancet 1996; 348: 850.
17. Schuenke KW, Cook RG, Rich RR. Binding specificity of a class II-restricted hepatitis B epitope by DR molecules from responder and nonresponder vaccine recipients. Human Immunol 1998; 59: 783.
18. Frasca L, Amendola A, Hornick P et al. Role of donor and recipient antigen-presenting cells and maintaining T cells with indirect allospecificity. Transplantation 1998; 66: 1238.
19. Papassavas AC, Iniotaki-Theodoraki A, Boletis J, Kostakis A, Stavropoulos-Giokas C. Epitope analysis of HLA class I donor specific antibodies in sensitised renal transplant recipients. Transplantation 2000; 70: 323.
20. Krensky AM, Clayberger C. The induction of tolerance to alloantigens using HLA-Based synthetic peptides. Curr Opin Immunol 1994; 6: 791.
21. Babbitt BP, Matsueda G, Haber E, Unanue ER, Allen PM. Antigenic competition at the level of peptide-Ia binding. Proc Natl Acad Sci USA 1986; 83: 4509.
22. Wraith DC, Smilek DE, Mitchell DJ, Steinman L, McDevit HO. Antigen recognition in autoimmune encephalomyelitis and the potential for peptide mediated immunotherapy. Cell 1989; 59: 247.
23. Gautam AM, Pearson CI, Smilek DE, Steinman L, McDevitt HO. A polyalanine peptide with only five native myelin basic protein residues induces autoimmune encephalomyelitis. J Exp Med 1992; 176: 605.
24. Adorini L, Appella E, Doria G, Nagy ZA. Mechanisms influencing the immunodominance of T cell determinants. J Exp Med 1988; 168: 2091
25. Taams LS, Wauben MHM. Anergic T cells as active regulators of the immune response. Hum Immunol 2000; 61: 633.
26. Lane PJL, McConnell FM, Schieven GL, Clark EA, Ledbetter JA. The role of class II molecules in human B cell activation. J Immunol 1990; 144: 3684.
27. Fuleihan R, Mourad W, Geha RS, Chatila T. Engagement of MHC-class II molecules by staphylococcal exotoxins delivers a comedogenic signal to human B cells. J Immunol 1991; 146: 1661.
28. Mourad W, Geha RS, Chatila T. Engagement of major histocompatibility complex class II molecules induces sustained lymphocyte function associated molecule I dependent cell adhesion. J Exp Med 1990; 172: 1513.
29. Yoshino T, Cao L, Nishiuchi R, et al. Ligation of HLA class II molecules promotes sensitivity to CD95 (Fas antigen, APO-1)-mediated apoptosis. Eur J Immunol 1995; 25: 2190.
30. Buus S. Description and prediction of peptide-MHC binding: the “human MHC project.” Curr Opin Immunol 1999; 11: 209.
31. Benichou G, Tam RC, Soares LRB, Fedoseyeva EV. Indirect T-cell allorecognition: perspectives for peptide-based therapy in transplantation. Immunol Today 1997; 18: 67.
32. Eynon EE, Parker DC. Small B cells as antigen presenting cells in the induction of tolerance to soluble protein antigens. J Exp Med 1992; 175: 131.
33. Eynon EE, Parker DC. Parameters of tolerance induction by antigen targeted to B lymphocytes. J Immunol 1993; 151: 2958.
34. Svetic A, Finkelman FD, Jian YC, et al. Cytokine gene expression after in vivo primary immunization with goat antibody to mouse IgD antibody. J Immunol 1991; 147: 2391.
35. Saudi A, Simmonds S, Huitinga I, Mason D. Prevention of experimental allergic encephalomyelitis in rats by targeting autoantigen to B cells: evidence that the protective mechanism depends on changes in the cytokine response and migratory properties of the autoantigen-specific T cells. J Exp Med 1995; 58: 335.
36. De Magistris MT, Alexander J, Coggeshall M, et al. Antigen analog-major histocompatibility complexes act as antagonists of the T cell receptor. Cell 1992; 68: 625.
37. Krensky AM, Clayberger C. The induction of tolerance to alloantigens using HLA-based synthetic peptides. Curr Opin Immunol 1994; 6: 791.
38. Clayberger C, Parham P, Rothbard J, Ludwig DS, Schoolnik GK, Krensky AM. HLA-A2 peptides regulate cytolysis by human allogeneic T lymphocytes. Nature 1987; 330: 763.
39. McDonald CM, Bolton EM, Jaques BC, Walker KG, Bradley JA. Reduction of alloantibody response to class I major histocompatibility complex by targeting synthetic allopeptides for presentation by B cells. Transplantation 1997; 63: 926.
40. Colovai AI, Liu Z, Harris PE, Cortesini R, Suciu-Foca N. Allopeptide-specific T cell reactivity altered by peptide analogs. J Immunol 1997; 158: 48.