Adult New Zealand white rabbits (1.5–2 kg) were purchased from Peking University Animals Science Research Center. All animals were housed and treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

Dulbecco’s modified Eagle’s medium (DMEM), Dulbecco’s modified Eagle’s medium/Ham’s F-12 nutrient mixture (1:1; DMEM/F-12), and Dispase II were purchased from Invitrogen-Gibco BRL (Invitrogen, Grand Island , NY). Fetal bovine serum (FBS) was purchased from Hyclone (Hyclone, Logan , UT). Human epidermal growth factor (hEGF) was purchased from PeproTech Inc. (PeproTech, Rocky Hill , NJ). Dimethyl sulfoxide (DMSO), insulin-transferrin-sodium selenite media supplement, Human type VI collagen, hydrocortisone, Percoll, 5-bromo-2-deoxyuridine (BrdU), anti-BrdU antibody, and FITC- and Cy3-conjugated secondary antibodies were purchased from Sigma-Aldrich (Sigma-Aldrich, St. Louis , MO). The primary antibodies against cytokeratin3 (AE5 clone), and integrin β1 were purchased from Chemicon International, Inc. (Chemicon, Temecula , CA). Anti-P63 monoclonal antibody and CD34 antibody were purchased from Neomarkers (Lab Vision Corporation, Fremont , CA). All plastic cell culture wares were from Corning Costar Co. (Corning Inc, Lowell , MA). Culture plate inserts used to construct the culture system were purchased from Millipore Co. (Millipore, Billerica , MA). Tissue-Tek OCT was purchased from Electron Microscopy Sciences (Electron Microscopy Sciences, Fort Washington , PA). The fibrin gel was provided by Professor Bin Han (Chief scientist of Yiling Biological Corporation kindly presented fibrin gel to us).

New Zealand white rabbits (n=20) were subjected to corneal damage in the right eye as previously described [15]. All procedures were performed under general anesthesia (2 mg/kg ketoprofen and 1 mg/kg acepromazine given subcutaneously). Corneal wounds were made in the right eye of each rabbit by contact with filter paper (15 mm diameter; 30 s contact time) soaked with 1 N NaOH and then washed with 0.9% NaCl. Antibiotic drops were applied to the wounded eye three times per day.

Rb-LSCs used for cell-suspension culture were isolated from New Zealand white rabbit limbal tissues by modification of a method previously described [16]. Rabbit limbal tissues were washed in PBS containing 100 U/ml penicillin and 50 μg/ml gentamicin. After carefully removing the corneal endothelium, iris, excessive sclera, conjunctiva, and subconjunctival tissue, the limbal rings were exposed to dispase II (1.2 IU/ml in Hanks’ balanced salt solution free of Mg2⁺ and Ca2⁺) at 37 °C for 1 h. The loosened epithelial sheets were removed with a cell scraper and separated into single cells by 0.25% trypsin + 0.02% EDTA for 10 min at 37 °C. Cells were pelleted at 1,000 rpm for 5 min and resuspended in SHEM. SHEM consisted of an equal volume of Dulbecco’s modified Eagle’s medium (DMEM) and Ham’s F12, supplemented with 5% fetal bovine serum, 5 μg/ml insulin, 5 μg/ml transferrin, 5 μg/ml sodium selenite, 10 ng/ml epidermal growth factor, 0.5% dimethyl sulfoxide, 0.5 μg/ml hydrocortisone, and 100 IU/ml penicillin-streptomycin. Cells were plated at 10⁴ cells/cm² in cell culture dishes containing MMC-treated 3T3 feeder layer and incubated at 37 °C. Immunofluorescence was used to check the molecular phenotype of Rb-LSCs.

Rb-MSCs were obtained from the femurs of adult New Zealand white rabbits as previously described [17]. In brief, 10 ml bone marrow was diluted 1:2 with phosphate buffer solution (PBS) and loaded over 5 ml Percoll (density, 1.077). Cells were harvested from the interface after centrifugation at 2,000 rpm for 20 min and washed with DMEM. Cells were resuspended in DMEM containing 10% fetal bovine serum, 100U/ml penicillin, 100 g/ml streptomycin, and incubated at 37 °C. Flow cytometric analysis was used to check the molecular phenotype of Rb-MSCs.

For adipogenic differentiation assays, Rb-MSCs were plated to a 6 well plate (5,000/cm²) and grown to 80% confluence. The adipogenic differentiation was induced by using Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% FBS, 0.5 mM isobutylmethylxanthine, 10 μM insulin, 1 μM dexamethasone, and 200 μM indomethacin for 2 weeks with full medium changes performed every 3 days. To detect fat deposition in the cells, cells were fixed with 4% paraformaldehyde for 30 min at room temperature and stained with oil red O.

To obtain cells for later identification expressing 5-bromo-20-deoxyuridine (BrdU), media containing 10 μmol/l BrdU was added to 70% confluent cultures for 24 h. To investigate its efficiency for cell labeling, at the completion of the incubation period, the cells were stained with antibodies against BrdU. Cultured cells on chamber slides were fixed with ice-cold acetone/methanol (1:1) at 20 °C for 30 min. After incubating with 2 N of HCl at 20 °C for 30 min and washed three times in PBS, the slides were incubated overnight with primary antibodies against BrdU at a 1:500 dilution at 4 °C. Negative control samples were incubated in PBS without the primary antibodies. After incubation with the proper secondary antibody at 37 °C for 45 min, slides were mounted with anti-fading mounting medium and examined using fluorescence microscope equipment (Olympus BX51, Olympus Co., Ltd., Tokyo, Japan).

Twenty eight days after the corneal injury, fibrin gels were transplanted onto the rabbit cornea. For the transplantation, the rabbits with damaged cornea were divided into two random groups. Group 1 underwent a graft of fibrin gel with labeled Rb-MSCs (the labeled Rb-MSCs were suspended in a fibrin gel). For Group 2, the transplanted to fibrin gel did not contain Rb-MSCs. For all the rabbits, the damaged corneal surface was carefully keratectomized under anesthesia. The surgery procedures were the same in both groups. A 360° limbal peritomy was performed with Vanass scissors and Bonn forceps. The corneal epithelium that remained was scraped off from the central cornea with a number 15 scalpel blade. The fibrin gel with (or without) Rb-MSCs was placed on the cornea with the conjunctiva covering the edges. The grafts were secured in place by four interrupted cardinal sutures of 10–0 nylon. To protect the fresh graft, a complete central tarsorrhaphy was achieved with a single 5–0 mattress suture. Seven days after surgery, the eyelid was opened. After the eyelid was opened, antibiotic drops were applied to the operated eye three times a day throughout 21 day.

Rabbit corneas were observed with a slit lamp microscope (SL-1600; Nidek Co., Ltd., Aithi, Japan) every day to evaluate reepithelialization, neovascularization, and transparency. On postoperative day 28, rabbits were euthanized with a 10% overdose of trichloroacetaldehyde monohydrate solution for histology and immunofluorescence study.

To induce differentiation in vitro, BrdU labeled Rb-MSCs were divided into two groups: (1) Group A: Rb-MSCs were established using 0.45 μm MILLICELL-HA membrane Transwell culture system. A suspension of Rb-LSCs was added to the upper membrane surface and inserts were positioned in the culture wells, which had been incubated with Rb-MSCs. (2) Group B: Rb-MSCs were cultured with supernatant medium. The supernatant medium used to culture Rb-MSCs had first been used to culture Rb-LSCs and were then filtered with a 0.45 μm filter. After induced treatment, the expression of cytokeratin 3 (CK3) was observed by immunofluorescence and flow cytometric analysis. The process was repeated three times in all groups. Pure Rb-MSCs were used as negative controls.

At the end of the experiment, rabbit corneas from the in vivo experiments were collected, fixed in 4% paraformaldehyde, and embedded in Tissue Tek OCT compound for sectioning. The cells grown in glass chamber slides were fixed with ice-cold acetone/methanol (1:1) for 30 min at room temperature. For hematoxylin and eosin (H&E) staining, consecutive sections (6 μm) were routinely washed, stained, and observed with a light microscope (Olympus BX51; Olympus Co., Ltd., Tokyo, Japan). For immunofluorescence, the cells and the sections were incubated with 3% BSA/0.1% Triton X-100/PBS for 30 min at room temperature to block unspecific binding sites. The cells were incubated with primary antibody (CK3 1:100, p63 1:50, integrinß1 1:50), and the sections were incubated overnight with anti-CK3 (1:100) antibody at 4 °C. After washing three times in PBS, the slides were incubated with proper secondary antibody at 37 °C for 45 min. Isotype-matched primary antibodies served as controls. Specimens were mounted with anti-fading mounting medium and examined using fluorescence microscope equipment (Olympus BX51; Olympus Co., Ltd.).

Double staining with CK3 and BrdU was performed on cells grown on coverslips and on the sections. After incubation with antibody directed against CK3 (1:100) and Cy3-conjugated secondary antibody, the slides were subjected to anti-BrdU (1:100) antibody and FITC-conjugated secondary antibody. After covering with anti-fading mounting medium, the specimens were examined by confocal microscopy (Zeiss 510 META; Carl Zeiss Co, Ltd, Jena , Germany).

Cells were harvested by trypsinization, and nonspecific binding was prevented by incubating the cells in 3% BSA for 30 min. Approximately 5x10⁵ cells were suspended in a final volume of 100 μl of flow cytometric (FC) buffer and incubated with the primary antibody (CK3 1:100, CD34 1:100, and CD29 1:100) or isotype contro1 at 4 °C for 30 min. After being washed twice, the cells were incubated with proper secondary antibody in FC buffer for 30 min at 4 °C. The cells were washed twice with cold FC buffer and suspended in a final volume of 500 μl for flow cytometric analysis. Fluorescent cell analysis was performed with FACSC alibur cytometer (Becton Dickinson Immunocytometry System, San Jose, CA) and data analyzed by FlowJo software (Tree Star, Inc., Ashland, OR).

Results are expressed as the mean±SD% from three replicate experiments. Where appropriate, differences were analyzed by ANOVA. A p<0.05 was regarded as statistically significant.

Within 24 h after seeding, most Rb-LSCs attached to plastic and started spreading. Small colonies, including approximately 8–10 cells, were formed day 3–4 after inoculation and extended very quickly thereafter. The cell morphology appeared to be compact, uniform, and small polygon in shape. By day 10–14, colonies began to fuse and formed a monolayer on confluence (Figure 1A). Immunofluorescence revealed that most Rb-LSCs were positive for p63 (Figure 1B), integrinß1 (Figure 1C), and some of the cultured cells expressed cytokeratin 3 (Figure 1D,E). MSCs were successfully established from bone marrow collected from 10 anesthetized rabbits (n=10). Most of the non-adherent cells were removed during the first media change at 24 h. Colonies of fibroblast-like cells attached to the plastic were evident at day 4–5 after initial seeding. Most cell lines were composed of cells with a characteristic spindle shape. The number and size of the colonies reached 80% confluence by day 14–15 (Figure 1F). After adipogenic induction, intracytoplasmic lipid vesicles were observed by oil red O staining in induced Rb-MSCs (Figure 1G). Immunofluorescence revealed that Rb-MSCs labeling by BrdU expressed green fluorescence in the cell nucleus (Figure 1H). Flow cytometry analyses confirmed that 97. 5% of the cells expressed CD29 antigen and 5.15% of the cells expressed CD34 antigen (Figure 2).