IL-17A-Mediated Protection against Acanthamoeba Keratitis

Abstract

Acanthamoeba keratitis (AK) is a very painful and vision impairing infection of the cornea that is difficult to treat. Although past studies have indicated a critical role of neutrophils and macrophages in AK, the relative contribution of the proinflammatory cytokine, IL-17A, that is essential for migration, activation and function of these cells into the cornea is poorly defined. Moreover, the role of the adaptive immune response, particularly the contribution of CD4⁺ T cell subsets, Th17 and Treg cells, in AK is yet to be understood. In this report, using a mouse corneal intrastromal injection-induced AK model, we show that Acanthamoeba infection induces a strong CD4⁺ T effector and regulatory T cell response in the cornea as well as local draining lymph nodes (dLN). We also demonstrate that corneal Acanthamoeba infection induces IL-17A expression and that IL-17A is critical for host protection against severe AK pathology. Accordingly, IL-17A neutralization in Acanthamoeba-infected wild-type mice or Acanthamoeba infection of mice lacking IL-17A resulted in a significantly increased corneal AK pathology, increased migration of inflammatory cells at the site of inflammation and a significant increase in the effector CD4⁺ T cell response in dLN. Thus, in sharp contrast to other corneal infections such as herpes and P. aeruginosa keratitis where IL-17A exacerbates corneal pathology and inflammation, findings presented in this manuscript suggest that IL-17A production after Acanthamoeba infection plays an important role in host protection against invading parasites.

Introduction

Acanthamoeba keratitis (AK) is a debilitating, extremely painful and vision-impairing infection of the cornea caused by parasites of genus Acanthamoeba (1–6). In immunocompetent individuals, cornea is the single tissue most susceptible to infection by Acanthamoebae. The mechanisms by which the parasite produces keratitis have not been fully elucidated. Contact lens wear is thought to be the leading risk factor (7–9). However, the disease also occurs, albeit less frequently, in non-contact lens wearers (10–13). Recently, AK was reported in a 5-year old boy without a history of contact lens usage or trauma (10). It is believed that minor corneal surface injury caused by contact lens wear or other deleterious agents and exposure to contaminated solutions, including lens care products and tap water, are two major factors in the pathogenesis of AK.

Given that over 25 million individuals in the United States alone wear contact lenses and the amoebae are ubiquitous in the environment, one wonders why the occurrence of AK is not much higher. A reasonable answer is that protective factors must be present in vivo. Both innate and acquired immune systems are thought to play a role in providing protection against AK (14). Seminal studies by Niederkorn and coworkers have suggested that specifically, the mucosal immune system plays an instrumental role in providing immunity to primary AK (15–17). Little is known about the involvement of the host immune response, particularly the role of CD4⁺ T cells, in the pathogenesis of AK. This is, in part, due to difficulty in developing a robust mouse model to study various critical events that occur after Acanthamoeba infection. Niederkorn and colleagues have developed self-limiting pig and Chinese hamster animal models to study AK pathogenesis (17, 18) and have demonstrated a critical role of innate immune cells, particularly neutrophils and macrophages, in providing protection against AK pathogenesis (14, 19, 20). It has been shown that neutrophils and macrophages infiltrate the cornea soon after infection and are essential for effective killing of the parasites post Acanthamoeba infection (14, 19, 20). On the other hand, neutrophils may also contribute to corneal tissue damage and AK lesion severity through release of various proteases (3, 21). The role of CD4⁺ T cells in AK pathogenesis is poorly understood. Past studies have shown the presence of CD4⁺ T cells in corneas from AK patients as well as from infected corneas of experimental animals (3, 22–24). However the migration of CD4⁺ T cells during ongoing AK and contribution of Th1 (IFN-γ⁺ CD4⁺ T cells), Th2 (IL-4⁺ CD4⁺ T cells), Th17 (IL-17A⁺ CD4⁺ T cells) and regulatory T cells (Foxp3⁺ CD4⁺ T cells) has not yet been reported. In this study, using corneal intrastromal injection of Acanthamoeba in mice, we demonstrate that corneal Acanthamoeba infection induces a strong CD4⁺ T effector and regulatory T cell response in both cornea and local draining lymph nodes (dLN).

IL-17A, a proinflammatory cytokine, plays a critical role in migration and activation of inflammatory cells such as neutrophils and macrophages at the site of inflammation (25–28). In this respect, IL-17A has been shown to exacerbate herpes simplex virus (HSV) and Pseudomonas keratitis lesion severity through increased production of various chemokines and cytokines essential for migration and activation of neutrophils into the cornea (29–32). However, a protective role of IL-17A in host defense against microbes has also been documented (33–38). Given the predominant contribution of neutrophils in AK and the emerging role of IL-17A in neutrophil function, in the current study we examined whether IL-17A contributes to corneal immunopathology or host protection after ocular Acanthamoeba infection. We demonstrate here that: (i) IL-17A expression is markedly upregulated during AK and that (ii) neutralization of corneal IL-17A using local subconjunctival injections of anti-IL-17A mAb in wild-type mice or infection of IL-17A knock-out (IL-17AKO) mice results in increased corneal opacity and AK lesion severity. Collectively, our data suggest the critical involvement of the previously unrecognized IL-17A-neutrophil-CD4⁺ T cell axis in host protection against corneal Acanthamoeba infection and associated AK pathogenesis.

Materials and Methods

Mice and Parasites

Six- to eight-week old female wild-type (C57BL/6) mice were purchased from the Jackson Laboratory (Bar Harbor, ME). IL-17AKO mice were obtained from Professor Yoichiro Iwakura, Tokyo University of Science, Japan. All animal treatments in this study conformed to the Association for Research in Vision and Ophthalmology guidelines on the use of Animals in Vision Research and the recommendations of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Animals were housed in animal facilities approved by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). An Acanthamoeba strain derived from an infected human cornea (MEEI 0184; Acanthamoeba castnelli) was used throughout this study. The parasites were axenically cultured in a protease peptone/yeast extract/glucose medium (39). In some experiments, for comparison purposes, amoebae were killed prior to use by overnight treatment with 4% formalin.

Reagents and Antibodies

CD4-APC (RM4-5), IL-17-PE (TC11-18H10), IFN-γ-FITC (XMG1.2), CD45-PerCp (30-F11), CD45-APC (30-F11), CD11b-PerCp (M170), Ly6G-PE (1A8), anti-CD3 (145.2C11), anti-CD28 (37.51), and GolgiPlug (Brefeldin A) were purchased from BD Bioscience (San Jose, CA). Foxp3-PE and ELISA kit for IL-4, IL-10 and IL-17A were purchased from eBioscience (San Diego, CA). Liberase TL was purchased from Roche Diagnostics, Indianapolis, IN.

Corneal Infection

Corneal infections of mice were induced by intrastromal injection of amoebae using the procedure described by Pearlman and colleagues for producing Fusarium keratitis (40). Briefly, mice were anesthetized by intraperitoneal injection of ketamine and xylazine, corneal epithelium was abraded using a 30-gauge needle and through the abrasion was delivered 1 μl suspension of parasites (2.5 × 10⁴ amoebae/μl PBS; > 95% trophozoites) into the stroma using a 33-gauge Hamilton syringe. The eyes were examined on different days post infection (pi) for the development of corneal opacity. The following scoring system, graded from 0 to 4, was used as previously described (41). 0- eye macroscopically identical to the uninfected control eye, 1- partial corneal opacity covering the pupil, 2- dense corneal opacity covering the pupil, 3- dense opacity covering the entire anterior segment, 4- perforation of the cornea, phthisis bulbi (shrinkage of the globe post inflammatory disease). To further assess the fate of the amoebae injected into the corneal stroma, the trophozoites were transfected with p-475TBP-m red fluorescent protein (RFP) using the Superfect reagent as described by Peng and colleagues (42). After transfection, parasites were allowed to recover for 24h in PYG media and were then transferred in selection media (PYG containing 50 μg of neomycin G418/ml [Cellgro, Hendon, VA]). Under a fluorescence microscope, nearly 100% of the surviving cells appeared bright red. The transfected cells were passaged in the same way as the control cells without the loss of RFP intensity, and were injected into the corneal stroma as described above. Immediately after injection and every other day thereafter, amoebae were visualized in the corneas of live mice under a fluorescence-dissecting microscope (Leica MZ FL III) equipped with a camera (Imaging, Ret, GA 2000R).

Isolation of Acanthamoebae from infected cornea

Infected corneas were triturated in Page’s amoeba saline (1 ml/cornea) to produce small pieces (<1 mm). Aliquots (100 μl) of the tissue suspension were pippeted onto various locations of a 100-mm petri dish containing a thin layer of 1.5% non-nutrient agar. Prior to application of the tissue sample, the surface of the agar plate was overlaid with a suspension of E. coli. Plates were incubated at 30°C and were examined every other day for Acanthamoeba growth under a phase contrast microscope.

Anti-IL-17A Neutralization

Subconjunctival injections of anti-IL-17A mAb in Acanthamoeba infected mice were performed using a 2-cm needle attached to a 32-gauge syringe (Hamilton, Reno, NV) to deliver anti-IL-17A (5 μg in 10 μl volume) into the subconjunctival space (31). The infected mice were scored for corneal opacity on day 1 pi and animals with similar corneal opacity were randomly divided into two groups. One group of mice was treated with anti-IL-17A by local subconjunctival injections every other day from day 1 until day 7 pi. Control mice were injected with 5 μg of isotype mAb.

Quantification of RFP⁺ Acanthamoebae from infected corneas

Corneas from infected mice were harvested at indicated time points pi, and digested with 60 U/ml Liberase for 35 minutes in a tissue culture incubator. After incubation, the corneas were disrupted with a syringe plunger on a cell strainer and single cell suspensions in complete RPMI 1640 medium were analyzed for quantification of RFP⁺ Acanthamoebae by flow cytometry.

Detection of corneal infiltrating cells and flow cytometry

Corneas from infected mice were harvested (6 to 8 corneas per group) at indicated time points pi, and digested with 60 U/ml Liberase for 35 minutes at 37°C in a tissue culture incubator. After incubation, the corneas were disrupted with a syringe plunger on a cell strainer and single cell suspensions were made in complete RPMI 1640 medium. The single cell suspensions obtained from corneas, draining lymph nodes (dLN) and spleen were stained for different cell surface molecules using conjugated antibodies as described below. Briefly, cells were blocked with an unconjugated anti-CD32/CD16 mAb for 30 min in FACS buffer (2% Fetal Bovine Serum in PBS), and were then incubated with different cocktails of antibodies depending on cell type (30 min on ice). For CD4⁺ T cells, CD4-APC (RM4-5) and CD45-PerCP (30-F11) were used. For neutrophils, CD45-APC (30-F11), CD11b-PerCP (M170), Ly6G-PE (1A8) mAbs were used. The stained cells were washed three times using FACS buffer and were then fixed in 1% paraformaldehyde.

To quantify the number of IFN-γ, IL-17A and IL-4 producing CD4⁺ T cells, intracellular cytokine staining was performed as previously described with minor modifications (29). Briefly, total corneal cells or one million cells each from dLN and spleen were stimulated with anti-CD3 (1μg/ml) and anti-CD28 (0.5 μg/ml) for 5 hrs in the presence of GolgiPlug (37°C in a tissue culture incubator). At the end of the stimulation period, cell surface staining for CD4 was performed as described above. Subsequently cell permeabilization and intracellular cytokine staining was performed using a Cytofix/Cytoperm kit (BD Biosciences). For intracellular staining, PE labeled IL-4, FITC labeled IFN-γ and PE labeled IL-17A antibodies were used. To quantify Foxp3⁺CD4⁺ T cells, cell surface staining was performed for CD4 (APC) followed by intranuclear staining using the fixation/permeabilization kit and an anti-Foxp3-PE mAb (eBioscience). Samples were acquired with a FACS Calibur (BD Biosciences) and the data were analyzed using FlowJo software (Tree Star, Ashland, OR).

ELISA

Homogenates of pooled corneal samples were centrifuged and supernatant was used for analyses. The concentrations of IL-17A were measured using a sandwich ELISA kit (eBioscience).

Statistical analysis

An unpaired two-tailed Student’s t test was used to determine statistical significance for disease severity between different groups. P values ≤ 0.05 were considered significant. For some experiments, as mentioned in figure legends, a one-way ANOVA test was applied. All experiments were repeated at least two times.

Results

Mouse model of Acanthamoeba keratitis

Past studies using corneal infection models in Chinese hamsters and pigs have shown the critical function of innate immune cells in host protection against corneal Acanthamoeba infection (3, 14). However, due to lack of immunological tools such as antibodies against various cell surface receptors, knockout animals and molecular tools for genetic analysis in these animal species, various critical events in the pathogenesis of AK have not been fully characterized. In the current study, to understand crucial pathophysiological events, and host innate and adaptive immune responses after corneal Acanthamoeba infection, we used a mouse model of AK. As shown in Figure 1A–B, intrastromal injection of Acanthamoebae into mouse corneas induced corneal opacity on day 1 postinfection (pi), with a peak disease severity around day 3 to 5 pi followed by a decline in corneal opacity during the subsequent course of AK. Live parasites were successfully isolated and grown on the lawn of E. coli in agar plates from corneas of day 3 as well as day 7 pi. Although corneal opacity and disease severity had regressed to baseline by day 9 pi, we were able to isolate and grow Acanthamoeba from infected mice corneas obtained as late as day 14 pi (data not shown). Generally, amoebae growth on the agar plates was seen in less than 72 hours from day 3 and day 7 pi corneas. However, a minimum of one week was required to detect the amoebae growth from day 14 pi corneas suggesting that they contained fewer parasites compared to the corneas with an overt, early stage disease. To assess the fate of injected amoebae, mice were infected with RFP-tagged amoebae and the corneas were imaged using a fluorescence dissecting microscope on day 0, 3 and 6 pi. As shown in Figure 1C, clusters of amoebae could be readily seen in corneal stroma immediately after corneal intrastromal injection (day 0). Numerous parasites were seen in the cornea of day 3 followed by clearance with a decreased parasite load seen on day 6 pi (Fig. 1C). In some experiments, mice were also infected with dead amoebae to ascertain that corneal pathology seen in mice infected with live amoebae was not due to surgical manipulation. Mice infected with dead amoebae did not develop corneal opacity or severe AK lesions (Fig. 1D–E). To further characterize the kinetics of Acanthamoeba in the cornea, we used RFP-tagged live and killed Acanthamoebae for intrastromal injections. Quantification of RFP⁺ Acanthamoebae by flow cytometry on day 3 and day 6 pi showed significantly reduced amoebae number in dead Acanthamoeba group on day 3 pi when compared to live Acanthamoeba group (Fig. 1F). However, there was no significant change in total amoebae numbers between these two groups on day 6 pi (Fig. 1F). Collectively, these results suggest that corneal intrastromal injection of Acanthamoeba induces acute but self-limiting AK. Consistent with published reports using the hamster model (14, 17), our observations also suggest that the host mounts a strong innate immune response against corneal Acanthamoeba infection which promotes rapid clearance of pathogen from corneal tissue. Furthermore, the absence of severe corneal opacity and AK lesion severity in mice infected with dead Acanthamoeba suggests: (i) that the corneal opacity and severe lesions seen in corneas injected with live amoebae were not due to surgical manipulation and (ii) virulence factors secreted by live Acanthamoeba play a critical role in mediating initial corneal tissue damage and opacity.

FIGURE 1. Mouse model of Acanthamoeba keratitis.

C57BL/6J mice corneas were intrastromally injected with a 1 μl suspension of Acanthamoebae (2.5 × 10⁴ amoebae/μl PBS, >95% trophozoites). In some experiments, for comparison purposes, the amoebae were killed by formaldehyde treatment prior to injection. The progression of AK was monitored by measuring corneal opacity scores on day 1, 3, 5 and 7 pi. (A) Corneal opacity scores on day 0 (uninfected) and day 1, 3, 5 and 7 pi. (B) Representative eye images showing the disease severity at various days pi. (C) Representative fluorescent micrographs of corneas injected with RFP-labeled live Acanthamoeba immediately after injection and on day 3 and 6 pi. (D) Representative eye images comparing the disease severity of mice infected with live and dead Acanthamoeba on day 3 and 6 pi. (E) Corneal opacity score comparison between mice infected with live and dead Acanthamoeba. (F) Total number of RFP⁺ Acanthamoebae as quantified by flow cytometry on day 3 and day 6 pi between live and dead Acanthamoeba infected group. Data are representative of two independent experiments and show mean values ± SEM. *p≤0.05, ***p<0.0001. Statistical levels of significance were analyzed by Student’s t test (unpaired).

Innate and adaptive immune response to corneal Acanthamoeba infection

Past studies using in vitro and in vivo models have shown that neutrophils and macrophages play an important role in Acanthamoeba killing and host protection (14, 19, 20). Although the role of the adaptive immune response, particularly CD4⁺ T cells, is not defined in corneal AK, studies have shown the presence of CD4⁺ T cells in corneas of human Acanthamoeba patients (3, 14, 22–24). Furthermore, intraperitoneal challenge of Acanthamoeba trophozoites as well as cysts in mice has shown the induction of the adaptive immune response against Acanthamoeba trophozoites and cyst antigens (3, 14, 43). To understand the kinetics of immune cell, particularly neutrophil and CD4⁺ T cell, migration in the cornea following Acanthamoeba infection, mice corneas infected with Acanthamoebae were collected and analyzed on day 1, 3, 5 and 8 pi for neutrophil and CD4⁺ T cell analysis by flow cytometry. The results in Figure 2A–B show the frequency of neutrophils (CD45⁺ CD11b⁺ Ly6G⁺ cells) and CD4⁺ T cells in Acanthamoeba infected cornea at various days pi. As shown in Figure 2A, neutrophils constitute approximately 55% of total leukocytes in the cornea at day 1, and 38% on day 5 pi. As infection regressed to a diminished keratitis severity on day 8 pi, there was a significant reduction in neutrophil infiltration in the cornea. In contrast, the percentage of CD4⁺ T cells in the cornea was significantly increased on day 8 pi when compared to day 5 pi (Fig. 2B). When total numbers of different cell types infiltrating the cornea were calculated at day 5 and 8 pi, we observed a significant increase in total numbers of CD45⁺ cells on day 5 pi as compared to day 1 and day 3 pi (Fig. 2C). Interestingly, there was no significant change in neutrophil numbers in the cornea on day 1, day 3 and day 5 pi. However, there was a significant decrease in total numbers of CD45⁺ cells and neutrophils on day 8 pi as compared to day 5 pi (Fig. 2C–D). In contrast, there was no significant difference in CD4⁺ T cells on day 8 as compared to day 5 pi (Fig. 2E). Taken together, these data suggest that the total number of CD45⁺ cells and neutrophils was higher during the early stages of keratitis, and decreased as infection regressed to a less severe form of keratitis. Although we observed an increase in the percent of CD4⁺ T cells on day 8 compared to day 5 pi (Fig. 2B), there was no significant difference in the total number of CD4⁺ T cells in corneas collected on day 5 versus day 8 pi (Fig. 2E).

FIGURE 2. Corneal neutrophil and CD4⁺ T cell response changes after Acanthamoeba infection.

C57BL/6J mice corneas were intrastromally injected with 2.5 × 10⁴ Acanthamoebae, and CD45⁺ CD11b⁺ Ly6G⁺ polymorophonuclear neutrophils and CD4⁺ T cells infiltrating the cornea on day 1, 3, 5 and 8 pi were analyzed by flow cytometry. (A) Representative FACS plots for CD11b⁺ Ly6G⁺ neutrophils gated on CD45⁺ cells from pooled corneal samples on day 1, 3, 5 and 8 pi are shown. (B) Representative FACS plots (left) and frequencies (right) for CD4⁺ T cells gated on CD45⁺ cells from pooled corneal samples on day 5 and 8 pi are shown. Total cell numbers of CD45⁺ cells (C), neutrophils gated on CD45⁺ cells (D) and CD4⁺ T cells gated on CD45⁺ cells (E) in the cornea on day 1, 3, 5 and day 8 pi are shown. Combined results of two separate experiments are shown as mean values ± SEM, 5 to 6 corneas/group were used in each experiment. *p≤0.05, **p≤0.001. Statistical levels of significance were analyzed by Student’s t test (unpaired) (Fig 2B & 2E) and one-way ANOVA test with Tukey’s multiple comparison test (Fig 2C–D).

Ocular Acanthamoeba infection induces a Th1, Th2 and Th17 cell response in the cornea and dLN

Past studies have shown that corneal infection with different pathogens such as P. aeruginosa or HSV results in the induction of Th1 and/or Th17 cells depending on the type of pathogen and the stage of infection (29, 32, 44). Different CD4⁺ effector T cell responses contribute to corneal pathology through their secretion of different cytokines in the cornea (29, 32, 44). Thus, we next examined whether corneal Acanthamoeba infection induces a selective Th1, Th2 or Th17 response at different stages of AK. To characterize the relative migration of different CD4⁺ T cell subsets in the cornea, Acanthamoeba infected mice corneas were collected on day 5 and 8 pi and single cell suspensions were prepared from collagen-digested pooled corneal samples for analysis of infiltrating cells by intracellular staining for IFN-γ, IL-17A and IL-4. As evident in Figure 3A–C, we noted the infiltration of IFN-γ, IL-17A and IL-4 producing CD4⁺ T cells in the cornea on day 5 pi. However, there was no significant change in the total cell number of CD4⁺ T cell subpopulations on day 8 pi. In addition, IFN-γ, IL-4 and IL-17A cytokine analysis by ELISA of the corneal samples from various days pi showed production of these cytokines in the cornea, suggesting induction of Th1, Th2 and Th17 cell response in the cornea after Acanthamoeba infection (Fig 3D–F). Moreover, since innate cells also produce IL-17A (38), we measured the levels of this cytokine during early stages of infection (d1 and d3 pi). IL-17A induction was seen as early as d3 pi. Based on recently published studies showing that neutrophils contribute to the innate source of IL-17A during bacterial and fungal infections of the cornea (37), it is tempting to speculate that neutrophils may be the source of IL-17A in our model. However, additional studies are needed to ascertain the source of IL-17A during early stages of AK.

FIGURE 3. Induction of Th1, Th2 and Th17 effector T cell response in the cornea and local dLNs after corneal Acanthamoeba infection.

C57BL/6J mice corneas were intrastromally injected with 2.5 × 10⁴ Acanthamoebae. Flow cytometry was used to detect IFN-γ, IL-17A and IL-4 production by CD4⁺ T cells (anti-CD3/anti-CD28 stimulated for 5 hrs) obtained from pooled corneas on day 5 and day 8 pi. Representative FACS plots (left), and cell numbers (right) for IFN-γ (A), IL-17A (B) and IL-4 (C) secreting CD4⁺ T cells from pooled corneas are shown. IFN-γ (D), IL-4 (E) and IL-17A (F) protein levels in pooled corneal samples at indicated days pi. Data are representative of at least two independent experiments and show mean values ± SEM. Combined results of two separate experiments are shown, 5 to 6 corneas/group were used in each experiment. *p≤0.05. Statistical levels of significance were analyzed by Student’s t test (unpaired). The total cell numbers of IFN-γ (G); IL-17A (H) and IL-4 (I) producing CD4⁺ T cells in dLN of uninfected (day 0) and Acanthamoeba infected (day 5 and day 8 pi) mice are shown. Data are representative of two independent experiments and show mean values ± SEM (n = 8 mice at each indicated time point p.i.). ***p ≤ 0.001,**p ≤ 0.01,*p ≤ 0.05, ns- non-significant. Statistical levels of significance were analyzed by one-way ANOVA test with Tukey’s multiple comparison test.

Previous studies on ocular inflammatory conditions have shown the induction of a CD4⁺ effector T cell subset response in the dLN and possible migration of the activated effector T cells from the dLN to the site of inflammation depending on the chemokine gradient during the course of inflammation (29, 32, 44, 45). Therefore, we next examined whether corneal Acanthamoeba infection induces a similar CD4⁺ effector T cell response in dLN as observed in the cornea at various days pi. The results in Figure 3G–I show the frequency and total number of Th1, Th17 and Th2 cells in dLNs from uninfected (Day 0), and Acanthamoeba infected mice on day 5 and day 8 pi. As shown in Figure 3G, there was a significant increase in the total cell number of Th1 cells on day 5 as well as day 8 pi as compared to uninfected mice. Similarly, the total number of Th17 cells was significantly higher on day 5 and day 8 pi as compared to uninfected mice (Fig. 3H). Furthermore, when the Th2 response was analyzed, there was a significant increase in total number of Th2 cells on day 8 pi dLN samples but not in day 5 pi dLN as compared to uninfected dLN samples (Fig. 3I). Interestingly, the Th2 response was dominant in dLN as compared to the Th1 and Th17 cell response on both day 5 and day 8 pi (Fig. 3G–H). Collectively these data suggest that corneal Acanthamoeba infection results in the expansion of Th1, Th17 and Th2 cells in dLN, and migration of these cells from dLN to the cornea could be possible during the course of Acanthamoeba infection. Our data also suggest that during the severe form of keratitis on day 5 pi, there is a significant increase in Th1 and Th17 response, with no change in Th2 response in the cornea. However, a significant induction of Th2 response in dLN with no change in Th1 and Th17 response on day 8 pi as compared to day 5 pi could be responsible for diminished lesion severity during the late stage of AK on day 8 pi.

Induction of Treg response in the cornea and dLN after Acanthamoeba infection

The discovery of Treg cells and their critical role in suppression of various Th1, Th2 and Th17 cell-mediated pathologies has revealed possible new therapeutic avenues for control of several autoimmune and inflammatory conditions (46–48). Indeed, it has been demonstrated that in various ocular inflammatory conditions, shifting the balance from Th1 and Th17 cells towards Treg alleviates lesion severity and associated immunopathology (32, 49–51). To examine whether corneal Acanthamoeba infection induces a Treg cell response in the cornea and dLN, we analyzed corneal as well as dLN Treg numbers on day 5 and day 8 pi. As evident in Figure 4A, Treg constituted a major proportion of total CD4⁺ T cells (approximately 40%) in the cornea during early and late stages of corneal Acanthamoeba infection. However, there was no significant change in percentage (not shown) or total number of Treg cells on day 8 pi as compared to day 5 pi (Fig. 4B). Likewise, there was a strong induction of Treg response in dLN after corneal Acanthamoeba infection as evident by a significant increase in Treg cell numbers on day 5 and day 8 pi when compared to dLN from uninfected mice (Fig. 4C). Collectively, our data suggest that corneal Acanthamoeba infection induces Treg cell response in the cornea and dLN on day 5 as well as day 8 pi.

FIGURE 4. Corneal Acanthamoeba infection induces Treg cell response in the cornea and lymphoid organs.

C57BL/6J mice corneas were intrastromally injected with 2.5 × 10⁴ Acanthamoebae for corneal infection. Flow cytometry was used to detect Foxp3-expressing CD4⁺ T cells obtained from corneas on day 5 and day 8 pi and dLN from uninfected (day 0) and Acanthamoeba-infected mice (day 5 and day 8 pi). Representative FACS plots (A) and cell numbers (B) for Foxp3⁺ CD4⁺ T cells obtained from pooled corneal samples are shown. (C) The total cell numbers of Foxp3⁺ CD4⁺ T cells in dLN of uninfected (day 0) and Acanthamoeba-infected (day 5 and day 8 pi) mice. Data are representative of two independent experiments and show mean values ± SEM (n = 8 mice at each indicated time point). ***p ≤ 0.001,**p ≤ 0.01,*p ≤ 0.05, ns- non-significant. Statistical levels of significance were analyzed by one-way ANOVA test with Tukey’s multiple comparison test.

Protective role of IL-17A: IL-17A neutralization post Acanthamoeba infection increases AK lesion severity

Although IL-17A has been indicated as a proinflammatory cytokine in various autoimmune and chronic inflammatory conditions (26), several studies have shown that IL-17A is an integral component of host protection against fungal pathogens such as Candida albicans and Fusarium (33, 37, 38, 52, 53). Since our earlier data suggested IL-17A induction in the cornea after Acanthamoeba infection (Fig. 3F), it was of interest to determine whether IL-17A exacerbates the AK disease or provides protection. To investigate the role of IL-17A in AK, the effect of IL-17A neutralization was studied over 8 days pi. To evaluate whether IL-17A exacerbates the disease or contributes to host protection against corneal Acanthamoeba infection, C57BL/6J mice were infected with Acanthamoeba, and anti-IL-17A mAb (5 μg in 10 μl/eye) was administered by subconjunctival injections every other day from day 1 pi to day 7 pi. The AK lesion severity was recorded every alternate day and compared with isotype mAb-treated control mice (Fig. 5A–C). As shown in Figure 5A and C, all infected mice on day 1 had a similar extent of AK lesion severity, which increased in both groups on day 3 pi. However, as infection progressed, animals in the anti-IL-17A mAb-treated group showed a significant increase in corneal opacity and AK lesion severity as compared to the normal course of AK observed in isotype mAb-treated mice (Fig. 5A–C). These findings are in sharp contrast to other corneal infections such as HSV and Pseudomonas keratitis where IL-17A contributes to corneal pathology through an increased inflammatory response in the cornea (29, 30, 32, 54). Taken together, these data suggest that increased IL-17A expression after corneal Acanthamoeba infection is essential for host protection against the severe form of corneal pathology as IL-17A neutralization promotes AK pathogenesis and delays resolution of corneal opacity. Next, we analyzed whether IL-17A neutralization had any effect on total number of amoebae as well as corneal infiltrating neutrophils during the early phase of infection. As shown in Figure 5D–E, IL-17A neutralization showed an increase in total amoebae and decrease in neutrophil numbers. However, the difference between the IL-17A neutralization and control group was not statistically significant.

FIGURE 5. IL-17A neutralization post Acanthamoeba infection exacerbates keratitis severity.

C57BL/6J mice corneas were intrastromally injected with 2.5 × 10⁴ Acanthamoebae for corneal infection. Infected mice were treated with anti-IL-17A (5 μg in 10 μl/eye) or isotype mAb (5 μg in 10 μl/eye) by local subconjunctival injections from day 1 to 7 pi. The progression of keratitis was monitored by measuring corneal opacity scores on day 1, 3, 5 and 7 pi. (A) The severity of Acanthamoeba keratitis was significantly increased in anti-IL-17A-treated animals compared to control mice treated with isotype mAb. Kinetic analysis for mean scores at each indicated day pi is shown. (B) Individual eye scores for corneal opacity on day 7 pi from anti-IL-17A and isotype control groups are shown. (C) Representative eye photos on day 1, 3, 5 and 7 pi show corneal lesion severity from anti-IL-17A-treated animals as compared to isotype mAb-treated mice. All eye pictures at different days pi are from the same eye of each group. (D) Total number of RFP⁺ Acanthamoebae on day 3 pi between anti-IL-17A and isotype mAb treated group. (E) Total number of neutrophils on day 3 pi between anti-IL-17A and isotype mAb treated group. Data are representative of two independent experiments and show mean values ± SEM (n = 22–25 mice/group; *p<0.05). Statistical levels of significance were analyzed by Student’s t test (unpaired).

Anti-IL-17A treatment shifts the balance between Teffector and Treg cell response in dLN

To assess the effect of IL-17A neutralization on increased AK lesion severity, CD4⁺ T cell subset composition in dLN of anti-IL-17A- and isotype mAb-treated animals was analyzed by FACS at day 8 pi. As shown in Figure 6A–B, anti-IL-17A treatment significantly increased the total number of Th1 and Th17 cells in dLN as compared to isotype mAb-treated group. Although the percentage of Th2 cells was unchanged between anti-IL-17A and isotype mAb-treated group, there was an increase in the total number of Th2 cells in the anti-IL-17A treated group as compared to isotype mAb treated mice (Fig. 6C). Anti-IL-17A treatment slightly decreased Treg percentage, but total Treg cell numbers were increased as compared to isotype-treated mice (Fig. 6D). Although there was a significant increase in total Treg cell number in anti-IL-17A-treated group (Fig. 6D), the total number of Treg cells per Th1 or Th17 cell was significantly diminished in the anti-IL-17A group as compared to isotype-treated mice (Fig. 6E–F). Collectively, these data suggest that increased AK lesion severity in anti-IL-17A-treated mice post Acanthamoeba corneal infection could result from an increase in total number of Th1 and Th17 cells and a significant decrease in the proportion of Treg per Th1 and Th17 cells as compared to isotype mAb-treated mice.

FIGURE 6. IL-17A neutralization after corneal Acanthamoeba infection promotes CD4⁺ T cell effector and regulatory T cell response in lymphoid organs.

Acanthamoeba-infected C57BL/6J mice were treated with anti-IL-17A or isotype mAb by local subconjunctival injections from day 1 to 7 pi. Flow cytometry was used to detect IFN-γ, IL-17A and IL-4 production by CD4⁺ T cells (anti-CD3/anti-CD28 stimulated for 5 hrs) obtained from dLN of anti-IL-17A- and isotype-treated animals on day 8 pi. Representative FACS plots (left) and total cell numbers (right) of IFN-γ (A), IL-17A (B), IL-4 (C) producing CD4⁺ T cells and Foxp3⁺ CD4⁺ T cells (D) in dLN on day 8 pi are shown. Cell ratios for total numbers of Treg per Th1 cell (E) and Treg per Th17 (F) in dLN from anti-IL-17A- (clear bar) and isotype mAb-treated (grey bar) groups. Data are representative of two independent experiments and show mean values ± SEM (n = 8 mice at each indicated time point p.i.). ***p ≤ 0.001,**p ≤ 0.01,*p ≤ 0.05, ns- non-significant. Statistical levels of significance were analyzed by one-way ANOVA test with Tukey’s multiple comparison test.

Anti-IL-17A treatment post Acanthamoeba infection promotes inflammatory cell migration into the cornea

To further characterize the mechanism by which anti-IL-17A treatment leads to increased corneal pathology, corneal infiltration of various immune cell types was examined on day 8 pi. The number of CD45⁺ cells on day 8 pi was ~6 fold higher in anti-IL-17A-treated mice than in the isotype-treated animals (Fig. 7A). Although the percentages of neutrophils and CD4⁺ T cells were similar in both groups, numbers of neutrophils (~10 fold) and CD4⁺ T cells (~4 fold) were higher in the anti-IL-17A-treated group when compared to isotype-treated mice (Fig. 7B–C). Additional experiments of the same design were conducted to compare the effector functions of the infiltrated CD4⁺ T cells in the anti-IL-17A- and isotype mAb-treated mice. Ex vivo analysis of infiltrating cells isolated from corneal tissues of anti-IL-17A and isotype-treated mice was carried out for CD4⁺ T cell production of IL-17A and IFN-γ by stimulating corneal cells with anti-CD3 and anti-CD28 for 5 hrs in the presence of Golgi plug. There was a ~2 fold increase in the percentage and ~10 fold increase in the total number of Th1 and Th17 cells in the cornea of IL-17A neutralized group over those in isotype-treated animals (Fig. 7D–E). Although the frequency of Treg was slightly reduced in the anti-IL-17A-treated group, Treg numbers were ~3 fold increased in anti-IL-17A-treated group (Fig. 7F). However, the ratio of Treg per Th1 and Treg per Th17 cell was ~2.5 fold lower in anti-IL-17A-treated group compared to the isotype- treated mice (Fig. 7G–H). Taken together, our results suggest that the inhibition of IL-17A response after corneal Acanthamoeba infection increases keratitis lesion severity along with an increase in migration of both innate immune cells and effector CD4⁺ T cells in the cornea during the late stage of AK pathogenesis.

FIGURE 7. IL-17A neutralization after corneal Acanthamoeba infection promotes migration of both innate and adaptive immune cells in the cornea.

Acanthamoeba-infected C57BL/6J mice were treated with anti-IL-17A or isotype mAb by local subconjunctival injections from day 1 to 7 pi. Immune cells infiltrating the corneas were quantified on day 8 pi by flow cytometry. Representative FACS plots (left) and total cell numbers (right) of CD45⁺ cells (A), CD11b⁺ Ly6G⁺ polymorphonuclear neutrophils gated on CD45⁺ cells (B) and, CD4⁺ T cells gated on CD45⁺ cells (C). Representative FACS plots (left) and total cell numbers (right) of IFN-γ (D) and IL-17A (E) producing CD4⁺ T cells harvested from the corneas on day 8 pi and stimulated with anti-CD3/anti-CD28 for 5 h. (F) Representative FACS plots (left) and total cell numbers (right) of Foxp3⁺ CD4⁺ T cells. Cell ratios for total numbers of Treg per Th1 cell (G) and Treg per Th17 (H) in the cornea from anti-IL-17A- (clear bar) and isotype mAb-treated (grey bar) groups. Results of a representative experiment of two with a minimum 8 mice per group are shown.

Increased susceptibility of IL-17AKO mice to AK pathogenesis

To further confirm the protective role of IL-17A during corneal Acanthamoeba infection, corneas of wild-type and IL-17AKO mice were infected with Acanthamoeba and the development and progression of AK pathology was monitored from day 1 through day 8 pi. As shown in Figure 8A–C, IL-17AKO mice exhibited significantly increased AK lesion severity as compared to wild-type mice. Indeed, corneas of some animals in the IL-17AKO group perforated due to increased disease severity by day 8 pi (Fig 8B, highest AK score of 4). To further understand the IL-17A-mediated protection during AK pathogenesis, we quantified amoebae numbers in wild-type and IL-17AKO mice on day 2, 3 and 8 pi (Fig 8D). Although we did not observe significant change in Acanthamoebae numbers between wild-type and IL-17AKO mice on day 2 pi, there was significantly less clearance of Acanthamoebae from corneas of IL-17AKO mice on day 3 pi as compared to wild-type mice (Fig 8D). However, by day 8pi, Acanthamoebae numbers reduced to slightly over background levels in the corneas of both wild type and IL-17A KO mice (Fig. 8D). Furthermore, on day 8pi, the number of corneal infiltrating leukocytes, neutrophils and CD4⁺ T cells was higher in IL-17AKO group as compared to wild-type mice (Fig 8-E–G). Similarly, dLN analysis for various CD4⁺ T cell subsets showed a significant increase in Th2, and Treg cell response in the IL-17AKO group as compared to the wild-type mice (Fig 8H–J). Taken together, these findings suggest that initial IL-17A response is critical for controlling Acanthamoebae numbers in the cornea during early stages of AK pathogenesis. Increased disease severity in IL-17AKO mice could result from Acanthameobae-mediated tissue damage during early stages of infection followed by uncontrolled immune response during late stages of AK pathogenesis.

FIGURE 8. Increased AK lesion severity in IL-17AKO mice.

Wild-type and IL-17AKO mice were infected with Acanthamoeba and progression of AK lesion severity from day 1 through day 8 pi was monitored followed by immune cell parameter analysis on day 8 pi in the cornea as well dLNs. (A) The severity of Acanthamoeba keratitis was significantly increased in IL-17AKO mice compared to wild-type animals. Kinetic analysis for mean scores at each indicated day pi is shown. (B) Individual eye scores for corneal opacity on day 7 pi from IL-17AKO and wild-type mice groups are shown. (C) Representative eye photos on day 1, 3, 5 and 7 pi showing corneal lesion severity from IL-17AKO mice as compared to wild-type mice. All eye pictures at different days pi are from the same eye of each group. (D) Total number of RFP⁺ Acanthamoebae on day 2 and 3 pi in corneas of IL-17AKO and wild-type group. Total cell numbers of CD45⁺ cells (E), CD11b⁺ Ly6G⁺ polymorphonuclear neutrophils gated on CD45⁺ cells (F), and CD4⁺ T cells gated on CD45⁺ cells (G) in the cornea on day 8 pi are shown. Total cell numbers of IFN-γ (H), IL-4 (I), and Foxp3⁺ CD4⁺ T cells (J) in dLN on day 8 pi are shown. Combined results of two separate experiments are shown as mean values ± SEM. *p≤0.05, **p≤0.001. Statistical levels of significance were analyzed by Student’s t test (unpaired).

Discussion

In this report, using a mouse corneal intrastromal injection-induced AK model, we have shown the critical role of IL-17A, neutrophils and CD4⁺ T cells in the pathogenesis of AK. Our data suggest that Acanthamoeba infection leads to increased IL-17A production in the cornea that is essential for host protection against invading corneal Acanthamoeba and associated corneal tissue damage. Specifically, we demonstrate that the severity of Acanthamoeba infection is significantly higher in IL-17A KO mice compared to the wild type mice. Likewise, neutralization of IL-17A in wild-type mice resulted in increased AK severity. Furthermore, our results also suggest that corneal Acanthamoeba infection induces both Teffector (Th1, Th2 and Th17) and Treg cell responses in the cornea as well as dLN and that sequential migration of innate and adaptive immune cells with the induction of IL-17A response is required for optimal clearance of Acanthamoeba infection.

Corneal transparency is a prerequisite for optimal vision. However, different injuries to the cornea including infections with microorganisms such as viruses (HSV), bacteria (Pseudomonas aeruginosa, Staphylococcus aureus), fungi (Candida albicans, Aspergillus) and amoeba (Acanthamoeba) induce various inflammatory events, particularly migration of immune cells and destruction of corneal architecture, that impede optimal transmission of light leading to vision impairment (44, 45, 55). Although the role of host and pathogen derived factors have been well studied in the pathogenesis of viral, bacterial and fungal corneal infections (44, 45, 55), the role of the host immune response, particularly contribution of adaptive immune cells and their effector molecules, in AK pathogenesis is poorly defined (14). This is, in part, due to the lack of availability of a mouse model of AK. As discussed earlier, Niederkorn and colleagues have developed Chinese hamster and pig animal models to study AK (14, 17, 18). In these studies, only Chinese hamsters and pigs were found to be susceptible to self-limiting AK. However, lack of molecular, cellular, genetic and immunological tools to study the role of the immune response in the above disease models has necessitated the development of AK mouse models using different routes to infect the cornea. Recently, three different methods including corneal scratch, corneal scratch covered with amoebae-loaded contact lenses, and direct intrastromal injection of Acanthamoeba have been used to establish rat and mouse models of AK pathogenesis (56). Comparison of these three models showed that direct intrastromal injection of Acanthamoeba is the most effective model to induce AK in rat and mouse (56). However, in this model, live amoebae were not isolated from infected cornea, and the impact of infecting with dead amoebae was not assessed. This made it difficult to dissect the effect of amoebae infection versus surgical trauma. In the current study, we demonstrate that live amoebae could be isolated and successfully cultured from infected cornea, and that when dead amoebae are injected, the disease does not progress beyond day 2 pi. Considering that contact lens wear and ocular surface injury are the two major risk factors in the pathogenesis of AK, it may be asked to what extent the model used in this study involving intrastromal injection of parasite is valid for understanding the pathogenic mechanisms of AK. The method used in the current study is clearly not suitable for studying initial steps of infection such as mechanisms mediating adhesion of parasites to host cells. However, the method is well suited for studying immune response because the AK infection does not proceed unless the parasites penetrate the stroma. Indeed, in corneas of patients with AK, the parasites are invariably seen in the stroma. Clinically, AK pathogenesis in the mouse was acute with peak disease severity around day 3 to 5 pi followed by a sharp decline in corneal lesion severity. Although corneal opacity diminished after day 7 pi, indicating the self-limiting course of AK pathogenesis, we were able to culture Acanthamoeba from corneal samples obtained on day 14 pi, suggesting that residual parasites resided much longer than the overt symptoms of the disease. Furthermore, our findings that infection with dead Acanthamoeba does not induce AK suggests the critical role of virulence factors induced by live amoebae in the corneal tissue damage. Thus, our data support the notion that both Acanthamoeba-induced factors and host immune response contribute to the induction and resolution of AK, respectively.

Past studies using hamster and pig models have shown the critical role of neutrophils and macrophages in AK pathogenesis (3, 14, 19, 20). It has been reported that depletion of macrophages or inhibition of neutrophil migration in the cornea following Acanthamoeba infection leads to a severe form of AK (14, 19, 20). In the present study using the mouse model, we also observed that the neutrophil is the predominant cell type during early stage of AK pathogenesis followed by a gradual decrease in both neutrophil numbers and AK lesion severity. Similarly, clinical samples of AK patients have shown the predominance of neutrophils in the cornea following Acanthamoeba infection (3, 14).

IL-17A plays a central role in migration and activation of neutrophils (26, 28, 57). Specifically, IL-17A stimulates a variety of cell types including epithelial cells, endothelial cells and fibroblasts to produce and release chemokines (CXCL1, MIP2), cytokines (IL-6, TNF-α), metalloproteases (MMP-1, -3, -9 and -13) and growth factors (G-CSF, GM-CSF) essential for migration and activation of innate cells such as neutrophils and macrophages (26, 28, 57, 58). The predominance of neutrophils and increased IL-17A expression we observed during corneal Acanthamoeba infection is consistent with the critical role of IL-17A in AK pathogenesis. Although past studies showed that IL-17A promotes corneal inflammation during HSV and P. aeruginosa infection (29, 30, 32), our data suggest that IL-17A contributes to host protection during Acanthamoeba infection. Thus, infection of IL-17AKO mice or IL-17A neutralization post Acanthamoeba infection resulted in increased corneal lesion severity as well as increased effector CD4⁺ T cell response in both dLN and cornea. Although IL-17A initially was shown to promote the proinflammatory response during the pathogenesis of various chronic inflammatory and autoimmune diseases (26), recent studies have indicated the critical role of IL-17A in host defense against bacterial, fungal and protozoan parasites (33–35). Furthermore, IL-17A has also been shown to play a protective role in T cell-mediated intestinal inflammation (59). Thus, it appears that the type of pathogen and inflammatory response based on tissue or organ system influences the function of IL-17A in host protection and inflammation. As stated earlier, multiple studies have reported that IL-17A promotes neutrophil activation and migration. However, we observed an increased migration of neutrophils into the cornea of anti-IL-17A-treated mice. This could result from either increased Acanthamoeba load in the cornea or reduced activation of neutrophils and macrophages in the absence of IL-17A facilitating further migration of these cells into the cornea. It is also possible that compensatory mechanisms could have induced increased synthesis of neutrophil chemokines in the absence of IL-17A in anti-IL-17A-treated mice. Moreover, IL-17A has also been shown to play an immunoregulatory role through reduction in accumulation and activity of neutrophils during inflammation (60).

Little is known about the role of CD4⁺ T cells in the pathogenesis of AK. Histopathological analyses have demonstrated infiltration of CD4⁺ T cells in human corneas from Acanthamoeba infected patients (3, 22–24). Similarly, our data also indicated the presence of CD4⁺ T cells in Acanthamoeba infected corneas as early as day 5 pi. Furthermore the data presented here suggest that in AK, there is a predominance of Treg followed by Th1, Th17 and Th2 cells in infected corneas as well as dLN. Corneal Acanthamoeba infection induced a strong Teffector (Th1, Th2 and Th17 cell) and Treg response in local dLN. Corneal HSV and P. aeruginosa infection also induce CD4⁺ effector T cell response, however it is predominantly Th1-mediated in HSV infection and Th17-mediated in P. aeruginosa infection (29, 32). Furthermore, these studies have indicated that both Th1 and Th17 responses contribute to corneal immunopathology and that skewing the balance from Th1 and Th17 towards Th2 and/or Treg inhibits corneal pathology (32, 49–51). Although corneal Acanthamoeba infection also showed the induction of the Treg and Th2 response, further studies are necessary to demonstrate the role of these cells in limiting corneal tissue damage and associated immunopathology after Acanthamoeba infection.

AK is a globally emerging, very painful condition of the cornea often leading to vision impairment in chronically infected patients (1, 61). Although once considered a rare disease, the incidence of AK has been increasing worldwide due to extensive use of contact lenses, better diagnosis and increased awareness of AK among clinicians as well as patients (1, 61). If the disease is not diagnosed early and treated aggressively, extensive ocular damage can occur, and enucleation (complete removal of eye ball) may be required (62–65). Currently, AK diagnosis is not straightforward and treatment is very demanding (66–68). Although hourly, around-the-clock, aggressive treatment involving the use of three or more drugs can contain the disease, patients must often be treated for months, and, in some cases, extensive treatment must be sustained for over a year (68, 69). Despite the aggressive treatment, infrequently, the disease fails to resolve, and some patients withdraw therapy because of excruciating burning sensation associated with medication instillation (66). Therefore, understanding various crucial events for induction of effective host immune response against Acanthamoeba is critical for the design of novel therapeutics to target this globally emerging, sight-threatening condition. Our data suggesting the essential role of IL-17A in minimizing corneal tissue damage and AK lesion severity after corneal Acanthamoeba infection is a novel finding that is in sharp contrast to published studies showing IL-17A-mediated corneal damage during P. aeruginosa and HSV keratitis. In summary, our data suggest that the IL-17A response is essential for protection against AK. These findings lay the foundation for additional studies to characterize the mechanisms by which IL-17A provides protection against AK. For example, it would be important to investigate whether lack of IL-17A response affects the phagocytic activity of neutrophils and macrophages that is essential for killing of Acanthamoeba trophozoites. It is our hope that with time our findings will lead to the development of strategies to manage the aspects of AK for the benefits of patients.

Abbreviations used in this article

AK

Acanthamoeba keratitis

dLN

draining lymph node

Foxp3

Forkhead box P3

pi

post infection

Treg

regulatory T cells