Highly concentrated collagen solutions leading to transparent scaffolds of controlled three-dimensional organizations for corneal epithelial cell colonization.

This study aimed at controlling both the organization and the transparency of dense collagen scaffolds making use of the lyotropic mesogen properties of collagen. Cholesteric or plywood-like liquid crystal phases were achieved using mixtures of acetic and hydrochloric acids as solvents. The critical pH at which the switch between the two phases occurred was around pH = 3. The use of the two acids led to fibrillated collagen I scaffolds, whose visual aspect ranged from opaque to transparent. Rheological investigations showed that viscoelastic properties of the plywood-like solutions were optimized for molding due to faster recovery. They also confirmed the correlation between the elastic modulus and the diameter of collagen fibrils obtained after fibrillogenesis under ammonia vapor. Human corneal epithelial cells, grown from donor limbal explants, were cultured both on transparent plywood-like matrices and on human amniotic membranes for 14 days. The development of corneal epithelium and the preservation of epithelial stem cells were checked by optical microscopy, colony formation assay, immuno-fluorescence and quantitative polymerase chain reaction. A higher level of amplification of limbal stem cells was obtained with collagen matrices compared with amniotic membranes, showing the high biocompatibility of our scaffolds. We therefore suggest that collagen solutions presenting both plywood-like organization and transparency might be of interest for biomedical applications in ophthalmology.

Easy xeno-free and feeder-free method for isolating and growing limbal stromal and epithelial stem cells of the human cornea.

Epithelial and stromal stem cells are required to maintain corneal transparency. The aim of the study was to develop a new method to isolate and grow both corneal stromal (SSC) and epithelial limbal (LSC) stem cells from small human limbal biopsies under culture conditions in accordance with safety requirements mandatory for clinical use in humans. Superficial limbal explants were retrieved from human donor corneo-scleral rims. Human limbal cells were dissociated by digestion with collagenase A, either after epithelial scraping or with no scraping. Isolated cells were cultured with Essential 8 medium (E8), E8 supplemented with EGF (E8+) or Green's medium with 3T3 feeder-layers. Cells were characterized by immunostaining, RT-qPCR, colony forming efficiency, sphere formation, population doubling, second harmonic generation microscopy and differentiation potentials. LSC were obtained from unscraped explants in E8, E8+ and Green's media and were characterized by colony formation and expression of PAX6, ΔNP63α, Bmi1, ABCG2, SOX9, CK14, CK15 and vimentin, with a few cells positive for CK3. LSC underwent 28 population doublings still forming colonies. SSC were obtained from both scraped and unscraped explants in E8 and E8+ media and were characterized by sphere formation, expression of PAX6, SOX2, BMI1, NESTIN, ABCG2, KERATOCAN, VIMENTIN, SOX9, SOX10 and HNK1, production of collagen fibrils and differentiation into keratocytes, fibroblasts, myofibroblasts, neurons, adipocytes, chondrocytes and osteocytes. SSC underwent 48 population doublings still forming spheres, Thus, this new method allows both SSC and LSC to be isolated from small superficial limbal biopsies and to be primary cultured in feeder-free and xeno-free conditions, which will be useful for clinical purposes.

Development of human corneal epithelium on organized fibrillated transparent collagen matrices synthesized at high concentration.

Several diseases can lead to opacification of cornea requiring transplantation of donor tissue to restore vision. In this context, transparent collagen I fibrillated matrices have been synthesized at 15, 30, 60 and 90 mg/mL. The matrices were evaluated for fibril organizations, transparency, mechanical properties and ability to support corneal epithelial cell culture. The best results were obtained with 90 mg/mL scaffolds. At this concentration, the fibril organization presented some similarities to that found in corneal stroma. Matrices had a mean Young's modulus of 570 kPa and acellular scaffolds had a transparency of 87% in the 380-780 nm wavelength range. Human corneal epithelial cells successfully colonized the surface of the scaffolds and generated an epithelium with characteristics of corneal epithelial cells (i.e. expression of cytokeratin 3 and presence of desmosomes) and maintenance of stemness during culture (i.e. expression of ΔNp63α and formation of holoclones in colony formation assay). Presence of cultured epithelium on the matrices was associated with increased transparency (89%).

Kinetics of expansion of human limbal epithelial progenitor cells in primary culture of explants without feeders.

The aims of this study were to determine whether human limbal explant cultures without feeder cells result in expansion of epithelial progenitors and to estimate the optimal expansion time for progenitor cells. Limbal explants from ten human corneas were cultured for 7, 9, 11, 14, 18, and 21 days. Limbal explants from two corneas were enzymatically dissociated or directly cultured for 14 days. Progenitor cells were characterized by their ability to form colonies, by immunocytochemistry, and by quantitative real-time polymerase chain reaction. Colonies were identified after 9, 11, 14, and 18 days of culture, but not after 21 days. The number of colonies per explant was significantly higher after 14 days than after 9 and 21 days. The mean percentage of seeded cells giving rise to clones was 4.03% after 14 days of culture and 0.36% for non-cultured dissociated limbal epithelial cells. The number of cells giving rise to clones per cornea significantly increased from an average of 2275 for non-cultured cells to 24266 for cells cultured for 14 days. Immunocytochemical analysis detected positive staining for cytokeratin (CK) 3, CK5/6/8/10/13/18, CK19, vimentin, p63, and p63α, in both cultures and clones. CK3 expression increased significantly with culture time. Transcript expression was observed for CK3, CK19, vimentin, and Delta N p63α at each culture time point, both in cultures and clones. The optimal culture time for limbal explants in cholera toxin-free Green medium without feeder cells was 14 days leading to the expansion of progenitors.

Effects of isoproterenol and cholera toxin on human limbal epithelial cell cultures.

PURPOSE: Cholera toxin and isoproterenol (ß-adrenergic receptor agonist) are largely used to enhance cell proliferation. The aim of the study was to assess the effects of cholera toxin and isoproterenol on growth and differentiation of cells cultured from human superficial limbal explants. METHODS: Limbal epithelial cells were cultured from superficial limbal explantsin basal medium either supplemented with cholera toxin or isoproterenol for 3 weeks. Growth kinetics and morphometry were studied by light and confocal microscopy. Progenitor and differentiated epithelial cell markers were studied by immunocytochemistry, flow cytometry, Colony Formation Assay, and reverse transcription and polymerase chain reaction. RESULTS: Cell proliferation was significantly higher with 0.5 µg/ml (p = 0.049), 1 µg/ml (p = 0.005), and 2 µg/ml (p = 0.008) isoproterenol whereas, cholera toxin and 4 µg/ml isoproterenol did not significantly increase cell proliferation. Multilayered epithelial cell sheets were obtained in all culture conditions. Addition of isoproterenol resulted in smaller cell size (p < 0.05) 14 days after cells were cultured, whereas cholera toxin had no effects. Strong expression of cytokeratins 3 and 4/5/6/8/10/13/18 and lower expression of cytokeratin 19, vimentin, and Delta N p63α were observed after 3 weeks of culture with no significant differences in the percentage of positive cells according to culture medium. Colony-forming efficiencies were observed after 2 weeks in all culture condition but not after 3 weeks. CONCLUSION: Isoproterenol was more efficient than cholera toxin for enhancing cell proliferation and resulted in smaller cell size. It appears to be useful and safe for growing human limbal epithelial progenitors from limbal explants with no feeders before transplantation to patients with limbal deficiency.

Infectious keratitis in severe limbal stem cell deficiency: characteristics and risk factors.

PURPOSE: To evaluate the incidence, clinical and microbiological characteristics and risk factors of infectious keratitis in patients with limbal stem cell deficiency (LSCD). METHODS: Retrospective comparative case series of 35 patients with severe LSCD. RESULTS: The mean follow-up time was 46 months. Infectious keratitis were mainly caused by Gram positive bacteria (94%). Only 7 infections (37%) healed under fortified adapted antibiotics. In 8 cases (42%), amniotic membrane transplantation was required and in 4 cases (21%) «à chaud¼ keratoplasty was performed. Significant risk factors associated with infectious keratitis were: soft contact lens extended wear, history of persistent epithelial defects, number of quadrants of corneal vascularization, re-epithelialization time after amniotic membrane or corneal transplantation, and use of corticosteroid or cyclosporin eye drops. CONCLUSION: Infectious keratitis in LSCD is frequent and severe. The restoration of the epithelial barrier integrity and a careful use of therapeutic contact lenses may help to prevent infection.