Background: Deceased human donors are currently the only source for corneal grafts, but the number procurable are far from supplying the demand. The availability and ability to use genetically modified pigs as donors make xenotransplantation more attractive. However, immune rejection still remains to be overcome despite the use of genetically engineered pig corneas, cells, and/or biomaterials. Although decellularization techniques aim to avoid this immune reaction, the problem with decellularization strategies is that almost all techniques have various adverse effects on the corneal tissue. Our study compares the three previously applied, conventional porcine corneal decellularization methods and demonstrates the importance of preserving the corneal limbus through decellularization.

Methods: Wild-type (WT) porcine corneas with or without limbus were freshly isolated and decellularized using three different methods: (i) sodium dodecyl sulfate (SDS), (ii) hypertonic saline (HS), and (iii) N2 gas (NG) treatments. Decellularized porcine corneas were evaluated by histology, remaining nucleic acid, and ultrastructural analysis by transmission electron microscopy (TEM). Glycerol was used to reduce the adverse effects of decellularization. The corneas were preserved for two weeks in a cornea storage medium.

Results: At macroscopic assessment, all the methods considerably reduced the transparency. All the groups regained transparency with 2 hours of glycerol treatment, except the SDS w/o limbus group. All three decellularization methods reduced the number of keratocytes at different rates in the stromal tissue (WT 162.50 ± 6.68, SDS w/ limbus 9.25 ± 2.10, SDS w/o limbus 0, HS w/ limbus 32.25 ± 3.38, HS w/o limbus 41.25 ± 4.05, NG w/ limbus 87.50 ± 7.24, NG w/o limbus 86.25 ± 3.66, p<0.0001). However, all methods, except SDS, resulted in retaining large numbers of cells and cell fragments. The SDS method (0.1% SDS, 48h) resulted in almost 100% decellularization in corneas without limbus, eliminating approximately 94.3% of nucleic acid in corneas w/ limbus. The low decellularization capacity of the NG method (< 50%) could make it unfavorable. Although the HS method had a more balanced damage-decellularization ratio, its decellularization capacity was lower than the SDS method. Preservation of the corneoscleral limbus could partially prevent structural damage and edema, but it would reduce the decellularization capacity.

Conclusion: Our results suggest that the NG-based method was not associated with significant damage to the corneal ultrastructure, but its low decellularization capacity (<50%) made it unfavorable. The HS method had a more balanced damage-decellularization ratio, but its decellularization capacity was lower than the SDS method. The SDS is a very powerful decellularization method, but it damages the cornea irreversibly. Preserving the corneoscleral limbus decreases the efficiency of decellularization but also reduces the damage.