Development of stromal corneal grafts using a novel decellularization method with sodium cocoyl glutamate on GGTA1/ CMAH/ β4GalNT2 knock-out porcine corneas

Abstract

Background: Diseases of the human cornea often necessitate corneal transplantation. However, donor corneas are not always readily available, leaving many patients waiting for donated corneas. Porcine corneas are a promising alternative to human donor corneas due to their close anatomical and physiological similarities. In this study, we produced GGTA1/CMAH/β4GalNT2 knockout pigs [triple knockout (TKO)] to minimize immune rejection. We investigated the efficacy and safety of a novel corneal decellularization process using sodium cocoyl glutamate (SCG) and supernuclease (SN).

Methods: We harvested cornea stromal grafts from 2-month-old TKO pigs, decellularized them using SCG and SN. The optical transparency, DNA content, collagen content, glycosaminoglycan content, and tensile strength of the decellularized corneas were measured. The in vivo safety and efficacy of the decellularized corneas were evaluated by transplanting them into the stromal pockets of rabbit corneas. Comparisons between wild type (WT) and TKO corneas, both decellularized and non-decellularized, were performed over a 4-week period post-transplantation.

Results: Compared to a previous method using sodium N-lauroyl glutamate (SLG), the method using 0.5% SCG and SN more effectively removed DNA from the corneal stroma without significantly changing tensile strength, transparency, collagen, or glycosaminoglycan content. When decellularized corneas were implanted into corneal stromal pockets of rabbits, at 4 weeks post-surgery, decellularized corneas from WT pigs showed significant corneal neovascularization and opacity. In contrast, those from TKO pigs with 0.5% SCG plus SN decellularization maintained good transparency with minimal vascularization.

Conclusions: Corneas from TKO pigs could be successfully decellularized using 0.5% SCG plus SN method, showing promising results after transplanting them into rabbit corneas.

Keywords: Cornea; decellularization; knock out; pig; sodium cocoyl glutamate (SCG).

Figure 1

Preparation of lamellar cornea graft and insertion into rabbit cornea. (A) A fresh porcine cornea is mounted in an artificial chamber (a) and the cover is tightened to create a watertight environment (b). Then pre-connected saline is infused at a constant pressure to maintain corneal pressure (c) and a keratome is used to perform lamellar dissection to obtain a stromal graft with a thickness of approximately 200 micrometers (d). (B) The stromal graft is decellularized and dried to create a 4-mm diameter circular graft in dry form. (C) After general anesthesia, a 30-gauge needle is used to inject saline into the corneal stroma of the rabbit to induce corneal edema, facilitating lamellar dissection. A stromal pocket was then created using a corneal dissector and the prepared stromal graft was inserted into the pocket, followed by corneal suturing. (D) The anterior segment photograph taken immediately after surgery shows corneal edema intentionally made for stromal pocket formation, the circular graft inserted into the stromal pocket, and corneal sutures.

Figure 2

Production of TKO pigs. (A) PCR and T7E1 analysis demonstrate that target regions of GGTA1, CMAH, and β4GalNT2 genes are significantly deleted in TKO pigs compared to WT pigs, indicating successful gene knock-out. The gel images were un-cropped whole gel images. The inserted graph compared the quantification of band intensity of T7E1 analysis. WT: wild type pig; Donor: triple knockout cells produced as cell donors; L1: a cloned animal produced from single donor cell by somatic cell nuclear transfer method. (B) The result of DNA sanger sequencing analysis of WT and L1. (C) Flow cytometry of peripheral blood mononuclear cells from TKO pigs shows markedly decreased expression of GGTA1, CMAH, and β4GalNT2, further confirming their knock-out at the cellular level. (D) Immunostaining of corneal tissue reveals that GGTA1, CMAH and β4GalNT2, which are present in WT corneas, are absent in TKO corneas, highlighting the effectiveness of the gene knock-out in corneal tissues. Scale bar indicates 100 µm. APC, allophycocyanin; FITC, fluorescein isothiocyanate; PCR, polymerase chain reaction; T7E1, T7 endonuclease 1; TKO, triple knockout; WT, wild-type.

Figure 3

Corneal thickness and tensile strength were compared between TKO pigs and WT pigs. (A) Corneal thickness of TKO pigs was significantly thinner than those of 2-month-old WT pigs. (B) There was no significant difference in tensile strength of cornea between 2-month-old TKO pigs and WT pigs. ***, P<0.001. d, days; mo, months; ns, not significant; TKO, triple knockout; wk, weeks; WT, wild-type.

Figure 4

Cornea evaluation after decellularization. (A) Comparison of corneal transparency after decellularization using SLG and SCG. The experiment was conducted using corneas of 2-month-old triple knock-out pigs. Compared to the method using 0.5% SLG and SN, the decellularization method combining 0.1% and 0.5% SCG with SN achieved similar corneal transparency. However, using 1.0% SCG with SN resulted in a noticeable decrease in transparency. (B) The total amount of residual DNA in corneal tissue after decellularization was compared. The combination of SN with either SLG or SCG effectively reduced residual DNA compared to using SLG or SCG alone. Additionally, 0.5% SCG was more effective in reducing residual DNA than 0.1% SCG. (C,D) The amounts of glycosaminoglycan and collagen showed no significant differences among all tested decellularization methods compared to pre-decellularization levels. (E) Hematoxylin and eosin staining of native and decellularized cornea shows intact corneal structures. Black scale bar: 50 µm. *, P<0.05; **, P<0.01; ***, P<0.001. GAG, glycosaminoglycan; ns, not significant; SCG, sodium cocoyl glutamate; SLG, sodium N-lauroyl glutamate; SN, supernuclease.

Figure 5

The decellularization method combining 0.5% SCG and SN did not significantly affect tensile strength of cornea. The experiment was conducted using corneas of 2-month-old triple knockout pigs. SCG, sodium cocoyl glutamate; SN, supernuclease.

Figure 6

In vivo evaluation of cornea graft. (A) Representative anterior segment photographs of rabbits after corneal stromal transplantation. Edema was commonly observed on the first day post-surgery. Severe neovascularization developed at the corneal limbus by the first week in rabbits transplanted with WT or decellularized WT corneas. By the second and fourth weeks, severe corneal opacity and inflammation around the graft or partial graft melting were observed. In contrast, rabbits with TKO and decellularized TKO corneas showed clear grafts without significant inflammation or neovascularization. (B-E) Anterior segment photograph and OCT image of a rabbit cornea at four weeks after transplantation with a WT (B), decellularized WT (C), TKO (D), and decellularized TKO (E) cornea. Severe neovascularization and opacity were observed, with OCT showing significant graft swelling and increased signal intensity both in WT and decellularized WT corneas. In contrast, TKO and decellularized TKO cornea grafts maintained transparency and minimal vascularization after four weeks. Histology (H&E staining; scale bar: 100 µm) revealed residual nuclear structures of WT and TKO graft (red arrows). Suspected infiltrating inflammatory cells (red arrowhead) were found in decellularized WT graft. Immunostaining confirmed strong CD3-positive cell infiltration (white arrows) in WT graft and significant CD3- or CD5-positive inflammatory cells in the decellularized WT graft (yellow arrows). Yellow squares were magnified to clearly demonstrate inflammatory cells in panel C. However, no inflammatory cells infiltration was observed both in TKO and decellularized TKO grafts. Scale bar: 100 µm. OCT, optical coherence tomography; SCG, sodium cocoyl glutamate; SN, supernuclease; TKO, triple knockout; WT, wild-type.

Figure 7

Comparison of corneal neovascularization and corneal opacity between groups. Four weeks after corneal stromal transplantation surgery, corneal neovascularization and corneal opacity were compared between groups in rabbit corneas. In rabbits transplanted with WT and decellularized WT corneas, severe neovascularization (A) and corneal opacity (B) were observed. However, in rabbits transplanted with TKO or decellularized TKO corneas, only mild neovascularization and corneal opacity were observed. No significant differences were found between TKO and decellularized TKO groups (n=5 for each group; ***, P<0.001). SCG, sodium cocoyl glutamate; SN, supernuclease; TKO, triple knockout; WT, wild-type.