Comparative analysis of ocular surface restoration following different techniques of allogeneic limbal transplantation in a rabbit model of limbal stem cell deficiency

The current study aimed to compare the efficacy of different allogeneic limbal transplantation techniques for promoting epithelial regeneration in a rabbit model of mechanically induced limbal stem cell deficiency (LSCD). Limbal stem cell deficiency was induced in the left eye of New Zealand white rabbits (n = 18) through mechanical debridement involving 360° removal of limbal epithelium. Following injury, the eyes were monitored for corneal opacity, vascularization, and epithelial defect using clinical examination, slit-lamp imaging, and anterior segment optical coherence tomography. Four weeks post-debridement, fifteen animals underwent one of three transplantation procedures: allogeneic simple limbal epithelial transplantation (allo-SLET) (n = 5), conjunctival limbal allograft (CLAL) (n = 5), and allogeneic cultivated limbal epithelial transplantation (allo-CLET) (n = 5), which were performed using donor tissue obtained from Dutch Belted rabbits. Post-transplantation outcomes were evaluated over a 12-week period. At the end of the study, corneal tissues were analyzed by immunofluorescence staining for corneal epithelial (CK3) and conjunctival epithelial (CK19) markers. qRT-PCR was performed for epithelial-specific markers (CK12, CK14, CK19), stem cell marker (ABCG2), and myofibroblast marker (α-SMA). The untreated LSCD corneas (n=3) demonstrated significant corneal opacity, vascularization, and reduced epithelial thickness compared to healthy control eyes (n=18). All three transplantation methods achieved complete re-epithelialization by week 12, with progressive reduction in corneal opacity and vascularization scores compared to untreated LSCD controls. Allo-SLET showed significantly greater central corneal epithelial thickness (CCET) recovery from week 3 onwards (p ≤ 0.01) and allo-CLET showed significant improvement at early post-transplantation timepoints (p ≤ 0.05) compared to LSCD controls, whereas CLAL did not achieve statistically significant CCET improvement at any timepoint. Immunofluorescence confirmed restoration of corneal epithelial phenotype with CK3 re-expression and spatial restriction of CK19 protein to the limbal periphery. qRT-PCR demonstrated significant upregulation of progenitor marker CK14 across all groups (p ≤ 0.05 to p ≤ 0.0001), with α-SMA significantly elevated only in allo-CLET (p < 0.01) and no significant change in ABCG2 in any group. No clinical signs of acute allograft rejection or recurrent epithelial defects were observed post-transplantation. This study provides the first direct comparison of allo-SLET, allo-CLET, and CLAL in a rabbit model of LSCD. The findings demonstrate that all three allogeneic approaches can restore ocular surface integrity; however, allo-SLET and allo-CLET demonstrated greater CCET recovery within the 12-week observation period compared to CLAL. These results highlight the potential advantages of epithelial-focused allogeneic limbal transplantation strategies for the treatment of bilateral LSCD and provide insights into the underlying biology of ocular surface restoration.