ACAID Induced by Allogeneic Corneal Tissue Promotes Subsequent Survival of Orthotopic Corneal Grafts

Abstract

purpose. To determine whether immune deviation is induced by allogeneic corneal tissue implanted in the anterior chamber and whether survival of subsequent orthotopic corneal allografts is thereby enhanced.

methods. Corneal tissue from C57BL/6 mice was implanted in the anterior chamber of eyes of BALB/c mice. The fate of these implants was assessed histologically, and the donor-specific immune response of recipient mice was tested for donor-specific delayed hypersensitivity and the capacity to accept or reject C57BL/6 corneas grafted orthotopically into the fellow eye.

results. C57BL/6 cornea implants in the anterior chamber failed to induce donor-specific delayed hypersensitivity but impaired donor-specific delayed hypersensitivity in a proportion of recipients with implants in place for 2 weeks. Mice with donor-specific delayed hypersensitivity rejected the intraocular implants. Mice bearing C57BL/6 cornea implants in the anterior chamber for 2 (but not 4) weeks accepted the C57BL/6 corneas grafted orthotopically into the fellow eye at a high rate and with few rejection reactions.

conclusions. Implantation of allogeneic corneal tissue in the anterior chamber of the eye has the transient capacity to alter the recipient alloimmune response in a manner that promotes survival of subsequent orthotopic corneal allografts.

The cornea has long been considered an immune privileged tissue. ^{1 2} The extraordinary success of penetrating keratoplasties in human beings testifies not only to the privileged nature of corneal tissue, but to the existence of immune privilege in the anterior chamber of the eye. ³ Nonetheless, corneal graft rejection remains a significant clinical problem, and experimental efforts to find methods to prevent immune destruction of orthotopic cornea grafts continue. In the recent past, various immunosuppressive agents have been explored in animals model systems, including corticosteroids, ⁴ cyclosporin A, ⁵ FK-506, ⁶ and antibodies directed at surface molecules thought to be important in T-cell–dependent graft rejection: intercellular adhesion molecule (ICAM)-1, ⁷ lymphocyte function–associated antigen (LFA)-1, ⁷ CD4, and CD8. ^{8 9} Corticosteroids and cyclosporin A are used clinically to suppress corneal graft rejection, but the other experimental therapies have yet to reach routine clinical application. 

Because corneal allografts express transplantation antigens that elicit specific immunity, the possibility exists that strategies could be designed to suppress the recipient’s specific immune response to these antigens (i.e., to induce a form of immunologic tolerance that would abort or nullify the rejection response). Orthotopically grafted corneas form the anterior wall of the anterior chamber, and therefore corneal endothelial cells express histocompatibility antigens within this immune privileged site. Anterior chamber associated immune deviation (ACAID) is a stereotypic systemic immune response that mammals generate when antigens are placed within the anterior chamber. ¹⁰ ACAID is a kind of “immunologic tolerance,” and therefore the possibility exists that antigens expressed on cells of orthotopic corneal allografts may induce donor-specific ACAID. If true, then the act of performing an orthotopic corneal allograft may in and of itself represent a “tolerizing” strategy that should promote the graft’s survival. 

To test this possibility, we examined the fate and systemic immune consequences of placing allogeneic corneal tissue in the anterior chambers of mouse eyes. Our findings indicate that corneal tissue placed in the anterior chamber could promote subsequent orthotopic corneal allograft success by suppressing donor-specific immunity. 

Materials and Methods

Animals

BALB/c and C57BL/6 mice 10 to 12 weeks of age were used in these experiments. These strains differ at the major histocompatibility complex locus (H-2) as well as at numerous minor histocompatibility loci. All procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Implantation of Corneal Fragments in the Anterior Chamber

Cornea buttons of 2.0-mm diameter (to be used as grafts) were carefully excised from eyes of C57BL/6 and BALB/c mice. When used as fragments implanted intracamerally, these corneal buttons were cut into full-thickness fragments of approximately 1.0 × 0.3 mm. Through a lateral incision of the recipient cornea, three corneal fragments were implanted into the left anterior chamber of eyes of BALB/c mice (see Fig. 1 ). The corneal wound was closed with an interrupted 11-0 nylon suture, which was removed 7 days later. 

Preparation of Splenocytes for Immunization

Spleens were harvested from euthanatized C57BL/6 donors, rendered into single-cell suspension by pressing through nylon mesh, and used for subcutaneous (SC; 10 × 10⁶ cells), anterior chamber (5 × 10⁵ cells), or intrapinnae (1 × 10⁶ cells) injections. 

Assay for Donor-Specific Delayed Hypersensitivity

At selected times after implantation of allogeneic fragments or splenocytes, 1 × 10⁶ irradiated (2000 rad) spleen cells/10 μm from C57BL/6 donors were injected into the right pinna. For a positive control, a similar number of irradiated spleen cells were injected into the ear pinnae of normal BALB/c mice that received as SC injection of 10 × 10⁶ C57BL/6 spleen cells 1 week previously. Twenty-four hours after intrapinnae injection, ear thickness was measured with a low-pressure engineer’s micrometer. Ear swelling, as an indication of delayed hypersensitivity, was expressed as follows: Specific swelling = [(24-hour numerical values of right ear − 0-hour numerical values of right ear) − (24-hour numerical values of left ear − 0 hr numerical values of left ear)] × 10⁻³ mm. Ear-swelling responses at 24 hours after injection are presented as individual values (10⁻³ mm) for each tested animal and as a group mean ± SEM. Naive mice that received only ear challenge served as negative control subjects. 

Assay for Suppression of Donor-Specific Delayed Hypersensitivity

BALB/c mice bearing intracameral fragments of either allogeneic or syngeneic corneas at 2 weeks after implantation received an SC injection of 10 × 10⁶ C57BL/6 splenocytes. One week later, the ears of these mice were challenged with 1 × 10⁶ irradiated C57BL/6 splenocytes, and ear swelling was assessed 24 hours later. As an ACAID control for suppression, one panel of BALB/c mice received an intracameral injection of 5 × 10⁵ C57BL/6 spleen cells. One week later, these mice received an SC immunization with 10 × 10⁶ C57BL/6 splenocytes, and the ears of the mice were challenged with 1 × 10⁶ irradiated C57BL/6 splenocytes 7 days later. All experimental and control panels consisted of at least five mice. 

Histopathologic Assessment of Corneal Fragments Implanted in the Anterior Chamber

Syngeneic and allogeneic cornea fragments were assessed histologically. Fragment-bearing eyes were enucleated at appropriate times after implantation, fixed and imbedded in paraffin, and sectioned and stained with hematoxylin and eosin. 

Orthotopic Corneal Transplantation

Orthotopic corneal transplantation was performed as described previously. ¹¹ Briefly, donor central corneas (2.0-mm diameter) were excised by Vannas scissors and placed in chilled phosphate-buffered saline. Recipients were anesthetized with intraperitoneal injections of ketamine and xylazine. Donor and recipient corneas were trephined with identical 2.0-mm diameters. Donor corneas were sutured in place with eight interrupted sutures of 11-0 nylon (Sharpoint; Vanguard, Houston, TX). All sutures were removed on day 8 after the procedure. Corneal opacity was examined weekly for 8 weeks to judge the appearance of rejection reactions and graft failure. Five experimental panels were established:

1.  
   Orthotopic transplantation of C57BL/6 cornea to BALB/c recipients.
2.  
   C57BL/6 corneal fragments implanted in the anterior chamber 2 weeks before orthotopic transplantation of C57BL/6 cornea in the contralateral eye.
3.  
   C57BL/6 corneal fragments implanted in the anterior chamber 4 weeks before orthotopic transplantation of C57BL/6 cornea in the contralateral eye.
4.  
   C57BL/6 corneal fragments implanted in the anterior chamber 2 weeks before orthotopic transplantation of C57BL/6 cornea in the contralateral eye; fragment-bearing eye enucleated at the time of orthotopic transplantation.
5.  
   C57BL/6 spleen cells injected into the anterior chamber 2 weeks before orthotopic transplantation of C57BL/6 cornea in the contralateral eye.

Evaluation and Scoring of Orthotopic Corneal Transplants

Grafts were examined with slit lamp microscopy at weekly intervals, as described previously. ¹¹ A scoring system was devised to describe semiquantitatively the extent of opacity (0–5+ ), as follows: 0, clear graft; 1+, minimal superficial (nonstromal) opacity; 2+, minimal deep stromal opacity; 3+, moderate deep stromal opacity; 4+, intense deep stromal opacity; 5+, maximum stromal opacity. Grafts with opacity scores of 3+ or greater until 3 weeks were regarded as displaying early rejection reactions. If these grafts never cleared subsequently (scores of 0 or 1+), a diagnosis of early rejection was made. Grafts that were initially clear but achieved opacity scores of 2+ or higher beyond 3 weeks were assigned a diagnosis of late rejection reaction. Grafts with opacity scores 2+ or greater at 8 weeks were considered irreversibly rejected. 

Statistical Analyses

Ear-swelling measurements were evaluated statistically by using a two-tailed Student’s t-test. P < 0.05 was deemed significant. Corneal graft rejection reactions were evaluated using a two-tailed Fisher’s exact test, with P < 0.05 deemed significant. 

Results

Clinical Fate of Corneal Fragments Implanted in the Anterior Chamber

Corneal fragments, containing epithelial, stromal and endothelial layers, were implanted in the anterior chamber of mouse eyes, immediately behind and adjacent to the center of the recipient corneal endothelium. Within 24 to 48 hours the grafts were sufficiently clear to enable the margins of the recipient pupil to be visualized easily. Syngeneic cornea fragments retained this clear appearance throughout the 8-week observation period. Approximately 50% of the allogeneic cornea fragments also retained a high level of clarity during this interval, but the remainder began to undergo pathologic changes within 2 to 3 weeks. As judged by slit lamp biomicroscopy, the edges of fragments became progressively more irregular and indistinct, whereas healthy fragments maintained crisp margins. The failing fragments appeared to “melt”—that is, they became swollen, increasingly opaque, and slipped into the angle of the anterior chamber. We refer to these rejected fragments as “melted.” From these clinical observations, we tentatively concluded that full-thickness syngeneic corneal fragments can survive in the anterior chamber for prolonged intervals, but that the survival of some allogeneic fragments can be curtailed, presumably by an alloimmune response to transplantation antigens expressed on the grafts. 

Induction of Donor-Specific Delayed Hypersensitivity by Intracameral Implants of Allogeneic Tissues and Cells

We first sought to determine whether allogeneic corneal fragments implanted in the anterior chamber of the eye generated systemic donor-specific delayed hypersensitivity. Panels of BALB/c mice received intracameral implants of C57BL/6 corneal fragments or splenocytes (5 × 10⁵). Positive control mice received an SC injection of 10 × 10⁶ C57BL/6 spleen cells. After 2 or 4 weeks (in the case of one panel of cornea fragment recipients) the ears of these mice were challenged with 1 × 10⁶ irradiated C57BL/6 spleen cells. As the results of a representative experiment displayed in Figure 2 show, allogeneic fragments in the anterior chamber for either 2 or 4 weeks failed to induce systemic, donor-specific delayed hypersensitivity. It is important to point out that donor-specific delayed hypersensitivity was examined in both mice bearing healthy allogeneic corneal fragments, as well as mice bearing those that“ melted.” In neither case was delayed hypersensitivity detected. These findings imply that allogeneic corneal fragments in the anterior chamber are incapable of instigating a donor-specific alloimmune rejection response, even though some of the fragment grafts appeared to have deteriorated. 

We next determined whether allogeneic corneal fragments implanted intracamerally were capable of inducing donor-specific ACAID. In this experiment, panels of BALB/c mice received either allogeneic or syngeneic corneal fragments in the anterior chamber. As in previous experiments, approximately 50% of the intraocular allogeneic cornea fragments displayed a melted phenotype, whereas the remainder were clear when examined at 2 weeks after engraftment. At this time, recipient mice received SC injections of 10 × 10⁶ C57BL/6 splenocytes. Positive control mice received an SC immunization with allogeneic spleen cell alone. After 1 week, the ears of these mice were challenged with irradiated C57BL/6 splenocytes. The results of this experiment are presented in Figure 3 . Mice bearing melted intracameral allogeneic cornea fragments displayed donor-specific delayed hypersensitivity similar to that of positive controls (that received syngeneic corneal fragments intraocularly) and of mice primed by SC immunization with allogeneic spleen cells. However, mice bearing clear allogeneic corneal fragments at the time of testing mounted only feeble delayed hypersensitivity responses to donor alloantigens. These findings indicate that allogeneic corneal implants within the anterior chamber are not uniformly ignored by recipient immune systems. In a significant number of cases, intraocular graft recipients acquire donor-specific ACAID. Moreover, the acquisition of ACAID correlates with the clinical condition of the graft fragment. Mice with melted grafts displayed no evidence of ACAID, whereas mice with healthy grafts displayed ACAID. 

Histopathologic Appearance of Allogeneic Cornea Fragment Implants in the Anterior Chamber

Eyes bearing anterior chamber implants of allogeneic and syngeneic cornea fragments were enucleated immediately after the delayed hypersensitivity assays were completed. The eyes were then prepared for histologic analysis. Microscopic evaluation of syngeneic cornea fragments revealed three-layered structures. On one edge, a thin layer of epithelium was present; on the opposite margin, an attenuated layer of endothelial cells rested on Descemet’s membrane. Keratocytes of a density and morphology similar to that found in recipient cornea stroma were readily apparent (Fig. 4A ). Allogeneic graft fragments that were clear when examined clinically strongly resembled syngeneic grafts. Epithelial and endothelial layers were present, and the stroma was intact. In some sections of these fragments, a mild infiltrate of inflammatory cells was observed at the graft margins (Fig. 4B) . By contrast, allogeneic cornea fragments that were judged clinically to be melted and whose recipients failed to acquire ACAID, displayed significant inflammatory and destructive changes (Fig. 4C) . The stroma was swollen, and the density of keratocytes was markedly reduced. Neither an epithelial layer nor an endothelial cell layer was visible. The margins of these grafts were adjacent to an intense inflammatory infiltrate, although the stroma of the graft itself was curiously devoid of either neo blood vessels or inflammation. We interpret these results to mean that allogeneic corneal fragments implanted in the anterior chamber can, but do not always, undergo an immunologically mediated deterioration. Allografts that avoided this fate survived in mice that also acquired donor-specific ACAID, implying that suppression of delayed hypersensitivity may have protected these grafts. 

Influence of Intracameral Allogeneic Cornea Fragments on Survival of Subsequent Orthotopic Corneal Allografts

The previous experiments support the hypothesis that implants of allogeneic corneal fragments in the anterior chamber promote immune deviation, at least in a significant number of experimental animals. We next wished to determine the fate of orthotopic corneal allografts placed in unmanipulated eyes of mice whose contralateral eyes already contained allogeneic implants. Several different experimental groups were tested. Group 1 contained the grafting control subjects: Normal BALB/c mice received orthotopic transplants of C57BL/6 corneas with no other experimental manipulations. In group 2, C57BL/6 corneas were transplanted orthotopically in one eye of BALB/c mice that contained C57BL/6 cornea fragments in the contralateral eye for 2 weeks. In group 3, similar grafts were placed in eyes of mice with allogeneic fragments in place for 4 weeks. In group 4, similar grafts were placed in eyes of mice with allogeneic fragments in place for 2 weeks; the fragment-containing eyes were enucleated at the time orthotopic grafts were placed in the fellow eye. In group 5, C57BL/6 corneas were grafted orthotopically into eyes of mice that had received C57BL/6 spleen cells (5 × 10⁵) in the contralateral eye 2 weeks previously. 

Orthotopic corneal grafts experienced one of three fates: accepted indefinitely, early rejection (x), or late rejection (Δ). The results are summarized in Table 1 . Approximately 80% of grafts in the unmanipulated control mice experienced rejection reactions, and of these 65% had irreversible rejection (data not shown). Early rejection reactions occurred in 47%, and late rejection reactions occurred in 32% of control mice. Among mice that received orthotopic corneal allografts 2 weeks after allogeneic fragments were implanted intracamerally (group 2, Fig. 5 ), only 36% of orthotopic grafts displayed rejection reactions (significantly less than controls, P = 0.02), and all these grafts suffered irreversible rejection. Only 2 of 11 cornea grafts in group 2 mice had rejection reactions of 3+ intensity or greater, whereas 19 (56%) of grafts in control mice displayed reactions equal to or greater than 3+ (P = 0.04). Orthotopic allografts in eyes of group 3 mice (4 weeks after fragment implantation in AC) experienced fates similar to those in the control group (Fig. 6) : Rejection reactions occurred in 89%, irreversible rejection in 78%, and rejection reactions with scores of 3+ or greater in 44%. These findings suggest that allogeneic corneal implants in place for 2 weeks reduce the risk of rejection in subsequent orthotopic corneal allografts, but that this effect wanes quickly, and by 4 weeks after implantation, no salutary effect on subsequent orthotopic corneal allografts is observed. 

This interpretation is supported by the results of corneal allografts in group 4—mice in which the fragment-containing eye was enucleated at the time the orthotopic graft was placed in the contralateral eye. As is shown in Figure 7 , very few (25%) of these orthotopic grafts showed rejection reactions compared with grafts in the control eyes (P = 0.006), and these were the only grafts that were permanently destroyed. None of these grafts showed reactions of 3+ intensity (compared with control, P = 0.004). These results suggest that the difference in test corneal allograft survivals between groups 2 and 3 may be related to persistence of the corneal fragments in the contralateral eye. Removal of that stimulus correlated in group 4 with improved test graft survival. 

Finally, BALB/c mice that first encountered C57BL/6 alloantigens by way of an intracameral injection of C57BL/6 splenocytes mounted intense rejection reactions when grafted subsequently with orthotopic C57BL/6 corneas (Fig. 8) . Ninety-two percent of these grafts experienced rejection reactions, and most of these reactions occurred early. Thus, intracameral exposure to alloantigens in the form of spleen cells had no mitigating effect on subsequent corneal allograft rejection. 

Discussion

Immune privilege is an important feature of the eye and undoubtedly contributes to the fact that penetrating keratoplasties are highly (albeit not uniformly) successful in humans. Multiple mechanisms have been identified as participating in the creation and maintenance of ocular immune privilege. ¹² Among those mechanisms is the phenomenon of ACAID in which antigens placed in the anterior chamber induce a deviant systemic immune response that is deficient in delayed hypersensitivity and complement-fixing antibodies but replete with other effectors of immunity, including cytotoxic T cells and non–complement-fixing antibodies. ¹⁰ ACAID has been elicited by a wide range of different types of antigens, including the polymorphic molecules called transplantation antigens. We have previously reported that ACAID develops in mice that have accepted allogeneic orthotopic corneal transplants for prolonged periods. ¹³ More recently, Yamada and Streilein ¹⁴ have demonstrated that insertion of segments of allogeneic cornea tissue into the anterior chamber of mouse eyes induced ACAID if the implants were permitted to remain in oculi for longer than 6 weeks. Together, these two sets of observations reveal, on the one hand, that allogeneic corneal tissue is capable of inducing ACAID, but on the other hand, that the tardiness with which ACAID is induced argues against immune deviation’s playing an important role in the early success of orthotopic corneal allografts. 

The results of our present experiments add information to this issue. Implants of allogeneic cornea tissue in the anterior chamber of naive BALB/c mice failed to induce donor-specific delayed hypersensitivity whether the recipients were tested at 2 or 4 weeks after implantation. However, when mice with implants in place for only 2 weeks were then immunized SC with donor alloantigens, a significant proportion of animals displayed only feeble delayed hypersensitivity. Suppressed delayed hypersensitivity of this type was not observed in any recipients immunized 4 weeks after fragment implantation. These findings indicate that the immune impairment that is evoked in approximately 50% of mice by intraocular corneal fragments is transient. The transient nature of the systemic effect and that it was observed in only a proportion of mice bearing intraocular implants may suggest that the systemic effects of intraocular allogeneic corneal implants were trivial. 

Two observations argue against this conclusion and raise the possibility that corneal tissue in the anterior chamber may have a salutary effect on subsequent orthotopic corneal allografts—that is, if the test orthotopic graft is placed on the fellow eye within 2 weeks of implantation of the original tissue. First, histologic examination of implants of mice that displayed vigorous donor-specific delayed hypersensitivity after SC immunization with donor spleen cells revealed that the implants were destroyed, presumably by immune rejection. Histologic evidence of graft deterioration was heralded by clinical evidence that these grafts had undergone stromal melting and were opaque. By contrast, microscopic study of implants in eyes of mice with impaired donor-specific delayed hypersensitivity revealed healthy corneal tissue, virtually indistinguishable from comparably healthy syngeneic corneal implants placed intraocularly. The clinical appearance of these grafts was one of clarity, indistinguishable from that of syngeneic control subjects. These findings argue strongly that impaired delayed hypersensitivity observed in some mice bearing intraocular allogeneic corneal implants could play an important role in the early acceptance and persistent survival of subsequent orthotopic corneal grafts. 

Second, orthotopic corneal allografts grafted in contralateral eyes of mice bearing donor-type cornea implants in the anterior chamber of one eye had an especially high rate of survival, compared with allogeneic corneas grafted in eyes of naive recipients. Moreover, the frequency and intensity of rejection reactions in orthotopic grafts of mice with anterior chamber implants was dramatically reduced compared with those in control subjects. The ability of intraocular allogeneic corneal fragments to promote subsequent orthotopic cornea graft survival was only observed if the orthotopic graft followed the implant by a 2-week interval. If the interval was extended to 4 weeks, survival of orthotopic grafts was not promoted; instead, these grafts showed an even higher incidence of rejection than did grafts placed orthotopically in eyes of control mice. Based on these findings, as well as the results described earlier, we infer that intraocular implants of allogeneic corneal tissue have a profound, albeit transient, capacity to modify the donor-specific host immune response in a direction that interferes with graft rejection. Whether this systemic effect is donor-specific ACAID or not is unclear, because others have reported that ACAID is usually very long lasting once it has been induced. 

The transient nature of the effect of allogeneic corneal implants on subsequent orthotopic corneal allograft survival warrants comment. One of our experiments, in which the fragment-containing eye was enucleated at the time of the orthotopic transplant in the contralateral eye, suggests that the influence of the allogeneic fragment in the anterior chamber that promotes graft survival during the first 2 weeks abruptly changes, and beyond 2 weeks the same fragment has a deleterious effect on subsequent orthotopic corneal allograft survival. This finding resembles those reported by Yamada and Streilein ¹⁴ in which allogeneic corneal fragments inserted in the anterior chamber induced donor-specific delayed hypersensitivity that was evident at 4, but not 2, weeks after implantation. They then produced evidence to suggest that the ability of the fragment to induce donor-specific delayed hypersensitivity was due to the presence of corneal epithelium. Donor-specific delayed hypersensitivity persisted in these recipients only as long as epithelium remained on the fragments. As the fragment epithelium deteriorated (beyond 4 weeks), donor-specific delayed hypersensitivity also disappeared. Of note, corneal fragments stripped of epithelium before implantation failed to induce delayed hypersensitivity at any time after implantation in the anterior chamber. Thus, corneal epithelium appears to promote delayed hypersensitivity induction, but only after the fragment or graft has been in oculus for more than 2 weeks. Emergence of delayed hypersensitivity thereafter interferes with engraftment of orthotopic corneal tissue performed subsequently. 

Whereas allogeneic corneal fragments in the anterior chamber promoted immune deviation and subsequent survival of orthotopic corneal allografts, injection of allogeneic spleen cells into the anterior chamber failed to promote subsequent graft survival. In fact, mice pretreated with C57BL/6 spleen cells through the anterior chamber rejected subsequent C57BL/6 orthotopic cornea grafts at a very high rate and with an intensity that exceeded the rate in unimmunized controls. This result requires comment, because several other laboratories have reported that allogeneic spleen cells injected into the anterior chamber of rats induce ACAID and promote orthotopic corneal allograft survival, ¹⁵ that multiple injections of allogeneic splenocytes into the anterior chamber improves the clinical score of keratoepithelioplasty in mice, ¹⁶ and that allogeneic peritoneal exudate cells treated in vitro with transforming growth factor-β and then injected intravenously promote survival of allogeneic corneas transplanted into high-risk eyes. ¹⁷ ACAID is by no means a complete failure of donor-specific immune responses; instead, it is a selective, antigen-specific immune deficiency in which the mediators of concomitant immunity remain intact. ACAID promotes graft survival only if delayed hypersensitivity is the primary effector mechanism, and antibodies, cytotoxic T cells, and/or other types of specific immune effectors are irrelevant. Rejection of orthotopic corneal grafts in low-risk eyes of immunologically naive mice appears to be mediated almost exclusively by delayed hypersensitivity T cells. ¹⁸ However, in presensitized mice, the precise immune effectors have not been elucidated. We suspect that intraocular injection of allogeneic splenocytes, while inducing ACAID, evokes other immune effectors that are as effective at graft rejection as if the recipient had been specifically sensitized. Our data support this suspicion. 

It may also be relevant that allogeneic corneal fragments that promote subsequent orthotopic corneal allograft survival remain within the chamber for 2 weeks, potentially delivering antigenic signals on a continual basis to the systemic immune system. By contrast, allogeneic spleen cells injected intracamerally are rapidly cleared from the eye (data not shown), and cannot, therefore, provide a continual antigenic stimulus. Yao et al. ¹⁹ have recently reported that maintenance of ACAID requires persistence of the spleen. Perhaps, maintenance of ACAID also requires persistence of antigen. Because allogeneic lymphocytes are rapidly removed from the body by an innate immune mechanism, ²⁰ the inability of allogeneic spleen cells injected intracamerally to sustain an antigenic signal may also help to explain their failure to promote, in a long-lasting fashion, survival of subsequent orthotopic corneal transplants. 

 Submitted for publication October 14, 1998; revised April 14 and September 17, 1999; accepted September 29, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Three fragments (each 1.0 × 0.3 mm) of full-thickness corneas from C57BL/6 donors implanted in the anterior chamber of BALB/c mouse eyes at 14 days after implantation.

Figure 2.

 

Capacity of allogeneic cornea fragments in the anterior chamber to induce delayed hypersensitivity. C57BL/6 cornea fragments were placed in the anterior chamber of normal BALB/c mice. Two or 4 weeks later, the ear pinnae of these mice received injections of 1 × 10⁶ irradiated C57BL/6 spleen cells. Positive control primed mice received 10 × 10⁶ C57BL/6 spleen cells 2 weeks before ear challenge. Anterior chamber control mice received 5 × 10⁵ C57BL/6 spleen cells intracamerally 2 weeks before ear challenge. Negative controls underwent only the ear challenge. Ear-swelling responses were assessed after 24 hours and are shown as mean ± SE. Mean responses compared with positive controls, *P < 0.005, **P < 0.0005.

Figure 3.

 

Capacity of allogeneic cornea fragments in the anterior chamber to induce ACAID. C57BL/6 cornea fragments were placed in the anterior chamber of normal BALB/c mice. Syngeneic controls received cornea fragments from BALB/c donors. Two weeks later these mice and the positive control animals received SC immunization with 10 × 10⁶ C57BL/6 spleen cells. One week later the ear pinnae were challenged with 1 × 10⁶ irradiated C57BL/6 spleen cells. Ear-swelling responses were assessed and segregated into high and low delayed hypersensitivity responses. Mean responses compared with positive control animals, *P < 0.005, **P < 0.0005.

Figure 4.

 

Histologic appearance of corneal fragments in the anterior chamber of mice tested for donor-specific ACAID. Cornea fragments obtained from eyes of mice tested as described in the legend to Figure 3 , fixed, sectioned and stained with hematoxylin and eosin. (A) BALB/c fragments; (B) C57BL/6 fragment from mouse with intact fragments. (C) C57BL/6 fragment from mouse with melted fragments. Top frame is the epithelial side, bottom is the endothelial side. Original magnification:× 200.

Table 1.

 

Summary of Fate of Orthotopic Corneal Allografts

	n	Rejection Reaction
				Early Rejection	Late Rejection	Total	Score 3+ or More at 8 weeks
Group 1	34	47.1	32.3	79.4	55.8
Group 2	11	36.3	0	36.3	18.2
Group 3	9	55.5	33.3	88.8	44.4
Group 4	8	12.5	12.5	25.0	0
Group 5	13	92.3	0	92.3	92.3

 Rejection reactions are a percentage of the total group (n) tested.

Figure 5.

 

Fate of orthotopic corneal allografts in mice with corneal fragments in the anterior chamber for 2 weeks. Two weeks after implantation of C57BL/6 cornea fragments in one eye, C57BL/6 corneas were grafted in the contralateral eye. Clinical score was assessed for each graft at weekly intervals and scored as accepted (•) or rejected earlier than 3 weeks (×).

Figure 6.

 

Fate of orthotopic corneal allografts in mice with corneal fragments in the anterior chamber for 4 weeks. Four weeks after implantation of C57BL/6 cornea fragments in one eye, C57BL/6 corneas were grafted in the contralateral eye. Clinical score was assessed at weekly intervals and scored as accepted (•), rejected earlier than 3 weeks (×), or rejected after 3 weeks (▵).

Figure 7.

 

Fate of orthotopic corneal allografts in mice with corneal fragments in the anterior chamber for 2 weeks. Effect of enucleation on keratoplasty success. Two weeks after implantation of C57BL/6 cornea fragments in one eye, C57BL/6 corneas were grafted in the contralateral eye. The fragment-containing fellow eye was enucleated at the time of keratoplasty. Clinical score was assessed at weekly intervals as in Figure 6 .

Figure 8.

 

Fate of orthotopic corneal allografts in mice that received allogeneic spleen cells in the anterior chamber 2 weeks previously. C57BL/6 spleen cells (5 × 10⁵) were injected into the AC of one eye of BALB/c mice. Two weeks later, C57BL/6 corneas were grafted orthotopically in the fellow eye. Clinical score was assessed at weekly intervals as in Figure 6 .