Ординатура / Офтальмология / Английские материалы / Ocular Periphery and Disorders_Dartt, Bex, Amore_2011

mechanisms driving activation and differentiation of these cells are currently unknown. Following ocular surface insults, it is suspected that Tregs function to suppress inflammation by dampening T-cell priming in the lymphoid organs, and inhibit T-cell effector function within inflamed ocular surface tissues. The immunosuppressive properties of Tregs are likely to occur by cell contact-dependent, for example, Treg:APC and/or Treg:T cell, and cell contactindependent mechanisms, that is, anti-inflammatory cytokine production. One possibility is that CD4þ Tregs temper the inflammatory response on the ocular surface by secreting TGF-b and IL-10, which may bias APC-mediated activation and differentiation of other regulatory lymphocytes in the lymphoid tissue and/or decrease Th1-mediated inflammation locally within ocular surface tissues.

Afferent arm of the immune response during autoimmune-based inflammation

During dry eye the afferent arm of the immune response is initiated following environmental stress-mediated desiccation and/or increased tear osmolarity (Figure 2). The initial response disrupts the protective barrier by promoting proteolysis of tight junction proteins and alters the pattern of epithelial differentiation towards squamous metaplasia and decreased mucus production. Osmotic stress activates signaling pathways in a variety of cell types, including the ocular surface epithelia. Exposure to increased osmolarity in vivo or in vitro activates mitogenactivated protein kinase (MAPK) pathways, including p38 and c-Jun N-terminal kinases, and nuclear factor (NF)-kB in the ocular surface epithelia; desiccating and osmotic

stress-mediated MAPK activation stimulates corneal epithelial cells to produce of a variety of proinflammatory mediators. In addition, the altered barrier facilitates diffusion of soluble inflammatory factors into the epithelium and stroma and inflammatory cell infiltration into the ocular surface tissues.

Exposure to desiccating stress (DS) initiates ocular surface epithelial cells to release proinflammatory cytokines and chemokines (e.g., IL-1b, TNF-a, IFN-g, IL-8, CXCL10, MMP-1, -3, -9, -10, and -13). For instance, IFN-g upregulates the adhesion molecule ICAM-1 expression within the epithelium, stromal fibroblasts and vascular endothelium, and increases vascular permeability in animal models and patients with non-Sjo¨gren’s or Sjo¨gren’s syndrome-mediated dry eye. Indeed, effector CD4þ T cells express high levels of lymphocyte function-associated antigen 1 (LFA-1), the cognate binding partner to ICAM-1. The upregulation of adhesion molecules on T cells and endothelial cells coupled with increased vascular permeability results in peripheral immune cell infiltration. The IL-1a and IL-1b levels are also elevated in the tears and conjunctiva from Sjo¨gren’s syndrome-associated tear deficiency patients and in a mouse model of dry eye. The IL-1 also induces upregulation of adhesion molecules localized on endothelial cells and stimulates expression of chemokines, which act in concert to facilitate leukocyte infiltration. The TGF-b- dependent suppression of DC activation is also compromised as TGF-b2-secreting conjunctival goblet cells undergo apoptosis during the immunopathogenesis of dry eye. Along these lines, IL-1 and TNF-a also activate immature APCs in the cornea and conjunctiva, which results in increased expression of MHC class II antigens, co-stimulatory molecules, VEGFr3, and CCR7. These molecules act together to coordinate APC migration to the draining lymph nodes and activation of effector T cells that drive the efferent arm of the immune response.

Efferent arm of the immune response during autoimmune-based inflammation

The efferent arm of the immune response is mediated by autoreactive T cells that (1) are activated within secondary lymphoid tissue via cell-to-cell contact with ocular surface-derived APCs, (2) are targeted to ocular surface tissues by acquisition of trafficking molecules and chemokine receptors, and (3) compromise the integrity of the ocular surface and contribute to prolonged inflammation (Figure 2). During the immunopathogenesis of dry eye, activation of autoreactive CD4þ T cells is thought to be driven within secondary lymphoid tissue by ocular surface-derived dendritic cells bearing self-antigen presented in the context of MHC class II. This theory is well supported by animal studies demonstrating that CD4þ T cells from the cervical lymph nodes of dry eye mice specifically home to the ocular surface tissue and mediate

dry eye disease when adoptively transferred to athymic nude recipient mice. Importantly, DS-specific CD4þ T cells are not detected in any other tissues indicating that these cells are targeted to the ocular surface during activation within the secondary lymphoid organs.

Tissue-specific targeting of autoreactive T cells occurs within secondary lymphoid organs and is dictated by interactions between adhesion molecules and chemokine receptors that bind and respond to ligands expressed locally within inflamed tissues. Expression of LFA-1 on the T cells from patients with non-Sjo¨gren’s or Sjo¨gren’s syndrome-mediated dry eye and high-level ICAM-1 expression within ocular surface tissues of these patients suggest that LFA-1:ICAM-1 binding contributes to T-cell infiltration. Chemokine receptor signaling also appears to contribute to efferent homing of autoreactive T cells during dry eye disease. Elevated expression of the chemokine receptors CCR5 and CXCR3 and the chemokine ligands, CCL3, CCL4, CCL5, CXCL9, and CXCL10 has been detected within the cornea and conjunctiva of dry eye mice. In addition, CCR5 is also expressed on cells within conjunctival epithelium of patients with dry eye and CCL5 and CXCL10 are upregulated in human conjunctival epithelium cells in response to cytokine stimulation. These results suggest that the CCL5:CCR5 and CXCL9/ CXCL10:CXCR3 signaling axes play a role in T-cell trafficking during dry eye disease.

The T-cell infiltration compromises the integrity of the ocular surface and drives chronic autoimmunemediated inflammation. In the mouse model of dry eye, accumulation of CD4þ T cells within ocular surface tissues correlates with increased cytokine production (e.g., IFN-g, IL-1b, TNF-a, and MMP-9), epithelial cell apoptosis, and decreased goblet cell density, tear production, and turnover. The presence of T cells in human dry eye patients also correlates with similar pathology. Indeed, CD4þ T cells are a prominent source of IFN-g, which can induce expression of a variety of proinflammatory factors, including trafficking molecules, such as ICAM-1, CCL5, and CXCL10, and the pro-apoptotic proteins, Fas-FasL. These factors may contribute to chronic inflammation on the ocular surface by attracting infiltrating T cells and macrophages and perpetuate bystander tissue damage, including epithelial cell apoptosis and nerve damage. The IFN-g was also shown to activate a cornified envelope precursor and genes involved in conjunctival epithelial differentiation, implicating CD4þ T cells as mediators of epithelial cell abnormalities and keratinization of the ocular surface. The CD4þ T cell accumulation is also associated with increased levels of IFN-g in the tears of dry eye mice and is inversely related to goblet cell density and conjunctival squamous metaplasia. Furthermore, exogenous administration of IFN-g to IFN-g- deficient mice results in goblet cell loss. These findings suggest that CD4þ T-cell-derived IFN-g is a major

component of chronic inflammation, epithelial cell metaplasia, and tissue destruction observed during dry eye. Emerging evidence demonstrating (1) the presence of Th17-specific CD4þ T cells within the draining cervical lymph nodes of mice with experimental dry eye and

(2) high-level expression of IL-6 and IL-17 in experimental dry eye and human patients suggests that in addition to IFN-g-producing Th1 cells, IL-17-producing Th17 cells also play a pivotal role in the underlying immunopathogenesis of dry eye.

Past, Future, and Current Therapies

Artificial Tears

Individuals with aqueous tear deficiency have decreased tear film stability and diminished tear volume. There are several therapeutic options for these patients. Artificial tears, applied topically to the ocular surface, are polymerbased. The type of polymer used determines the artificial tear viscosity, retention time, and adhesion to the ocular surface. For example, artificial tears containing hyaluronic acid exhibit non-Newtonian properties and relatively long retention times. Other types of polymers used in different artificial tears include cellulose esters (increases tear viscosity), polyvinyl alcohol (provides optimal wetting), povidone (superior wetting), and carbomers (longer retention times). There are tear gels made with polyacrylic acid that provide greater retention times than artificial tears. Some formulations add a lipid component, such as castor oil, that helps to prevent evaporation. Many tears contain electrolytes and buffers to help normalize the tear pH. Artificial tears offer provisional relief of eye irritation but do not reverse conjunctival squamous metaplasia. Preservatives in artificial tears such as benzalkonium chloride can induce ocular surface epithelial toxicity if frequently applied on patients with low tear turnover or individuals that have punctual occlusion. Preservativefree artificial tears may be a consideration for patients using artificial tears more than four times per day. Artificial tears provide a palliative therapy to dry eye patients, but do not prevent the underlying cause of disease. Inflammation!

Corticosteroids

Dry eye is an immune-based inflammatory disease. Chronic dry eye requires topical treatment with therapies designed to manage inflammation. Corticosteroids effectively block multiple inflammatory pathways including proinflammatory cytokine and chemokine secretion, synthesis of matrix metalloproteinases and prostaglandins, and cell adhesion molecule expression. Activated steroid receptors bind to DNA and control gene expression and impede transcriptional regulators (AP-1 and NFkB) of proinflammatory genes. Topical corticosteroid use, while effective, is

normally prescribed for short-term use (up to 4 weeks) due to the plethora of potential side effects including glaucoma, cataracts, and ocular infection. Topical nonpreserved methylprednisolone (1%) treatment of 15 Sjo¨gren’s syndrome patients three times daily for 2 weeks followed by punctual occlusion resulted in moderate to complete relief of disease symptoms.

Cyclosporine

Restasis (topical CsA, 0.5%) is the only FDA-approved therapeutic for dry eye syndrome. A fungal-derived peptide, CsA, inhibits nuclear translocation of cytoplasmic transcription factors that are necessary for T-cell activation and the production of pro-inflammatory cytokines. The CsA was first identified as therapy for dry eye in dogs with spontaneous keratoconjunctivitis sicca. In human dry eye patients, treatment with topical CsA reduced conjunctival cellular infiltration, IL-6 levels, and increased conjunctival goblet cell numbers. In two 6-month independent FDA Phase III clinical trials CsA treatment resulted in a significant (p 0.05) improvement in corneal fluorescein staining and anesthetized Schirmer test values in patients treated with CsA (0.05 or 0.1%) compared to patients treated with vehicle alone. There was no indication of a dose-dependent effect with a safety profile of 0.05% and 0.1%. Moreover, patients did not show any serious adverse effects other than occasional burning and stinging. In addition, CsA also increased tear production in patients. There were no detectable levels of CsA in the blood of patients treated with CsA for 12 months, suggesting that topical CsA does not reach high enough levels to impact the systemic immune response. In addition to the clinical improvement in CsA-treated patients, there was also a marked decrease in expression of HLA-DR and IL-6 by conjunctival epithelial cells. Infiltration of CD3þ, CD4þ, and CD8þ T cells was also decreased in the conjunctiva of patients treated with CsA, but increased in those treated with vehicle alone.

Mucin Secretagogues

Secretagogues induce tear production by the lacrimal glands and the ocular surface epithelia. Pilocarpine and cevimeline are cholinergic agonists approved for oral administration. Patients taking Pilocarpine 5 mg four times daily reported a significantly greater overall improvement in ocular problems. Patients taking cevimeline had improvement in ocular irritation symptoms and aqueous tear production. A potential future secretagogue, Diquafosol tetrasodium, is an agonist for the P2Y2 receptor. Small molecule tyrosine kinase receptor agonists that induce mucin MUC5AC secretion from conjunctival goblet cells have also been developed.

Tetracyclines

Tetracyclines, traditionally used as antibiotics, have been reported to have a number of anti-inflammatory properties including inhibition of proinflammatory cytokines, MMP production, and nitric oxide production. On the ocular surface, tetracyclines reduce human corneal epithelial production of IL-1 and MMPs, preserving the ocular surface epithelial barrier, and blocking the activation of destructive cytokines.

See also: Adaptive Immune System and the Eye: Mucosal Immunity; Conjunctiva Immune Surveillance; Conjunctival Goblet Cells; Defense Mechanisms of Tears and Ocular Surface; Lacrimal Gland Hormone Regulation; Lacrimal Gland Overview; Lacrimal Gland Signaling: Neural; Meibomian Glands and Lipid Layer; Ocular

Mucins; Overview of Electrolyte and Fluid Transport Across the Conjunctiva; Tear Film.

Further Reading

Lam, H., Bleiden, L., de Paiva, C. S., et al. (2009). Tear cytokine profiles in dysfunctional tear syndrome. American Journal of Ophthalmology

147(2): 198–205.

Niederkorn, J. Y., Stern, M. E., Pflugfelder, S. C., et al. (2006). Desiccating stress induces T cell-mediated Sjo¨gren’s syndrome-like lacrimal keratoconjunctivitis. Journal of Immunology 176(7): 3950–3957.

Siemasko, K. F., Gao, J., Calder, V. L., et al. (2008). In vitro expanded CD4þCD25þFoxp3þ regulatory T cells maintain a normal phenotype and suppress immune-mediated ocular surface inflammation. Investigative Ophthalmology and Visual Science

49(12): 5434–5440.

Stern, M. E., Beuerman, R., and Pflugfelder, S. C. (2009). Dry Eye and the Ocular Surface. New York: Marcel Dekker.

Glossary

Afferent and efferent components of allogeneic response – Afferent arm of the immune response is the inductive stage in which antigens are presented to T lymphocytes in lymph nodes. Efferent arm of the allogeneic response is the stage in which T cells and antibodies are generated and are available to mediate graft rejection or protection from infectious agents.

Alloantigen – An antigen present in some, but not all, individuals of the same species.

Alloreactive – The reaction of lymphocytes or antibodies with alloantigens.

Anterior chamber-associated immune deviation –

Systemic downregulation of antigen-specific cell-mediated immunity that is induced when antigens are introduced into the anterior chamber of the eye.

Delayed-type hypersensitivity – Immune response consisting mostly of T cells that develops 24–72 h after exposure to antigen.

Immune privilege – Condition in which immune responses are suppressed or downregulated.

Indirect and direct allorecognition – Two pathways whereby immune responses to histocompatibility antigens on a corneal allograft elicit an immune response and ultimately allograft rejection. Indirect pathway occurs when host-derived antigen presenting cells reprocess antigens from the allograft. Direct pathway involves direct stimulation of the host’s T lymphocytes by major histocompatibility antigens expressed on the corneal allograft.

Penetrating Keratoplasty: Indications

and Survival

Penetrating keratoplasty, or full thickness corneal transplantation, is the surgical procedure most commonly used in management of blinding corneal disease. Overall graft survival rates in corneal transplantation are similar to those in cadaveric renal transplantation: this indicates that the impact of corneal graft failure, due in most patients to allogeneic rejection, is significant, even if very few of the cornea patients receive oral immunosuppression as prophylaxis. In this way, corneal transplantation is one

clinical circumstance in which immune homeostasis of the eye might seem undermined. However, in those patients in whom irreversible transplant rejection occurs, the selective targeting of immune-mediated injury to donor cells is a striking example of protection of host ocular tissue from bystander injury – afferent and efferent components of the allogeneic response in tandem maintaining immune homeostasis of the eye.

Large cohort outcome studies have identified a number of factors on multivariate analysis which, if present, have a statistically significant detrimental effect on corneal graft survival. These are a previous ipsilateral failed graft, ipsilateral ocular inflammation, vascularization of the recipient cornea, and the primary corneal diagnosis. In most patients, these factors increase risk of graft failure by immune rejection.

Corneal Immune Privilege

Apart from the specific diagnoses of keratoconus and Fuchs endothelial disease, few of the common indications for corneal transplantation can truly be considered low rejection risk. Nevertheless, it is clear that corneal transplants do enjoy comparative immune privilege. Early investigators attributed the immune privilege of the cornea entirely to its lack of vascularity, that is, sequestration of alloantigen from the immune response. There is no doubt that this is an important factor. Animal models and large human cohort studies have identified recipient corneal vascularization as the most significant factor conferring high rejection risk in multivariate analyses. In fact, the cornea is an immune-privileged tissue situated in and transplanted to an immune-privileged site.

Immune-Privileged Tissue

A number of factors are known to contribute to the relative immune privilege of the cornea as a tissue.

. On a gross level, the normal cornea is isolated from circulating immune cells due to (1) its avascular nature and (2) the blood–aqueous barrier.

. Until relatively recently, the cornea was thought to contain no passenger antigen-presenting cells (APCs). Recent work has established that it does, in fact, contain APCs but that they are immature and do not express major histocompatibility complex (MHC) class II in the normal setting. Subsequently, the normal cornea is

devoid of lymphatics to transport APCs. Experimental studies which increase the numbers of passenger APCs in the donor cornea have been shown to erode immune privilege and increase the direct component of allorecognition. Because in clinical transplantation normal avascular donor cornea is grafted, high rejection risk status is conferred by vascularization or inflammation of the residual host cornea.

. The stroma and endothelium have low immunogenicity.

. The endothelium, which is the most important target in rejection, expresses (i) low levels of MHC class I and II and (ii) high levels of Fas ligand which induces apoptosis in alloreactive T-lymphocyte cells and protects the graft.

Immune-Privileged Site

The cornea, or specifically its endothelium, constitutes the anterior boundary of the anterior chamber which has been shown to be an immune-privileged site by Medawar in one of the earliest studies of this concept. The aqueous humor in contact with the endothelial cells contains high levels of immunoregulatory proteins such as transforming growth factor beta (TGF-b). In addition, antigen placed in the anterior chamber of the eye alters the immune response (anterior chamber-associated immune deviation or ACAID) to subsequent exposure to the antigen even at a different site. Antigen from the anterior chamber leaves the eye via several pathways but at least some leaves via the conventional aqueous outflow pathway and travels to the spleen. There the interaction of antigen, natural killer (NK), T cells, B cells, and gd T cells induces a type of operational tolerance.

Clinical Features of Corneal Graft

Rejection

Selective rejection signs can be observed in the epithelium, stroma, and endothelium of donor cornea in the human. The corneal endothelial cell monolayer controls stromal hydration, which has an essential role in transparency and transmission of light (Figure 1). As (1) these cells do not have mitotic capability and (2) following uncomplicated corneal transplantation the endothelial cell monolayer density declines at an even faster rate than in health, rejection of the endothelial layer is a terminal threat to graft transparency unless reversed sufficiently early. Even if reversed by treatment, a proportion of endothelial cells is lost.

Patients with isolated epithelial rejection are often asymptomatic. In epithelial rejection, an elevated curvilinear opaque line is seen on the epithelium. However, epithelial rejection is direct evidence that the recipient has been sensitized to the graft and may progress to rejection of the deeper corneal layers. The clinical symptoms

and signs of stromal rejection have two patterns. It more frequently manifests as subepithelial opacities. These infiltrates have a similar appearance to those seen in adenovirus viral keratitis but are seen only in the donor cornea. This type of stromal rejection is often asymptomatic, but like epithelial rejection it may precede the onset of a more severe and visually significant process. Rejection of the deeper stroma results in spreading graft opacification and decreased visual acuity.

Endothelial rejection usually presents with discomfort and decreased visual acuity. A line of leukocytes may, over a period of days, be seen migrating across the donor endothelium leaving edematous stroma in its wake (Figure 2). This line often spreads out like a wave from an area of deep vascularization to the graft host junction. Alternatively, a

Figure 1 Transparent corneal transplant 8 months following surgery. The donor cornea is fully transparent, indicating normal endothelial function. Both recipient and donor cornea are avascular.

Figure 2 Endothelial rejection is indicated by the demarcated region of donor corneal opacification bordered by a horizontal linear opacity. Vascularization to the superior graft periphery can also be seen.

3Exposure of circulating primed lymphocytes and other leukocytes to graft

4Recognition of alloantigen (efferent allorecognition)

5Recruitment of other effector cells

more diffuse corneal edema may be seen with diffuse keratitic precipitates of variable density. In all cases, there are visible cells in the anterior chamber and the signs of inflammation and of endothelial dysfunction are limited to the donor tissue, demonstrating specificity of the allogeneic response.

Pathogenesis of Rejection

Current understanding of the pathway to rejection of donor cornea following penetrating keratoplasty will be described in terms of the afferent component or sensitization, in which donor transplant antigens are recognized and processed, and the effector component. Components of the allogeneic response in low and high rejection risk corneal transplantation are summarized in Table 1.

Afferent Mechanisms and Components

Histocompatibility antigens

Otherwise known as transplantation antigens, these are proteins and peptides derived from donor cells; in most forms of transplantation, the most potent are class I and class II molecules of the MHC. Additional transplantation antigens termed minor histocompatibility antigens also induce allogeneic tissue rejection. The designation of

major or minor refers to the relative importance of these antigens in vascularized organ transplants and may be quite inaccurate in corneal transplantation, in which, for example, minor H antigen appears to be relatively more important than MHC antigens.

Published evidence from several laboratories indicates that following transplantation recipient APCs enter the graft and endocytose exogenous alloantigen. APCs then migrate to the local lymph node where the alloantigen is presented on MHC class II molecules to naive CD4 cells and on MHC class I to naive CD8 cells. APC-associated antigen may be detected in the draining lymph node within hours of corneal transplantation. The exogenous antigen in question could be either a donor minor histocompatibility antigen or part of a donor major histocompatibility antigen. One self APC activates both CD4 and CD8 cells in what is known as the three-cell model of alloantigen presentation.

Direct and indirect alloantigen recognition

The type of antigen presentation by recipient APCs outlined above is termed indirect antigen presentation. Another form of antigen presentation, unique to the transplantation, is direct antigen presentation mediated by donor APCs transplanted within the graft (passenger leukocytes). According to the concept of self-restriction, alloantigen presented by these cells should not be recognized

by host T cells if the major histocompatibility antigens are not matched. In reality, the alloantigens are recognized by a significant number of host T cells. Some of these lymphocytes may recognize, and be primed, by the alloantigenic MHC molecule itself regardless of the peptide it bears. This method of antigen presentation is consistent with the three-cell model and in vascularized organ grafts directly primed T cells constitute 90% of the alloreactive cells during acute graft rejection.

With its lack of mature resident APCs and lymphatics, the cornea would appear to be poorly equipped to facilitate antigen presentation via the direct route. This is confirmed by studies which show that the indirect route of antigen presentation plays a more prominent role in low-risk corneal transplantation. This lack of influence of the direct route may explain the following findings which are on first consideration counterintuitive:

. In human studies, MHC class I and particularly class II matching of corneal grafts has shown no convincing survival benefit.

. In animal studies, mismatches in minor rather than major histocompatiblity antigens have been shown to be equally or more important in influencing graft survival.

Further experimental evidence suggests that some of the factors which confer immune privilege do so by minimizing or preventing direct antigen presentation and that in erosion of immune privilege (i.e., in high-risk grafts) the direct route of antigen presentation becomes relatively more prominent. Nevertheless, the indirect route of antigen presentation is sufficient to induce sensitization.

T-lymphocyte activation

In the lymph node the MHC–peptide complex interacts with the T cell receptor (TCR), a complex cell surface receptor. The vigor of the T cell response following presentation of antigen is variable and dependent on the nature of the dendritic cell itself, the affinity of the clonotypic TCR for the MHC–peptide complex in question, the state of the T cell ( naive, memory), soluble factors in the immediate microenvironment (cytokines and chemokines), and the interactions between accessory (adhesion and costimulatory) molecules.

Once a T cell is activated in the lymph node, there is rapid clonal expansion of alloantigen-specific T cells which enter the circulation. The lifespan of these cells is limited and it follows that there is a limited window of opportunity for these cells to bring about graft destruction in the absence of continuous antigenic stimulation. The rapid expansion of T cells is followed by contraction as many effector T cells undergo apoptosis. Memory (central and effector) cells make up part of the T cell repertoire thereafter.

Effector Mechanisms

Once primed in the regional lymph nodes, activated lymphocytes enter the peripheral circulation. The avascular nature of the cornea and the blood–aqueous barrier provide barriers to immune cell infiltration and endothelial cell destruction. In the case of vascularized corneas, immune cells have easier access to graft antigens/cells.

The nature of graft-infiltrating cells in corneal allograft rejection has been studied in human and animal pathological specimens. The cell types which appear in the highest numbers and with the greatest consistency are CD4+ and CD8+ lymphocytes of the adaptive immune system, and innate immunity component macrophages and NK cells. The presence of a cell in a tissue during rejection does not prove that the cell has an effector functional role: the important questions as to which cells cause endothelial cell destruction and by what mechanism(s) remain poorly understood. For instance, the mechanisms of allorecognition in the effector stage of graft rejection are unclear. A particular conundrum has been the question of how indirectly primed T cells (host MHC(+mH) molecules) can recognize antigen on donor cells (donor MHC(+mH) molecules). The discovery of a semidirect pathway of antigen presentation, whereby recipient APCs present whole donor MHC molecules as well as their own MHC molecules, provides a possible explanation for this but has not been demonstrated experimentally in corneal transplantation.

T lymphocytes

Because both CD4 and CD8 have been found in pathological specimens of rejected corneal grafts, much interest has fallen on the roles of these cells in corneal graft rejection. Several studies have demonstrated the presence of two distinct lymphocyte populations in response to a corneal allograft. One group appear to be CD4+ and are activated by indirect presentation of alloantigen. The other group are CD8+ cells with direct specificity for alloantigen. CD8+ cells act directly on target cells and are cytotoxic but it appears that CD8+ cells are less important in corneal graft rejection than in other organs. While CD4+ cells have the capacity using FasL to be directly cytotoxic, their primary modus operandi in corneal graft rejection appears to be via delayed-type hypersensitivity (DTH), by secreting cytokines and recruiting other cells such as macrophages. The evidence for role of CD4+ cells is supported by the finding that DTH responses rather than cytotoxic responses are found in rejectors of corneal grafts. We may conclude from findings in several laboratories that CD4+ cells play a more important role than CD8+ cells in graft rejection under most conditions but that either cell type may mediate rejection and that neither is essential for the process. However, there is considerable redundancy within the allogeneic response, with several lines of investigation supporting alternative cellular pathways for graft destruction.

Macrophages

The heavy mononuclear cell infiltrate in rejected grafts is in keeping with a DTH reaction. Depletion of corneal macrophages with clodronate liposomes prolongs corneal graft survival in rats, and macrophages have been shown to be necessary as APCs rather than as effector cells. However, cells of monocyte/macrophage lineage have been shown to be the dominant cell type in human aqueous humor during acute endothelial rejection. The function(s) of these cells in the effector arm of the rejection process remains unknown.

NK cells

NK cells of the innate immune system have been found in rejected corneal grafts and in the aqueous humor of experimental animals with corneal allograft rejection. These cells usually function in the elimination of virally infected cells. The default function of an NK cell is to kill any cell with which it comes in contact. Only the presence of self MHC class I on the cell inhibits this process. In vitro studies have demonstrated the capacity of NK cells to kill allogeneic corneal endothelial cells, so these cells are likely to be functionally active in the efferent arm of rejection.

Breakdown of immune privilege?

In a high rejection risk transplant there is a preexisting erosion of immune privilege at one or more of these steps mentioned in section ‘‘Immune-privileged Tissue’’. Low rejection risk grafts that reject later may be thought of as grafts that have acquired high rejection risk characteristics due to breakdown of immune privilege. In nonvascularized corneas, immune rejection occurring months or years after transplantation is often seen to be preceded by a local episode of alloantigen-independent inflammation (e.g., loosened transplant suture, bacterial suture-associated infection, and recurrent herpetic infection) which may lead to recruitment of immune-competent cells, angiogenesis, lymphangiogenesis, and upregulation of MHC molecules on the graft cells. Each step and the factors within it contributing to immune privilege are reasonably well understood but the extent to which one step inevitably follows the preceding one is less clear. We may ask of grafts which are not rejected, whether the recipient has not been sensitized due to the immune system not seeing the antigen (ignorance of the alloantigen), whether the immune system has seen the antigen but does not or cannot mount a response (tolerance of the antigen), or whether the immune system has seen the antigen and been sensitized but its effector cells cannot see the target antigen due to sequestration of the graft in its avascular bed.

While it is tempting to speculate that a single step exists, manipulation of which would induce tolerance or absolute immune privilege in all cases, it is far more likely that the relative contributions to immune privilege at each step is different for each person and for each graft and

there is no factor contributing to immune privilege that cannot be overcome by one of the many redundant cellular pathways and mechanisms known to bring about rejection.

Treatment of Rejection

The mainstay of treatment for established rejection is intensive topical corticosteroid treatment. The most commonly used regimen is prednisolone acetate 1% or dexamethasone 0.1% hourly. This treatment effectively suppresses graft inflammation but once inflammation has been suppressed, the question of whether graft clarity will return depends on the extent to which the endothelium has been damaged. Topical corticosteroids influence effector cells such as T cells and macrophages in the cornea chiefly by inducing the expression of anti-inflammatory genes (Annexin-1, SLPI) and repressing the expression of proinflammatory genes (cytokines, chemokines, adhesion molecules, and MHC molecules). Inhibition of IL-2 receptor production inhibits T cell proliferation but this may not be an important effect of topical treatment as Tcell proliferation occurs quite distal to the site of application in the regional lymph nodes. Corticosteroids also affect dendritic cell (DC) function and have been shown to alter cytokine production, to induce apoptosis in DCs and to delay DC maturation with resultant impairment of antigen presentation. Corticosteroids inhibit angiogenesis but this is unlikely to be relevant in setting of acute rejection. While some clinicians treat endothelial rejection with systemic as well as topical corticosteroid, a trial of intravenous methylprednisolone in addition to intensive topical treatment did not show an improvement in outcome compared with topical treatment alone.

Prevention of Rejection

The key to minimizing immune-mediated graft failure in patients is a dual strategy of (1) identifying preand posttransplant those at high risk of rejection, tailoring their management appropriately and (2) educating graft recipients as to the signs and symptoms of rejection. Preoperative risk factors for rejection include unmodifiable factors such as a previously rejected ipsilateral graft or previous herpetic keratitis and factors which are modifiable to a greater or lesser degree such as corneal vascularization or active external eye inflammation. All ocular inflammation should be brought under control where possible before elective corneal transplantation. A degree of regression of corneal vessels may be induced by topical steroid treatment particularly in an inflamed cornea. More established vessels may be difficult to treat. Of most concern are deep vessels and vessels close to the (projected) graft–host interface. Inhibition of formation

of lymphatic vessels from preexisting lymphatic vessels (lymphangiogenesis) in the host cornea has been shown to prevent or delay graft rejection by inhibiting APC egress and sensitization to alloantigen. Antivascular endothelial growth factor (anti-VEGF) treatment is a promising area for future investigation as it may help to mitigate both the afferent (lymphangiogensis) and efferent (angiogenesis) arms of the immune response to allogeneic cornea.

One rational approach to management of transplants at high rejection risk is the use of systemic immunosuppression with calcineurin inhibitors to prevent alloreactive T cell clonal expansion. Unfortunately, there is no robust evidence favoring any such regimes and complete absence of randomized trials. In grafts not at high risk of rejection, very long-term local immunosuppression with topical corticosteroid may be useful in preventing rejection but the benefit must be weighed against such risks as glaucoma, susceptibility to infection, and impaired corneal wound healing. In the postoperative period, patients presenting with alloantigen-independent ocular surface inflammation, such as suture loosening, should be treated promptly and aggressively.

See also: Adaptive Immune System and the Eye: T CellMediated Immunity; Antigen-Presenting Cells in the Eye and Ocular Surface; Dynamic Immunoregulatory Processes that Sustain Immune Privilege in the Eye;

Immunosuppressive and Anti-Inflammatory Molecules that Maintain Immune Privilege of the Eye.

Further Reading

Barnes, P. J. (2006). How corticosteroids control inflammation: Quintiles Prize Lecture 2005. British Journal of Pharmacology 148: 245–254.

Ferguson, T. A. and Griffith, T. S. (2006). A vision of cell death: Fas ligand and immune privilege 10 years later. Immunological Reviews 213: 228–238.

Fu, H., Larkin, D. F. P., and George, A. J. T. (2008). Immune modulation in corneal transplantation. Transplantation Reviews 22: 105–115.

Hamrah, P., Liu, Y., Zhang, Q., and Dana, M. R. (2003). The corneal stroma is endowed with a significant number of resident dendritic cells. Investigative Ophthalmology and Visual Science 44: 581–589.

Herrera, O. B., Golshayan, D., Tibbott, R., et al. (2004). A novel pathway of alloantigen presentation by dendritic cells. Journal of Immunology 173: 4828–4837.

Katami, M. (1991). Corneal transplantation – immunologically privileged status. Eye 5: 528–548.

Niederkorn, J. Y. (2006). See no evil, hear no evil, do no evil: The lessons of immune privilege. Nature Immunology 7: 354–359.

Rogers, N. J. and Lechler, R. I. (2001). Allorecognition. American Journal of Transplantation 1: 97–102.

Streilein, J. W., Wilbanks, G. A., Taylor, A., and Cousins, S. (1992). Eye-derived cytokines and the immunosuppressive intraocular microenvironment: A review. Current Eye Research 11(supplement): 41–47.

Williams, K. A. and Coster, D. J. (2007). The immunobiology of corneal transplantation. Transplantation 84: 806–813.

Williams, K. A., Esterman, A. J., Bartlett, C., et al. (2006). How effective is penetrating corneal transplantation? Factors influencing long-term outcome in multivariate analysis. Transplantation 81: 896–901.