Using Amniotic Membrane as the Carrier for ex Vivo Cell Culture for Cell Therapy in Ocular Surface Reconstruction

Human amniotic membrane (AM) is composed of three layers: a single epithelial layer, a thick basement membrane, and an avascular stroma. The amniotic membrane was first used along with the chorion as a biologic membrane to promote healing of skin burns in 1910. In ophthalmology, it was used in 1940 in the management of conjunctival defects. Its revival in the 1990s was due to its ability to reduce ocular surface inflammation and scarring, promote rapid epithelialization due to the presence of growth factors, and antimicrobial properties.1,2 The amniotic membrane is a biologic tissue that has been used as a graft for corneal and conjunctival reconstruction in a variety of ocular surface diseases. It is avascular and possesses anti-angiogenetic, anti-scarring and antiinflammatory properties. It is not a substitute but rather a substrate upon which cells can migrate and regenerate, forming new and healthy tissue.2 Amniotic membrane has several properties and is extremely useful in ocular surgery2,3:

1. Promotes Epithelialization:

   Amniotic membrane acts like a basement membrane and facilitates the migration of epithelial cells.4 It reinforces adhesion of basal epithelial cells, promotes epithelial differentiation, prevents epithelial apoptosis, and improves corneal sensitivity and tear film stability,5 although the exact mechanism is not known. It also produces growth factors that promote epithelial cell growth.6 It is believed to expand and maintain epithelial progenitor cells ex vivo.7

2. Inhibits Fibrosis:

   Fibroblasts are normally responsible for the scarring associated with wound healing and are activated by transforming growth factor TGF-β. Amniotic membrane downregulates TGF-βand the receptor expression by fibroblasts and thus reduces fibrosis. Choi and Tseng 8,9 implanted dispase-treated amniotic membrane with and without cultured rabbit corneal epithelial cells intrastromally into rabbit corneas. They observed that the amniotic membrane alone did not induce keratocyte differentiation into myofibroblasts and the cornea remained clear. Amniotic membrane also suppressed the TGF-β signaling in corneal and limbal fibroblasts and in conjunctival and pterygial fibroblasts.10

3. Antiinflammatory and Antiangiogenic Factors:

   There are several reports of reduction of inflammation with amniotic membrane.11 The exact mechanism is not known. The amniotic membrane probably acts as a barrier against the tear film resulting in a reduced amount of inflammatory cells and hence the amount of inflammatory mediators.12 In addition, the presence of proteinase inhibitors may promote healing.

4. Antimicrobial and Antiviral Properties:

   Amniotic membrane seems to have antimicrobial properties that decrease the risk of postoperative infection. It also contains cystatin E, an analogue of cysteine proteinase inhibitors, which has complementary antiviral properties.13 Its antimicrobial and possible antiviral properties warrant further studies.

5. High Hydraulic Conductivity:

The different layers of the placenta have been studied, and it has been found that the amniotic membrane has a high hydraulic conductivity, thus facilitating its use in bleb repair following glaucoma-filtering surgery.14 Amniotic membrane transplantation is currently being used for a continuously widening spectrum of ophthalmic indications. Amniotic membrane transplantation has been shown to be effective in the reconstruction of the corneal surface in the setting of persistent epithelial defects, sterile corneal ulcerations, and partial limbal stem cell (LSC) deficiency states, including those secondary to chemical or thermal burns. The amniotic membrane has several properties that promote epithelialization and reduce inflammation, and these have prompted its use in the management of persistent epithelial defects,15,16 deep ulcers,17,18 neurotrophic ulcers,19 and perforations,20,21 with initial success rates varying from 50% to 90%. Multiple layers of AM restore stromal thickness in deep and perforated noninfectious ulcers and probably provides a substrate for collagens and growth factors for epithelial healing. AMT was also used for huge pterygium or recurrent pterygium.27 The limitations of AMT include continuous tissue destruction beneath the graft, the need for adequate stem cells at the limbus and normal keratocytes in the surrounding tissue, and intact sensory innervation for healing.15 In mild to moderate chemical injuries, AMT restores corneal and conjunctival surfaces. It prevents symblepharon formation in severe burns.22,23 In severe cases, owing to extensive ocular surface inflammation with deep stromal ischemia and near-total destruction of the limbal stem cells, the amniotic membrane may at best reduce inflammation, prevent further stem cell damage, and prevent symblepharon formation in the acute stages. Joseph et al 24 describe the complete failure of AMT in 4 cases of acute grade 4 chemical burns with complete loss of vision in all.

With the development of cell therapy, amniotic membrane can be also used as a carrier of limbal stem cells or their progeny, cultivated in vitro.2,25,26 Most recently, ex vivo cultivation and expansion of limbal epithelial cells has been performed utilizing AM as a matrix. However, the superiority of AMT over other treatment modalities in many of these settings needs to be substantiated by controlled clinical trials.5 The use of amniotic membrane in ophthalmic surgery has been shown to provide an alternative for corneal and conjunctival reconstruction in many clinically challenging situations. It has without doubt provided an option where previously none existed in some conditions. Nevertheless, it is has its limitations and is not universally successful for every indication. The indications for its use are ever increasing, often on anecdotal evidence. In some instances where it is successfully used, existing and at times simpler options are equally effective. The collective global experience with the membrane is now vast, and the field should now move into proper randomized controlled studies to evaluate its true potential.1,2,28