IOVS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hori, J.
Right arrow Articles by Streilein, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hori, J.
Right arrow Articles by Streilein, J. W.
(Investigative Ophthalmology and Visual Science. 2001;42:720-726.)
© 2001 by The Association for Research in Vision and Ophthalmology, Inc.

Role of Recipient Epithelium in Promoting Survival of Orthotopic Corneal Allografts in Mice

Junko Hori and J. Wayne Streilein

From the Schepens Eye Research Institute, Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
PURPOSE. To determine whether epithelium-deprived corneal allografts covered with syngeneic epithelium display immune privilege in orthotopic transplantation and whether syngeneic epithelium containing antigen-presenting cells nullifies this effect.

METHODS. Epithelium-deprived allogeneic corneas (C57BL/6) and epithelium-deprived allogeneic corneas reconstituted with syngeneic (BALB/c) epithelium (containing or deprived of Langerhans’ cells) were transplanted orthotopically into normal eyes of BALB/c mice. Graft survival was assessed clinically and evaluated histologically.

RESULTS. Epithelium-deprived corneal grafts survived in syngeneic recipients but were swiftly rejected in allogeneic recipients. These allografts incited intense stromal inflammation and neovascularization. Epithelium-deprived allografts that were resurfaced in vivo by syngeneic epithelium derived from immune-incompetent SCID mice also underwent intense acute rejection when placed in normal eyes of BALB/c mice. The epithelium of in vivo resurfaced grafts was replete with Langerhans’ cells. By contrast, most of the epithelium-deprived allografts reconstituted in vitro with fresh, normal BALB/c corneal epithelium survived indefinitely when placed in eyes of BALB/c mice. Similar grafts reconstituted with BALB/c epithelium containing Langerhans’ cells were swiftly rejected.

CONCLUSIONS. Replacement of donor epithelium with Langerhans’ cell-deficient syngeneic epithelium protects orthotopic allogeneic cornea grafts (stroma plus endothelium) from immune-mediated rejection. The presence of an intact, histocompatible layer of corneal epithelium has two important effects on orthotopic corneal allografts: It suppresses nonspecific inflammation and neovascularization within the graft, and it blunts the alloimmunogenicity of the histoincompatible stroma and endothelium.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The capacity of an orthotopic corneal allograft to induce immunity is an expression of the tissue’s inherent alloimmunogenicity. Whereas orthotopic corneal allografts elicit alloimmunity directed at both major and minor transplantation antigens,1 all three layers of the cornea are not believed to contribute equally to the cornea’s alloimmunogenicity. We have recently reported that corneal epithelium confers potent immunogenicity when full-thickness corneal allografts are placed beneath the kidney capsule.2 By contrast, corneal allografts deprived of the epithelial layer are neither immunogenic nor susceptible to rejection when placed at this heterotopic site. Expression of CD95 ligand on corneal endothelium is central to the ability of epithelium-deprived corneal allografts to resist immunologic rejection beneath the kidney capsule.3 Thus, inclusion of the epithelial layer with heterotopic corneal allografts routinely leads to sensitization of the recipient and rapid graft rejection, thereby canceling the protective effect of CD95L on the corneal endothelium.2 We wondered whether the immunologic fate of heterotopic as well as orthotopic corneal allografts might be improved if the epithelium’s contribution to a corneal allograft’s immunogenicity could be blunted.

It has been more than 20 years since it was first reported that cornea as a tissue is virtually devoid of bone-marrow–derived cells bearing class II major histocompatibility antigens.4 At the time, the deficiency of Langerhans’ cells in the corneal epithelium was linked to the reduced ability of corneal allografts to sensitize recipients to donor class II alloantigens.5 More recently, it has been appreciated that sensitization to donor MHC alloantigens plays a relatively minor role in causing rejection of orthotopic corneal allografts, further evidence that the deficit of passenger leukocytes within normal cornea reduces the tissue’s inherent immunogenicity.1 6 7 Thus, the deficit of donor-derived antigen-presenting cells in the normal corneal epithelium argues against the notion that this layer of the tissue serves as a potent immunizing layer of the cornea. Yet, laboratory and clinical experiences point in the opposite direction—to the corneal epithelium’s being central to a graft’s immunogenicity.8

To gain more insight into this issue, we conducted a series of experiments in mice that received orthotopic corneal allografts that were deprived of epithelium or that were reconstituted with corneal epithelium obtained from normal or lightly cauterized donor eyes of syngeneic or allogeneic origin. The results favor the conclusion that the capacity of corneal epithelium to promote sensitization to graft alloantigens resides primarily in the ability of the allogeneic epithelium, directly or indirectly, to induce neovascularization and inflammation in the graft at its attachment to the graft bed. In our experiments, corneal epithelium (syngeneic) that did not possess this allostimulatory capacity actually protected allogeneic grafts comprising stroma and endothelium from the threat of immune rejection.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice and Anesthesia
Male BALB/c (H-2d), C57BL/6 (B6, H-2b), and C.B17 SCID (H-2d) mice were purchased from Taconic Farms (Germantown, NY). BALB/c and C.B17 SCID mice are histocompatible at the major histocompatibility (MHC) locus and at most of the minor histocompatibility loci, whereas both strains of mice differ from C57BL/6 at the MHC and at multiple minor histocompatibility loci. All mice were used at 8 to 12 weeks of age and treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Each mouse was anesthetized by intramuscular injection of a mixture of 3.75 mg ketamine and 0.75 mg xylazine before all surgical procedures.

Preparation of Epithelial Sheet and Epithelium-Deprived Cornea
The corneal epithelial cell layer was peeled off, as an intact sheet, from full-thickness cornea of normal BALB/c and C57BL/6 mice after 1 hour of incubation in 20 mM EDTA at 37°C and then washed with phosphate-buffered saline (PBS). Stroma together with the endothelium component was also washed with PBS, and then it was used as a graft.

Composite Grafts Created by Reconstitution of the Cornea In Vitro
Corneal epithelium as intact sheets were prepared from full-thickness normal or cauterized (2 weeks previously) corneas of BALB/c or C57BL/6 mice. Two-millimeter-diameter corneal stroma plus endothelium prepared from full-thickness normal corneas from C57BL/6 mice was gently covered by similarly sized epithelium sheet under a dissecting microscope. The following reconstituted corneas were used as grafts for BALB/c recipients: reconstituted corneas with normal syngeneic BALB/c-derived epithelium and normal allogeneic C57BL/6-derived stroma plus allogeneic endothelium; reconstituted corneas with normal syngeneic epithelium and normal syngeneic stroma plus endothelium; reconstituted corneas with normal allogeneic epithelium and normal allogeneic stroma plus endothelium; and reconstituted corneas with syngeneic epithelium which had been cauterized previously (and therefore contained Langerhans’ cells) and normal allogeneic stroma plus endothelium. The reconstituted corneas were transplanted to BALB/c recipients orthotopically.

Chimeric Grafts Created by Reconstitution of the Cornea In Vivo
Two-millimeter-diameter full-thickness or epithelium-deprived corneas of C57BL/6 were transplanted orthotopically to C.B17 immunodeficient mice. Three or 8 weeks later, grafted corneas were harvested from C.B17 mice that had accepted full-thickness or epithelium-deprived C57BL/6 orthotopic corneal allografts and then transplanted orthotopically to normal BALB/c mice.

Orthotopic Corneal Transplantation and Graft Evaluation
Penetrating keratoplasty was performed as described previously. Briefly, 2-mm-diameter donor corneas were placed in the same sized recipient bed with eight interrupted sutures (11-0 nylon). Sutures were removed at 8 days after grafting. Orthotopic grafts were observed by slit lamp microscopy at weekly intervals, and assessment of orthotopic corneal graft survival was performed in masked fashion according to a previously described scoring system: 0, clear graft; 1+, minimal superficial nonstromal opacity; 2+, minimal deep stromal opacity with the pupil margin and iris vessels visible; 3+, moderate deep stromal opacity with only the pupil margin visible; 4+, intense deep stromal opacity with the anterior chamber visible; and 5+, maximum stromal opacity with total obscuration of the anterior chamber. Grafts with opacity scores of 2+ or higher after 3 weeks were considered to have been rejected. Each experiment was repeated more than twice.

Statistical Analyses
Corneal graft survival was compared using Kaplan–Meier survival curves and the Breslow–Gegan Wilcoxon test. P < 0.05 were deemed significant.

Histology of Reconstituted Corneal Grafts
Eyes bearing the reconstituted corneal grafts were removed for histologic analysis 4 weeks after transplantation, fixed with 10% formalin, imbedded in paraffin, sectioned, and stained with hematoxylin and eosin.

Assessment of Antigen-Presenting Cells in Chimeric Corneal Tissue
To discern the chimeras of recipient-derived infiltrating cells in the grafts at 3 or 8 weeks after the initial orthotopic corneal grafting from C57BL/6 donors to C.B17 recipients, immunohistochemical studies for I-Ad, F4/80, CD11b, and CD45 were performed on accepted corneal allografts, using fluorescein isothiocyanate (FITC)–labeled rat anti-mouse I-Ad, FITC-labeled rat anti-mouse F4/80, FITC-labeled rat anti-mouse CD11b, and phycoerythrin (PE)-labeled mouse anti-mouse CD45 monoclonal antibodies (PharMingen, San Diego, CA). Graft-bearing whole corneas were removed at 3 or 8 weeks after initial orthotopic corneal grafting from C57BL/6 donors to C.B17 recipients, fixed in acetone for 10 minutes, and incubated in the monoclonal antibody, diluted to 4 µg/ml for 2 hours at room temperature. After a wash in PBS, the sample was mounted on a slide with mounting medium according to the manufacturer’s instruction (Vectastain; Vector, Burlingame, CA), and each layer of the corneal grafts was observed by confocal microscopy.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Fate of Orthotopic Epithelium-Deprived Corneal Grafts
Our first goal was to determine the fate in mouse eyes of corneal grafts deprived of surface epithelium. Corneas were excised from eyes of BALB/c and C57BL/6 donors. The excised tissues were incubated for 1 hour in 20 mM EDTA, after which the intact epithelial layer was easily removed with forceps. Epithelium-deprived corneas as well as full-thickness corneas as control specimens were then grafted orthotopically into the right eyes of BALB/c recipients. The fate of the grafts was assessed clinically at weekly intervals and thereafter for 8 weeks. Although all syngeneic grafts elicited a circumferential neovascular response in the recipient bed, only epithelium-deprived syngeneic grafts showed strong neovascularization responses with the graft stroma. Neovascularization of the stroma of epithelium-deprived grafts was evident as early as 1 week after grafting and reached peak intensity at 2 weeks. At 5 to 6 weeks after grafting, stroma ghost vessels were still evident in these grafts. Despite the intrusion of neovascularization in the stromal of syngeneic epithelium-deprived grafts, the grafts remained perfectly clear throughout the 8-week observation period. Unlike epithelium-deprived syngeneic grafts, full-thickness syngeneic corneal grafts acquired no evidence of stromal neovascularity, and these grafts remained perfectly clear throughout the observation period. Thus, in the absence of alloantigen expression on corneal tissue, epithelium-deprived grafts elicited a vigorous neovascularization response within their own stroma.

The fate of epithelium-deprived corneal allografts also differed from the fate of their epithelium-intact counterparts. Orthotopic corneal allografts deprived of epithelium evoked a brisk stromal neovascular response, similar to epithelium-deprived syngeneic corneal grafts. By contrast, much less neovascularization developed within the stroma of allogeneic grafts with intact epithelial layers. Moreover, epithelium-deprived allografts were rejected more rapidly and more often than allogeneic grafts with an intact epithelium (Fig. 1) . Thus, although epithelium-deprived corneal allografts carry a quantitatively lesser burden of transplantation antigens than full-thickness grafts, epithelium-deprived grafts had an enhanced rate of rejection. These results led us to design experiments to create corneal allografts in which donor epithelium was replaced by epithelium of recipient origin. We wanted to test the hypothesis that corneal allografts covered with syngeneic epithelium at the time of grafting would experience enhanced survival compared with full-thickness corneal allografts.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 1. Fate of full-thickness and epithelium-deprived C57BL/6 corneal allografts in eyes of BALB/c mice. Corneal buttons harvested from eyes of BALB/c and C57BL/6 mice were incubated in EDTA, after which the epithelium was stripped away, leaving stroma and endothelium. These epithelium-deprived grafts were placed orthotopically in eyes of BALB/c mice and their fate assessed clinically. Full-thickness C57BL/6 corneal grafts served as positive control specimens. Epithelium-deprived BALB/c corneal grafts served as a control for the surgical manipulation. (*) Survival of epithelium-deprived allografts was significantly less than survival of full-thickness allografts (P < 0.002).

 
Fate of Epithelium-Deprived Orthotopic Corneal Allografts Reconstituted in Vivo by Epithelium Syngeneic with Recipient
The epithelial surface of orthotopic corneal grafts, whether full-thickness or epithelium-deprived, is rapidly replaced by cells that migrate centripetally from the limbus of the recipient eye.9 The process of resurfacing the epithelium of orthotopic corneal grafts is presumed to be an exaggeration of the normal process by which the normal corneal epithelial surface is constantly renewed.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Based on this knowledge, we generated chimeric corneal grafts by parking epithelium-deprived corneas orthotopically in recipient eyes until the epithelial surface was replenished from the recipient limbus. Thus, epithelium-deprived corneas were prepared from C57BL/6 eyes and placed orthotopically in eyes of C.B17 SCID mice. These recipients have two features that were important in the experiment: First, they are genetically very similar to BALB/c mice (although the strains differ at one locus that may encode a minor antigen), and second, they do not have immunologic competence. As a consequence, they are unable to reject allogeneic grafts, yet they resurface the graft with epithelium histocompatible with BALB/c. Control SCID mice received full-thickness corneal grafts from C57BL/6 donors. Sample grafts were removed at 2 weeks and examined histologically. All grafts were found to be completely covered with an epithelium that had the distinctive features of corneal epithelium—that is, the epithelium-deprived grafts were chimeric, comprising C.B17 SCID epithelium and C57BL/6 stroma and endothelium. After 3 or 8 weeks in residence, the parked corneas were removed, trimmed of any remnant donor tissue at the periphery, and grafted orthotopically into eyes of normal BALB/c mice.

The fates of the grafts, defined clinically, are displayed in Figure 2 . Epithelium-deprived C57BL/6 corneas placed for 3 weeks in C.B17 SCID recipients showed a high rate of rejection in BALB/c eyes. The majority of these grafts were destroyed by 3 weeks, with all grafts rejected by 7 weeks. Similarly, full-thickness C57BL/6 corneas that had been parked orthotopically in C.B17 SCID eyes for 3 and 8 weeks, when grafted orthotopically into eyes of normal BALB/c mice, were the objects of intense and complete immune rejection. Similar chimeric grafts placed orthotopically on C.B17 SCID recipients showed no evidence of rejection (data not shown). Thus, in apparent refutation of the hypothesis being tested, chimeric corneal grafts comprising syngeneic epithelium and allogeneic stromal and endothelium were not protected from immune rejection. On the contrary, these chimeric grafts were more vulnerable to immune rejection than full-thickness cornea allografts prepared from normal eyes (Fig. 1) .



View larger version (23K):
[in this window]
[in a new window]
  		Figure 2. Fate of in vivo–generated chimeric corneal allografts in eyes of BALB/c mice. Epithelium-deprived and full-thickness C57BL/6 corneas were parked for 3 or 8 weeks as orthotopic transplants in eyes of C.B17 recipients. The grafts were then removed and placed orthotopically in eyes of normal BALB/c recipients, and graft outcome was assessed. Differences between survival of three types of grafts are statistically indistinguishable.

 
Description of Bone-Marrow–Derived Recipient Cells in Chimeric Corneal Grafts
Many traumatic stimuli to the ocular surface lead to local inflammation and to infiltration of Langerhans’ cells into the epithelium. Specifically, Yamada et al.16 and Sano et al.17 have recently documented that the epithelia of corneal allografts in residence for 2 weeks or more contain substantial numbers of recipient-derived class II MHC–positive cells. We suspected that the chimeric grafts used in our experiments may have been similarly contaminated with recipient bone-marrow–derived cells and that this may have accounted for their enhanced immunogenicity. To evaluate this possibility, chimeric grafts of C57BL/6 corneas in residence for 3 and 8 weeks on C.B17 SCID mice were removed and evaluated by immunohistochemistry for the presence of I-Ad, F4/80, CD11b, and CD45. The tissues were counterstained with propidium iodide to identify nuclei and were examined by confocal microscopy. When examined at 3 weeks, chimeric grafts contained numerous I-Ad cells with dendritic morphology within the epithelium and in the superficial stroma. A similar pattern was observed in grafts that had been in residence for 8 weeks.

In addition, I-Ad-bearing cells were present within the stroma and the endothelium of these grafts (Figs. 3A 3B, 3C ). Many bone-marrow–derived cells that expressed F4/80 and CD11b were also found in the stroma of chimeric cornea grafts (data not shown). When the tissues were simultaneously stained for CD45 and I-Ad, F4/80, or CD11b, all cells with the latter markers expressed CD45, indicating that the cells detected were of bone marrow origin. Thus, epithelium-deprived corneal grafts that were placed orthotopically in eyes of C.B17 SCID mice not only acquired a covering of recipient epithelium, but that epithelium, as well as the underlying stroma, became infiltrated with bone-marrow–derived cells of recipient origin. It is known that full-thickness corneal allografts that already contain such cells at the time of grafting induce donor-specific delayed hypersensitivity very briskly and undergo early, acute rejection18 19 Therefore, our efforts to create chimeric corneas in vivo by allowing epithelium-deprived allogeneic corneal grafts to acquire recipient corneal epithelial cells failed. Grafts parked in SCID eyes accumulated significant numbers of potent antigen-presenting cells and underwent acute rejection.



View larger version (94K):
[in this window]
[in a new window]
  		Figure 3. I-Ad-positive dendritic cells were present in all layers of the in vivo–generated chimeric corneal allografts. Full-thickness C57BL/6 corneas that were parked for 8 weeks as orthotopic transplants in eyes of C.B17 recipients were harvested and stained with FITC anti-mouse I-Ad antibody and propidium iodide, and each layer was observed by confocal microscopy. I-Ad-positive cells were present in epithelium (A), stroma (B), and endothelium (C).

 
Fate of Epithelium-Deprived Orthotopic Corneal Allografts Reconstituted in Vitro with Syngeneic Epithelium
As a means of avoiding the contamination of chimeric corneal grafts with Langerhans’ cells, we turned to an in vitro approach in an effort to create antigen-presenting-cell–deficient corneal grafts comprising allogeneic stroma and endothelium plus syngeneic epithelium. Normal corneal epithelium is known to be virtually devoid of Langerhans’ cells and therefore is incapable of contributing antigen-presenting cells to composite grafts formed in vitro by uniting normal epithelial sheets with epithelium-deprived stroma and endothelium. To create grafts of this type, corneal buttons were removed from normal eyes of C57BL/6 and BALB/c mice. The grafts were incubated in EDTA-containing medium, and the epithelium was separated as an intact sheet from the stroma. With the use of microsurgical instruments, BALB/c corneal epithelial sheets were then floated gently onto epithelium-deprived C57BL/6 stroma plus endothelium. After 15 minutes, these composite grafts were then sutured orthotopically into normal eyes of BALB/c recipients. As controls for the surgical manipulations, two additional types of composite grafts were prepared: In one, BALB/c epithelium was floated onto BALB/c stromal plus endothelium; in the second, C57BL/6 epithelium was floated onto C57BL/6 stroma plus endothelium. Full-thickness C57BL/6 corneas grafted into normal eyes of BALB/c mice also served as conventional positive control specimens.

The fate of these grafts was observed clinically, and the results are displayed in Figure 4 . Composite grafts composed of BALB/c epithelium and BALB/c stroma plus endothelium displayed a clinical appearance virtually indistinguishable from full-thickness syngeneic grafts. The epithelium of these grafts remained firmly united to the stroma when the sutures were removed at 8 days, and the grafts remained completely clear throughout the 8-week observation period. As expected, a subset of full-thickness C57BL/6 corneas began to experience rejection reactions between 2 to 4 weeks, and approximately 50% of these grafts were judged to be rejected at 8 weeks. A similar fate was observed with composite grafts prepared from C57BL/6 epithelium and C57BL/6 stroma plus endothelium. By contrast, few of the composite grafts comprising BALB/c epithelium and C57BL/6 stroma plus endothelium showed evidence of rejection reactions. At 8 weeks after grafting, only 15% of these grafts were judged to have been rejected.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 4. Fate of epithelium-deprived C57BL/6 corneal allografts reconstituted in vitro with BALB/c (syngeneic) epithelium. Epithelium-deprived BALB/c and C57BL/6 corneas were reconstituted in vitro by layering BALB/c or C57BL/6 epithelium over the stromal surface. These chimeric corneas were grafted orthotopically to eyes of BALB/c recipients. For comparison, the results of full-thickness C57BL/6 corneal allografts in BALB/c eyes, presented in Figure 1 , are included in this figure as a positive control. (*) Survival of epithelium-deprived C57BL/6 corneas reconstituted with BALB/c epithelium was significantly greater than survival of full-thickness C57BL/6 corneas or epithelium-deprived C57BL/6 corneas reconstituted with C57BL/6 epithelium (P < 0.01). Differences between survival of full-thickness C57BL/6 corneas and epithelium-deprived C57BL/6 corneas reconstituted with C57BL/6 epithelium are statistically indistinguishable at 8 weeks.

 
Chimeric corneal allografts that were accepted at 4 weeks were subjected to histologic examination. As revealed by the photomicrographs in Figure 5 , healthy grafts of BALB/c epithelium plus C57BL/6 stroma plus endothelium displayed no evidence of inflammatory infiltrates in the stroma, and no evidence of neovessels was found. By contrast, composite grafts composed of C57BL/6 epithelium layered over C57BL/6 stroma plus endothelium, which were rejected within 4 weeks, revealed intense stromal inflammation and neovascularization, with loss of endothelium. Thus, allogeneic stroma and endothelium grafts that were provided with a superficial covering of fresh, syngeneic epithelium were less vulnerable to rejection in normal eyes of BALB/c mice than composite grafts similarly created in vitro by layering C57BL/6 epithelium over C57BL/6 stroma plus endothelium.



View larger version (78K):
[in this window]
[in a new window]
  		Figure 5. Clinical and histologic appearance of in vitro–generated chimeric cornea grafts. Epithelium derived from C57BL/6 or BALB/c corneas was layered in vitro onto the denuded surface of C57BL/6 stroma–endothelium corneas. These chimeric corneas were grafted orthotopically in eyes of BALB/c recipients. (A) Clinical appearance of chimeric graft (BALB/c epithelium plus C57BL/6 stroma-endothelium) at 4 weeks. Histologic appearance of chimeric grafts at 4 weeks: (B) BALB/c epithelium plus C57BL/6 stroma–endothelium; (C) BALB/c epithelium plus C57BL/6 stroma-endothelium; (D) C57BL/6 epithelium plus C57BL/6 stroma-endothelium. (*), Recipient-graft junction. Magnification, (B, D) x10; (C) x40.

 
Influence of Langerhans Cells on Fate of Composite Corneal Allografts Reconstituted In Vitro
In the previous experiments, the syngeneic epithelium used in vitro to resurface allogeneic stroma plus endothelium grafts was obtained from eyes of normal BALB/c mice. To illuminate the role of Langerhans’ cells in promoting allograft immunogenicity and vulnerability to rejection, the next experiments attempted to recreate for in vitro–generated chimeric grafts the enhanced alloimmunogenicity found with in vivo–generated chimeric grafts. Our strategy was to use epithelium removed from donor corneas prepared from BALB/c eyes that had been lightly cauterized, a procedure that draws Langerhans’ cells into the corneal epithelium within 7 to 14 days.20 21 22 Light cautery was applied to the central corneal surface of the right eyes of BALB/c mice. Two weeks later (when the epithelium of these corneas contained numerous I-A+ Langerhans’ cells), the eyes were removed, central corneal buttons were prepared, and the buttons were incubated in EDTA-containing medium. Epithelial sheets were then removed and floated onto epithelium-deprived corneas prepared from eyes of normal C57BL/6 mice. For control, epithelial sheets from normal, noncauterized BALB/c eyes were placed on epithelium-deprived C57BL/6 corneas. These composite grafts were placed orthotopically on BALB/c eyes, and their clinical fate was evaluated. As presented in Figure 6 , composite grafts in which the epithelium was obtained from cauterized donor eyes were rejected in an acute fashion, and most of these grafts were opaque by 6 weeks. By contrast, grafts composed of normal BALB/c epithelium layered on C57BL/6 stroma plus endothelium had a very low incidence of rejection during the 8-week observation interval. We conclude that antigen-presenting cells within the epithelium can rob a composite cornea graft (syngeneic epithelium layered onto allogeneic stroma plus endothelium) of its ability to avoid immune rejection.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 6. Comparison of fates of epithelium-deprived C57BL/6 corneal allografts reconstituted in vitro with normal or Langerhans’-cell–containing BALB/c epithelium. Central corneas of eyes of BALB/c mice were lightly cauterized. After 2 weeks, the epithelial layer was removed and layered in vitro onto the surface of denuded C57BL/6 stroma–endothelium corneas. Similar grafts were prepared with epithelium obtained from normal BALB/c eyes. (This experiment was performed simultaneously with the experiment described in Fig. 4 .) These chimeric grafts were placed orthotopically in eyes of normal BALB/c recipients and their survival assessed clinically. (*), Survival pattern significantly different from that of chimeric grafts comprising normal BALB/c epithelium and C57BL/6 stroma–endothelium (P < 0.03).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Epithelial cells, as components of solid-tissue allografts have long been known to be potently immunogenic. The cornea as a graft is no exception. Corneal epithelial cells strongly express class I antigens encoded by genes within the MHC,23 24 25 and they also sensitize recipients to minor transplantation antigens.6 7 When the epithelium is removed from normal corneal tissue, and the tissue is then implanted as an allograft beneath the kidney capsule, it fails to sensitize its recipient and resists immune rejection indefinitely.3 In theory, removal of the epithelium from an orthotopic corneal allograft should similarly improve its fate. The results of experiments reported here substantiate this theoretical implication, and at the same time they reveal unanticipated dimensions of epithelial cell contribution to corneal immunogenicity.

Our most dramatic finding is that composite corneal grafts, composed of syngeneic normal epithelium layered in vitro on allogeneic stroma plus endothelium, displayed a significantly reduced vulnerability to rejection when placed in eyes of normal BALB/c mice. More than 85% of these grafts remained perfectly clear after 8 weeks in residence, compared with an acceptance rate of only 40% to 50% for full-thickness corneal allografts. In fact, allogeneic grafts of stroma plus endothelium that were covered in vitro with epithelium syngeneic with the recipient survived as well in eyes as did similar grafts placed heterotopically beneath the kidney capsule. Although the reasons for the improved fate of these chimeric grafts are undoubtedly multifactorial, our results emphasize the importance of an intact covering of epithelium in determining the fate of corneal allografts.

Replacing donor epithelium with recipient epithelium necessarily reduces the alloantigenic load of the graft, which would be expected to express itself in reduced vulnerability to rejection. If this were the only factor responsible for the improved survival of chimeric grafts, however, then corneal allografts comprising stroma and endothelium alone (referred to as epithelium-deprived) should have been similarly well tolerated. On the contrary, epithelium-deprived corneal allografts placed in normal BALB/c eyes had exceptionally rapid and universal rejection, far more intense than that experienced by full-thickness allografts. Clinical inspection of epithelium-deprived orthotopic cornea grafts revealed that these grafts, even when syngeneic with the recipient, incited intense neovascularization in the recipient bed. This indicates that an intact epithelial layer can inhibit the development of inflammation and neovascularization within a graft by a mechanism that is unrelated to immunity. Nonetheless, once intense inflammation and neovascularization is elicited in an epithelium-deprived graft, we suspect that this situation results in an enhanced ability of immune effector cells and molecules to gain access to the graft and to initiate its destruction. Stulting et al.26 showed similar findings in their clinical trial to determine the effect of epithelial removal on graft survival. Their results indicate the possibility of enhanced graft failure by epithelial removal.

Our experiments do not indicate either the nature or the source of factors that promote neovascularization in epithelium-deprived grafts. Because in vitro–prepared composite grafts of syngeneic epithelium layered over syngeneic stroma and endothelium failed to promote neovascularization to a similar degree when grafted orthotopically, it is possible that the epithelium may be a source of factors that can suppress neovascularization in the stroma. It is well known that removal of epithelium from the normal cornea has deleterious effects on the underlying stroma, including the triggering of apoptosis among superficial keratocytes.27 Injury to stromal cells after removal of the epithelium may lead to the secretion of factors that promote angiogenesis,28 29 and this may account for the strong neovascularization observed in epithelium-deprived corneal grafts. If these considerations are relevant, we surmise that by simply covering the raw stroma with a layer of epithelium before grafting suffices to suppress subsequent angiogenesis and inflammation and eventually inhibits immunity and graft rejection.

Clinical and histologic inspection of epithelium-deprived corneal grafts after residing in eyes of SCID mice for 3 or 8 weeks revealed that the surface had been repopulated with recipient-derived epithelium. It also revealed that this epithelial layer was replete with class II MHC-bearing Langerhans’ cells. There is ample evidence to indicate that corneal allografts that contain Langerhans’ cells in the epithelium are intensely alloimmunogenic, inducing vigorous delayed hypersensitivity that renders them susceptible to acute, irrevocable rejection.18 19 We suspect that the capacity of infiltrating recipient Langerhans’ cells to promote delayed hypersensitivity to graft-derived antigens accounts for the enhanced vulnerability of these in vivo parked grafts to rejection when they were grafted back onto normal BALB/c eyes. The fate of in vitro–created chimeric grafts composed of epithelium derived from cauterized eyes supports this contention. We observed that composite grafts created in vitro by layering syngeneic epithelium over allogeneic stroma plus endothelium were rapidly rejected if the epithelium was harvested from donor eyes into which Langerhans’ cells had already infiltrated in response to prior cauterization of the corneal surface. When prepared in this manner, in vitro–created composite grafts were found to be as susceptible to immune rejection as chimeric grafts generated in vivo. These results emphasize the power of infiltrating bone-marrow–derived cells to prejudice the fate of orthotopic cornea grafts. Corneas deficient in bone-marrow–derived cells are much less vulnerable to immune rejection.

There is a need to improve the rate of survival of orthotopic corneal transplants in the clinic. Recent advances in our understanding of the pathogenesis of corneal allograft failure in murine model systems are beginning to point toward strategies with potential clinical merit. In the present study, as well as in many other studies, the evidence points to the corneal epithelium as a key factor in dictating graft outcome. First, the epithelium is potently immunogenic and has the capacity to eclipse the inherent immune privilege residing in the stroma and endothelium.2 Even when the epithelium is devoid of Langerhans’ cells, its immunogenicity is overwhelming. Therefore, elimination of an epithelium from a graft significantly reduces its immunogenicity and thereby its vulnerability to immune rejection—-a result recently observed when epithelium-deprived grafts were placed at a heterotopic site.2 3 Second, the epithelium, as an intact surface on a cornea graft, is a potent inhibitor of inflammation and neovascularization of the underlying stroma, as well as of the graft bed. In the absence of an intact epithelium, both the graft stroma and the graft bed become intolerably inflamed and neovascularized. However, when the surface of a corneal graft is provided by a layer of epithelium that is syngeneic with the recipient, inflammation in the graft stroma and in the graft bed is much reduced, and grafts of this type enjoy a high degree of success. These realizations suggest that modification of human corneas for grafting in such a way that donor epithelium is replaced with bone-marrow-cell–deficient epithelium syngeneic with the recipient may have a salutary effect on the outcome of penetrating keratoplasty.


   Acknowledgements
 
The authors thank Jacqueline M. Doherty and Jian Gu for their support.


   Footnotes
 
Supported in part by National Institutes of Health Grant EY10765. JWS is a recipient of a Research to Prevent Blindness Senior Scientific Investigator Award.

Submitted for publication June 22, 2000; revised November 1, 2000; accepted November 8, 2000.

Commercial relationships policy: N.

Corresponding author: J. Wayne Streilein, Schepens Eye Research Institute, 20 Staniford Street, Boston, MA 02114. waynes{at}vision.eri.harvard.edu


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sonoda, Y, Streilein, JW (1992) Orthotopic corneal transplantation in mice: evidence that the immunogenic rules of rejection do not apply Transplantation 54,694-704[Medline][Order article via Infotrieve]
  2. Hori, J, Joyce, N, Streilein, JW (2000) Immune privilege and immunogenicity reside among layers of the mouse cornea Invest Ophthalmol Vis Sci 41,3032-3042[Abstract/Free Full Text]
  3. Hori, J, Joyce, N, Streilein, JW (2000) Epithelium-deficient corneal allografts display immune privilege beneath the kidney capsule Invest Ophthalmol Vis Sci 41,443-452[Abstract/Free Full Text]
  4. Streilein, JW, Toews, GB, Bergstresser, PR (1979) Corneal allografts fail to express Ia antigens Nature 282,326-327[Medline][Order article via Infotrieve]
  5. Rubsamen, PE, McCulley, J, Bergstresser, PR, Streilein, JW (1984) On the Ia immunogenicity of mouse corneal allografts infiltrated with Langerhans’ cells Invest Ophthalmol Vis Sci 25,513-518[Abstract/Free Full Text]
  6. Sonoda, Y, Sano, Y, Ksander, B, Streilein, JW (1995) Characterization of cell-mediated immune responses elicited by orthotopic corneal allografts in mice Invest Ophthalmol Vis Sci 36,427-434[Abstract/Free Full Text]
  7. Sano, Y, Ksander, BR, Streilein, JW (1996) Minor H, rather than MHC, alloantigens offer the greater barrier to successful orthotopic corneal transplantation in mice Transplant Immunol 4,53-56[Medline][Order article via Infotrieve]
  8. Alldredge, OC, Krachmer, JH (1981) Clinical types of corneal transplant rejection: their manifestations, frequency, preoperative correlates, and treatment Arch Ophthalmol 99,599-604[Abstract]
  9. Hori, J, Streilein, JW (1981) Fate of donor cells and replacement by recipient cells after orthotopic corneal transplantation in mice. [ARVO Abstract] Invest Ophthalmol Vis Sci 41(4),S668Abstract nr 3551
  10. Cenedella, RJ, Fleschner, CR (1990) Kinetics of corneal epithelium turn over in vivo: studies of lovastatin Invest Ophthalmol Vis Sci 31,1957-1962[Abstract/Free Full Text]
  11. Lamprecht, J. (1990) Symmetric and asymmetric cell division in rat corneal epithelium Cell Tossu Kinet 23,203-216
  12. Sharma, A, Coles, WH (1989) Kinetics of corneal epithelial maintenance and graft loss: a population balance model Invest Ophthalmol Vis Sci 30,1962-1971[Abstract/Free Full Text]
  13. Rodrigues, M, Ben-Zvi, A, Krachmer, J, Schermer, A, Sun, TT (1987) Suprabasal expression of 64-kilodalton keratin (no. 3) in developing human corneal epithelium Differentiation 34,60-67[Medline][Order article via Infotrieve]
  14. Cotsarelis, G, Cheng, SZ, Dong, G, Sun, TT, Lavker, RM (1989) Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells Cell 57,201-209[Medline][Order article via Infotrieve]
  15. Thoft, RA, Friend, J. (1983) The x, y, z hypothesis of corneal epithelial maintenance Invest Ophthalmol Vis Sci 24,1442-1443[Free Full Text]
  16. Yamada, J, Dana, MR, Zhu, SN, Alard, P, Streilein, JW (1998) Interleukin 1 receptor antagonist suppresses allosensitization in corneal transplantation Arch Ophthalmol 116,1351-1357[Abstract/Free Full Text]
  17. Sano, Y, Ksander, BR, Streilein, JW (2000) Langerhans cells, orthotopic corneal allografts, and direct and indirect pathways of T-cell allorecognition Invest Ophthalmol Vis Sci 41,1422-1431[Abstract/Free Full Text]
  18. Niederkorn, JY (1995) Effects of cytokine-induced migration of Langerhans cells on corneal allograft survival Eye 9,215-218
  19. Peeler, JS, Niederkorn, JY (1986) Antigen presentation by Langerhans cells in vivo: donor derived Ia+ Langerhans cells are required for induction of delayed type hypersensitivity but not for cytotoxic T lymphocyte responses to alloantigens J Immunol 136,4362-4371[Abstract]
  20. Williamson, JSP, DiMarco, S, Streilein, JW (1987) Immunobiology of Langerhans cells on the ocular surface Invest Ophthalmol Vis Sci 28,1527-1532[Abstract/Free Full Text]
  21. Dana, MR, Dai, R, Zhu, S, Yamada, J, Streilein, JW (1998) Interleukin-1 receptor antagonist suppresses Langerhans cell activity and promotes ocular immune privilege Invest Ophthalmol Vis Sci 39,70-77[Abstract/Free Full Text]
  22. Dekaris, I, Zhu, SN, Dana, MR (1999) TNF-alpha regulates corneal Langerhans cell migration J Immunol 162,4235-4239[Abstract/Free Full Text]
  23. Whitsett, CF, Stulting, RD (1984) The distribution of HLA antigens on human corneal tissue Invest Ophthalmol Vis Sci 25,519-524[Abstract/Free Full Text]
  24. Wang, HM, Kaplan, HJ, Chan, WC, Johnson, M. (1987) The distribution and ontogeny of MHC antigens in murine ocular tissue Invest Ophthalmol Vis Sci 28,1383-1389[Abstract/Free Full Text]
  25. Treseler, PA, Sanfilippo, F. (1986) The expression of major histocompatibility complex and leukocyte antigens by cells in the rat cornea Transplantation 41,248-252[Medline][Order article via Infotrieve]
  26. Stulting, RD, Waring, GO, Bridges, WZ, Cavanagh, HD (1988) Effect of donor epithelium on corneal transplant survival Ophthalmology 95,800-812
  27. Wilson, SE, Liu, JJ, Mohan, RR (1999) Stromal-epithelium interactions in the cornea Prog Retinal Eye Res 18,293-309[Medline][Order article via Infotrieve]
  28. Kruse, FE, Volcker, HE (1997) Stem cells, wound healing, growth factors, and angiogenesis in the cornea Curr Opin Ophthalmol 8,46-54
  29. Edelman, JL, Castro, MR, Wen, Y. (1999) Correlation of VEGF expression by leukocytes with the growth and regression of blood vessels in the rat cornea Invest Ophthalmol Vis Sci 40,1112-1123[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
J. Hori, M. Wang, M. Miyashita, K. Tanemoto, H. Takahashi, T. Takemori, K. Okumura, H. Yagita, and M. Azuma
B7-H1-Induced Apoptosis as a Mechanism of Immune Privilege of Corneal Allografts
J. Immunol., November 1, 2006; 177(9): 5928 - 5935.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
C. Cursiefen, L. Chen, M. Saint-Geniez, P. Hamrah, Y. Jin, S. Rashid, B. Pytowski, K. Persaud, Y. Wu, J. W. Streilein, et al.
Nonvascular VEGF receptor 3 expression by corneal epithelium maintains avascularity and vision
PNAS, July 25, 2006; 103(30): 11405 - 11410.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Ophthalmol.Home page
R Maini and J K G Dart
The management of severe sclerokeratitis following corneal transplantation surgery
Br. J. Ophthalmol., July 1, 2006; 90(7): 918 - 920.
[Full Text] [PDF]

Home page
 Br. J. Ophthalmol.Home page
J Plskova, L Kuffova, M Filipec, V Holan, and J V Forrester
Quantitative evaluation of the corneal endothelium in the mouse after grafting
Br. J. Ophthalmol., September 1, 2004; 88(9): 1209 - 1216.
[Abstract] [Full Text] [PDF]

Home page
 IOVSHome page
H. Osawa, K. Maruyama, and J. W. Streilein
CD95 Ligand Expression on Corneal Epithelium and Endothelium Influences the Fates of Orthotopic and Heterotopic Corneal Allografts in Mice
Invest. Ophthalmol. Vis. Sci., June 1, 2004; 45(6): 1908 - 1915.
[Abstract] [Full Text] [PDF]

Home page
 IOVSHome page
J. Hori and J. W. Streilein
Survival in High-Risk Eyes of Epithelium-Deprived Orthotopic Corneal Allografts Reconstituted In Vitro with Syngeneic Epithelium
Invest. Ophthalmol. Vis. Sci., February 1, 2003; 44(2): 658 - 664.
[Abstract] [Full Text] [PDF]

Home page
 IOVSHome page
F. C. Figueiredo, S. M. Nicholls, and D. L. Easty
Corneal Epithelial Rejection in the Rat
Invest. Ophthalmol. Vis. Sci., March 1, 2002; 43(3): 729 - 736.
[Abstract] [Full Text] [PDF]

Home page
 IOVSHome page
J. Hori and J. W. Streilein
Dynamics of Donor Cell Persistence and Recipient Cell Replacement in Orthotopic Corneal Allografts in Mice
Invest. Ophthalmol. Vis. Sci., July 1, 2001; 42(8): 1820 - 1828.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hori, J.
Right arrow Articles by Streilein, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hori, J.
Right arrow Articles by Streilein, J. W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS