TABLE 1.
MSC source | Delivery mechanism | Procedure | Cell passage | Animal model | Study length | Key findings | Ref. |
---|---|---|---|---|---|---|---|
Human AD‐MSCs | Topical | 2.5E+05 cells seeded on amniotic membrane; eyelids sutured | 3/4 | Rabbit partial and total LSCD | 11 weeks | MSCs migrated to inflamed tissues, reduced inflammation, inhibited neovascularization, and corneal opacification, and expressed CK3 in the corneal epithelium, demonstrating partial restoration of epithelial phenotypes | 75 |
Human AD‐MSCs | Topical | 1.25E+05/mm2 cells seeded on a scleral contact lens | 3 | Rabbit severe acute alkaline burn | 4 weeks | MSCs prevented corneal melting and symblepharon, reduced the inflammatory and fibroblastic response, and significantly reduced epithelial defects | 76 |
Human BM‐MSCs | Topical | 1E+05 cells/cm2 seeded on amniotic membrane and cultured to 90% confluence. Eyelids sutured for 10 day | 1 | Rat corneal chemical burn | 4 weeks | MSCs enhanced repairmen of injured ocular surface and epithelial integrity. They inhibited inflammation and inflammation‐induced neovascularization, decreasing levels of IL‐2 and CD45, therefore improving transparency | 77 |
Mouse BM‐MSCs | Topical | 4E+04 cells on polyamide 6/12 nanofiber scaffold. Eyelids sutured closed. Also seeded with LESCs | Unknown | Mouse corneal mechanical injury | 2 weeks | MSCs significantly reduced immune response, through suppression of IFN‐y, iNOS, and IL‐2 gene expression in local corneal cells | 78 |
Rabbit BM‐MSCs | Topical | Unknown cell dose seeded in a fibrin gel, sutured to corneal surface. Eyelids sutured for 7 days | Unknown | Rabbit corneal alkali burn | 4 weeks | MSCs did not improve corneal epithelium integrity, neovascularization, or corneal opacity. They expressed CK3, demonstrating differentiation into corneal‐like cells | 79 |
Rabbit BM‐MSCs and AD‐MSCs | Topical | 3E+05 cells seeded on PLA nanofiber scaffold, sutured to the conjunctiva. Eyelids closed | 3 | Rabbit corneal alkali burn | 15 days | MSCs caused suppression of MMP9 and iNOS, reduced levels of αSMA, TGF‐B, and VEGF, leading to reduced corneal opacification, neovascularization, and corneal thickness | 50 |
Rat MSC line. Source unknown | Topical | 2E+06 cells in media applied for 2 hours a day for three consecutive days. Eyelids sutured | Unknown | Rat corneal chemical burn | 3 weeks | MSC anti‐inflammatory potency through IL‐6 suppressing maturation of DCs, and anti‐angiogenic through upregulation of TSP‐1 | 37 |
Human and mouse C‐MSCs | Topical | 5E+03 cells in a fibrinogen (Ethicon) gel, eyelids sutured | 3 | Mouse pathologic corneal vascularization | 3 days | Inhibition of corneal neovascularization, likely through secreted sFLT‐1 and PEDF | 80 |
Human immature DPSCs | Topical | Cell sheet held in place with sutured amniotic membrane | 6 to 7 | Rabbit mild and severe chemical induced LSCD | 3 months | MSCs lead to reconstruction of the corneal epithelia in the mild model, but not severe. Cells in both models adopted an epithelial‐like phenotype | 64 |
Rabbit limbal‐MSC | Topical | 0.5E+05 cells/cm2 on human AM. Also seeded with LESCs. AM sutured into place | 3/4 | Rabbit epithelial debridement and limbal keratectomy | 12 weeks | MSCs promoted epithelialization, however neovascularization was seen when L‐MSC were applied without LESC. Cell did not migrate into the healing epithelium | 81 |
Human BM‐MSCs | Intravenous injection | 1E+06 cells in balanced salt solution | 2 | Mouse suture induced corneal neovascularization | 1 week | MSCs reduced neovascularization, through a TSG‐6 dependent mechanism. They reduced inflammation through reduction of IL‐1B, IL‐6, and TNF‐a and suppression of infiltrating immune cells | 40 |
Mouse BM‐MSCs | Intravenous injection | 5E+05 cells in saline. Eyelids sutured closed for 3 days | 2 | Mouse corneal transplant model | 2 months | MSCs inhibited corneal leukocyte infiltration, maturation of APCs and generation of Th1 cells to promote graft survival. Data show HGF as key paracrine factor | 38 |
Mouse BM‐MSCs | Intravenous injection | 5E+05 cells in saline | 2 | Mouse corneal mechanical injury | 3 days | MSCs lead to increased levels of HGF at the ocular surface, which helped to restore corneal transparency and suppress TGF‐B‐induced α‐SMA expression | 39 |
Mouse BM‐MSCs | Intravenous injection | 1E+06 cells | 5 to 6 | Mouse corneal transplant model | 2 weeks | MSCs homed directly to the inflamed ocular surface, inhibited APC maturation, suppress allosensitization, and promote allograft survival | 31 |
Rat BM‐MSCs | Intravenous injection | 1E+06 cells in PBS | 2 | Rat high‐risk corneal transplant model | 37 days | MSCs increased rejection‐free survival, reducing inflammation through increasing regulatory T cells and release of immunomodulatory mediators including PGE2 | 82 |
Rabbit AD‐MSCs | Intravenous injection | 2E+06 cells in HBSS. 4 injections; D‐7, D0, D3, and D14‐15 | 3 to 4 | Rabbit high‐risk corneal allograft rejection model | 19 days | MSCs did not home to cornea or engraft. MSCs increased edema and neovascularization and had no effect on infiltration of immune cells | 53 |
Human BM‐MSCs | Subconjunctival injection | 2E+05 cells in PBS | 3 | Mouse GVHD | 18 days | MSCs did not engraft but prevented T lymphocyte infiltration and reduced inflammatory gene markers TNF, PAX6, and Sprr1b and reduced keratinization of the cornea | 83 |
Mouse BM‐MSCs | Subconjunctival injection | 5E+04 cells in PBS | 3 to 5 | Diabetic mouse model of corneal epithelial injury | 3 days | Homing of MSCs to wound edge of cornea, with TSG‐6 secretion responsible for enhanced wound healing, increased epithelial stem cell proliferation, and reduction of inflammatory infiltrates and inflammatory markers; MPO, TNF‐α, and IL‐1β | 84 |
Rat BM‐MSCs | Subconjunctival injection | 2E+06 cells in PBS. Used polysaccharide hydrogel as bandage | 3 | Rat corneal alkali burn | 4 weeks | MSCs promoted epithelial recovery, corneal clarity, reduced neovascularization, and reduced MIP‐1a and MCP‐1. All results were enhanced with hydrogel | 85 |
Human and mouse BM‐MSCs | Periorbital injection | 1E+03 or 1E+05 cells in balanced salt solution | 2 | Mouse inflammation‐induced dry eye | 1 week | MSCs did not engraft, but increased tear production, reduced CD4+ IFN‐γ secreting cell infiltration and restored goblet cells in the conjunctiva | 51 |
Human UC‐MSCs | Multiple | 2E+06 cells in PBS | 5 | Rabbit corneal alkali burn | 4 weeks | MSCs lead to reduced neovascularization, corneal opacification, and VEGF and α‐SMA in the cornea. They also resulted in increased re‐epithelialization and proliferation of keratocytes | 86 |
Mouse BM‐MSCs | Multiple | 5E+05 cells in PBS | 3 | Mouse corneal mechanical injury | 4 days | MSCs administered through intravenous and subconjunctival injection significantly reduced inflammation, corneal opacity, fibrosis, and restored epithelial integrity and tissue architecture. No significant difference observed for topical and intraperitoneal administration | 87 |
Human UC‐MSC | Multiple |
2E+04 cells in alpha‐MEM (intrastromal injection) 2E+04 cells in a fibrin gel carrier |
Unknown (after P4) | Mouse keratectomy wound | 2 weeks | MSCs increased corneal transparency and increase collagen fiber organization | 88 |