High risk corneal allografts and why they lose their immune privilege

Abstract

Purpose of review

Corneal allografts are routinely performed without HLA typing or systemic immunosuppressive drugs. However, certain conditions create high risks for immune rejection. This review discusses recent insights into the mechanisms that rob the corneal allograft of its immune privilege.

Recent findings

Studies in mice have revealed that stimuli that induce new blood vessel growth in the cornea also elicit proliferation of lymph vessels. Lymph vessels facilitate migration of antigen presenting cells to regional lymph nodes where they induce alloimmune responses. The presence of blood vessels in the corneal graft bed creates a unique chemokine milieu that stimulates recruitment of sensitized lymphocytes into the corneal allograft. Other data indicate that corneal allograft survival is closely associated with Foxp3 expression in CD4⁺CD25⁺Foxp3⁺ T regulatory cells (Tregs), while reduced expression of Foxp3 in T regs creates a high risk for graft rejection. Recent evidence indicates that allergic diseases have a profound impact on the immune response and produce a dramatic increase in corneal allograft rejection.

Summary

Understanding the underlying mechanisms that create “high risk” hosts may provide important therapeutic targets for restoring immune privilege of corneal allografts and enhancing their survival.

Introduction

Cornea grafting has been performed on humans for over 100 years and remains the most common and arguably, the most successful form of organ transplantation [1–4]. First-time corneal transplants are routinely performed without HLA matching or systemic immunosuppressive drugs. Patients who require transplants because of corneal developmental anomalies, such as keratoconus, do not have inflamed or vascularized corneal graft beds and have an exceptionally high acceptance rate that often exceeds 90%. In these patients topical application of corticosteroids is usually all that is needed to keep the immune system at bay [5].

Immune privilege of corneal allografts

lmmune privilege of the eye was recognized over 50 years ago by Billingham and Medawar who noted the prolonged survival of skin allografts placed into the anterior chamber of the eye and the acceptance of orthotopic corneal allografts placed onto the eyes of rabbits [6,7]. Medawar understood the significance of these observations and coined the term “immune privilege” to reflect the cornea’s apparent exemption from the laws of transplantation [7].

Tangible evidence of immune privilege is shown in rodents in which long-term survival of fully allogeneic corneal allografts often exceeds 50% even in the absence of topical corticosteroids [2–4]. By contrast, other categories of allografts, such as skin transplants, experience a 100% incidence of immune rejection.

Although corneal transplantation has been performed on experimental animals since 1837 [8], it was not until the development of rodent models of penetrating keratoplasty that the immunological basis of corneal allograft rejection was fully appreciated [9,10]. The past 20 years of animal research have provided keen insights into the processes that provide corneal allografts with immune privilege [1–3,11–13]. Immune privilege of corneal allografts is the product of three fundamental adaptations that: a) block the induction of destructive alloimmune responses; b) deviate alloimmune responses toward a tolerogenic pathway; and c) block expression of immune effector elements at the graft/host interface (Table 1).

Table 1.

Conditions that promote immune privilege of corneal allografts

Condition	Mode of action	Effect on immune privilege
Absence of lymph vessels	Prevents APC migration to regional lymph nodes	Prevents induction of immune responses
FasL	Induces apoptosis of activated T cells at the graft–host interface	Blocks expression of immune responses
PD-L1	Inhibits T cell proliferation and induces T cell apoptosis at the graft–host interface	Blocks expression of immune responses
ACAID	Suppresses DTH responses and deviates alloantibody responses to noncomplement-fixing isotypes	Prevents both induction and expression of immune responses
Tregs	Suppress immune responses	Prevent both induction and expression of immune responses

ACAID, anterior chamber-associated immune deviation; DTH, delayed-type hypersensitivity; FasL, Fas ligand (CD95L); PD-L1, programmed death ligand 1; Tregs, T regulatory cells.

Role of immune deviation and T regulatory cells in corneal allograft survival

Antigens placed into the anterior chamber (AC) elicit a form of systemic immune tolerance called anterior chamber-associated immune deviation (ACAID), which is characterized by the antigen-specific down-regulation of delayed-type hypersensitivity (DTH) responses and a shifting of the antibody response from complement-fixing to non-complement fixing isotypes [13–16]. Orthotopic corneal allografts are in direct contact with the AC and it has been suggested that corneal antigens are sloughed into the AC during corneal transplantation and induce ACAID [13]. In support of this is the observation that long-term corneal allograft survival in mice closely correlates with the presence of antigen-specific suppression of DTH responses to donor alloantigens that closely resembles ACAID [13,16,17]. Likewise, manipulations that block the induction of ACAID result in a significant increase in corneal allograft rejection [13,18–21].

Recent findings have shed light on the role of T regulatory cells in maintaining immune privilege of corneal allografts. Examination of the draining lymph nodes in mice that had either rejected or accepted orthotopic corneal allografts revealed that both categories of mice expressed equal numbers of CD4⁺CD25⁺Foxp3⁺ T regulatory cells (T regs) [22]. However, T regs isolated from mice with accepted corneal allografts (= acceptors) displayed ~50% higher levels of Foxp3 than T regs in rejector mice. Moreover, Tregs from acceptors were significantly more effective in suppressing T cell proliferation and produced up to 3-fold more suppressive cytokines compared to T regs from rejectors. Adoptively transferring T regs from acceptors into naïve recipients resulted in 67% long-term corneal allograft survival compared to 33% graft survival in mice that received T regs from graft rejectors. Thus, long-term survival of corneal allografts rests on the development of CD4⁺CD25⁺Foxp3⁺ and the level of Foxp3 expression in those T regs.

Effect of lymph and blood vessels on the immune privilege of corneal allografts

One of the most recognizable properties of the cornea is the conspicuous absence of blood vessels. Less apparent, but equally important, is the exclusion of lymph vessels in the corneal graft bed [23–26]. It was widely believed that the presence of blood vessels increased the risk for immune rejection by providing conduits for egression of alloantigens to the peripheral immune apparatus and by also facilitating the migration of circulating effector immune elements into the corneal allograft. However, recent findings indicate that many of the stimuli that induce new blood vessels in the graft bed coincidentally stimulate migration and penetration of lymph vessels along with the blood vessels.

Inserting sutures into the cornea is a powerful stimulus for inducing new blood and lymph vessels. However, administration of an antagonist of α5β1 integrin selectively inhibits the induction of lymphangiogenesis while preserving hemangiogenesis in response to intracorneal sutures [26]. Importantly, blocking lymphangiogenesis reduces the incidence of corneal allograft rejection to levels found in mice with avascular corneal graft beds, even though the corneal allografts in the former experiment were placed in graft beds containing blood vessels.

VEGF-C and VEGF-D bind to VEGF receptor 3 (VEGFR3) and induce blood and lymph vessel formation in the cornea [24,27]. Soluble VEGF receptors suppress both hemangiogenesis and lymphangiogenesis [25]. It has recently been shown that corneal epithelial and stromal cells secrete a soluble form of VEGFR2, which blocks VEGF-C and prevents lymphangiogenesis in the cornea without affecting hemangiogenesis [23]. Local production of VEGFR2 by the corneal epithelium and stroma preserves the avascularity of the cornea. The importance of the soluble form of VEGFR2 in preventing lymph vessel invasion of the cornea was demonstrated in studies in which tissue-specific expression of soluble VEGFR2 was ablated in the cornea using Cre-loxP technology. Mice unable to express soluble VEGFR2 at birth developed corneas that were densely supplied with lymphatic vessels but were devoid of blood vessels [23]. Moreover, administration of soluble VEGFR2 inhibited lymph vessel formation in normal mice and enhanced corneal allograft survival even if the corneal graft beds had a dense network of blood vessels that were induced by the sutures [23].

Lymph vessels rob the corneal allograft of its immune privilege by providing conduits for antigen presenting cells to traffic from the graft bed to the regional lymph node where they induce the activation and clonal expansion of alloantigen-specific T cells. Activated T cells subsequently migrate to the graft bed and initiate graft rejection. Recruitment of allospecific effector T cells to the allograft is a crucial step in the rejection of vascularized organ allografts and is influenced by the chemokines that emanate from the allograft [28–34]. Amescua and coworkers hypothesized that a similar condition might occur in “high-risk” corneal allografts in which the insertion of sutures induced a luxuriant growth of blood vessels [35]. They found that “high-risk” vascularized corneal allografts produced the T cell chemokine, CXCL1/KC, which in turn stimulated the production of other T cell chemokines, CXCL9/Mig and CXCL10/IP10, which are intimately involved in recruitment of allospecific T cells into vascularized grafts [28–34]. Moreover, increased levels of CXCL-1/KC that were found in vascularized “high risk” corneal allografts were not present in non-vascularized, normal-risk corneal allografts. “High-risk” hosts treated with neutralizing anti-CXCL-1/KC antibody behaved like normal risk hosts with avascular graft beds. Likewise, administration of CXCL-1/KC into the corneal allograft converted low-risk hosts to a high-risk phenotype resulting in 100% corneal allograft rejection, thereby confirming the important effect of the chemokine milieu in abrogating the immune privilege of corneal allografts. Thus, the presence of blood and lymph vessels in “high-risk” hosts also ablates immune privilege of corneal allografts by promoting migration, recruitment, and infiltration of immune effector elements into the corneal allograft and thus, enhances the efferent arm of the immune response.

Allergic diseases as newly recognized risk factors for corneal allograft rejection

It was previously believed that the rejection of corneal transplants, was solely mediated by CD4⁺ Th1 cells and that tilting the immune response toward a Th2 pathway would promote graft survival. However, recent evidence not only refutes this proposition, but suggests that the opposite may occur. Deviating the alloimmune response to a Th2 immune response by elimination of the Th1 cytokine, interferon-γ (IFN-γ), or by inducing allergic diseases, denies the corneal allograft its immune privilege and promotes corneal allograft rejection [36–40]. IFN-γ knockout (KO) mice or wild-type mice treated with anti-IFN-γ antibody reject 90–100% of their corneal allografts compared to a 50% incidence of rejection in normal mice [38].

Clinicians have noted that patients with severe ocular allergies have an elevated risk for corneal graft rejection [41–43]. It was thought that the increased rejection of corneal allograft that was associated with conjunctivitis was due to the effects of a “hot eye” and inflammation produced by local allergic responses. However, studies in mice indicate that allergic diseases do in fact exacerbate corneal allograft rejection, but the mechanism is through perturbation in the systemic alloimmune response and not to local effect as previously suspected [36,37,39]. Mice with allergic conjunctivitis that was intentionally limited to one eye, experienced a >90% incidence of corneal allograft rejection when a corneal allograft was placed into the contralateral eye that was not expressing allergic conjunctivitis [36]. Moreover, mice with airway hyperreactivity (AHR), which is a model of allergic asthma, reject 90–100% of the corneal allografts compared to a 50% incidence of rejection in non-allergic mice [40]. The increased corneal allograft rejection in mice with allergic diseases in different organs (i.e., lungs or contralateral eye) is further evidence that allergic diseases exert a systemic, rather than a local effect in corneal allograft rejection.

Recent investigations have shown that the increased rejection of corneal allografts in mice with allergic conjunctivitis and AHR is limited to allergic diseases of mucosal tissues, as it does not occur in mice with cutaneous immediate hypersensitivity [39]. The notion that classical Th2 immune responses, such as allergic disease, are monolithic and occur in the absence of the Th1 cytokine, IFN-γ, is losing favor and there is mounting evidence that IFN-γ is necessary for full expression of Th2 diseases [44–46]. The exacerbation of corneal allograft rejection that occurs in allergic conjunctivitis appears to also require Th1 cells. Adoptively transferring unfractionated CD4⁺ T cells from allergic mice to T cell-deficient nude mice induced 100% corneal allograft rejection compared to 0% rejection in nude mice that did not receive CD4⁺ T cells [39]. However, adoptive transfer of CD4⁺ Th1 cells alone or CD4⁺ Th2 cells alone resulted in 70% and 20% rejection respectively, while combining Th1 and Th2 cells produced 100% graft rejection. Interestingly, administration of exogenous IFN-γ could substitute for Th1 cells and produced 100% corneal allograft rejection when combined with CD4⁺ Th2 cells. Thus, allergic diseases of mucosal surfaces exacerbate corneal allograft rejection by activating both Th1 and Th2 alloimmune responses and represent a new risk factor for corneal allograft rejection (Table 2). The mechanisms responsible for the increased incidence and tempo of rejection remain to be elucidated, but once understood, could provide important clues for restoring the immune privilege of corneal allografts in atopic patients.

Table 2.

Factors that abolish immune privilege of corneal allografts

Factor	Mode of action	Effect on alloimmune response
Lymph vessels in graft bed	Provide conduits for APC trafficking to regional lymph nodes	Promotes induction of alloimmune response
Blood vessels in graft	Elevation of chemokine expression	Promotes recruitment and activation of primed immune T cells into the corneal allograft
Mucosal allergic diseases	Unknown	Activates both Th1 and Th2 alloimmunity
Splenectomy	Prevents induction of ACAID	Prevents generation of ACAID Tregs
Depletion of γδ T cells	Prevents induction of ACAID	Prevents generation of ACAID Tregs

ACAID, anterior chamber-associated immune deviation; APC, antigen-presenting cell; Tregs, T regulatory cells.

Conclusions

Mouse studies have provided a wealth of information regarding the immunobiology and immune privilege of corneal allografts. The association between vascularized graft beds and increased corneal allograft rejection has been recognized for over a half century. However, it was only recently that the role of lymph vessels was demonstrated. Lymph vessels enhance the induction of immune responses, while the presence of blood vessels facilitates recruitment and migration of sensitized immune effector cells into the graft. Recognition of this dichotomy opens the door for targeted therapy for enhancing corneal allograft survival.

The role of CD4⁺CD25⁺Foxp3⁺ T regs in preventing corneal allograft rejection has been recently demonstrated. However, the mere expression of Foxp3 alone is not sufficient for promoting corneal allograft survival; instead it is the intensity of Foxp3 expression and the functional properties of T regs that correlate with corneal allograft survival. These insights may provide clues for developing new strategies for restoring immune privilege in the high-risk host.

The last 2–3 years have witnessed the characterization of allergic diseases as a new risk factor for corneal allograft rejection. These studies have revealed that Th2-based immune responses in mucosal tissues dramatically increase the risk for corneal graft rejection. It will be important to determine the underlying mechanisms that are responsible for the increased tempo and incidence of corneal allograft rejection that occurs in hosts with mucosal allergic diseases.