Experimental Induction of Acute Acanthamoeba castellanii Keratitis in Cats

Abstract

Purpose: To develop a feline model of acute Acanthamoeba keratitis using methods that replicate natural routes of infection transmission.

Methods: Corneal Acanthamoeba castellanii inoculation was performed by three methods: topical inoculation with Acanthamoeba solution following corneal abrasion, placement of a contaminated contact lens for 7 days, and placement of a contaminated contact lens for 7 days following corneal abrasion. Sham inoculations with parasite-free medium and sterile contact lenses were also performed. Cats were monitored by ocular examination and in vivo corneal confocal microscopy for 21 days post-inoculation. Corneal samples were collected at intervals for microbiologic assessment, histopathology, and immunohistochemistry.

Results: All cats in the corneal abrasion groups developed clinical keratitis. Clinical ocular disease was inconsistently detected in cats from the contaminated contact lens only group. Initial corneal lesions were characterized by multifocal epithelial leukocyte infiltrates. Ocular lesions progressed to corneal epithelial ulceration and diffuse stromal inflammation. After 14 days, corneal ulcerations resolved, and stromal inflammation consolidated into multifocal subepithelial and stromal infiltrates. Corneal amoebae were detected by culture, in vivo confocal microscopy, histopathology, and immunohistochemistry in cats with keratitis. Neutrophilic and lymphocytic keratoconjunctivitis with lymphoplasmacytic anterior uveitis were identified by histopathology. Coinfection with aerobic bacteria was detected in some, but not all, cats with keratitis. Ocular disease was not detected in the sham inoculation groups.

Conclusions: Feline Acanthamoeba keratitis is experimentally transmissible by contaminated contact lenses and topical inoculation following corneal epithelial trauma.

Translational Relevance: Experimentally induced acute Acanthamoeba keratitis in cats is clinically and histopathologically similar to its human counterpart.

Introduction

Acanthamoeba spp. are free-living amoebae that are widely distributed in diverse natural and manmade habitats.¹ Acanthamoeba keratitis is a vision-threatening ocular infection that is frequently recalcitrant to therapy and associated with prolonged ocular morbidity.^2,3 Contact lens wear is the most common risk factor associated with the development of Acanthamoeba keratitis; however, ocular infection may also develop subsequent to corneal trauma with exposure to contaminated water or soil.^4,5 

Acanthamoeba keratitis can be experimentally induced in pigs, rabbits, hamsters, rats, and mice; however, naturally acquired corneal Acanthamoeba infection is not reported in these host species.^6–11 To consistently induce keratitis, many of the previously described animal models of Acanthamoeba keratitis utilize methods that do not replicate natural infection pathogenesis in human patients, including techniques such as corneal stromal injection of amoebae, concurrent treatment with immunosuppressive medications, or delivery of amoebae to the cornea in wax films.^9–11 The potential pathophysiologic contribution of corneal coinfection with other pathogens has also received limited investigation in most experimental animal models of Acanthamoeba keratitis. 

In contrast to the previously described animal models of ocular Acanthamoeba infection, naturally acquired Acanthamoeba keratitis is described in the domestic cat (Felis catus).^12,13 The feline cornea is also highly susceptible to in vitro binding and invasion by Acanthamoeba castellanii, a characteristic that is typically predictive of in vivo susceptibility to corneal infection.¹⁴ Acanthamoeba castellanii binding to the feline cornea in vitro is enhanced by, but does not require, a prior epithelial defect.¹⁴ 

The purpose of the present study was to develop a feline model of acute Acanthamoeba keratitis using experimental methods that replicate natural routes of infection transmission in human patients. In addition, this study evaluated the potential contributions of coinfections with other infectious agents in cats with experimental Acanthamoeba keratitis. 

Materials and Methods

Animals

All protocols were approved by the Animal Care and Use Committee of Cornell University (Protocol #2009-0104) and were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Fifteen 9-month-old, male laboratory cats were maintained in individual cages in an Association for Assessment and Accreditation of Laboratory Animal Care Accredited facility. All cats received complete ophthalmic and physical examinations upon arrival at the facility. Blood was collected from each cat by peripheral venipuncture for hemogram and serum biochemistry panel analysis. 

Acanthamoeba and Contact Lens Preparation

A human corneal Acanthamoeba castellanii isolate (ATCC #30868) was cultured axenically in peptone–yeast–glucose (PYG) medium as previously described.¹⁵ The parasite suspension was cultivated to a concentration of 3 × 10⁶ amoebae/mL and contained approximately 90% trophozoites and 10% cysts. Hydrophilic feline soft contact lenses (Acrivet Pat Classic; Bausch & Lomb, Vaughan, ON, Canada) were individually fitted to study cats by measurement of the corneal diameter with a caliper. For the infection groups, the contact lenses were incubated submerged in the amoeba suspension for 24 hours at 35°C. For the sham inoculation group, the contact lenses were incubated with sterile PYG medium in the same manner. 

Study Design and Acanthamoeba Inoculations

After a 2-week acclimation period, cats were randomly assigned into five study groups (three cats per group) divided by method of inoculation. Cats were sedated with intravenous ketamine (3.0 mg/kg) and diazepam (0.3 mg/kg). Proparacaine 0.5% ophthalmic solution (Akorn, Lake Forest, IL) was applied to the right eye of each cat for topical anesthesia. Corneal Acanthamoeba inoculation was performed in the right eye of cats by one of three methods: (1) topical inoculation with the amoeba suspension (0.2 mL) following light corneal abrasion, (2) placement of a contaminated contact lens for 7 days, and (3) placement of a contaminated contact lens for 7 days following light corneal abrasion. Sham inoculations were performed in the right eye of cats by one of two methods: (1) topical inoculation of parasite-free PYG medium (0.2 mL) following light corneal abrasion, and (2) placement of a contact lens incubated with sterile PYG medium for 7 days following light corneal abrasion. 

Corneal abrasions were performed with a 25-gauge needle and light manual pressure. Ten parallel lines of slight epithelial disruption were made in the horizontal and vertical meridians to form a grid pattern in the central cornea (10 mm × 10 mm). A partial tarsorrhaphy was performed with a single horizontal mattress suture of 4-0 silk through the lateral eyelid margins in all cats (to facilitate contact lens retention). The contact lenses and temporary tarsorrhaphy sutures were removed after 7 days. All cats received subcutaneous buprenorphine (0.01 mg/kg) every 12 hours for the duration of the study for analgesia. 

Cats were monitored for 21 days post-inoculation by clinical ophthalmic examinations performed daily. At 7-day intervals, cats were sedated with intravenous ketamine and diazepam, as previously described, and corneal in vivo confocal microscopy (IVCM) examinations were performed following application of topical anesthetic. Corneal swab samples were collected from all cats for microbiological assessment. One randomly selected cat in each study group was euthanized by intravenous pentobarbital (120 mg/kg) administration at 7-day intervals following examination and sample collection. Whole globes were collected immediately postmortem for histopathological and immunohistochemical evaluation, and a complete necropsy was performed. 

Clinical Examinations and Microbiologic Sample Collections

Complete physical examination and ophthalmic examination, including Schirmer I tear tests, slit-lamp biomicroscopy (Kowa SL-15; Kowa Company, Tokyo, Japan) before and after application of fluorescein stain (FUL-GLO fluorescein sodium strips; Moore Medical LLC, Farmington, CT), applanation tonometry (Tono-Pen XL; Reichert, Inc., Depew, NY), and indirect ophthalmoscopy (Heine Omega 500; Heine Optotechnik, Herrsching, Germany) were performed on both eyes of each cat prior to amoeba or sham inoculation. Abbreviated ophthalmic examinations were then performed on both eyes of each cat daily throughout the study. These abbreviated ophthalmic examinations included slit-lamp biomicroscopy before and after corneal application of fluorescein stain. The presence of corneal epithelial ulceration and re-epithelialization was determined by the presence or absence of fluorescein retention as observed with slit-lamp biomicroscopy. A previously reported⁷ clinical scoring system for experimental Acanthamoeba keratitis was used to quantify examination findings and scores were calculated every 2 days. With this clinical scoring system, epithelial defects, stromal edema, vascularity, and stromal opacities were each scored on a scale of 0 to 4 as previously described.⁷ 

Corneal sample collections for microbiological analysis were collected following clinical and IVCM examinations. Sterile, polyester-tipped swabs were brushed against the corneal surface, avoiding contact with eyelid skin and cilia, for amoeba culture, aerobic bacterial culture, fungal culture, and feline herpesvirus-1 (FHV-1) polymerase chain reaction (PCR) assay. 

In Vivo Confocal Microscopy Examinations

Corneal IVCM examinations were performed with a modified Heidelberg Retina Tomograph II and Rostock Cornea Module (Heidelberg Engineering, Heildelberg, Germany) using a 63× objective (Carl Zeiss Meditec, Jena, Germany). Confocal microscopic examinations were performed on the right eye of each cat. Several drops of contact gel (GenTeal tear gel; Novartis Pharmaceuticals, East Hanover, NJ) were applied to the front of the microscope lens and ocular surface. A sterile, single-use polymethyl methacrylate cap (TomoCap; Heidelberg Engineering) mounted on the microscope lens was positioned perpendicular to, and in slight contact with, the corneal surface. The polymethyl methacrylate caps were changed between each cat examined. Multipoint IVCM examination of the axial, paraxial, and peripheral cornea was performed in a systematic manner. Full-thickness images of corneal lesions were captured using manual and automated image acquisition modes. Following examinations, digitized corneal images were analyzed for pathology. 

Histopathology and Immunohistochemistry

Sections (4 µm thick) of formalin-fixed, paraffin-embedded corneal tissue were prepared. Sections for histopathology were stained with hematoxylin and eosin (H&E) or periodic acid–Schiff. Sections for immunohistochemistry were deparaffinized in xylene and rehydrated in graded ethanol. Antigen retrieval was performed by heating sections in citrate buffer (0.01 mol/L, pH 6.0) for 2 × 10 minutes. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in methanol for 10 minutes. Nonspecific staining was blocked with a mixture of 10% goat serum and 2× casein for 30 minutes at room temperature. The primary antibody (rat anti-Acanthamoeba castellanii polyclonal antibody; Antibody Systems, Inc., Hurst, TX) was diluted to 1:500 in phosphate-buffered saline (PBS) containing 1× casein, and incubated with the sections for 1.5 hours at 37°C. The secondary antibody (biotinylated goat anti-rabbit IgG; Vector Laboratories, Burlingame, CA) was diluted to 1:200 in PBS and incubated with the sections for 30 minutes at room temperature. 

An immunoperoxidase method of Acanthamoeba visualization was used and the avidin–biotin–peroxidase complex method was followed. NovaRED or 3-amino-9-ethylcarbazole (AEC; Invitrogen, Carlsbad, CA) was used as chromogen, and the sections were lightly counterstained with hematoxylin. PBS with 0.05% Tween 20 was used for washing throughout the procedure. Rat sera at the same final protein concentration were used as a negative control. Histology and immunohistochemical results were examined using a compound microscope. 

Microbiologic Assays

Corneal samples for amoeba culture were cultured onto non-nutrient agar inoculated with Escherichia coli strain JM109. Samples were divided and incubated at 25°C and 35°C. Cultures were examined for cysts and trophozoites daily for 14 days using inverted microscopy. The FHV-1 PCR assays were performed in a real-time PCR detection system (Applied Biosystems, Waltham, MA) using a Platinum SYBR Green qPCR SuperMix-UDG kit (Invitrogen). The thermocycling conditions included preincubation for 2 minutes at 50°C to prevent the amplification of carryover PCR products, denaturation for 2 minutes at 95°C, 40 cycles of denaturation at 95°C for 15 seconds, and annealing and extension at 60°C for 1 minute. The primers employed were: FHV-1 TK and TK rev for the FHV-1 thymidine kinase (TK) gene. 

For aerobic bacterial cultures, an automated system (Sensititre; Trek Diagnostic Systems, Cleveland, OH) was employed for bacteriological identifications (New York State Animal Health Diagnostic Center, Ithaca, NY). Automated bacterial identifications were supplemented with bacterial phenotypic classification tests and reverse-transcription PCR (RT-PCR) speciation to confirm identifications when necessary. Fungal cultures were performed by direct inoculation onto Sabouraud dextrose Emmons agar plates and inhibitory mold agar plates and enrichment in aerobic blood culture bottles. Agar plates and enrichment bottles were incubated at 27°C to 31°C in sealed bags without carbon dioxide for 14 days and examined regularly for fungal growth. 

Statistical Analysis

The frequency of clinical keratitis development and corneal Acanthamoeba detection by the various diagnostic modalities was compared between the infection and control study groups using Fisher's exact test. The frequency of bacterial growth on culture was compared between cats in the infection and control study groups using Fisher's exact test. Statistical significance was defined by P ≤ 0.05 for all comparisons. 

Results

Clinical Findings

Physical and complete ophthalmic examinations, including tonometry and Schirmer tear tests, performed prior to the beginning of the study were unremarkable for each cat. Values for hemograms and serum biochemistry panels were within reference ranges for each cat. All six cats in the corneal abrasion groups (including those infected by topical inoculation and placement of a contaminated contact lens) clinically developed keratitis. Keratitis was detected in a single cat from the contaminated contact lens only group without corneal abrasion. The remaining two cats from the contaminated contact lens only group remained clinically normal throughout the duration of the study. The ocular disease followed a similar clinical course in the first 2 weeks post-inoculation in all cats that developed keratitis. 

Ocular lesions on post-inoculation days 1 to 4 included mild conjunctivitis, intermittent blepharospasm, mucopurulent ocular discharge, and corneal lesions characterized by an irregular epithelium and multifocal epithelial leukocyte infiltrates (Figs. 1, 2). Corneal infiltrates assumed a linear pattern in cats from the corneal abrasion groups and were located in the previous areas of manual abrasion. Multifocal punctate areas of corneal ulceration that retained fluorescein were present and confined to the linear abrasion tracks. In the single cat with keratitis from the contaminated contact lens only group, the corneal infiltrates were present in a multifocal punctate pattern in the central cornea. On post-inoculation days 5 to 7, geographic regions of the axial corneal epithelium ulcerated, epithelial and subepithelial leukocyte infiltrates progressed, and focal central corneal edema developed (Figs. 1, 2). Lesions progressed during post-inoculation days 8 to 12 to moderate conjunctivitis, corneal epithelial ulceration, mild peripheral corneal vascularization, diffuse stromal inflammation, diffuse corneal edema, and anterior uveitis (Figs. 1, 2). Large superficial geographic corneal ulcers were present that encompassed the central 25% to 50% of the axial corneal surface in each cat with keratitis. Anterior uveitis was associated with mild-to-moderate aqueous flare and miosis. Clinical corneal lesions were similar between all cats that developed keratitis in the first 12 post-inoculation days, and only mild variation in lesion severity was noted between the cats (Fig. 3). No subjective clinical differences were detected between cats inoculated by the three different methods. 

Figure 1.

 

Clinical photographs of cats with experimentally induced Acanthamoeba castellanii keratitis. (A, B) Post-inoculation day 3, corneal lesions were characterized by an irregular epithelium and multifocal epithelial leukocyte infiltrates, (C, D) Post-inoculation day 6, geographic regions of superficial corneal ulceration developed (corneas stained with fluorescein in the photographs), and epithelial and subepithelial leukocyte infiltrates had progressed. (E, F) Post-inoculation day 10, large regions of corneal epithelial ulceration (corneas stained with fluorescein in the photographs), diffuse stromal inflammation, diffuse corneal edema, peripheral stromal vascularization, and anterior uveitis were present. (G, H) Post-inoculation day 14 (G) and post-inoculation day 18 (H), the regions of corneal ulceration epithelialized, and the diffuse stromal inflammation consolidated into multifocal subepithelial and mid-stromal infiltrates associated with progressive peripheral corneal stromal vascularization. In A, C, and H, the cats were topically inoculated following light corneal abrasion. In B, D, and E, the cats were inoculated by placement of a contaminated contact lens following corneal abrasion. In F and G, the cats were inoculated by placement of a contaminated contact lens only.

Figure 2.

 

Sequential clinical photographs of two cats with experimentally induced Acanthamoeba castellanii keratitis demonstrating progression of the keratitis over time. (A–D) A cat inoculated topically following corneal abrasion: (A) Post-inoculation day 3, ocular lesions were characterized by mucopurulent discharge, chemosis, and multifocal corneal epithelial leukocyte infiltrates (a temporary tarsorrhaphy suture is present). (B) Post-inoculation day 7, superficial corneal ulceration, progressive epithelial and subepithelial leukocyte infiltrates, and peripheral corneal vascularization developed. (C) Post-inoculation day 14, the diffuse stromal inflammation consolidated into multifocal subepithelial and mid-stromal infiltrates. (D) Post-inoculation day 21, continued consolidation and reduction of the subepithelial and mid-stromal infiltrates associated with progressive corneal vascularization were seen. (E–H) A cat inoculated by placement of a contaminated contact lens following corneal abrasion: (E) Post-inoculation day 3, ocular lesions were characterized by mucopurulent discharge and multifocal corneal epithelial leukocyte infiltrates (a temporary tarsorrhaphy suture and contact lens are present). (F) Post-inoculation day 7, superficial corneal ulceration, progressive epithelial and subepithelial leukocyte infiltrates, and central corneal edema developed. (G) Post-inoculation day 14, the diffuse stromal inflammation consolidated into multifocal subepithelial and mid-stromal infiltrates. (H) Post-inoculation day 21, continued consolidation and reduction of the subepithelial and mid-stromal infiltrates associated with progressive corneal vascularization were seen.

Figure 3.

 

Mean keratitis clinical scores (±SDs) calculated every 2 days for cats with experimentally induced Acanthamoeba castellanii keratitis: group 1, topical inoculation with the amoeba suspension following corneal abrasion; group 2, placement of an amoeba contaminated contact lens only; group 3, placement of an amoeba-contaminated contact lens following corneal abrasion; group 4, topical inoculation of parasite-free PYG medium following light corneal abrasion; and group 5, placement of a contact lens incubated with sterile PYG medium following corneal abrasion.

During post-inoculation days 13 to 21, the regions of corneal ulceration epithelialized, and the diffuse stromal inflammation consolidated into multifocal subepithelial and mid-stromal infiltrates associated with the progression of peripheral corneal stromal vascularization (Figs. 1, 2). Conjunctival inflammation and anterior uveitis progressively reduced during this period and the blepharospasm resolved. By post-inoculation day 21, two clinical patterns were observed in the remaining cats with keratitis. In one cat that was inoculated topically following corneal abrasion, multifocal corneal stromal abscesses persisted with mild concurrent conjunctivitis and uveitis. The other remaining cat with keratitis that was inoculated by placement of a contaminated contact lens following corneal abrasion had only mild linear stromal infiltrates with resolution of active conjunctivitis and uveitis. 

No cats in the control groups that received sham inoculations developed clinical keratitis (P ≤ 0.01). In the cats from the control groups, the corneal abrasions had re-epithelialized by post-inoculation day 2, and only faint linear subepithelial corneal opacities that did not retain fluorescein stain were visible. These opacities slowly cleared over the following 14 days. 

In Vivo Confocal Microscopy

The IVCM examinations performed 7 days post-inoculation revealed marked epithelial and anterior stromal leukocyte infiltrates in all cats that clinically developed keratitis (Fig. 4). Clumps of structures consistent with Acanthamoeba were present within the basal epithelium and anterior stroma in all seven cats with clinical keratitis. At 14 and 21 days post-inoculation, the leukocytes persisted and were associated with small clusters of Acanthamoeba within the anterior and mid-stroma (Fig. 4). Langerhans cell infiltration was also present in the central corneal epithelium after 14 days post-inoculation. No leukocytes or structures consistent with Acanthamoeba were identified in cats from the control groups that received sham inoculations or in the cats from the contaminated contact lens only group that did not develop keratitis. The only abnormality detected by IVCM in cats from the control groups were linear hyperreflective opacities in the subepithelial stroma presumed to be associated with the previous corneal abrasions. These linear opacities were only detected in cats from the corneal abrasion groups. 

Figure 4.

 

In vivo corneal confocal photomicrographs of cats with experimentally induced Acanthamoeba castellanii keratitis. (A, B) At 7 days post-inoculation, leukocytes (arrowheads) and amoebae (arrows) were present within the corneal epithelium (A) and subepithelial stroma (B). (C, D) At 14 and 21 days post-inoculation, small clusters of leukocytes (arrowheads) and amoebae (arrows) were present in the anterior and mid-corneal stroma. In A and C, the cats were topically inoculated following light corneal abrasion. In B and D, the cats were inoculated by placement of a contaminated contact lens following corneal abrasion. Scale bars: 50 µm.

Histopathology and Immunohistochemistry

Consistent histopathologic findings in the seven feline globes from cats with clinical keratitis included neutrophilic and lymphocytic conjunctivitis, keratitis, and scleritis (Fig. 5). Corneal epithelial hyperplasia and fibrosis with lymphoplasmacytic anterior uveitis were also present. Corneal amoebae were detected by histopathology and immunohistochemistry in all cats with keratitis (Fig. 5). At 7 days post-inoculation, Acanthamoeba trophozoites were identified in the deep corneal epithelium and anterior stroma. At 14 and 21 days post-inoculation, Acanthamoeba trophozoites and cysts were present in the anterior and mid-stroma. Globe histopathology was unremarkable in cats from the sham inoculation control groups, except for vertical columns of keratocytes in the subepithelial stroma oriented parallel to the corneal surface in 2 cats. This finding was interpreted as fibroplasia secondary to the mechanical corneal abrasion. Amoebae were not detected in the sham inoculation groups (P ≤ 0.01) or cats from the contact lens infection group that did not develop clinical ocular disease. 

Figure 5.

 

Photomicrographs of cats with experimentally induced Acanthamoeba castellanii keratitis. (A) Low-magnification view of neutrophilic and lymphocytic keratitis in a cat on post-inoculation day 21 (H&E stain). Scale bar: 50 µm. (B) Acanthamoeba trophozoite (arrow) in the anterior corneal stroma on post-inoculation day 7 (H&E stain; original magnification, 200×). (C) Acanthamoeba trophozoite in the anterior corneal stroma on post-inoculation day 7; amoeba were visualized using an immunoperoxidase method (appears dark brown) with hematoxylin counterstain (original magnification, 200×). (D) Acanthamoeba cyst (arrow) in the deep corneal stroma on post-inoculation day 21 (H&E stain; original magnification, 200×). In A and B, the cats were inoculated by placement of a contaminated contact lens following corneal abrasion. In C, the cat was topically inoculated following light corneal abrasion. In D, the cat was inoculated by placement of a contaminated contact lens only.

Microbiologic Assays

Acanthamoeba were detected by culture on ≥1 sampling day in six of seven cats that clinically developed keratitis (Fig. 6). Culture detection of Acanthamoeba was most consistent in cats from the study group infected by placement of a contaminated contact lens following corneal abrasion. All of the cats from this group had positive cultures on each sampling day. There were no positive amoeba cultures from the samples collected from cats infected with a contaminated contact lens only or from cats in any of the control groups. 

Figure 6.

 

Phase-contrast photomicrographs of corneal amoeba cultures from cats with experimentally induced Acanthamoeba castellanii keratitis. (A, B) Acanthamoeba trophozoites and cysts are present. Scale bars: 50 µm.

Corneal bacterial cultures were positive for one or more bacteria in six of 18 samples collected from cats in the infection groups. Isolated bacteria included Klebsiella pneumonia (n = 3 isolates), Staphylococcus spp. (n = 3), Corynebacterium spp. (n = 1), and Enterococcus spp. (n = 1). Bacteria were cultured on at least one sampling day from the corneal samples of four of seven cats that clinically developed keratitis, including two of three cats topically inoculated after corneal abrasion and two of three cats inoculated with a contaminated contact lens following corneal abrasion. In cats topically inoculated after corneal abrasion, isolated bacteria included Staphylococcus spp. (n = 2 isolates) and Klebsiella pneumonia (n = 1). In cats inoculated with a contaminated contact lens following corneal abrasion, isolated bacteria included Klebsiella pneumonia (n = 2 isolates), Corynebacterium spp. (n = 1), Enterococcus spp. (n = 1), and Staphylococcus spp. (n = 1). No subjective difference in clinical lesions were detected between cats with and without bacterial coinfection. Bacterial cultures were positive for a single bacterium in two of 12 samples collected from cats in the control groups. All isolated bacteria from control groups were Staphylococcus spp. (n = 2). The frequency of positive aerobic bacterial corneal cultures was similar (P = 0.4) between the infection and control groups. All FHV-1 PCR assays and fungal cultures were negative. 

Necropsy

No evidence of systemic or extraocular Acanthamoeba infection was identified in any cat during necropsy. Mild findings found in some cats from both the infection and control study groups included splenic lymphoid hyperplasia, eosinophilic gastroenteritis, large intestinal lymphofollicular hyperplasia, mesenteric lymphoid hyperplasia, interstitial nephritis, and multifocal myocardial mineralization. These changes were considered incidental or attributed to the young age of the cats and potential food hypersensitivities. 

Discussion

The present study describes successful experimental induction of Acanthamoeba keratitis in cats. The methods used to induce infection, including placement of contaminated contact lenses, simulate the events that can lead to Acanthamoeba keratitis development in human patients.¹⁶ This large animal model of Acanthamoeba keratitis is unique, as domestic cats are the only animal species described to develop naturally acquired ocular Acanthamoeba infections.^13,14 This experimental model, integrated with investigations of naturally acquired feline Acanthamoeba infections, could advance research into the pathophysiology of ocular amoebic infections.¹⁷ Evaluation of the efficacy and toxicity of novel anti-amoebic drugs in animal models of ocular Acanthamoeba infection is an integral step in the development of novel clinical treatment strategies and another potential application of the described feline model of Acanthamoeba keratitis.¹⁸ Management of patients with Acanthamoeba keratitis is challenging, and few effective medications are available to clinicians. The described feline model could be used in the development of more effective therapeutics for Acanthamoeba infections. 

All cats in the corneal abrasion study groups, including those infected by topical ocular inoculation and contaminated contact lenses, developed Acanthamoeba keratitis. Interestingly, only a single cat from the study group where infection induction was attempted by the placement of a contaminated contact lens alone without corneal abrasion developed keratitis. This suggests that compromise of the ocular surface is a critical event during the initiation of Acanthamoeba keratitis in cats. Trauma is believed to increase the expression of the glycoproteins exploited by Acanthamoeba for binding to the human corneal epithelium, predisposing to the development of infections.^19,20 A previous in vitro study utilizing feline corneal buttons determined that the feline cornea was highly susceptible to Acanthamoeba castellanii binding and invasion.¹⁴ Although it was not a requirement, prior epithelial trauma enhanced Acanthamoeba binding to the feline cornea in vitro, and it may lower the threshold for fulminant keratitis development in vivo.¹⁴ 

Small animal models of experimentally induced Acanthamoeba keratitis that are described include rabbits, hamsters, rats, and mice.^7–11 Due to their lower purchase and housing costs, these models have the advantage of permitting more animals to be used in studies. The feline model of ocular Acanthamoeba infection offers several advantages compared to these small animal model systems. These benefits include a larger anatomic ocular surface facilitating detailed clinical examination, sample collections, diagnostic imaging, and surgical interventions. 

The described Yucatan micropig model of Acanthamoeba keratitis possesses several similarities to the novel feline model detailed in this report.⁶ Infection induction is consistently achieved in both models with corneal abrasion and placement of a parasite-laden soft contact lens. Both models are associated with acute keratitis that is both clinically and histopathologically similar to what is observed in human patients.⁶ Viable Acanthamoeba can also be cultured from diseased corneas of both animal species. High cost of purchase, large body size requiring specialized animal housing facilities, animal handling challenges, and the absence of a reported naturally occurring disease state in pigs are among the potential disadvantages of the swine model relative to the feline model.²¹ 

Corneal coinfections in human patients reported clinically with Acanthamoeba keratitis include bacteria, fungi, Pythium, other protozoa, adenovirus, and herpes simplex virus.^22–28 The pathophysiologic relationships between these infectious agents and Acanthamoeba are currently unclear. Epithelial disruption associated with infectious keratitis, including viral keratitis, might enhance amoebic binding to the cornea and facilitate the development of Acanthamoeba keratitis as a secondary event.^26,27 The presence of concurrent bacterial or fungal keratitis, or the occurrence of these microorganisms as amoebic endosymbionts, might enhance the pathogenicity of the Acanthamoeba and promote coinfection with both infectious agents as simultaneous events.²⁹ The intracellular growth of some bacteria within Acanthamoeba can also enhance the virulence of the bacteria while protecting them from antibiotics and immune responses.^30,31 A study evaluating experimental Acanthamoeba keratitis in rabbits induced by intrastromal injection found that corneal inoculation of Acanthamoeba alone failed to produce clinical disease, but keratitis was reliably induced by inoculation of corneas with Acanthamoeba that had been co-incubated with Pseudomonas aeruginosa.³² In a separate study, antibiotic pretreatment of proven pathogenic Acanthamoeba isolates to remove their bacterial endosymbionts resulted in a loss of pathogenicity in rabbits.³³ 

The potential contributions of coinfection in most animal models of Acanthamoeba keratitis have not been investigated. A previous publication describing experimental induction of Acanthamoeba keratitis in pigs briefly mentioned that bacterial cultures of corneoconjunctival specimens were performed during all stages of corneal infection and revealed what was described as normal bacterial flora, but no further information was provided, and the culture data were not published.⁶ The present study attempted to define potential corneal coinfections in experimental Acanthamoeba keratitis in cats. FHV-1, which is among the most common etiologies of ulcerative keratitis in cats, was not detected in corneal samples from any cat.³⁴ Concurrent keratomycosis was not identified in the study cats by fungal culture. Several bacteria were detected by corneal culture from the cats with keratitis. The isolated bacteria are frequent etiologies of bacterial keratitis in cats and relatively common components of the feline ocular surface microflora.^35–38 The bacterial isolates might have been incidental to the keratitis or represent secondary infections of the compromised corneas. Alternatively, these bacteria may have enhanced the corneal pathogenicity of the Acanthamoeba. Bacterial endosymbionts are theorized to increase Acanthamoeba pathogenicity, virulence, and resistance to anti-amoebic drugs in human infections.³⁹ The role of the cultured bacteria in the development of keratitis in the cats is not clear, but they do not appear to be critical to the development of keratitis, as not all cats with Acanthamoeba infection had positive cultures. Bacteria were also cultured from cats that did not develop clinically evident keratitis but at a much reduced frequency. 

Limitations of the present study include the relatively small number of cats included in each study group. Inclusion of a higher number of cats might have permitted the statistical detection of different rates of successful induction of corneal infection between the evaluated transmission methods and identification of potential variations in ocular lesions that developed between the study groups during experimental infection. Inclusion of a control group of cats that were treated with placement of a contact lens incubated with sterile PYG medium only would have been ideal and could have helped determine the relative effects of the contact lens versus the corneal abrasions. The potential presence of amoebic endosymbionts was also not evaluated in this study. The methods used to identify corneal co-infections were more likely to identify extracellular microbes than endosymbionts.⁴⁰ The duration of clinical ocular disease and ocular Acanthamoeba recovery by culture or other diagnostic methods was not determined in the present study that was restricted to the evaluation of acute infection. By 21 days post-inoculation, the ocular lesions were improving in the remaining cats with keratitis. The ultimate outcome of the experimental infection, and if the clinical lesions and corneal amoebae would have cleared completely, was not determined. Acanthamoeba keratitis is typically a chronic condition in human patients and can require months of treatment to achieve clinical resoluion.^2–4 It is currently unknown if the described feline experimental infections can replicate the chronic corneal infections that are observed in human patients or if these methods will be restricted to a model of acute infection. 

In conclusion, Acanthamoeba keratitis is experimentally transmissible in cats by contaminated contact lenses and topical inoculation following corneal epithelial trauma. Experimentally induced acute Acanthamoeba keratitis in cats is clinically and histopathologically similar to its human counterpart. 

Acknowledgments

Supported by the Rochester/Finger Lakes Eye & Tissue Bank Medical Research Program. 

Disclosure: E.C. Ledbetter, (P); E. Capistrano da Silva, None; L. Dong, None; S.P. McDonough, None