Development of a surface-modified contact lens for the transfer of cultured limbal epithelial cells to the cornea for ocular surface diseases.

Our aim was to develop an improved cell transfer system for delivering laboratory-cultured human limbal epithelial cells to the cornea, which would be low risk for the patient and convenient to use for the surgeon. We took a standard contact lens and developed a plasma polymer layer for coating this for attachment of cells to the lens and subsequent transfer of cells to the cornea. A range of plasma polymer surfaces were examined for initial cell attachment using three different combinations of human and rabbit epithelial and stromal cells, initially expanding cells both with and without bovine serum. The most promising surfaces, based on acrylic acid, were then coated onto contact lenses. Cell transfer from the lenses to the denuded surface of a 3D rabbit organ culture model was then used to make a second selection of substrates, which permitted reliable cell transfer. Primary rabbit and human corneal cells attached and proliferated well on acrylic acid-coated surfaces. Reliable transfer of primary epithelial cells from the coated contact lenses to a rabbit cornea was achieved by coating lenses with acrylic acid at 5 W/10 cm(3)/min and using cell densities of 1 x 10(5)/lens and above.

Randomized, controlled, single-blind study on use of autologous keratinocytes on a transfer dressing to treat nonhealing diabetic ulcers.

AIM: To compare the rate of healing of diabetic neuropathic ulcers using cultured autologous keratinocytes delivered on chemically defined transfer discs (Myskin) (active treatment) versus healing obtained with cell-free discs (placebo). MATERIALS AND METHODS: After a 4-week lead-in period patients (randomly assigned) received active or placebo treatments weekly for 6 weeks. All patients then received active treatments for a maximum of 12 treatments where required. Altogether, 16 patients with a total of 21 ulcers resistant to conventional therapy were recruited from four specialist diabetic centers in three cities. RESULTS: All 21 ulcers were treated and of these ten healed and eight improved, with two failing to respond (one ulcer was lost due to autoamputation). For analysis according to the study criteria, however, only the 12 patients with 12 index ulcers who completed treatment protocols were eligible - five in the placebo group and seven in the active group. Of these, five ulcers healed completely and seven were reduced by more than 50%. Complete healing took a median of ten active applications. CONCLUSIONS: Repeated regular applications of the patient's keratinocytes, delivered on the carrier dressing, initiated wound healing in ulcers resistant to conventional therapy, with 18 out of 21 ulcers responding. The healing observed did not appear attributable to patient recruitment or the cell-free carrier dressing but to the delivery of the cultured cells.

A xenobiotic-free culture system for human limbal epithelial stem cells.

AIMS: Murine 3T3 feeder cells are commonly used for stem cell expansion. Although 'feeder-free' systems are being developed for a variety of stem cells including embryonic, the use of feeder cells currently remains optimal for the expansion of epithelial stem cells. In this study, MRC-5, a human embryonic fibroblast cell line, has been investigated for its potential use as a feeder layer in human limbal epithelial (HLE) cell expansion under serum-free conditions, with the aim of developing a xenobiotic-free culture system for therapeutic corneal regeneration applications. MATERIALS AND METHODS: MRC-5 feeder cells were compared with J2 3T3 mouse fibroblasts, in both serum-supplemented and serum-free conditions, in terms of their relative ability to support HLE cell metabolic activity, expression of the putative stem cell markers ABCG2 and P63 alpha, cell differentiation using the cornea-specific cytokeratin 3 antibody and colony-forming efficiency. RESULTS: The proportion of HLE stem cells maintained was determined by functional colony-forming efficiency assays. The metabolic activity results showed that HLE cells cultured on MRC-5 fibroblasts under serum-free conditions proliferated as well as cells cultured on J2 cells under serum-free conditions. Moreover, the HLE cultured on MRC-5 fibroblasts under serum-free conditions expressed high levels of putative stem cell markers ABCG2 and P63 alpha and low levels of the differentiation marker CK3, indicating that they retained poorly differentiated 'stem cell-like' characteristics under those culture conditions. Clonal analysis of HLE cells cultured on growth-arrested feeder layers of J2 and MRC-5 fibroblasts showed that cells expanded on MRC-5 and J2 fibroblasts in serum-free conditions had a colony-forming efficiency of approximately 1.5%, indicating the maintenance of stem cells. CONCLUSIONS: These results demonstrate feasibility of expanding HLE cells for clinical purposes by using a human fibroblast cell line as a feeder layer, avoiding the use of bovine serum, while preserving the proliferative potential and stem cell characteristics of HLE cells.

Plasma polymer coated surfaces for serum-free culture of limbal epithelium for ocular surface disease.

The potential use of plasma polymer coatings as substrates for serum-free expansion of limbal epithelial cells was investigated. Preliminary studies using a human corneal epithelial cell line showed that acrylic acid-coated surfaces performed better than allyl amine and allyl alcohol coated surfaces in terms of cell metabolic activity and confluence as assessed using the MTT assay. Subsequently, the proliferation and maturity of primary human limbal epithelial cells in co-culture with growth arrested 3T3 fibroblasts on a range of acrylic acid plasma coated surfaces, octadiene plasma coated surfaces and tissue culture plastic was investigated using MTT and cytokeratin 3 immunostaining. The cells performed better in the presence of serum on all surfaces. However, the acrylic acid coated surfaces successfully sustained a serum-free fibroblast/epithelial cell co-culture. The metabolic activity of the epithelial cells was superior on the acrylic acid coated surfaces than on tissue culture plastic in serum-free conditions and their levels of differentiation were not significantly higher than in the presence of serum. These results suggest that these surfaces can be used successfully for the serum-free expansion of human limbal epithelial cells.

Clinical experience using cultured epithelial autografts leads to an alternative methodology for transferring skin cells from the laboratory to the patient.

We report a 10-year audit using cultured epithelial autografts (CEAs) for patients with extensive burns. Clinical take using CEAs averaged only 45% (as has been reported by others) but over half of all cells cultured for these patients had to be discarded owing to difficulties of timing the production of CEA sheets to the needs of the patients. CEAs could not be used until they had reached confluence and formed an integrated sheet, which took, on average, 12 days. However, once formed, they needed to be used within 2-3 days or they lost the ability to attach to wound beds. In response to this we developed a simpler carrier dressing methodology for transferring cultured subconfluent keratinocytes from the laboratory to the wound bed. This methodology offers an increase in speed of delivery but its major contribution is the greater flexibility in timing the transfer of cells from the laboratory to the changing needs of the patients.