Ocular Microbiology and Immunology Group
Back to OMIG Main Page

< Previous | 2026 Agenda and Abstracts | Next >

2026 OMIG Abstract

Longitudinal Three-Dimensional Quantitative Analysis of Corneal Epithelial Nerve Layers in a Neurotrophic Keratopathy Mouse Model

Gabriella Mezzich¹, Ali Khodor¹, Seitaro Komai¹, Rosa Alvarado-Villacorta¹, Hooman Ahmadzadeh¹, Yasufumi Tomioka¹, Carolina Moreira¹, Leonardo de Castro¹, Alain Chedotal², Daniel Adesse¹,
Midhat Abdulreda³, and Victor L. Perez¹

¹Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida; ²Institut de la Vision, Sorbonne Université, INSERM, CNRS, Paris, France; ³Diabetes Institute, University of Miami Miller School of Medicine, Miami, Florida

Purpose: To establish a persistent mouse model of neurotrophic keratopathy using benzalkonium chloride (BAK).

Methods: Corneal epithelial wounds were induced in the right eyes of (CGRP):GFP transgenic mice (C57BL/6). Group 1 (G1; n=7) received 20% ethanol on the central cornea for 30 seconds, followed by epithelial debridement with a spatula. Group 2 (G2; n=4) underwent the same procedure, followed by application of 10 µL of 0.1% BAK for 10 seconds. Mice were followed for 10 weeks. Corneal mechanosensation was assessed twice weekly using Cochet-Bonnet esthesiometry. Wound healing and corneal surface integrity were evaluated clinically at baseline, 24 hours, and weeks 1, 4, and 10. In vivo single-photon confocal microscopy was performed at the same time points, and layer-specific three-dimensional nerve analysis was conducted using Volocity software. At week 10, corneal flat mounts were analyzed by immunofluorescence to assess immune-cell infiltration. Statistical comparisons were performed using the Mann-Whitney test.

Results: At 24 hours, wound closure was significantly delayed in G2 compared with G1 (60.5+20.4% vs. 81.0+20.3%, p<0.05). Both groups exhibited complete loss of corneal sensation and an absence of detectable GFP-positive nerve signal within the wound area. By week 10, epithelial nerve skeletal length recovered to 70+38% of baseline in G1 but only 21+21% in G2 (p<0.05). A similar difference was observed in epithelial nerve volume (100+62% vs. 34+42%, p<0.05). Functional recovery paralleled these structural findings. Corneal sensation was significantly greater in G1 than in G2 at week 4 (2.18+0.72 vs. 1.19+0.62, p<0.05), and this difference persisted through week 10 (3.31+0.70 vs. 2.15+0.78, p<0.05). At week 10, fluorescein staining scores demonstrated significantly greater corneal surface damage in G2 (G1: 10.0+3.5 vs. G2: 0.6+1.0, p<0.05). Preliminary immunofluorescence findings also suggested increased central corneal macrophage infiltration in the BAK-treated group. Together with the persistent reductions in nerve length, volume, and function, these findings suggest that inflammation may contribute to impaired corneal nerve regeneration.

Conclusions: This study establishes a persistent BAK-induced neurotrophic keratopathy model that demonstrates long-term impairment of corneal nerve function and architecture. The combined use of esthesiometry, clinical scoring, and in vivo imaging provides a comprehensive assessment of these functional and structural deficits, offering a robust platform for studying chronic neurotrophic keratopathy and potential therapeutic approaches.