- Email: [email protected]
Inhibition of Murine Corneal Allograft Rejection by Treatment with Antibodies to CD80 and CD86
Exp. Eye Res. (2002) 74, 131±139 doi:10.1006/exer.2001.1109, available online at http://www.idealibrary.com on
Inhibition of Murine Corneal Allograft Rejection by Treatment with Antibodies to CD80 and CD86 F U M I E K A G AYA a, JU N KO H O R I a, K A Z U TA K A K A M I YA a, Y U IC H I K A JI a, T E T S U RO O S H I K A a, S H IRO A M A N O a*, S ATO R U YA M A G A M I b, TA D A H I KO T S U R U b, S U M I YO S H I TA N A K A c, H I RO N O R I M ATS U D A d, H ID E O YA G ITA d A N D KO O K U M U R A d a
Department of Ophthalmology, University of Tokyo School of Medicine, Tokyo, Japan, bDepartment of Ophthalmology, Jichi Medical School, Tochigi, Japan, cDepartment of Ophthalmology, Teikyo University Ichihara Hospital, Chiba, Japan and dDepartment of Immunology, Juntendo University School of Medicine, Tokyo, Japan (Received Seattle 4 June 2001 and accepted in revised form 19 September 2001) The purpose of this study was to investigate the role of CD80 and CD86 costimulatory molecules in corneal allograft rejection. Anti-CD80 and anti-CD86 monoclonal antibodies (mAbs) were administered after orthotopic corneal allograft transplantation. Graft rejection was observed by biomicroscopy. Population and localization of CD80 and CD86 cells in the cornea, cervical lymph nodes, and spleen were examined by ¯ow cytometry and immunohistochemistry. The combined use of anti-CD80 and antiCD86 mAbs was effective in prolonging corneal allograft survival. In the untreated mice bearing rejected graft, many CD86 and CD80 cells were found around the host-graft junctional area in the cornea, and CD86high cells were found in the cervical lymph node and spleen. In contrast, few CD86 or CD80 cells were observed in the cornea, cervical lymph node, and spleen from the mice treated with antiCD80/CD86 mAbs. These results demonstrated a signi®cant role of CD80 and CD86 costimulatory # 2002 Elsevier Science Ltd molecules in corneal allograft rejection. Key words: corneal transplantation; CD80; CD86; costimulatory signal; allogeneic rejection.
1. Introduction Antigen-speci®c T cell activation is a critical step in the rejection of transplanted allografts. To activate T cells completely, two kinds of signal are necessary: the antigen-speci®c signal mediated by T cell receptor (TCR)/CD3 complex and the costimulatory signal mediated by some cell surface adhesion molecules (Springer et al., 1990). Blockade of such a costimulatory signal by administration of monoclonal antibodies (mAbs) to LFA-1/ICAM-1, VLA-4/VCAM-1, or CD2/CD48 led to the long-term acceptance of cardiac, pancreatic, kidney, and corneal allografts (Cosimi, 1990; Isobe et al., 1992; Qin et al., 1994; Yang, Issekutz and Wright, 1995; Yamagami et al., 1996; Hori et al., 1997). To be more signi®cant than these molecules, CD28 is the major receptor on T cells delivering a costimulatory signal critical for determining antigen-speci®c activation or inactivation of T cells (Harding, 1992). Several studies have reported that the blockade of this pathway with mAbs to CD80 and CD86 (ligands for CD28) succeeded in prolonging allograft survival in pancreatic islet or cardiac transplantation models (Lenschow et al., 1995; Bashuda et al., 1996; Kano et al., 1998). These results * Address correspondence to: Shiro Amano, Department of Ophthalmology, University of Tokyo Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: [email protected]
0014-4835/02/01013109 $35.00/0
prompted the authors to investigate the effect of antiCD80 and anti-CD86 mAbs in inhibiting corneal allograft rejection in mice. 2. Materials and Methods Mice BALB/c (H-2d) and C3H/He (H-2k) male mice, 6±8 weeks of age, were obtained from Clea Japan Co. (Tokyo, Japan). This combination represents a complete mismatch at major and minor histocompatibility loci. The experiments were designed according to the guideline of the Association for Research in Vision and Ophthalmology Resolution on the use of animals in research. Orthotopic Corneal Transplantation BALB/c mice were used as the recipients and C3H/He mice were used as the donors for allograft transplantation. BALB/c mice served as both the recipients and the donors for syngeneic transplantation. Penetrating keratoplasty was performed as previously described (Sonoda and Streilein, 1992). Brie¯y, 2 mm diameter donor corneas were placed in the same size recipient bed with 8 interrupted sutures (11-0 nylon, Alcon, Fort Worth, TX, U.S.A.). O¯oxacin ointment (Santen Pharmaceutical, Osaka, Japan) # 2002 Elsevier Science Ltd
132
was applied and one suture of blepharorrhapy was performed (8-0 silk, Alcon, Fort Worth, TX, U.S.A.). The suture for blepharorrhapy was removed at day 2 and corneal sutures were removed at day 8 post grafting. Eyes complicated by cataract, infection, or iris herniation were excluded from this study.
F. K A G AY A E T A L .
protected least signi®cant difference (PLSD). P values less than 0.05 were considered as signi®cant. Histological and Immunohistochemical Examinations
The recipient mice were administrated intraperitoneally (i.p.) with 0.1 mg of anti-CD80 (RM80, n 8) or anti-CD86 mAb (PO3.1, n 8) alone, or 0.25 mg of anti-CD80 or CD86 mAb (n 6, each) alone, or the combination of 0.1 mg (n 10) or 0.25 mg (n 8) each of anti-CD80/CD86 mAbs at 0, 2, 4, 6, and 8 days after grafting. For a longer term treatment, recipients were i.p. administrated with 0.1 mg each of anti-CD80/CD86 mAbs every other day for 3 weeks (n 9). As a control, untreated recipient mice (n 10) were observed.
The graft bearing recipient BALB/c mice were subjected to histological and immunohistochemical studies. For histological studies, eyes were obtained from the untreated rejecting mice or the recipients treated with anti-CD80/CD86 mAbs (0.25 mg day 0± 8) at 2 weeks after grafting, ®xed with 10 % formalin, and processed for light microscopic examination. The specimens were stained with hematoxylin and eosin. For immunohistochemical staining, the graft bearing eyes, regional cervical lymph nodes, and spleen were obtained from the untreated or the treated (0.25 mg, day 0±8) recipients at 2 weeks after grafting. The eyes, cervical lymph nodes, and spleen from naive mice were also prepared. These tissues were frozen in Tissue-Tec O.C.T. compound (Miles Scienti®c, Naperville, IL, U.S.A.). Cryostat sections (4±6 mm) were air dried and ®xed with cold acetone. Since a standard two-step peroxidase method detected no CD80 and only weak expression of CD86, the staining for CD80 and CD86 was performed using a highly sensitive DAKO CSA (Catalyzed Signal Ampli®cation) system. The sections were incubated with biotinylated mAbs for 20 min at room temperature, peroxidase containing ABC complex, biotin conjugated tylamide, and then peroxidase conjugated strept-avidin for 15 min in each step. The sections were rinsed with Trisbuffered saline containing 0.05 % Tween 20 after each step. The slides were stained with diaminobenzidine and counterstained with Mayer's hematoxylin and then observed by light microscopy.
Clinical Evaluation
Flow Cytometry
The corneal grafts were observed by microscopy at weekly intervals until the 12th week or the graft was determined as rejected. Assessment of orthotopic corneal graft survival was performed according to the scoring system previously described (Sonoda and Streilein, 1992): 0 clear graft; 1 minimal super®cial non stromal opacity; 2 minimal deep stromal opacity with pupil margin and iris vessels visible; 3 moderate deep stromal opacity with only pupil margin visible; 4 intense deep stromal opacity with the anterior chamber visible; 5 maximum stromal opacity with total obscuration of the anterior chamber. Grafts with opacity scores of 2 or greater for 2 weeks were considered as rejected.
Cervical lymph node and spleen were harvested from naive mice and the untreated or treated (0.25 mg each of anti-CD80/CD86 mAbs, day 0±8) recipients at two weeks after grafting. Lymph node cells and splenocytes were preincubated with antiCD16/CD32 mAb (2.4G2, PharMingen, San Diego, CA, U.S.A.) to block non-speci®c binding of mAb to Fc g receptor and then incubated with FITC conjugated 1G10, GL1, or isotype-matched control mAb, for 20 min at 48C. The stained cells were analysed on FACScan (Becton Dickinson, San Jose, CA, U.S.A.) and the data were analysed using Cell Quest software (Becton Dickinson).
Statistical Analyses
Effect of Anti-CD80/CD86 MAbs on Survival of Corneal Allografts
Monoclonal Antibodies Hybridomas producing anti-mouse CD80 (RM80, rat IgG2a) and CD86 (PO3.1, rat IgG2b) mAbs were generated as previously described (Nakajima et al., 1995). These mAbs were puri®ed from ascites by standard procedures with caprylic acid, and the purity was checked by SDS±PAGE analysis. The activity of these mAbs was veri®ed by ¯ow cytometry. Biotinylated RM80 and PO3.1 were used for immunohistochemical study. FITC-conjugated anti-CD80 (1G10, rat IgG2a) and anti-CD86 (GL1, rat IgG2b) mAbs were obtained from PharMingen (San Diego, CA, U.S.A.). Antibody Treatment Regimen
Corneal graft survival was compared using KaplanMeier survival curves and by the Mantel-Cox log-rank test. Inter-group comparison was done with Fisher's
3. Results
The corneal allografts from C3H/He mice were rejected within 3 weeks in the untreated BALB/c
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
133
F IG . 1. Survival of corneal allografts in mice untreated or treated with anti-CD80 and anti-CD86 monoclonal antibodies (mAbs). (A) The Kaplan-Meier analysis disclosed a signi®cant difference among four groups [Fig. 1(A), Mantel-Cox log-rank test, P 5 0.001]. The treatment with 0.25 mg of anti-CD80/CD86 mAbs for 8 days was more effective than the untreated group (P 0.0093). The treatment with 0.1 mg of anti-CD80/CD86 mAbs for 3 weeks was more effective than that with the same dose of anti-CD80/CD86 mAbs for 8 days (P 0.0002) and also with a high dose for 8 days (P 0.0129). (B) The Kaplan-Meier analysis disclosed signi®cant differences among four groups (Mantle-Cox log-rank test, P 5 0.001). The administration of 0.25 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days was more effective than to treatment (P 5 0.0001), anti-CD mAb alone (P 0.0003), and anti-CD86 mAb alone (P 0.0037).
recipient mice [Fig. 1(A)], while the syngeneic BALB/ c grafts survived over 12 weeks (data not shown). The Kaplan-Meier analysis disclosed signi®cant differences among four groups [Fig. 1(A), MantleCox log-rank test, P 5 0.001]. The administration of 0.1 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days did not signi®cantly prolong the survival of allografts [Fig. 1(A), P 0.159]. A higher dose (0.25 mg each) of anti-CD80/CD86 mAbs for the same period was effective in prolonging the allograft survival compared to the untreated group [Fig. 1(A), P 0.0093]. A more extended treatment with 0.1 mg each of anti-CD80/CD86 mAbs for the initial 3 weeks was more effective than
the treatment with the same dose for the initial 8 days (P 0.0002) and also with a high dose for the initial 8 days (P 0.0129) [Fig. 1(A)]. The Kaplan-Meier analysis disclosed signi®cant differences among four groups [Fig. 1(B), Mantle-Cox log-rank test, P 5 0.001]. The administration of either anti-CD80 mAb (0.25 mg) or anti-CD86 mAb (0.25 mg) alone for the initial 8 days did not signi®cantly prolong the allograft survival [Fig. 1(B)]. The administration of 0.25 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days was more effective than no treatment (P 5 0.0001), anti-CD mAb alone (P 0.0003), and anti-CD86 mAb alone (P 0.0037) [Fig. 1(B)].
134
F. K A G AY A E T A L .
F IG . 2. Flow cytometric analysis of CD80 and CD86 expression on lymph node cells and splenocytes from naive mice and allografted mice at 2 weeks after transplantation. On the lymph node cells, expression of CD86 was observed in naive mice and increased in rejected mice. CD80 was not detected in naive or the rejected mice. On the splenocytes, expressions of both CD80 and CD86 were found in naive mice and increased in the rejected mice. The expression of CD86 on lymph node cells and that of CD80 and CD86 on splenocytes were suppressed in the anti-CD80/CD86 mAb-treated mice. Thin lines indicate the background staining with isotype-matched control antibodies.
Flow Cytometric Analysis of CD80 and CD86 Expression on Regional Lymph Node Cells and Spleen Cells In naive mice, the cervical lymph node cells expressed low level of CD86 but not CD80, and the
spleen cells expressed both CD80 and CD86 at low levels (Fig. 2, left). In the untreated mice rejecting corneal allografts, the CD86 expression on cervical lymph node cells and the CD80 and CD86 expression on spleen cells were markedly increased (Fig. 2,
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
135
F IG . 3. Histology and immnohistochemical staining for CD80 and CD86 expression in the grafts at 2 weeks after transplantation. H: host, G: graft, asterisk: host±graft junction. The allografts in the untreated recipient at 2 weeks after transplantation showed marked edema and massive in®ltration of mononuclear cells (A). The allografts from the recipients with anti-CD80/CD86 mAbs treatment, much less stromal edema and in®ltration were observed (B). In the rejected graft from the untreated mice, CD86 (C) and CD80 (D) were found on the in®ltrating cells around the host-graft junction. In the surviving graft from the anti-CD80/CD86 mAb treated mice, only a few CD86 cells were observed at the host-graft junction (E). (F) represents negative staining of the rejected graft with control mAb.
middle). The expression of CD80 and CD86 on both lymph node cells and spleen cells was rather reduced in the anti-CD80/CD86 mAb treated mice bearing surviving allografts (Fig. 2, right). Graft Histology and Immunohistochemical Staining for CD80 and CD86 The allografts in the untreated recipients at 2 weeks after transplantation showed marked edema and massive in®ltration of mononuclear cells [Fig. 3(A)].
In contrast, much less stromal edema and in®ltration were observed in the allografts from the recipients treated with 0.25 mg each of anti-CD80/CD86 mAbs for the initial 8 days [Fig. 3(B)]. No signi®cant staining with either anti-CD80 mAb or anti-CD86 mAb was observed in the normal cornea (data not shown). Both CD86 and CD80 were found on the in®ltrating cells around the host-graft junction in the rejected corneas from the untreated recipients at 2 weeks after grafting [Fig. 3(C) and (D)]. CD86 was more intensely expressed than CD80. There were
136
F. K A G AY A E T A L .
F IG . 4. Immunohistochemical staining for CD80 and CD86 expression in cervical lymph nodes at 2 weeks after transplantation. A few CD86 cells were detected in the cortex area of cervical lymph nodes from naive mice (A). In the lymph nodes from the allografted mice without treatment, massive CD86 cells were present in the germinal center (B), whereas a few CD80 cells were observed (C). In the lymph nodes from the allografted mice with anti-CD80/CD86 mAb treatment, CD86 cells were as few as in naive mice (D). (E) represents negative staining of the lymph node from rejecting mice.
few CD80 or CD86 cells in the surviving grafts from the anti-CD80/CD86 mAb treated recipients [Fig. 3(E) and data not shown]. In cervical lymph node from naive mice, cells expressing low level of CD86 were sparsely found in the cortex region [Fig. 4(A)], but CD80 cells were not found (data not shown). In the lymph node from the untreated recipients bearing rejected allograft, cells expressing CD86 at high level were found in the germinal center [Fig. 4(B)], whereas CD80 cells were few [Fig. 4(C)]. In the lymph nodes from the antiCD80/CD86-treated recipients bearing surviving allo-
graft, CD86 cells and CD80 cells were as few as in naive mice [Fig. 4(D)]. In the spleen from naive mice, CD86 cells were sparsely found in the cortex region [Fig. 5(A)], whereas CD80 cells were not found (data not shown). In the spleen from the untreated recipients bearing rejected allograft, cells highly expressing CD86 were found at the marginal zone [Fig. 5(B)], while cells highly expressing CD80 were not found [Fig. 5(C)]. In the spleen from the anti-CD80/CD86-treated recipients bearing surviving allografts, CD86 cells were rather fewer than in that from naive mice [Fig. 5(D)].
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
137
F IG . 5. Immunohistochemical staining for CD80 and CD86 expression in the spleen at 2 weeks after transplantation. In the spleen from naive mice, CD86 cells were sparsely found in the cortex region (A). In the spleen from rejecting recipients, the cells highly expressing CD86 were found at the marginal zone (B). CD80 cells were not apparent in the spleen from naive mice (not shown) or rejecting mice (C). In the spleen from the allografted mice with anti-CD80/CD86 mAb treatment, CD86 cells were fewer than in that from naive mice (D). (E) represents negative staining of the spleen from rejecting mice with control mAb.
4. Discussion The present results show that the combined use of anti-CD80 and anti-CD86 mAbs is effective in prolonging corneal allograft survival, demonstrating a signi®cant role for the CD80/CD86-CD28 costimulatory pathway in the induction of allogeneic responses after corneal transplantation. However, the most effective regime of anti-CD80/CD86 mAbs administration tested in the present study could induce allograft survival over 12 weeks in only 44 % cases, while inde®nite survival over 100 days was
induced after the administration of the same antiCD80/CD86 mAbs in all cases of murine cardiac allografts (Bashuda et al., 1996). This may be due to the fact that the cornea is an avascular tissue and antigen -presenting mechanism after corneal transplantation is possibly different from the vascularized cardiac grafts. In this situation, some costimulatory pathways other than CD80/CD86-CD28, such as CD40-CD40 ligand (CD40L), may be involved in the corneal allograft rejection as was the case with skin allograft rejection (Larsen et al., 1996). Further studies are underway to address this possibility.
138
While both CD80 and CD86 can act as the ligands for CD28, their roles in immune response are signi®cantly different. Previous studies reported that CD86 plays a more predominant role than CD80 in allograft rejection (Lenschow et al., 1995; Bashuda et al., 1996; Fumin et al., 1996). Consistently, also in the present study, CD86 cells predominated over CD80 cells in the graft, lymph node, and spleen from the recipients bearing rejected grafts. However, the fact that the administration of anti-CD86 mAb alone could not signi®cantly prolong the allograft survival suggests a substantial contribution of CD80 to the rejection. In this respect, it is noteworthy that CD80 cells were found only in the graft but not in the lymph node or spleen of the rejecting recipients. Therefore, the CD80 expression in the graft may play a critical role in rejection. The local and systemic expression of CD80 and CD86 before and after corneal transplantation were examined. CD86 cells were increased in the cervical lymph node and spleen in the untreated mice but not in the anti-CD80/CD86-treated mice. It is possible that the number of CD80 and CD86 cells in the antibody treated animals decreased because the administered antibodies remained in the animals and con¯icted with the antibodies for immunohistochemistry and ¯ow cytometry. However, this possibility was denied since no remaining antibody was detected in the same specimens using anti-rat IgG antibody as a secondary binding antibody (data not shown). Nakajima et al. (1995) reported similar results in the serum of CD80/ CD86-treated mice of lupus model. Then what mechanisms explain the depletion of CD80 or CD86 cells in mice treated with anti-CD80/CD86? One possibility is that anti-CD80/CD86 mAbs blocked the up-regulation of CD80/CD86 that takes place in APCs in spleen or lymph nodes when APCs bind to T cells. Another possible explanation is that anti-CD80/ CD86 mAbs blocked the T cell activation, suppressing the in®ltration of CD80 and CD86 in¯ammatory cells into the grafts. Further studies are necessary to clarify these points. In the spleen from the untreated mice bearing rejected graft, CD86 cells were found predominantly in the marginal zone where APCs present antigens to T cells. These results suggest that CD86 expressed on APCs play a signi®cant role in activating T cells in lymph nodes and spleen, leading to allograft rejection. While neither CD80 nor CD86 was expressed in the naive cornea, CD80 and CD86 cells were found around the host-graft junction in the rejected cornea. CD80 and CD86 on APCs are induced when CD40L on T cells binds to CD40 on APCs (June et al., 1994; Niederkorn, 1999). This suggests that CD80 and CD86 were induced on donor APC when in®ltrating recipient T cells directly recognized alloantigens on donor APCs. Alternatively, the CD80 and CD86 cells were in®ltrating recipient APCs that indirectly present donor derived alloantigens. Further studies
F. K A G AY A E T A L .
are needed to characterize the CD80 and CD86 cells in the graft. In summary, the combined use of anti-CD80 and anti-CD86 mAbs was effective in prolonging corneal allograft survival. In mice bearing rejected grafts, CD86 and CD80 cells were found around the hostgraft junction area, but only CD86 cells were found in the cervical lymph node, and spleen. These results demonstrated a signi®cant role for CD80/CD86-CD28 costimulatory pathway in corneal allograft rejection. References Bashuda, H., Seino, K., Kano, M., Sato, K., Azuma, M., Yagita, H. and Okumura, K. (1996). Speci®c acceptance of cardiac allografts after treatment with antibodies to CD80 and CD86 in mice. Trans. Pro. 28, 1039±41. Cosimi, A. B. (1990). In vivo effects of monoclonal antibody to ICAM-1 in nonhuman primates with renal allografts. J. Immunol. 14, 4604±12. Fumin, F., Li, Y., Qian, S., Lu, L., Chambers, F., Starzl, T. E., Fung, J. J. and Thomson, W. (1996). Costimulatory molecule-de®cient dendritic cell progenitors (MHC class II , CD80dim, CD86 ) prolong cardiac allograft survival in nonimmunosuppressed recipients. Transplantation 62, 659±65. Harding, F. A., McArthur, J. G., Gross, J. A., Raulet, D. H. and Allison, J. P. (1992). CD28-mediated signalling costimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356, 607±9. Hori, J., Isobe, M., Yamagami, S., Mizuochi, T. and Tsuru, T. (1997). Speci®c immunosuppression of corneal allograft rejected by combination of anti-VLA-4 and anti-LFA-1 monoclonal antibodies in mice. Exp. Eye Res. 65, 89± 98. Isobe, M., Yagita, H., Okumura, K. and Ihara, K. (1992). Speci®c acceptance of cardiac allograft treatment with antibodies to ICAM-1 and LFA-1. Science 225, 1125±7. June, C. H., Bluestone, J. A., Nadler, L. M. and Thompson, C. B. (1994). The B7 and CD28 receptor families. Immunol. Today 15, 321±31. Kano, M., Bashuda, H., Yagita, H., Okumura, K. and Morishita, Y. (1998). A crucial role of host CD80 and CD86 in rat cardiac xenograft rejection in mice. Transplantation 65, 837±43. Larsen, C. P., Elwood, E. T., Alexander, D. Z., Rotchie, S. C., Hendrix, R. and Tucker-Burden, C. (1996). Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathway. Nature 381, 434±8. Lenschow, D. J., Zeng, Y., Hathcock, K. S., Zuckerman, L. A., Freeman, G., Thistlethwaite, R., Gray, G. S., Hodes, R. J. and Bluestone, J. A. (1995). Inhibition of transplant rejection following treatment with anti-B7-2 and antiB7-1 antibodies. Transplantation 60, 1171±8. Nakajima, A., Azuma, M., Kodera, S., Nuriya, S., Terashi, A., Abe, M., Hirose, S., Shirai, T., Yagita, H. and Okumura, K. (1995). Preferential dependence of autoantibody production in murine lupus on CD86 costimulatory molecule. Eur. J. Immunol. 25, 3060±9. Niederkorn, J. Y. (1999). The immune privilege of corneal allografts. Transplantation 67, 1503±8. Nuriya, S., Yagita, H., Okumura, K. and Azuma, M. (1996). The differential role of CD86 and CD80 co-stimulatory molecules in the induction and the effector phases of contact hypersensitivity. Int. Immunol. 8, 917±26. Qin, L., Chavin, K. D., Lin, J., Yagita, H. and Bromberg, J. S. (1994). Anti-CD2 receptor and anti-CD2 ligand (CD48)
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
antibodies synergize to prolong allograft survival. J. Exp. Med. 179, 341±6. Springer, T. A. (1990). Adhesion receptors of the immune system. Nature 346, 425±33. Sonoda, Y. and Streilein, J. W. (1992). Orthotopic corneal transplantation in mice. Evidence that the immunogenetic rules of rejection do not apply. Transplantation 54, 694±704.
139
Yamagami, S., Tsuru, T., Isobe, M., Obata, H. and Suzuki, J. (1996). The role of cell adhesion molecules in allograft rejection after penetrating keratoplasty in mice. Graefe's Arch. Clin. Exp. Ophthalmol. 234, 382±7. Yang, H., Issekutz, T. B. and Wright, J. R. (1995). Prolongation of rat islet allograft survival by treatment with monoclonal antibodies against VLA-4 and LFA-1. Transplantation 60, 71±6.
Inhibition of Murine Corneal Allograft Rejection by Treatment with Antibodies to CD80 and CD86 F U M I E K A G AYA a, JU N KO H O R I a, K A Z U TA K A K A M I YA a, Y U IC H I K A JI a, T E T S U RO O S H I K A a, S H IRO A M A N O a*, S ATO R U YA M A G A M I b, TA D A H I KO T S U R U b, S U M I YO S H I TA N A K A c, H I RO N O R I M ATS U D A d, H ID E O YA G ITA d A N D KO O K U M U R A d a
Department of Ophthalmology, University of Tokyo School of Medicine, Tokyo, Japan, bDepartment of Ophthalmology, Jichi Medical School, Tochigi, Japan, cDepartment of Ophthalmology, Teikyo University Ichihara Hospital, Chiba, Japan and dDepartment of Immunology, Juntendo University School of Medicine, Tokyo, Japan (Received Seattle 4 June 2001 and accepted in revised form 19 September 2001) The purpose of this study was to investigate the role of CD80 and CD86 costimulatory molecules in corneal allograft rejection. Anti-CD80 and anti-CD86 monoclonal antibodies (mAbs) were administered after orthotopic corneal allograft transplantation. Graft rejection was observed by biomicroscopy. Population and localization of CD80 and CD86 cells in the cornea, cervical lymph nodes, and spleen were examined by ¯ow cytometry and immunohistochemistry. The combined use of anti-CD80 and antiCD86 mAbs was effective in prolonging corneal allograft survival. In the untreated mice bearing rejected graft, many CD86 and CD80 cells were found around the host-graft junctional area in the cornea, and CD86high cells were found in the cervical lymph node and spleen. In contrast, few CD86 or CD80 cells were observed in the cornea, cervical lymph node, and spleen from the mice treated with antiCD80/CD86 mAbs. These results demonstrated a signi®cant role of CD80 and CD86 costimulatory # 2002 Elsevier Science Ltd molecules in corneal allograft rejection. Key words: corneal transplantation; CD80; CD86; costimulatory signal; allogeneic rejection.
1. Introduction Antigen-speci®c T cell activation is a critical step in the rejection of transplanted allografts. To activate T cells completely, two kinds of signal are necessary: the antigen-speci®c signal mediated by T cell receptor (TCR)/CD3 complex and the costimulatory signal mediated by some cell surface adhesion molecules (Springer et al., 1990). Blockade of such a costimulatory signal by administration of monoclonal antibodies (mAbs) to LFA-1/ICAM-1, VLA-4/VCAM-1, or CD2/CD48 led to the long-term acceptance of cardiac, pancreatic, kidney, and corneal allografts (Cosimi, 1990; Isobe et al., 1992; Qin et al., 1994; Yang, Issekutz and Wright, 1995; Yamagami et al., 1996; Hori et al., 1997). To be more signi®cant than these molecules, CD28 is the major receptor on T cells delivering a costimulatory signal critical for determining antigen-speci®c activation or inactivation of T cells (Harding, 1992). Several studies have reported that the blockade of this pathway with mAbs to CD80 and CD86 (ligands for CD28) succeeded in prolonging allograft survival in pancreatic islet or cardiac transplantation models (Lenschow et al., 1995; Bashuda et al., 1996; Kano et al., 1998). These results * Address correspondence to: Shiro Amano, Department of Ophthalmology, University of Tokyo Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: [email protected]
0014-4835/02/01013109 $35.00/0
prompted the authors to investigate the effect of antiCD80 and anti-CD86 mAbs in inhibiting corneal allograft rejection in mice. 2. Materials and Methods Mice BALB/c (H-2d) and C3H/He (H-2k) male mice, 6±8 weeks of age, were obtained from Clea Japan Co. (Tokyo, Japan). This combination represents a complete mismatch at major and minor histocompatibility loci. The experiments were designed according to the guideline of the Association for Research in Vision and Ophthalmology Resolution on the use of animals in research. Orthotopic Corneal Transplantation BALB/c mice were used as the recipients and C3H/He mice were used as the donors for allograft transplantation. BALB/c mice served as both the recipients and the donors for syngeneic transplantation. Penetrating keratoplasty was performed as previously described (Sonoda and Streilein, 1992). Brie¯y, 2 mm diameter donor corneas were placed in the same size recipient bed with 8 interrupted sutures (11-0 nylon, Alcon, Fort Worth, TX, U.S.A.). O¯oxacin ointment (Santen Pharmaceutical, Osaka, Japan) # 2002 Elsevier Science Ltd
132
was applied and one suture of blepharorrhapy was performed (8-0 silk, Alcon, Fort Worth, TX, U.S.A.). The suture for blepharorrhapy was removed at day 2 and corneal sutures were removed at day 8 post grafting. Eyes complicated by cataract, infection, or iris herniation were excluded from this study.
F. K A G AY A E T A L .
protected least signi®cant difference (PLSD). P values less than 0.05 were considered as signi®cant. Histological and Immunohistochemical Examinations
The recipient mice were administrated intraperitoneally (i.p.) with 0.1 mg of anti-CD80 (RM80, n 8) or anti-CD86 mAb (PO3.1, n 8) alone, or 0.25 mg of anti-CD80 or CD86 mAb (n 6, each) alone, or the combination of 0.1 mg (n 10) or 0.25 mg (n 8) each of anti-CD80/CD86 mAbs at 0, 2, 4, 6, and 8 days after grafting. For a longer term treatment, recipients were i.p. administrated with 0.1 mg each of anti-CD80/CD86 mAbs every other day for 3 weeks (n 9). As a control, untreated recipient mice (n 10) were observed.
The graft bearing recipient BALB/c mice were subjected to histological and immunohistochemical studies. For histological studies, eyes were obtained from the untreated rejecting mice or the recipients treated with anti-CD80/CD86 mAbs (0.25 mg day 0± 8) at 2 weeks after grafting, ®xed with 10 % formalin, and processed for light microscopic examination. The specimens were stained with hematoxylin and eosin. For immunohistochemical staining, the graft bearing eyes, regional cervical lymph nodes, and spleen were obtained from the untreated or the treated (0.25 mg, day 0±8) recipients at 2 weeks after grafting. The eyes, cervical lymph nodes, and spleen from naive mice were also prepared. These tissues were frozen in Tissue-Tec O.C.T. compound (Miles Scienti®c, Naperville, IL, U.S.A.). Cryostat sections (4±6 mm) were air dried and ®xed with cold acetone. Since a standard two-step peroxidase method detected no CD80 and only weak expression of CD86, the staining for CD80 and CD86 was performed using a highly sensitive DAKO CSA (Catalyzed Signal Ampli®cation) system. The sections were incubated with biotinylated mAbs for 20 min at room temperature, peroxidase containing ABC complex, biotin conjugated tylamide, and then peroxidase conjugated strept-avidin for 15 min in each step. The sections were rinsed with Trisbuffered saline containing 0.05 % Tween 20 after each step. The slides were stained with diaminobenzidine and counterstained with Mayer's hematoxylin and then observed by light microscopy.
Clinical Evaluation
Flow Cytometry
The corneal grafts were observed by microscopy at weekly intervals until the 12th week or the graft was determined as rejected. Assessment of orthotopic corneal graft survival was performed according to the scoring system previously described (Sonoda and Streilein, 1992): 0 clear graft; 1 minimal super®cial non stromal opacity; 2 minimal deep stromal opacity with pupil margin and iris vessels visible; 3 moderate deep stromal opacity with only pupil margin visible; 4 intense deep stromal opacity with the anterior chamber visible; 5 maximum stromal opacity with total obscuration of the anterior chamber. Grafts with opacity scores of 2 or greater for 2 weeks were considered as rejected.
Cervical lymph node and spleen were harvested from naive mice and the untreated or treated (0.25 mg each of anti-CD80/CD86 mAbs, day 0±8) recipients at two weeks after grafting. Lymph node cells and splenocytes were preincubated with antiCD16/CD32 mAb (2.4G2, PharMingen, San Diego, CA, U.S.A.) to block non-speci®c binding of mAb to Fc g receptor and then incubated with FITC conjugated 1G10, GL1, or isotype-matched control mAb, for 20 min at 48C. The stained cells were analysed on FACScan (Becton Dickinson, San Jose, CA, U.S.A.) and the data were analysed using Cell Quest software (Becton Dickinson).
Statistical Analyses
Effect of Anti-CD80/CD86 MAbs on Survival of Corneal Allografts
Monoclonal Antibodies Hybridomas producing anti-mouse CD80 (RM80, rat IgG2a) and CD86 (PO3.1, rat IgG2b) mAbs were generated as previously described (Nakajima et al., 1995). These mAbs were puri®ed from ascites by standard procedures with caprylic acid, and the purity was checked by SDS±PAGE analysis. The activity of these mAbs was veri®ed by ¯ow cytometry. Biotinylated RM80 and PO3.1 were used for immunohistochemical study. FITC-conjugated anti-CD80 (1G10, rat IgG2a) and anti-CD86 (GL1, rat IgG2b) mAbs were obtained from PharMingen (San Diego, CA, U.S.A.). Antibody Treatment Regimen
Corneal graft survival was compared using KaplanMeier survival curves and by the Mantel-Cox log-rank test. Inter-group comparison was done with Fisher's
3. Results
The corneal allografts from C3H/He mice were rejected within 3 weeks in the untreated BALB/c
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
133
F IG . 1. Survival of corneal allografts in mice untreated or treated with anti-CD80 and anti-CD86 monoclonal antibodies (mAbs). (A) The Kaplan-Meier analysis disclosed a signi®cant difference among four groups [Fig. 1(A), Mantel-Cox log-rank test, P 5 0.001]. The treatment with 0.25 mg of anti-CD80/CD86 mAbs for 8 days was more effective than the untreated group (P 0.0093). The treatment with 0.1 mg of anti-CD80/CD86 mAbs for 3 weeks was more effective than that with the same dose of anti-CD80/CD86 mAbs for 8 days (P 0.0002) and also with a high dose for 8 days (P 0.0129). (B) The Kaplan-Meier analysis disclosed signi®cant differences among four groups (Mantle-Cox log-rank test, P 5 0.001). The administration of 0.25 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days was more effective than to treatment (P 5 0.0001), anti-CD mAb alone (P 0.0003), and anti-CD86 mAb alone (P 0.0037).
recipient mice [Fig. 1(A)], while the syngeneic BALB/ c grafts survived over 12 weeks (data not shown). The Kaplan-Meier analysis disclosed signi®cant differences among four groups [Fig. 1(A), MantleCox log-rank test, P 5 0.001]. The administration of 0.1 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days did not signi®cantly prolong the survival of allografts [Fig. 1(A), P 0.159]. A higher dose (0.25 mg each) of anti-CD80/CD86 mAbs for the same period was effective in prolonging the allograft survival compared to the untreated group [Fig. 1(A), P 0.0093]. A more extended treatment with 0.1 mg each of anti-CD80/CD86 mAbs for the initial 3 weeks was more effective than
the treatment with the same dose for the initial 8 days (P 0.0002) and also with a high dose for the initial 8 days (P 0.0129) [Fig. 1(A)]. The Kaplan-Meier analysis disclosed signi®cant differences among four groups [Fig. 1(B), Mantle-Cox log-rank test, P 5 0.001]. The administration of either anti-CD80 mAb (0.25 mg) or anti-CD86 mAb (0.25 mg) alone for the initial 8 days did not signi®cantly prolong the allograft survival [Fig. 1(B)]. The administration of 0.25 mg each of anti-CD80 and anti-CD86 mAbs for the initial 8 days was more effective than no treatment (P 5 0.0001), anti-CD mAb alone (P 0.0003), and anti-CD86 mAb alone (P 0.0037) [Fig. 1(B)].
134
F. K A G AY A E T A L .
F IG . 2. Flow cytometric analysis of CD80 and CD86 expression on lymph node cells and splenocytes from naive mice and allografted mice at 2 weeks after transplantation. On the lymph node cells, expression of CD86 was observed in naive mice and increased in rejected mice. CD80 was not detected in naive or the rejected mice. On the splenocytes, expressions of both CD80 and CD86 were found in naive mice and increased in the rejected mice. The expression of CD86 on lymph node cells and that of CD80 and CD86 on splenocytes were suppressed in the anti-CD80/CD86 mAb-treated mice. Thin lines indicate the background staining with isotype-matched control antibodies.
Flow Cytometric Analysis of CD80 and CD86 Expression on Regional Lymph Node Cells and Spleen Cells In naive mice, the cervical lymph node cells expressed low level of CD86 but not CD80, and the
spleen cells expressed both CD80 and CD86 at low levels (Fig. 2, left). In the untreated mice rejecting corneal allografts, the CD86 expression on cervical lymph node cells and the CD80 and CD86 expression on spleen cells were markedly increased (Fig. 2,
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
135
F IG . 3. Histology and immnohistochemical staining for CD80 and CD86 expression in the grafts at 2 weeks after transplantation. H: host, G: graft, asterisk: host±graft junction. The allografts in the untreated recipient at 2 weeks after transplantation showed marked edema and massive in®ltration of mononuclear cells (A). The allografts from the recipients with anti-CD80/CD86 mAbs treatment, much less stromal edema and in®ltration were observed (B). In the rejected graft from the untreated mice, CD86 (C) and CD80 (D) were found on the in®ltrating cells around the host-graft junction. In the surviving graft from the anti-CD80/CD86 mAb treated mice, only a few CD86 cells were observed at the host-graft junction (E). (F) represents negative staining of the rejected graft with control mAb.
middle). The expression of CD80 and CD86 on both lymph node cells and spleen cells was rather reduced in the anti-CD80/CD86 mAb treated mice bearing surviving allografts (Fig. 2, right). Graft Histology and Immunohistochemical Staining for CD80 and CD86 The allografts in the untreated recipients at 2 weeks after transplantation showed marked edema and massive in®ltration of mononuclear cells [Fig. 3(A)].
In contrast, much less stromal edema and in®ltration were observed in the allografts from the recipients treated with 0.25 mg each of anti-CD80/CD86 mAbs for the initial 8 days [Fig. 3(B)]. No signi®cant staining with either anti-CD80 mAb or anti-CD86 mAb was observed in the normal cornea (data not shown). Both CD86 and CD80 were found on the in®ltrating cells around the host-graft junction in the rejected corneas from the untreated recipients at 2 weeks after grafting [Fig. 3(C) and (D)]. CD86 was more intensely expressed than CD80. There were
136
F. K A G AY A E T A L .
F IG . 4. Immunohistochemical staining for CD80 and CD86 expression in cervical lymph nodes at 2 weeks after transplantation. A few CD86 cells were detected in the cortex area of cervical lymph nodes from naive mice (A). In the lymph nodes from the allografted mice without treatment, massive CD86 cells were present in the germinal center (B), whereas a few CD80 cells were observed (C). In the lymph nodes from the allografted mice with anti-CD80/CD86 mAb treatment, CD86 cells were as few as in naive mice (D). (E) represents negative staining of the lymph node from rejecting mice.
few CD80 or CD86 cells in the surviving grafts from the anti-CD80/CD86 mAb treated recipients [Fig. 3(E) and data not shown]. In cervical lymph node from naive mice, cells expressing low level of CD86 were sparsely found in the cortex region [Fig. 4(A)], but CD80 cells were not found (data not shown). In the lymph node from the untreated recipients bearing rejected allograft, cells expressing CD86 at high level were found in the germinal center [Fig. 4(B)], whereas CD80 cells were few [Fig. 4(C)]. In the lymph nodes from the antiCD80/CD86-treated recipients bearing surviving allo-
graft, CD86 cells and CD80 cells were as few as in naive mice [Fig. 4(D)]. In the spleen from naive mice, CD86 cells were sparsely found in the cortex region [Fig. 5(A)], whereas CD80 cells were not found (data not shown). In the spleen from the untreated recipients bearing rejected allograft, cells highly expressing CD86 were found at the marginal zone [Fig. 5(B)], while cells highly expressing CD80 were not found [Fig. 5(C)]. In the spleen from the anti-CD80/CD86-treated recipients bearing surviving allografts, CD86 cells were rather fewer than in that from naive mice [Fig. 5(D)].
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
137
F IG . 5. Immunohistochemical staining for CD80 and CD86 expression in the spleen at 2 weeks after transplantation. In the spleen from naive mice, CD86 cells were sparsely found in the cortex region (A). In the spleen from rejecting recipients, the cells highly expressing CD86 were found at the marginal zone (B). CD80 cells were not apparent in the spleen from naive mice (not shown) or rejecting mice (C). In the spleen from the allografted mice with anti-CD80/CD86 mAb treatment, CD86 cells were fewer than in that from naive mice (D). (E) represents negative staining of the spleen from rejecting mice with control mAb.
4. Discussion The present results show that the combined use of anti-CD80 and anti-CD86 mAbs is effective in prolonging corneal allograft survival, demonstrating a signi®cant role for the CD80/CD86-CD28 costimulatory pathway in the induction of allogeneic responses after corneal transplantation. However, the most effective regime of anti-CD80/CD86 mAbs administration tested in the present study could induce allograft survival over 12 weeks in only 44 % cases, while inde®nite survival over 100 days was
induced after the administration of the same antiCD80/CD86 mAbs in all cases of murine cardiac allografts (Bashuda et al., 1996). This may be due to the fact that the cornea is an avascular tissue and antigen -presenting mechanism after corneal transplantation is possibly different from the vascularized cardiac grafts. In this situation, some costimulatory pathways other than CD80/CD86-CD28, such as CD40-CD40 ligand (CD40L), may be involved in the corneal allograft rejection as was the case with skin allograft rejection (Larsen et al., 1996). Further studies are underway to address this possibility.
138
While both CD80 and CD86 can act as the ligands for CD28, their roles in immune response are signi®cantly different. Previous studies reported that CD86 plays a more predominant role than CD80 in allograft rejection (Lenschow et al., 1995; Bashuda et al., 1996; Fumin et al., 1996). Consistently, also in the present study, CD86 cells predominated over CD80 cells in the graft, lymph node, and spleen from the recipients bearing rejected grafts. However, the fact that the administration of anti-CD86 mAb alone could not signi®cantly prolong the allograft survival suggests a substantial contribution of CD80 to the rejection. In this respect, it is noteworthy that CD80 cells were found only in the graft but not in the lymph node or spleen of the rejecting recipients. Therefore, the CD80 expression in the graft may play a critical role in rejection. The local and systemic expression of CD80 and CD86 before and after corneal transplantation were examined. CD86 cells were increased in the cervical lymph node and spleen in the untreated mice but not in the anti-CD80/CD86-treated mice. It is possible that the number of CD80 and CD86 cells in the antibody treated animals decreased because the administered antibodies remained in the animals and con¯icted with the antibodies for immunohistochemistry and ¯ow cytometry. However, this possibility was denied since no remaining antibody was detected in the same specimens using anti-rat IgG antibody as a secondary binding antibody (data not shown). Nakajima et al. (1995) reported similar results in the serum of CD80/ CD86-treated mice of lupus model. Then what mechanisms explain the depletion of CD80 or CD86 cells in mice treated with anti-CD80/CD86? One possibility is that anti-CD80/CD86 mAbs blocked the up-regulation of CD80/CD86 that takes place in APCs in spleen or lymph nodes when APCs bind to T cells. Another possible explanation is that anti-CD80/ CD86 mAbs blocked the T cell activation, suppressing the in®ltration of CD80 and CD86 in¯ammatory cells into the grafts. Further studies are necessary to clarify these points. In the spleen from the untreated mice bearing rejected graft, CD86 cells were found predominantly in the marginal zone where APCs present antigens to T cells. These results suggest that CD86 expressed on APCs play a signi®cant role in activating T cells in lymph nodes and spleen, leading to allograft rejection. While neither CD80 nor CD86 was expressed in the naive cornea, CD80 and CD86 cells were found around the host-graft junction in the rejected cornea. CD80 and CD86 on APCs are induced when CD40L on T cells binds to CD40 on APCs (June et al., 1994; Niederkorn, 1999). This suggests that CD80 and CD86 were induced on donor APC when in®ltrating recipient T cells directly recognized alloantigens on donor APCs. Alternatively, the CD80 and CD86 cells were in®ltrating recipient APCs that indirectly present donor derived alloantigens. Further studies
F. K A G AY A E T A L .
are needed to characterize the CD80 and CD86 cells in the graft. In summary, the combined use of anti-CD80 and anti-CD86 mAbs was effective in prolonging corneal allograft survival. In mice bearing rejected grafts, CD86 and CD80 cells were found around the hostgraft junction area, but only CD86 cells were found in the cervical lymph node, and spleen. These results demonstrated a signi®cant role for CD80/CD86-CD28 costimulatory pathway in corneal allograft rejection. References Bashuda, H., Seino, K., Kano, M., Sato, K., Azuma, M., Yagita, H. and Okumura, K. (1996). Speci®c acceptance of cardiac allografts after treatment with antibodies to CD80 and CD86 in mice. Trans. Pro. 28, 1039±41. Cosimi, A. B. (1990). In vivo effects of monoclonal antibody to ICAM-1 in nonhuman primates with renal allografts. J. Immunol. 14, 4604±12. Fumin, F., Li, Y., Qian, S., Lu, L., Chambers, F., Starzl, T. E., Fung, J. J. and Thomson, W. (1996). Costimulatory molecule-de®cient dendritic cell progenitors (MHC class II , CD80dim, CD86 ) prolong cardiac allograft survival in nonimmunosuppressed recipients. Transplantation 62, 659±65. Harding, F. A., McArthur, J. G., Gross, J. A., Raulet, D. H. and Allison, J. P. (1992). CD28-mediated signalling costimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356, 607±9. Hori, J., Isobe, M., Yamagami, S., Mizuochi, T. and Tsuru, T. (1997). Speci®c immunosuppression of corneal allograft rejected by combination of anti-VLA-4 and anti-LFA-1 monoclonal antibodies in mice. Exp. Eye Res. 65, 89± 98. Isobe, M., Yagita, H., Okumura, K. and Ihara, K. (1992). Speci®c acceptance of cardiac allograft treatment with antibodies to ICAM-1 and LFA-1. Science 225, 1125±7. June, C. H., Bluestone, J. A., Nadler, L. M. and Thompson, C. B. (1994). The B7 and CD28 receptor families. Immunol. Today 15, 321±31. Kano, M., Bashuda, H., Yagita, H., Okumura, K. and Morishita, Y. (1998). A crucial role of host CD80 and CD86 in rat cardiac xenograft rejection in mice. Transplantation 65, 837±43. Larsen, C. P., Elwood, E. T., Alexander, D. Z., Rotchie, S. C., Hendrix, R. and Tucker-Burden, C. (1996). Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathway. Nature 381, 434±8. Lenschow, D. J., Zeng, Y., Hathcock, K. S., Zuckerman, L. A., Freeman, G., Thistlethwaite, R., Gray, G. S., Hodes, R. J. and Bluestone, J. A. (1995). Inhibition of transplant rejection following treatment with anti-B7-2 and antiB7-1 antibodies. Transplantation 60, 1171±8. Nakajima, A., Azuma, M., Kodera, S., Nuriya, S., Terashi, A., Abe, M., Hirose, S., Shirai, T., Yagita, H. and Okumura, K. (1995). Preferential dependence of autoantibody production in murine lupus on CD86 costimulatory molecule. Eur. J. Immunol. 25, 3060±9. Niederkorn, J. Y. (1999). The immune privilege of corneal allografts. Transplantation 67, 1503±8. Nuriya, S., Yagita, H., Okumura, K. and Azuma, M. (1996). The differential role of CD86 and CD80 co-stimulatory molecules in the induction and the effector phases of contact hypersensitivity. Int. Immunol. 8, 917±26. Qin, L., Chavin, K. D., Lin, J., Yagita, H. and Bromberg, J. S. (1994). Anti-CD2 receptor and anti-CD2 ligand (CD48)
M U R I N E C O R N E A L A L LO G R A F T R E J E C T I O N
antibodies synergize to prolong allograft survival. J. Exp. Med. 179, 341±6. Springer, T. A. (1990). Adhesion receptors of the immune system. Nature 346, 425±33. Sonoda, Y. and Streilein, J. W. (1992). Orthotopic corneal transplantation in mice. Evidence that the immunogenetic rules of rejection do not apply. Transplantation 54, 694±704.
139
Yamagami, S., Tsuru, T., Isobe, M., Obata, H. and Suzuki, J. (1996). The role of cell adhesion molecules in allograft rejection after penetrating keratoplasty in mice. Graefe's Arch. Clin. Exp. Ophthalmol. 234, 382±7. Yang, H., Issekutz, T. B. and Wright, J. R. (1995). Prolongation of rat islet allograft survival by treatment with monoclonal antibodies against VLA-4 and LFA-1. Transplantation 60, 71±6.