Corneal Graft Rejection in the Inbred NIH Minipig

Abstract

Purpose: : To develop a pre-clinical model of corneal graft rejection in the semi-inbred NIH minipig and to validate the model clinically, by immunohistology and by assessment of the capacity of corneal endothelium to proliferate in vivo.

Methods: : Corneas of swine leukocyte antigen (SLA)^dd strain or SLA^cc strain were transplanted to SLA^cc strain recipients aged 3-5 months. SLA^cc autografts were performed as controls. No immunosuppression was administered. Grafts undergoing rejection, non-rejected grafts and contralateral corneas were excised post-mortem, frozen and sectioned for immunofluorescence histology. Infiltrating antigen presenting cells (CD14⁺, CD16⁺, CD163⁺, MHC class II⁺), T cells (CD3⁺, CD4⁺, CD8⁺) and ingress of blood vessel were quantified by digital capture of 3-colour-channel images. Numbers of positive pixels for each fluorochrome were counted using ImageJ software, percentage areas of labeling were calculated and analyzed by factorial ANOVA. Excised corneas from non-transplanted pigs were subjected to aseptic freezing injury and cultured in vitro to monitor wound repair. Grafts of some minipigs were monitored for up to 30 days after rejection to determine whether opacity and edema resolved.

Results: : Post-operative complications were minimal. Initial post-transplant edema cleared within 7-10 days. Autografts (n=5) and SLA^cc to SLA^cc allografts (minor mismatches, n=5) remained clear thereafter. Rejection of SLA^dd allografts in SLA^cc recipients (MHC and minor mismatches, n=10) was characterized by graft opacity and edema, an epithelial rejection line and neovascularization to the centre of the graft. Median graft survival was 57 days (range 30->90 days). There was a significant increase in leukocyte subsets in transplanted vs contralateral eyes (p<0.001) and in clinically rejected compared with non-rejected grafts (p<0.001). In vitro wound repair of injured corneas was minimal over an 8-day period. Ki67 labeling of scattered endothelial cells (<50 per cornea), indicative of cell division, was evident after 2 days, whether or not corneas had been injured. However, confocal imaging indicated that such cells were detached from the endothelial cell layer. Grafts did not recover clarity in vivo.

Conclusions: : Rejection in the NIH minipig is an excellent pre-clinical model of human rejection. The capacity for corneal endothelial repair in vivo appears to be limited, as in man.

Keywords: transplantation • cornea: endothelium • immunohistochemistry