Stable topical application of antimicrobials using plumbing rings in an ex vivo porcine corneal infection model

Abstract

Microbial keratitis (MK) is a substantial cause of clinical blindness worldwide. Pseudomonas aeruginosa is an opportunistic Gram-negative bacterium and is the leading cause of MK. Infection models are vital tools in understanding host-pathogen interactions and the development of novel therapies. As well as ethical and practical advantages, ex vivo infection models enable researchers to study host-pathogen interactions with greater accuracy and physiological relevance compared to traditional cell culture systems. The versatility of porcine corneal ex vivo models have been employed to study various pathogens (for example Staphylococcus aureus and Acanthamoeba) and has enabled innovation of novel MK therapies. Here, we describe an improved porcine corneal ex vivo protocol, which uses plumbing rings and medical adhesive to circumvent several distinct limitations and challenges. The application of a 10 mm plumbing ring to the center of the cornea allows localized inoculation of pathogens of interest, maintaining them at the site of infection, rather than running the risk of "run off" of topically added aqueous solutions. The second important advantage is that topically applied therapeutic agents can be properly maintained on the cornea within the plumbing ring reservoir, allowing more accurate study of antimicrobial effects. In this contextualized protocol, we infected porcine corneas with the P. aeruginosa strain PA103 with topical treatments of moxifloxacin. PA103 colony-forming unit (CFU) quantification, spectrophotometric measurement of corneal opacity, and histological analysis of stromal edema using hematoxylin and eosin staining were employed to assess infection over 48 hours. Moxifloxacin treatment demonstrated a dose-dependent reduction in infection and corneal damage. These findings have contributed to the development of an improved and standardized ex vivo infection model for evaluating therapeutic interventions, potentially supporting clinical translation to alleviate the burden of microbial keratitis.

Fig 1. Application of plumbing ring during porcine cornea preparation.

The key steps in the preparation of the porcine cornea and the assembly of the plumbing ring reservoir are shown. (A) A scalpel was used to make an incision into the sclera. (B) Scissors were used to cut around the cornea to yield (C) a corneal scleral button. (D) Two pairs of forceps were used to remove the uveal tissue. (E) The corneas were immersed in 10% (v/v) iodinated povidone for 2 minutes and then washed 3 times in PBS (F). (G) LIQUIBAND adhesive was applied to a sterile 10 cm petri dish, and a 10 mm plumbing ring was immersed in this. (H) Using forceps, the epithelial side of the porcine corneas was pressed onto the adhesive-coated side of the plumbing ring, ensuring an even seal.

Fig 2. PA103 CFU detection in ex vivo porcine corneas after 48h infection and moxifloxacin treatment.

Ex vivo porcine corneas were infected with 1 x 10⁵ CFU PA103 for 48 h in the indicated concentrations of moxifloxacin. Whole corneas were homogenized, diluted in PBS followed by plating, overnight incubation at 37 ºC prior to CFU counting. Results plotted are mean ±  SD from 3 independent experiments. For statistical analysis, one-way ANOVA with multiple comparisons (to 0 µ M moxifloxacin) was performed in Prism (*<0.05, **<0.01, ***<0.001, ****<0.0001).

Fig 3. Opacity measurement to determine moxifloxacin efficacy in PA103 infected ex vivo porcine corneas.

The plumbing rings reservoir on the ex vivo porcine corneas was filled with 150 µ L PBS containing 10⁵ CFU PA103 and varying concentrations of moxifloxacin. After 48h incubation the corneas were imaged and their opacity measured. (A) Images of the corneas were acquired with a Bio-Rad Chemidoc. (B) The opacity of the corneas was determined by measuring their optical absorbance at 400 nm in a 12-well plate using a spectrophotometer. Results of 3 independent experiments were plotted and one-way ANOVA with multiple comparisons (to 0 µ M moxifloxacin) was performed in Prism (*<0.05, **<0.01, ***<0.001, ****<0.0001).

Fig 4. Histological analysis of PA103 infected porcine ex vivo corneas.

Ex vivo porcine corneas were infected with 1 x 10⁵ CFU of PA103 for 48h with and without 20 moxifloxacin. After parafilm embedding and sectioning, slides were stained with H&E. (A) Stained corneal sections were imaged using a Ventana DP 200 slide scanner. Black arrows indicate epithelial ulceration and white arrows stromal edema. (B) The percentage of space within the corneal stroma in response to PA103 infection was assessed using ImageJ. The stroma of the corneal sections was designated as the region of interest (ROI) and converted to 8-bit grayscale for analysis. A threshold range of 230–255 was applied to identify and quantify white space within the stroma. The resulting data were plotted using GraphPad Prism, and statistical significance was evaluated using one-way ANOVA with multiple comparisons (*p <  0.05, **p <  0.01, ***p <  0.001, ****p <  0.0001).