Immunologic Mechanisms of Corneal Allografts Reconstituted from Cultured Allogeneic Endothelial Cells in an Immune-Privileged Site

Abstract

purpose. To analyze outcomes and immunologic features after cultured corneal endothelial cell (CEC) transplantation in a murine model.

methods. CEC-deprived BALB/c corneas were reconstituted in vitro with an immortalized C3H-CEC cell line and then transplanted orthotopically into recipient BALB/c mice with experimental bullous keratopathy. Graft survival rates, donor-specific delayed hypersensitivity (DTH), and mixed lymphocyte reactions were evaluated in recipient mice after grafting. Fates of CEC transplantation were assessed after adoptive transfer, regrafting, and immunization with C3H splenocytes.

results. Chimeric CEC allografts composed of cultured allogeneic CECs did not provoke rejection reaction, DTH, or mixed-lymphocyte reactions, unlike the high rejection rate that occurred in full-thickness corneal allografts. Adoptive transfer of splenocytes from mice that had accepted chimeric CEC allografts did not increase the graft survival rate after full-thickness corneal transplantation, and the rejection rate of a second full-thickness graft was not improved in these mice, suggestive of no active immunosuppression. Pre-sensitization by subcutaneous injection of splenocytes with the same haplotype as cultured CECs induced systemic DTH to the same allogeneic antigens but did not promote the rejection of chimeric CEC allografts, suggesting that chimeric CEC allografts are ignored by the host immune system.

conclusions. These findings indicate that immunologic ignorance rather than active immunosuppression is important for the rejection-free acceptance of chimeric CEC allografts. Transplantation of corneal grafts formed with allogeneic CECs could be an ideal treatment strategy to overcome postoperative rejection.

Full-thickness corneal transplantation has become the standard surgical intervention for corneal endothelial cell (CEC) dysfunction, such as bullous keratopathy and endothelial graft failure. ^{1 2 3} CEC transplantation with Descemet membrane and the posterior stroma has been introduced. This technique has the benefit of minimizing irregular astigmatism and suture-related problems that are inevitable after full-thickness corneal transplantation. ^{4 5 6 7 8} However, these corneal transplantation techniques require fresh human corneas, and there is a severe worldwide shortage of donors. To overcome donor cornea shortage and to avoid postoperative refractive errors, surgical procedures—among them cell and sheet transplantation techniques in animal models—that replace the endothelium with cultured CECs have been reported. ^{9 10 11 12 13 14} Allotransplantation with cultured CECs will probably come into clinical use in the near future, but the immunologic features of allogeneic CEC (allo-CEC) transplantation are thus far unknown. 

The anterior chamber, including the cornea, is an immunologically privileged site, and this status is maintained by multiple mechanisms, such as the blood-aqueous barrier; the lack of blood vessels, lymphatics, and mature antigen-presenting cells in the central cornea; the presence of immunomodulatory factors in the ocular fluids; and the constitutive corneal expression of Fas-ligand. ^{15 16} Anterior chamber-associated immune deviation (ACAID) contributes to this immunologic privilege by suppressing the priming and elicitation of adaptive immune responses, which represents an ocular form of peripheral tolerance. After full-thickness allogeneic corneal transplantation, however, significant numbers of grafts are rejected in spite of this immune privilege. ² The corneal epithelium, stroma, and CECs have different roles in the sensitization of graft recipients. ^{15 16 17 18} In particular, CECs are the major target of allograft rejection through the mechanism of apoptosis in the early postoperative phase, ^{19 20 21} and CEC rejection greatly influences the fate of the graft because these cells are critical for maintaining corneal transparency. ²²  

We previously established a cultured allo-CEC transplantation model in mice with bullous keratopathy. ²³ In the present study, we analyzed the outcomes and immunologic changes after CEC transplantation in the same murine model to determine the immunologic features of this novel transplantation technique. 

Methods

Animals

BALB/c (H-2^d), C3H/HeN (C3H) (H-2^k), and C57BL/6(H-2^b) male mice weighing 25 to 30 g (8–12 weeks old) were purchased from Clea Japan Co., Ltd. (Tokyo, Japan). Mice were anesthetized by intramuscular injection of a mixture of 4 mg ketamine and 1 xylazine before all surgical procedures. Animals were treated in accordance with the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. 

Evaluation of the Antigenicity of Cultured Mouse CECs

A C3H-CEC cell line was immortalized by infecting with SV40. ^{24 25 26 27} To assess the antigenicity of cultured CECs, we examined the delayed-type hypersensitivity (DTH) reaction to these cells in an ear-swelling assay. BALB/c mice were used as recipients, and C3H/HeN or C57/BL6 (third party) mice were used as donors. 

C3H-spleen cells (1 × 10⁷) or immortalized C3H-CECs (1 × 10⁶ or 1 × 10⁴) were injected subcutaneously. After 1 week, C3H-spleen cells (1 × 10⁶) treated with 0.1 mg/mL mitomycin C (MMC; Kyowa-Hakko-Kogyo Co., Ltd., Tokyo, Japan) were injected into the left ear of each mouse for immunization, whereas the right ear was treated only with phosphate-buffered saline (PBS). Ear thickness was measured with an engineer’s micrometer (Mitsutoyo Corp., Tokyo, Japan) before and 24 hours after this challenge. Each group consisted of at least five mice. Specific ear swelling was calculated as follows: (24 hours–0 hours in the challenge ear) − (24 hours–0 hours in the control ear) × 10⁻³ mm ± SD. Immortalized C3H-CECs, C3H spleen cells, and C57/BL6 spleen cells were used for subcutaneous immunization. C3H-spleen cells or C57BL/6-spleen cells were injected into the ear to trigger a DTH reaction in preimmunized mice. 

Cell Culture and Labeling of CECs

The procedures followed in this study have been reported. ²³ Briefly, CECs were cultured in Eagle’s minimum essential medium (EMEM; Sigma, St. Louis, MO) with 10% FBS, after which CECs were labeled with the fluorescent cell linker (Zynaxis PKH26-GL; Zynaxis, Malvern, PA) and fluorescence was observed under a fluorescence microscope (BH2-RFLT3 or BX50; Olympus, Tokyo, Japan). The dye used for staining cannot transfer fluorescence to other cells; hence, all the fluorescent cells represented donor cells repopulating Descemet membrane. 

Reconstitution of Denuded Corneas with Fluorescein-Labeled CECs

The corneas of BALB/c or C3H mice were removed with a 2-mm trephine and were denuded of CECs by scraping. Then the corneas were immersed for 5 minutes in saline containing 50 mg/mL gentamicin (Sigma). Next, immortalized C3H-CECs stained with a fluorescent dye in 100 μL low-glucose EMEM containing 6% dextran (Sigma) were applied to each cornea in the wells of a 24-well plate, which was then centrifuged at 1000 rpm for 10 minutes. The corneas were subsequently maintained in the culture medium for another 2 days. Denuded BALB/c corneas that had been reconstituted with immortalized C3H-CECs (chimeric CEC allograft) were used as grafts for BALB/c and C3H recipients, and denuded C3H corneas reconstituted with immortalized C3H-CECs were used as sham-operated control grafts and control grafts for C3H recipients. 

Corneal Transplantation

All surgical procedures were performed under an operating microscope. Corneal endothelial injury was created in mice, as described previously with some modifications. ^{23 28} Briefly, 0.05% benzalkonium chloride was injected into the anterior chamber of recipient animals through a 30-gauge needle (Terumo, Tokyo, Japan) to suppress CEC regeneration. After 10 seconds, the anterior chamber was rinsed with saline for 20 seconds. Full-thickness corneal grafts were excised from normal mouse cornea with a 2-mm trephine (Inami, Tokyo, Japan). Each graft was transplanted into a 1.5-mm diameter recipient corneal bed and was attached with 8 or 10 interrupted 11-0 nylon sutures (Mani, Tochigi, Japan), which were removed 9 days after surgery. 

Corneal Transplantation and Exclusion Criteria

The following four types of grafts were transplanted into recipient BALB/c mice with experimental bullous keratopathy: full-thickness isografts from BALB/c mice (isograft group, n = 21); full-thickness allografts from C3H mice (allograft group, n = 22); reconstituted corneas formed by immortalized C3H-CECs attached to C3H corneas that had been denuded of endothelium (sham-operated group, n = 10); and chimeric CEC allografts (CEC/aG group, n = 20). To test whether surgical technique and corneal reconstitution with immortalized CECs affect graft survival rates, C3H (n = 6) or BALB/c (n = 6) corneas reconstituted with immortalized C3H-CECs were transplanted into C3H recipients. As a control for the nonreconstituted corneas, BALB/c mouse corneas denuded of endothelium were preserved under the same conditions and were transplanted. Because graft edema of denuded endothelium in BALB/c mice was not recovered by the proliferation of host corneal endothelium within 6 weeks, the follow-up period for clinical observation in this study was determined as 6 weeks. Donor–recipient combinations in the experimental groups are shown as Table 1 . 

Corneal grafts were observed twice a week by slit lamp biomicroscopy for 6 weeks after transplantation. Eyes with complications, such as corneal wound rupture or loss of the anterior chamber, were excluded. There were no cases of hyphema, infection, or cataract after transplantation. If fluorescein-tagged cells did not diffusely cover the entire posterior surface of the corneal graft on fluorescein microscopy 1 week after surgery in the CEC/aG group, the procedure was defined as unsuccessful and such eyes were also excluded from analysis. One hundred ten eyes were treated, among which 20 were excluded from analysis because of death of the recipient caused by overdose of anesthesia (n = 5), corneal wound rupture resulting from severe bullous keratopathy created by benzalkonium chloride injection (n = 15), or unsuccessful grafting in the CEC/aG group (n = 4). 

Histologic and Immunohistochemical Evaluation

Immunohistochemical studies were performed on frozen sections of the grafts, and each group included at least three samples. Phycoerythrin-conjugated rat anti–mouse CD4 and fluorescein isothiocyanate-conjugated anti–mouse CD8 monoclonal antibodies (eBioscience, Tokyo, Japan) were used as the primary antibodies. Eyes were enucleated at 4 weeks after transplantation, fixed in 4% paraformaldehyde, and frozen in optimal cutting temperature compound (Sakura Finetek, Tokyo, Japan) in liquid nitrogen. Then the frozen specimens were cut into 8-μm sections. After they were washed with PBS, the sections were incubated with the primary antibody diluted to 2.5 μg/mL for 2 hours. Then sections were stained with 4′,6-diamidine-2′-phenylindole dihydrochloride (DAPI; Roche Applied Science, Penzberg, Germany), mounted, and observed under a fluorescence microscope. At least five sections were analyzed for each tissue specimen derived from each eye. Mean numbers of infiltrating CD4 or CD8 T cells per square millimeter ± SD from 5 to 10 fields of corneas were analyzed. 

Donor-Specific DTH Response after Corneal Transplantation

Induction of a donor-specific DTH response after corneal transplantation was assessed by the ear-swelling assay. Each BALB/c recipient of a normal or a reconstituted cornea received an injection of 1 × 10⁶ C3H MMC-treated spleen cells into the left ear 4 and 8 weeks after grafting. Positive control mice received subcutaneous injection of 1 × 10⁷ C3H splenocytes 1 week before ear challenge. As a control for the effect of benzalkonium chloride, DTH was also tested in mice with benzalkonium chloride injection alone. Each group consisted of at least five mice. 

Mixed-Lymphocyte Reaction

BALB/c splenocytes (2.5 × 10⁵) were stimulated with MMC-treated C3H splenocytes or BALB/c splenocytes (2.5 × 10⁵) in 50 μL medium, consisting of RPMI 1640 with 10% FBS (Sigma) and ITS⁺ culture supplement (1 μg/mL iron-free transferrin, 10 ng/mL linoleic acid, 0.3 ng/mL Na₂Se, and 0.2 μg/mL Fe(NO₃)3; Collaborative Biomedical Products, Bedford, MA). ²⁹ The negative control was splenocytes from naive BALB/c mice, and the positive control was splenocytes from BALB/c mice that had received 1 × 10⁷ C3H-splenocytes subcutaneously 1 week earlier. Cell proliferation was monitored by determining the amount of formazan produced from water-soluble tetrazolium salt (WST-1) over 72 hours, and the mean value for five mice per group was calculated. ³⁰ Allospecific optical density was calculated as follows: (allogeneic optical density) − (syngeneic optical density). 

Test of Active Suppression after Adoptive Transfer and Corneal Regrafting

To evaluate whether active immunosuppression occurs after allo-CEC transplantation, we performed adoptive transfer and regrafting. Adoptive transfer was performed as described elsewhere. ^{29 31 32} Briefly, splenocytes obtained from the CEC/aG group 8 weeks after grafting (5 × 10⁷ splenocytes in 0.5 mL PBS) were injected into the tail veins of naive BALB/c mice. On the following day, full-thickness C3H corneal grafts were transplanted into the eyes of these mice. As a control group, full-thickness C3H corneal grafts were transplanted into the eyes of naive BALB/c mice. 

Regrafting of full-thickness corneas in eyes bearing CEC allografts was performed. Full-thickness C3H corneas were grafted into the same eyes 8 weeks after initial grafting in the CEC/aG group, as described elsewhere. ²⁹  

Influence of Pre-sensitization on CEC Allograft Survival

Pre-sensitization of BALB/c mice was performed by subcutaneous injection of 1 × 10⁷ C3H splenocytes 1 week before the transplantation of chimeric CEC allografts. Graft rejection rates and C3H-specific DTH responses were evaluated 8 weeks after surgery. 

Statistical Analysis

The unpaired t-test was used to compare mean values as appropriate. All analyses were performed with a statistical software package (StatView; Abacus Concepts, Berkeley, CA). The level of significance was set at P < 0.05. Graft survival rates for each experimental group were compared by the log-rank test. Data on the alloantigen-specific DTH response were analyzed with the Mann-Whitney U test to assess the differences between two independent groups. Statistical significance was set at P < 0.05, as appropriate. 

Results

Antigenicity of Cultured CECs

To confirm whether immortalized C3H-derived CECs have a common C3H-specific antigen, induction of an antigen-specific DTH response was evaluated in BALB/c mice after alloantigen immunization. Immortalized C3H-CECs, but not C57BL/6 spleen cells, induced a strong response to the injection of C3H-spleen cells into the ear (Fig. 1A) . In contrast, immunization with 1 × 10⁶ or 1 × 10⁴ C3H-CECs and C3H-spleen cells did not lead to any response after challenge with 1 × 10⁶ C57/BL6 spleen cells (Fig. 1B) . These findings imply that immortalized C3H-CECs have common C3H-specific antigenic features. 

Rejection-Free Rate of CEC Allografts

BALB/c mice were used as recipients. Up to the sixth week after surgery, the isograft group showed corneal transparency in all cases (n = 21), and the rejection-free rate was 100%. The rate was only 18% in the allograft group (n = 22) and 20% in the sham-operated group (n = 10) but increased to 100% in the CEC/aG group (n = 20), which showed significantly better acceptance of grafts compared with the allograft group (P < 0.01; Fig. 2A ). Normal C3H corneas with native CECs were accepted in C3H recipients (n = 6; not shown). All C3H donor corneas reconstituted with immortalized C3H-CECs (n = 6) were transparent when transplanted into C3H recipients, indicating that surgical technique and donor corneal reconstitution with immortalized CECs do not affect the graft survival rate (Fig. 2B) . In contrast, all BALB/c donor corneas reconstituted with immortalized C3H-CECs (n = 6) were rapidly rejected in C3H recipients, suggesting corneal epithelium and stroma are major components for allosensitization and allograft rejection (Fig. 2B) . Representative anterior segment photographs obtained 4 weeks after transplantation are shown in Figure 3 . In contrast to the opaque corneal grafts in the allograft group (Fig. 3A)and the sham-operated group (not shown), corneal grafts were transparent in the isograft group and in the CEC/aG group (Fig. 3B) , with most of the graft covered by transplanted fluorescein-labeled CECs (Fig. 3B , inset). Four weeks after transplantation, no edema and no infiltrating cells were present in the corneas of the isograft group. In particular, no CD4⁺ or CD8⁺ cells were present in the cornea or anterior chamber. In the allograft group (Fig. 3C)and the sham-operated group (not shown), numerous mononuclear cells were detected in the grafts, and many of them were CD4⁺ cells (Fig. 3D) . In the CEC-a/G group, neither infiltrating cells nor corneal edema were observed (Fig. 3E) , and no CD4⁺ (Fig. 3F)or CD8⁺ cells were detected. Figure 3Gshows the mean CD4⁺ and CD8⁺ cell numbers per section in each group. Significantly low numbers of CD4⁺ and CD8⁺ infiltrating cells in the graft of the CEC/aG group were detected compared with those in the allograft and sham-operated groups (Fig. 3G) . CECs were detected on Descemet membrane of the reconstituted corneas and formed a monolayer 4 weeks after transplantation (Fig. 3H) . These cells were shown to be PKH26⁺ by fluorescence microscopy, indicating that they were transplanted immortalized CECs and not host-derived CECs (Fig. 3I) . 

Donor-Specific DTH and Proliferative Response

Strong DTH responses were detected in the allograft and sham-operated groups, but not in the CEC/aG group, at 4 weeks (Fig. 4A) . Treatment of eyes with benzalkonium chloride alone did not affect the donor-specific DTH response (data not shown). To examine the alloantigen-specific proliferative response, splenocytes were obtained 4 weeks after grafting and were stimulated for 72 hours with MMC-treated splenocytes. The splenocytes from the CEC/aG group showed a weak response similar to that of naive BALB/c-splenocytes, but strong responses were seen in the allograft and sham-operated groups (Figs. 4B) . Similar findings of DTH and proliferative responses were obtained at 8 weeks (data not shown). Proliferative responses of draining lymph nodes also reproduced those of splenocytes (data not shown). These results indicate that in the present study, CEC allograft transplantation did not induce an alloantigen-specific immune response during the follow-up period. 

Adoptive Transfer and Regrafting of Full-Thickness Corneas

After full-thickness corneal transplantation, regulatory T cells in spleen cells are thought to have a role in long-term graft acceptance. ^{29 31 33} Therefore, we next tested whether regulatory T cells were produced by mice of the CEC/aG group. We used splenocytes, but not draining lymph nodes, because regulatory function was not detected in draining lymph nodes after full-thickness corneal transplantation in vivo. ³⁴ Full thickness corneal grafts from C3H mice were transplanted into the eyes of BALB/c mice (n = 8), which then received splenocytes from the CEC/aG group after grafting. As a result, the graft rejection rate did not decrease by adoptive transfer within the observation period of this study (8 weeks; Fig. 5 ). To evaluate whether active immunosuppression was induced after allo-CEC transplantation, full-thickness C3H corneas were grafted into the same eyes 8 weeks after initial grafting in the CEC/aG group (n = 11), but there was no improvement in graft failure rate compared with that in normal control mice (Fig. 5) . These findings suggest that active immunosuppression was not induced in the CEC/aG group, even though active immunosuppression has been detected in mice with long-term graft survival after full-thickness corneal transplantation. ^{29 31 33}  

Influence of Pre-sensitization by Alloantigens on CEC Allograft Transplantation

Finally, the possible influence on graft rejection of preimmunization with allogeneic splenocytes was investigated. Surprisingly, the graft survival rate was 100% in the preimmunized CEC/aG group (n = 10) and was comparable with that of the nonimmunized CEC/aG group (n = 15; Fig. 6A ). An antigen-specific systemic DTH response was detected 8 weeks after immunization in the preimmunized mice with CEC allograft acceptance (Fig. 6B) . Subcutaneous injection of splenocytes syngeneic to the chimeric CEC allografts did not affect the rejection rates of CEC allografts while showing a systemic high DTH response. These findings indicate that immunosuppression or T-cell anergy was not induced despite successful grafting and suggest that immunologic ignorance can explain the rejection-free acceptance of chimeric CEC allografts. 

Discussion

Most full-thickness corneal grafts transplanted into patients with the sequelae of ocular inflammation, who often have neovascularization and lymphangiogenesis, ³⁵ lead to irreversible graft failure, with multiple mechanisms involved in the effector phase of graft rejection. Acute rejection of histoincompatible corneal grafts involves the infiltration of CD4⁺ T cells and of CD8⁺ T cells, macrophages, dendritic cells, and neutrophils. CD4⁺ Th1 cells are the primary effectors of corneal graft rejection through mediation of an acute immune reaction that targets the CECs. ^{19 20 36 37 38} Other possible effectors include B cells, CTLs, ^{39 40 41 42 43} and the Th2-mediated immune response. ^{44 45} Compared with the rejection mechanisms operating after full-thickness corneal transplantation, the processes of sensitization and rejection after allo-CEC transplantation are unknown. 

In this study, we used a mouse CEC deficiency model for allo-CEC transplantation to mimic clinical bullous keratopathy. C3H-derived immortalized CECs evoked a C3H-specific antigenic response when injected subcutaneously into BALB/c mice. In contrast, based on clinical and histologic findings, chimeric CEC allograft (composed of allogeneic cultured CECs attached to syngeneic epithelium and stroma) did not evoke any rejection reaction in BALB/c recipients, unlike the high rejection rates in the allograft and sham-operated groups. Conversely, BALB/c donor corneas (allogeneic epithelium and stroma) reconstituted with immortalized C3H CECs) were swiftly rejected when transplanted into C3H recipients. These findings indicate that CEC is not an inducer of allograft rejection and suggest that CEC transplantation has advantages over conventional full-thickness corneal transplantation. 

Several possible mechanisms can explain the successful acceptance of CEC allografts: the induction of active immunosuppression including the induction of regulatory T cells (CD1d-reactive NKT cells), the induction of T-cell anergy, and immunologic ignorance of draining lymph node T-inducer priming cells. ^{33 46 47 48 49} Active immunosuppression is induced in mice that show long-term acceptance of full-thickness corneal transplantation. ^{29 31} However, CEC allografts failed to induce active immunosuppression, as evidenced by adoptive transfer of splenocytes from mice that had accepted chimeric CEC allografts and regraft challenge in CEC allograft-accepted mice. In our findings, ACAID is induced after the injection of CECs into the anterior chamber. The induction of ACAID after the injection of CECs into the anterior chamber is consistent with the published data. ²⁷ However, ACAID was not induced after CEC allograft transplantation (Hayashi T, unpublished observation, 2007). These findings suggest that CEC graft survival without rejection did not result from the induction of immunosuppression or T-cell anergy. Given that subcutaneous injection of splenocytes syngeneic to the chimeric CEC allografts did not increase the rejection rates of CEC allografts while showing a high systemic DTH response to the same antigen, immunologic ignorance seems to be the main reason for the rejection-free acceptance of chimeric CEC allografts. 

The mechanism underlying the occurrence of immunologic ignorance in the present CEC allograft model remains elusive. CECs are located between the aqueous humor and Descemet membrane of the cornea, which is composed primarily of unstructured type IV collagen and is not supposed to allow cell infiltration. Accordingly, antigen-presenting cells should be able to reach CECs only through the aqueous humor, and this may minimize the possibility of sensitization to CEC antigens and subsequent rejection. In addition to the benefit of an anatomically privileged site, the expression of these protective molecules by CECs may inhibit the activities of antigen-presenting cells and thus prevent rejection mediated by activated T cells when only CECs (without allogeneic corneal stroma and epithelium) are transplanted. CECs express Fas ligand (CD95L) ⁵⁰ and B7-H1. ^{51 52} Cell death factor FasL induces the apoptosis of infiltrating Fas-positive cells and contributes to the prolonged acceptance of full-thickness corneal grafts, ^{50 53 54} whereas B7-H1 has a critical role in regulating T-cell activation. ^{52 55} Moreover, CECs express an immunosuppressive factor, indoleamine 2,3-dioxygenase. ^{56 57 58} All these factors may directly or indirectly contribute to the immunologic ignorance after CEC transplantation in an immune-privileged site. 

The limitations of this study were as follows. We performed observation and immunologic evaluation at 6 and 8 weeks after transplantation, respectively, because CECs eventually regenerate in mice even after benzalkonium chloride is administered to suppress host CEC regeneration. Corneal allografts have a high acceptance rate over a short term, but this may become lower over a longer period, so further long-term evaluation is required to assess the chronic phase of allograft rejection. In addition, the anatomy and physiology of the rodent eye are not entirely comparable with those of the human eye. Furthermore, full-thickness trephination was performed in our mouse CEC transplantation model, but this technique is not consistent with the ideal method of CEC transplantation. Studies using larger animals, such as rabbits or monkeys, may allow allo-CEC transplantation to be performed without full-thickness trephination. Recently, Chauhan and Dana ⁵⁹ reported that Foxp3-expressing CD4⁺CD25⁺ regulatory T cells in draining lymph nodes can prevent corneal allograft rejection with an adoptive transfer assay. We could not completely deny the possibility, in our study, of active suppression by regulatory T cells in draining lymph nodes after CEC transplantation. Regulatory T cells in draining lymph nodes are, however, unlikely to be the main regulators for the rejection-free acceptance of chimeric CEC allografts. Whole draining lymph nodes do not have regulatory function after full-thickness corneal transplantation by lymph node transplantation experiments, ³⁴ and allospecific regulatory responses are generated in the spleen after corneal transplantation, as evidenced by experiments using mice as hosts after splenectomy. ³⁴ Moreover, the findings that preimmunization with allogeneic splenocytes do not increase rejection rates of CEC allografts imply impaired allosensitization in the CEC transplantation group and may noticeably decrease the possibility of producing significant numbers of regulatory T cells in draining lymph nodes. 

In summary, the transplantation of chimeric CEC allografts did not provoke immunologic rejection in mice, unlike the high rejection rate that occurs with full-thickness corneal allograft. Active immunosuppression was not responsible for the acceptance of the CEC allograft. Pre-sensitization by alloantigens did not increase the rejection of the transplanted CEC allograft but induced systemic DTH to the same allogeneic antigens, indicating that immunologic ignorance could explain the lack of rejection of these chimeric grafts. Thus, cultured CECs could be transplanted to retain more of the recipient’s own cornea and thereby decrease the foreign antigen load, which may help to reduce the incidence of rejection and may be a useful strategy to overcome the worldwide corneal donor shortage. 

 Supported by a grant-in-aid from the Ministry of Health, Labor and Welfare of Japan; a grant-in-aid from the Japan Medical Association; and a grant from the Johnson & Johnson K. K. Vision Care Company.

 Submitted for publication July 5, 2008; revised December 2, 2008, and January 12, 2009; accepted April 27, 2009.

 Disclosure: T. Hayashi, None; S. Yamagami, None; K. Tanaka, None; S. Yokoo, None; T. Usui, None; S. Amano, None; N. Mizuki, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Experimental Groups with the Donor/Host Combinations

Group	Donor Cornea			Host
		Epithelium-Stroma	Endothelium
Isograft	BALB/c (H-2d)	BALB/c (H-2d)	BALB/c (H-2d)
Allograft	C3H (H-2k)	C3H (H-2k)	BALB/c (H-2d)
Sham-operated	C3H (H-2k)	C3H-CECs (H-2k)	BALB/c (H-2d)
CEC/aG	BALB/c (H-2d)	C3H-CECs (H-2k)	BALB/c (H-2d)

Figure 1.

 

Antigenicity of cultured CECs. Immortalized C3H-CECs, C3H spleen cells, and C57/BL6 spleen cells were used as the antigens for subcutaneous immunization, and (A) C3H spleen cells or (B) C57BL/6 spleen cells were injected into the ear 1 week after immunization. BALB/c mice were used as recipients. After immunization with C3H-CECs and C3H spleen cells, but not C57/BL6 spleen cells, a significant DTH response occurred to C3H spleen cells. (B) After immunization with C57/BL6 spleen cells, a strong DTH response occurred to C57/BL6 spleen cells (positive control), whereas immortalized C3H-CECs did not induce a strong DTH response to C57BL/6 spleen cells (compared with negative control). *P < 0.001. **P = NS. Data represent mean ± SD.

Figure 2.

 

Rejection-free rates after transplantation. (A) Full-thickness isografts from BALB/c mice, full-thickness allografts from C3H mice, reconstituted corneas formed by immortalized C3H-CECs attached to C3H corneas, and chimeric CEC allografts were transplanted on BALB/c mice recipients. Each group was followed up for 6 weeks. The rejection-free rate was significantly higher in the CEC/aG group (n = 20) than in the allograft group (n = 22; *P < 0.01) or the sham-operated group (n = 10; **P < 0.01) and was comparable to that in the isograft group (n = 21). (B) In C3H recipients, all donor corneas reconstituted with immortalized C3H-CECs were transparent, whereas BALB/c donor corneas reconstituted with immortalized C3H-CECs were rapidly rejected (*P < 0.01).

Figure 3.

 

Anterior segment photographs and histologic findings 4 weeks after transplantation. (A) Corneal opacity with edema can be seen in the allograft group, whereas (B) corneal grafts are transparent and neither opacity nor edema is seen in the CEC/aG groups. Inset: most of the graft is covered with fluorescein-labeled CECs in the CEC/aG group. Green lines: graft margin. Black lines and white dotted line: eye margin. (C) Numerous mononuclear cells infiltrate the graft in the allograft group. (D) CD4⁺ (red) cells dominantly infiltrate the allograft. (E) No stromal edema or infiltrating cells and (F) no CD4⁺ (red) cells are present in the CEC/aG group. (G) More infiltrating CD4⁺ T cells (*P < 0.01) and CD8⁺ T cells (**P < 0.01) are present in the allograft group and the sham-operated group than in the CEC/aG group. (H) Transplanted CECs are detected on Descemet membrane of reconstituted corneas and form a monolayer 4 weeks after transplantation. (I) Monolayer of transplanted CECs stained with PKH-26 on Descemet membrane of a reconstituted cornea under fluorescein microscopy. Nuclei are stained blue with DAPI. Scale bars, 100 μm. Original magnifications: (A, B) ×10; (C–F) ×100; (H, I) ×400.

Figure 4.

 

Weaker donor-specific DTH response and MLR response by spleen cells in CEC/aG recipients. (A) Donor-specific DTH was tested in BALB/c mice 4 weeks after transplantation. C3H spleen cells were injected into the ear, and the ear-swelling assay was performed. Strong DTH responses were detected in the positive control, allograft, and sham-operated groups. The DTH response of the CEC/aG group was significantly lower than that of the positive control group. (B) Spleen cells harvested 4 weeks after transplantation were stimulated in vitro for 72 hours with MMC-treated spleen cells to examine the alloantigen-specific proliferative response. The response of cells from the CEC/aG group was as weak as that of naive C3H spleen cells, whereas the allograft and sham-operated groups had strong responses. Data represent the mean ± SD (compared with negative control). *P < 0.01. **P = NS.

Figure 5.

 

Lack of active immunosuppression in the CEC/aG group. The rejection-free graft survival rate was evaluated after transplantation of full-thickness corneal grafts into naive BALB/c mice after adoptive transfer of spleen cells derived from the CEC/aG group (LAD group; n = 8) and after corneal grafting in the CEC/aG group (regrafting group; n = 11). In both groups, the rejection-free graft survival rate showed no significant difference from that of the control group (20%; n = 10).

Figure 6.

 

No effect of preimmunization on the acceptance of chimeric grafts. (A) Preimmunization by subcutaneous injection of C3H spleen cells was performed 1 week before CEC allograft transplantation. The rejection-free graft survival rate was 100% in the preimmunized CEC/aG and the nonimmunized CEC/aG groups, indicating no difference between the two groups. (B) A strong DTH response was detected 8 weeks after immunization in the preimmunized mice with accepted CEC allograft (compared with negative control). *P < 0.01. Data represent the mean ± SD.

The authors thank Jun Yamada for critical comments, and Hiroe Morita and Shin Hato for assistance.