Abstract
Purpose:
To study microbial keratitis, cultured cells and in vivo animal models are commonly used but, these are either not representative of the in vivo situation or involve the use of many animals. Recently, there is increased use of ex vivo corneal models to replace or reduce animal use but there is no direct comparison of bacterial and fungal keratitis in these models. Accordingly, we aimed to establish reproducible models of bacterial and fungal infections in rabbit and human ex vivo corneal models to aid studies of keratitis.
Methods:
Wild brown rabbit and donated human corneas were maintained in corneal organ culture as previously described (Deshpande et al., Biomater. 2013). Corneas were wounded with a scalpel, and exposed to, or injected intrastromally with, 108Staphylococcus aureus, Pseudomonas aeruginosa or Candida albicans for 24 or 48h at 37°C. Corneas were lysed, the resulting suspension serially diluted and spotted onto agar plates for colony enumeration. Corneal tissue was histologically processed and sections were Gram-stained. Corneas not exposed to microbes were used as controls.
Results:
After 24h, using the scalpel wounding method, S.aureus, P.aeruginosa and C.albicans were recovered at 6.4±1.5x105, 5.5±0.8x106 and 2.2±0.8.1x104 CFU/rabbit cornea respectively and 3.8±0.8x106, 4.4±0.6x108 and 1.9±0.3x105 CFU/human cornea respectively. No difference in the CFU/cornea after 48h was observed compared with 24h. The injection method yielded a 10-fold increase (p<0.05) in detectable organisms for P.aeruginosa but no difference for S.aureus or C.albicans, compared with the scalpel method. Histology of the scalpel-wounded and injection models indicated extensive infiltration of P.aeruginosa, throughout the entire cornea, with less infiltration observed for S.aureus and C.albicans.
Conclusions:
Bacterial and fungal infections were initiated in ex vivo corneal models after 24 and 48h, with both scalpel wounding and injection methods suitable for inducing infection. Differences between the CFU/cornea for rabbit and human corneas may be due to the expression of different antimicrobial peptides or surface receptors influencing colonisation of the tissue. These simple and reproducible ex vivo models will be useful as an alternative to monolayer cells and in vivo models for investigating microbial keratitis.