Immunological rejection remains the leading cause of corneal graft failure (1) (2) * (3, 4)
It has been recognized for over 80 years that oral administration of antigens is an effective method for desensitizing hosts previously immunized with various antigens (5) (6) (7) (8) (9) (10, 11)
Although oral immunization results in specific immunologic unresponsiveness in many autoimmune diseases, there is limited evidence of its effect in promoting allograft survival. Oral administration has been shown to suppress the generation of DTH responses to class II major histocompatibility complex (MHC) antigens in rats (12-14) (15)
MATERIALS AND METHODS
Animals . Female NZB (H-2d ), C3H (H-2k ), and CB6F1 (H-2b/d ) mice were purchased from The Jackson Laboratories (Bar Harbor, ME) and used between the ages of 2 and 8 months.
Alloantigen preparation . Tissue cultured murine corneal epithelial and endothelial cells were used as alloantigens for the induction of oral tolerance. Cell cultures were established from freshly dissected corneal explants (16, 17) (18)
Keratinocyte suspensions were prepared from abdominal and back skin of euthanized C3H and NZB mice. Excised skin was cut into strips, placed dermal side up in a Petri plate, and excess fat and blood vessels were removed by scraping with a sterile scalpel. Skin strips were incubated in 0.25% trypsin (Sigma Chemical Co., St. Louis, MO), 0.5 mg/ml hyaluronidase (Worthington Biochemical Co., Freehold, NJ), and 2.0 mg/ml collagenase (GIBCO BRL, Grand Island, NY) in phosphate-buffered saline (pH = 7.2) overnight at 4°C. Epithelial sheets were scraped free of underlying dermis and were suspended in Hanks' balanced salt solution (HBSS). Cells were disaggregated by vigorous vortexing, washed twice in HBSS, and resuspended in HBSS at a concentration of 1 × 107 cells/ml for oral administration.
Oral tolerance induction . Cultured murine corneal cells and freshly isolated keratinocytes were used for inducing oral tolerance. For each immunization, either 1 × 107 freshly isolated keratinocytes or a mixture of 5 × 106 epithelial cells plus 5 × 106 endothelial cells was given directly into the stomach using a gavage tube. The hosts were fed daily for 8 to 12 days. These numbers were selected based on our preliminary experiments indicating that feeding less than 8 oral doses was not effective in inducing tolerance.
Orthotopic corneal transplants . Full-thickness penetrating NZB (H2d ) or C3H (H2k ) corneal grafts (2.5-mm diameter) were transplanted orthotopically onto anesthetized CB6F1 (H2b/d ) mice using a procedure previously described by She et al. (19) (20)
Clinical observations . Grafted eyes were examined with a slitlamp biomicroscope at least twice a week throughout the entire study period. Graft opacity, edema, and neovascularization were scored as minimal, moderate, or severe, as previously described (20)
Langerhans cell induction . The central corneal epithelium of the mouse is normally devoid of resident Ia+ Langerhans cells (LC), which are usually situated in the epithelium of the limbus (21) (21) (22)
Heterotopic corneal transplantation . To determine the effect of oral immunization in preventing corneal allograft rejection in preimmune, high-risk hosts, full-thickness C3H or NZB corneal grafts along with the peripheral limbus (total 3.0-3.5-mm diameter) were transplanted heterotopically onto vascularized subdermal graft beds on the lateral thorax of CB6F1 mice as previously described (23) (24)
Cell-mediated cytotoxicity assay . Selected CB6F1 recipient mice were killed 60 days after transplantation. Single-cell suspensions were prepared from spleens removed from individual animals and were used as effector cells in conventional in vitro cytotoxic T lymphocyte (CTL) assays as previously described (21) 51 Chromium-labeled NZB Con A blasts (1 × 104 cells) were added to the various wells to produce effector to target ratios ranging from 100:1 to 12.5:1. Plates were centrifuged at 100 × g for 2 min and incubated at 37°C for 4 hr in a humidified 5% CO2 atmosphere. Plates were then centrifuged, 100 μl of each supernatant was collected, and the counts per minute were determined by counting the specimens in a gamma counter (Tracor Analytical, Atlanta, GA). Cytotoxicity was calculated according to the formula : Delayed-type hypersensitivity assay . Delayed-type hypersensitivity (DTH) responses to alloantigens were measured by a conventional footpad swelling assay (21) 1 mice were immunized by applying two heterotopic NZB allografts to the lateral thorax of each mouse as described earlier (23) 1 mice that were grafted heterotopically but not fed alloantigens. Normal age-matched CB6F1 mice served as negative controls. In other experiments, the DTH responses were assessed in orally immunized hosts that either bore long term clear orthotopic corneal grafts or orally immunized mice that had previously rejected orthotopic corneal allografts. In these groups, DTH responses were assessed 60 days after orthotopic corneal transplantation. Both hind footpads of each mouse were measured with an engineer's micrometer (Mitutoyo Corp., Tokyo, Japan) immediately before footpad challenge. An eliciting dose of 1 × 107 gamma-irradiated (3000 cGy) NZB splenocytes suspended in 25 μl of Hanks' balanced salt solution (HBSS) was injected into the subcutaneous tissue of the right hind footpad. The left hind footpad served as a background control and received 25 μl of HBSS without splenocytes. Both footpads were measured 24 hr later and the difference in footpad swelling size was used as a measure of DTH. Results were expressed as specific footpad swelling, which equals: ([24 hr rt hind foot measurement - 0 hr rt hind foot measurement] - [24 hr left hind foot measurement - 0 hr left hind foot measurement])×10-4 ± SD in inches. Student's t test was used to evaluate the statistical significance of the results.
RESULTS
Effect of oral immunization on systemic allommune responses . The capacity of oral immunization to desensitize cell-mediated alloimmune responses of preimmunized hosts was evaluated. CB6F1 mice were immunized with heterotopic NZB corneal allografts (2 grafts/host). Two weeks later, mice received 10 daily doses of NZB corneal cell suspensions. DTH and CTL responses were evaluated 24 hr after the tenth oral dose of corneal cells. As expected, heterotopic corneal grafts induced potent CTL and DTH responses to donor alloantigens (Fig. 1) . By contrast, mice immunized orally with allogeneic corneal cells failed to develop DTH responses that were any greater than negative controls (Fig. 1) . Although oral immunization produced a precipitous reduction in the CTL responses of preimmune hosts, considerable cell-mediated cytotoxicity remained and was significantly greater than normal control mice (Fig. 1) .
Effect of orally induced tolerance on the incidence and tempo of rejection for fully allogeneic corneal allografts in naive hosts . C3H corneal allografts transplanted to CB6F1 hosts represents a complete mismatch at both major histocompatibility complex (MHC) loci and multiple minor histocompatibility (H) loci. In this combination, the incidence of corneal graft rejection in untreated naive hosts was 100%, with an MST of approximately 18 days (Fig. 2) . Oral immunization with cultured corneal cells reduced the incidence of rejection and prolonged the survival time of corneal allografts compared with untreated controls (Fig. 2) . Although oral immunization failed to prevent the ultimate rejection of orthotopic C3H corneal allografts in preimmune CB6F1 hosts, it did produce a modest prolongation in the MST of the orthotopic grafts (MST = 15 to MST = 20 days).
Survival of MHC-matched multiple minor H-mismatched corneal allografts in naive and presensitized hosts subjected to orally induced tolerance . Studies in rat and mouse models of orthotopic corneal transplantation have demonstrated that the rejection rate for MHC-matched, multiple minor H-mismatched corneal grafts in immunologically naive hosts is approximately 25-50% (25-28) 1 hosts represents allodisparity at multiple minor H loci but identity at the MHC. In this donor-host combination, the incidence of orthotopic corneal graft rejection was 29% (MST = 43.5 days) in naive hosts when LC-free, central corneal grafts were used, and thus was in the general range reported in other rodent models of corneal graft rejection across multiple minor H barriers (25-28) + (i.e., latex bead-treated) corneal grafts were used (Fig. 3) . The presence of donor-specific LC also reduced the MST of the grafts to approximately 18 days. Thus, the presence of donor-specific LC greatly increases the immunogenicity of MHC-matched, minor H-mismatched corneal allografts. Accordingly, the efficacy of oral immunization in promoting the acceptance of highly immunogenic (i.e., latex bead-treated) minor H-mismatched corneal grafts was examined and revealed that oral immunization prior to the transplantation of LC+ NZB corneal allografts significantly reduced the incidence of rejection and prolonged graft survival (Fig. 3) . Oral immunization also extended the survival of NZB corneal grafts in preimmunized CB6F1 hosts.
Antigen specificity of orally induced tolerance . The antigen specificity of orally induced tolerance was examined by feeding prospective hosts with alloantigens from one MHC haplotype and challenging with corneal allografts from an unrelated MHC haplotype. Accordingly, CB6F1 mice were orally immunized with cultured C3H corneal cells prior to receiving LC+ orthotopic NZB corneal allografts. As shown in Figure 4 , there was no difference in either the incidence or tempo of NZB corneal allograft rejection in unfed CB6F1 mice and CB6F1 mice fed C3H corneal cells. By contrast, only 30% of the LC+ NZB corneal allografts underwent rejection in CB6F1 mice fed NZB cells.
Adoptive transfer of orally induced tolerance . Previous studies have shown that oral immunization induced antigenspecific suppressor cells in gut lymphoid tissues and the spleen (5) 1 mice fed 10 doses of NZB corneal cells. Lymphoid cells were pooled and immediately injected intravenously into normal CB6F1 mice (5 × 107 cells per mouse). One to two days later, the lymphoid cell recipients were grafted orthotopically with LC+ NZB corneal grafts. Corneal allografts remained clear in 56% of the recipients of regulatory cells from orally immunized donors while only 2 of 10 grafts survived in the untreated hosts (Fig. 5) . Moreover, the survival time was significantly prolonged in lymphoid cell recipients compared with untreated mice (MST = 28 days compared with MST = 18 days).
Induction of oral tolerance with freshly isolated keratinocytes . Although orally induced tolerance was remarkably effective in desensitizing preimmune hosts and in promoting corneal graft survival, the use of cultured corneal cells in keratoplasty patients is not feasible. However, organ banks often have significant quantities of skin from cornea donors which can be cultured relatively easily. Therefore, the capacity of freshly isolated keratinocytes to induce oral tolerance was examined. CB6F1 mice were presensitized with heterotopic NZB corneal allografts. Fourteen days later, the heterotopically grafted mice received 10 daily doses of NZB keratinocytes. On the eleventh day, the CB6F1 mice were challenged with orthotopic NZB corneal allografts. As in previous experiments, over 90% (13/14) of the grafts were rejected by preimmune hosts (MST = 17 days) (Fig. 6) . By contrast, only 42% (5/12) of the hosts that were fed keratinocytes rejected their grafts (MST = 19 days). This reduction of graft rejection is comparable to that produced by feeding cultured corneal cells (i.e., 36% incidence of rejection; P = 0.4; Mann-Whitney U test). Thus, keratinocytes, like corneal cells, are capable of inducing oral tolerance within a relatively short period of time and promoting graft acceptance in a “high-risk” setting.
Correlation between graft survival and orally induced suppression of DTH and CTL alloimmune responses . Although oral administration of alloantigenic cells dramatically promoted corneal allograft survival, a significant number of grafts underwent rejection. The failure to promote graft acceptance in all of the orally immunized hosts was presumably due to the emergence of cell-mediated immune responses to donor alloantigens in those hosts that went on to reject their grafts. This hypothesis was tested by comparing the CTL and DTH responses in orally immunized hosts that rejected their corneal allografts with the responses of orally tolerized hosts bearing long-term corneal grafts. Orally immunized hosts were categorized as “rejecters” or “acceptors” according to the fate of their orthotopic corneal grafts at day 60 post-transplantation. DTH and CTL responses to donor (NZB) alloantigens were assessed in orally immunized mice, as well as normal CB6F1 control mice. Interestingly, orally immunized hosts that rejected their orthotopic corneal allografts expressed significant cell-mediated cytotoxicity as well as DTH responses (Fig. 7) . By contrast, hosts bearing clear corneal grafts (“acceptors”) did not develop DTH and CTL responses that were significantly greater than naive controls.
DISCUSSION
Although corneal allografts enjoy a remarkable success rate, approximately 10% of the 40,000 keratoplasties performed each year in the United States will fail due to immunological rejection (29) (29)
The results demonstrate that oral administration of alloantigens produced a significant enhancement of corneal allograft survival, even with corneal allografts representing maximal allodisparity (i.e., MHC plus minor H-mismatched grafts). Although 100% of the fully allogeneic corneal grafts were rejected in immunologically naive hosts, oral administration of alloantigens reduced the rejection rate by almost 50%. In a preliminary report, Feder et al. (30)
Oral antigen presentation also proved beneficial in promoting the acceptance of LC+ MHC-matched, multiple minor H-disparate corneal allografts. In naive hosts, 80% of the LC+ minor H disparate grafts underwent rejection; however, in orally tolerized hosts, the rejection rate fell to 30%. The inability of oral immunization to prevent corneal graft rejection in all of the hosts could be attributed to the presence of the E6/E7 oncoproteins that are known to be expressed in corneal endothelial cells transformed with E6/E7 oncogenes (18) (31)
The present findings also reveal interesting nuances about the characteristics and limitations of orally induced tolerance. Oral administration of alloantigens prevented the induction of alloimmunity and rejection of fully allogeneic corneal allografts in naive hosts but failed to prevent rejection in preimmune hosts. By contrast, antigen feeding was effective in desensitizing hosts previously immunized with minor H alloantigens.
Orally induced tolerance was also effective in preventing the induction of alloimmunity by Ia+ passenger cells (i.e., Langerhans cells). The central cornea is normally devoid of Ia+ LC, and grafts prepared from the central cornea and transplanted across multiple minor H barriers are rejected in only 20% of the normal hosts. However, instillation of sterile latex beads into the donor's central cornea induces a rapid centripetal migration of peripheral LC (21) + corneas are highly immunogenic and are rejected in approximately 80% of the naive recipients. However, the present results revealed that oral administration of donor alloantigens reduced the rejection rate from 80% down to approximately 30%. Thus, orally induced tolerance is an effective method for preempting the “passenger cell” effect.
Murine models of orthotopic corneal transplantation allow prospective studies for evaluating the clinical feasibility of novel strategies for promoting graft acceptance. With this in mind, we examined the capacity of keratinocytes to induce oral tolerance since organ banks routinely obtain significant amounts of skin from the same donors that provide corneal buttons used for keratoplasty. Moreover, human keratinocytes can be maintained in vitro and provide quantities of cells comparable to those used to promote corneal allograft survival in the present study. It is noteworthy that oral tolerance was successfully induced with 10 daily doses of antigen. Corneal buttons retain their viability and can be successfully transplanted after being held in storage medium for up to two weeks. Thus, oral tolerance can be induced within a realistic time frame relative to the limitations of corneal graft storage.
The facility with which oral tolerance can be induced against minor H antigens in the preimmune host has important clinical implications. Since corneal grafts can be HLA matched with the prospective “high-risk” recipient, it is feasible to transplant corneal allografts that represent only minor H barriers and thereby maximize the therapeutic benefit of orally induced tolerance.
In a broader sense, these results clearly demonstrate that oral administration of alloantigens can desensitize previously immunized hosts. Both CTL and DTH alloimmune responses are profoundly depressed as a result of antigen feeding. Although the success rate is high, a significant number of grafts fail in hosts subjected to oral antigen feeding. In these cases, graft failure consistently correlates with the emergence of allospecific CTL and DTH responses.
Orally administered antigens can induce tolerance by active suppression, clonal anergy, or clonal deletion (5, 27, 28, 29, 32)
Figure 1: Effect of oral immunization on cell-mediated alloimmune responses. CB6F1 mice were immunized with heterotopic NZB LC+ corneal allografts (2 grafts/mouse). Mice received 10 daily doses of NZB corneal cell suspensions commencing two weeks after heterotopic grafting. DTH (panel A) and CTL (panel B) responses against NZB spleen cells were assessed one day after the final oral immunization. There were five mice in each group. Bar represents mean ± SEM.
Figure 2: Survival of fully allogeneic (MHC + minor H disparities) C3H orthotopic corneal allografts on orally immunized CB6F1 hosts. Naive or preimmunized CB6F1 mice were orally immunized with 10 daily doses of C3H corneal cells prior to receiving orthotopic C3H corneal allografts. Naive (n = 7); naive + oral (n = 9); preimmune (n = 10); preimmune + oral (n = 9).
Figure 3: Survival of LC+ orthotopic multiple minor H-mismatched corneal allografts on naive hosts and LC- corneal allografts on preimmune hosts. Naive CB6F1 mice were orally immunized with 10 daily doses of NZB corneal cells prior to receiving LC+ NZB corneal allografts. In other experiments, CB6F1 mice were immunized with heterotopic NZB LC+ corneal allografts (2 grafts/mouse) 2 weeks prior to oral immunization. All mice were challenged with orthotopic NZB corneal allografts 1-2 days after final oral immunization. Allogeneic (LC+ )(n = 10); Allogeneic (LC+ ) + oral (n = 10); preimmune (n = 10); preimmune + oral (n = 11)
Figure 4: Antigen specificity of oral tolerance. CB6F1 mice were orally immunized with 10 daily doses of either NZB or C3H corneal cells. Mice were challenged with orthotopic NZB corneal allografts 1-2 days after final oral immunization. NZB-> CB6F1 (n = 10); NZB-> CB6F1 + oral NZB (n = 10); NZB-> CB6F1 + oral C3H (n = 13)
Figure 5: Adoptive transfer of oral tolerance. CB6F1 mice were orally immunized with 10 daily doses of NZB corneal cells. Spleen cells were collected 1-2 days after the final oral immunization and transferred intravenously (5 × 107 cells/recipient). One to two days later, the spleen cell recipients were challenged with orthotopic LC+ NZB corneal allografts. Allogeneic (LC+ ) = normal CB6F1 mice challenged with LC+ NZB graft (n = 10); Allogeneic (LC+ ) + oral = CB6F1 mice immunized orally and challenged with LC+ NZB corneal graft (n = 10); Adoptively transferred + (LC+ ) = CB6F1 mice that received spleen cells and were challenged with LC+ NZB corneal graft (n = 9).
Figure 6: Induction of oral tolerance with freshly isolated keratinocytes. CB6F1 mice were presensitized with heterotopic NZB corneal allografts. Fourteen days later mice received 10 daily doses of NZB keratinocytes. On the eleventh day, the orally immunized, presensitized mice (n = 12) and the untreated, presensitized mice (n = 14) were challenged with orthotopic NZB corneal allografts. Bar represents mean ± SEM.
Figure 7: Cell-mediated alloimmune responses in CB6F1 mice immunized orally with NZB corneal cells and challenged with LC+ NZB orthotopic corneal grafts. CTL and DTH responses were assessed at day 60. Mice were categorized as “rejectors” or “acceptors” based on the status of the corneal grafts at day 60. Survivors (n = 6); Rejectors (n = 6); Negative (n = 5); Positive (n = 5).
Footnotes
Supported in part by NIH Grant EY07641 and an unrestricted grant from Research to Prevent Blindness, Inc., New York.
Abbreviations: CTL, cytotoxic T lymphocyte; CsA, cyclosporine; DTH, delayed-type hypersensitivity; H, histocompatibility; LC, Langerhans cells; MHC, major histocompatibility complex; MST, mean survival time.
REFERENCES
1. Shapiro MB, Mandel MR, Krachmer JH. Adult penetrating keratoplasty. In: Brightbill FL, ed. Rejection in corneal surgery theory, technique and tissue. St. Louis: Mosby, 198: 310.
2. Lindstrom RL. Advances in corneal transplantation. N Engl J Med 1986; 315: 1.
3. Belin MW, Bouchard CS, Philips TM. Update on topical cyclosporin A: background, immunology and pharmacology. Cornea 1990; 9: 184.
4. Dupon E, Wybran J, Tonssaint C. Glucocorticosteroids and organ transplantation. Transplantation 1984; 37: 331.
5. Friedman A, Al-Sabbagh A, Santos LMB, et al. Oral tolerance: a biologically relevant pathway to generate peripheral tolerance against external and self antigens. In: Granstein RD, ed. Mechanisms of immune regulation. Basel: Karger, 1994: 259.
6. Higgins PJ, Weiner HL. Suppression of experimental autoimmune encephalomyelitis by oral administration of myelin basic protein and its fragments. J Immunol 1988; 140: 440.
7. Nussenblatt RB, Caspi RR, Mahdi R, et al. Inhibition of S-antigen induced experimental autoimmune uveoretinitis by oral induction of tolerance with S-antigen. J Immunol 1989; 144: 1689.
8. Zhang Z, Lee C, Weiner H. Suppression of adjuvant arthritis by oral administration of type II collagen. J Immunol 1980; 145: 2489.
9. Zhang Z, Davidson L, Elisenbarth G, Weiner H. Suppression of diabetes in nonobese diabetic mice by oral administration of porcine insulin. Proc Natl Acad Sci USA 1991; 88: 10252.
10. Weiner HL, Mackin GA, Matsui M, et al. Double-blind pilot trial of oral tolerization with myelin antigens in multiple sclerosis. Science 1993; 259: 1321.
11. Trentham DE, Dynesius-Trentham RA, Orav EJ, et al. Effects of oral administration of type II collagen on rheumatoid arthritis. Science 1993; 261: 1727.
12. Sayegh MH, Khoury SJ, Hancock WW, Weiner HL, Carpenter CB. Induction of immunity and oral tolerance with polymorphic class II major histocompatibility complex allopeptides in the rat. Proc Nat Acad Sci USA 1992; 89: 7762.
13. Sayegh MH, Zhang ZJ, Hancock WW, Kwok CA, Carpenter CB, Weiner HL. Orally administered alloantigen down-regulates the immune response to histocompatibility antigens and prevents sensitization by skin allografts. Transplantation 1992; 53: 163.
14. Hancock WW, Khoury SJ, Carpenter CB, Sayegh MH. Differential effects of oral versus intrathymic administration of polymorphic MHC class II peptides on mononuclear and endothelial cell activation and cytokine expression during a delayed-type hypersensitivity response. Am J Pathol 1994; 144: 1149.
15. Hancock WW, Sayegh MH, Kwok CA, Weiner HL, Carpenter CB. Oral, but not intravenous alloantigen prevents accelerated allograft rejection by selective intragraft Th2 cell activation. Transplantation 1993; 55: 1112.
16. He YG, McCulley JP. Growing human corneal epithelium on collagen shield and subsequent transfer to denuded cornea in vitro. Curr Eye Res 1991; 10: 851.
17. Yue BYJT, Sugar J, Gilboy JE, Elvart JL. Growth of human corneal endothelial cells in culture. Invest Ophthalmol Vis Sci 1989; 30: 248.
18. Wilson SE, Weng J, Blair S, He Y-G, Lloyd S. Expression of E6/E7 or SV40 large T antigen-coding oncogenes in human corneal endothelial cells indicates regulated high-proliferative capacity. Invest Ophthalmol Vis Sci 1995; 36: 32.
19. She SC, Steahly LP, Moticka EJ. A method for performing full-thickness, orthotopic, penetrating keratoplasty in the mouse. Ophthalm Surg 1990; 21: 781.
20. He YG, Ross J, Niederkorn JY. Promotion of murine orthotopic corneal allograft survival by systemic administration of anti CD4 monoclonal antibody. Invest Ophthalmol Vis Sci 1991; 32: 2723.
21. Peeler JS, Niederkorn JY. Antigen presentation by Langerhans cells in vivo: donor derived IA
+ Langerhans cells are required for induction of delayed-type hypersensitivity but not for cytotoxic T lymphocyte responses to alloantigens. J Immunol 1986; 136: 4362.
22. Niederkorn JY, Peeler JS, Mellon J. Phagocytosis of particulate antigens by corneal epithelial cells stimulates interleukin-1 secretion and migration of Langerhans cells into central cornea. Regional Immunol 1989; 2: 83.
23. Streilein JW, McCulley JP, Niederkorn JY. Heterotopic corneal grafting in mice: a new approached to the study of corneal alloimmunity. Invest Ophthalmol Vis Sci 1982; 23: 489.
24. Peeler J, Matoba AY, Niederkorn JY. Corneal allografts induce cytotoxic T cell but not delayed hypersensitivity responses in mice. Invest Ophthalmol Vis Sci 1985; 26: 1516.
25. Sonoda Y, Streilein JW. Impaired cell-mediated immunity in mice bearing healthy orthotopic corneal allografts. J Immunol 1993; 150: 1727.
26. Katami M, White DJG, Watson PG. An analysis of corneal graft rejection in the rat. Transplant Proc 1989; 21: 3147.
27. Niederkorn JY. Effect of cytokine-induced migration of Langerhans cells on corneal allograft survival. Eye 1995; 9: 215.
28. Ross J, He Y-G, Pidherney M, Mellon J, Niederkorn JY. The differential effects of donor versus host Lengerhans cells in the rejection of MHC-matched corneal allografts. Transplantation 1991; 52: 857.
29. The Collaborative Corneal Transplantation Studies Research Group. The collaborative corneal transplantation studies (CCTS): effectiveness of histocompatibility matching in high-risk corneal transplantation. Arch Ophthalmol 1992; 110: 1392.
30. Feder JM, Antonysamy MA, Moticka EJ. Oral administration of alloantigens enhances corneal allograft survival. Invest Ophthalmol Vis Sci (suppl) 1995; 36: S1010.
31. Tevethia SS. Recognition of simian virus 40 T antigen by cytotoxic T lymphocytes. Mol Biol Med 1990; 7: 83.
32. Chen Y, Inobe J-I, Marks R, Gonnella P, Kuchroo VK, Weiner HL. Peripheral deletion of antigen-reactive T cells in oral tolerance. Nature 1995; 376: 177.
