Human corneas with donor consent for research were obtained from the Queensland Eye Bank and used with ethics approval from the Metro South Hospital and Health Service's Human Research Ethics Committee (HREC/07/QPAH/048). Sheep corneas were obtained from euthanized animals at the Herston Medical Research Facility of the University of Queensland under a tissue sharing agreement.
1. Preparation of dissection tools
2. Preparation of culture medium and tissue culture plates
3. Explant dissection and cell culture procedure
4. Continuous production of corneal endothelial cells by serial explant culture
NOTE: Explants can be transferred to fresh tissue culture plates after 10 days to establish additional corneal endothelial cell cultures.
5. Growing corneal endothelial cells on glass coverslips for immunofluorescence analyses
NOTE: Cell cultures that are destined to be analyzed using immunofluorescence should be established on glass coverslips that can be mounted onto glass microscope slides following the staining procedure.
6. Subculture of corneal endothelial cells using Dispase II
NOTE: Large fibroblastic cells can be selectively removed from explant cultures in 6-well plates before subculturing using this procedure. If all cells are to be subcultured, do not perform steps 6.2 to 6.4. The aim of this procedure is to transfer the cells to fresh plates while maintaining their cell-to-cell contacts as much as possible. The cells should be handled gently. Completely confluent wells should be passaged at a ratio of 1:2, while subconfluent wells should be passaged at a ratio of 1:1 or less.
7. Growth of corneal endothelial cell layers on biomaterial membranes
NOTE: The following procedure describes the steps involved in mounting a membranous biomaterial in a custom-made mounting device—called a micro-Boyden chamber—for cell culture. Please refer to our recent publication6 for further information about the device and for purchasing details.
The method for isolating and expanding corneal endothelial cells from human or sheep corneas is summarized in Figure 1 and Figure 2. Most explants that are derived from the corneas of 1 to 2-year-old sheep or human donors of less than 30 years of age will attach to Attachment Factor-coated tissue culture plates within a week, however, it is not unusual to find that up to one third of explants fail to attach within this time. These 'floating' explants can be removed from the cultures. Explants from human donors older than 30 years are less likely to attach to the plate and also less likely to produce cell cultures. Representative images of corneal endothelial cell cultures generated from sheep and human explants are shown in Figure 3 and Figure 4. The cells that emerge from the explants generally remain in contact with each other as they migrate out onto the plate. This kind of migration is known as collective cell migration, and it is a feature of epithelial cells7. By 2 weeks of culture, patches of small, tightly-packed cells will have formed immediately next to many of the explants from both sheep and human corneas8. These patches of cells do not exhibit morphological characteristics of EMT and expand slowly over time. Larger cells with more irregular, fibroblastic shapes can be found outside of these patches. Once the cultures have been established, the explants can be removed using forceps and placed into fresh plates to establish new cultures.
Small, tightly packed cells within the corneal endothelial cell cultures are very resistant to digestion with TrypLE, while the larger fibroblastic cells are more sensitive to it. This difference in TrypLE resistance can be exploited to selectively remove large cells from the cultures before transferring the smaller cells to new plates. Representative images of human corneal endothelial cell cultures throughout the subculturing process using TrypLE and Dispase II are shown in Figure 4.
Immunofluorescence analyses can be conducted on corneal endothelial cell cultures to locate specific proteins in the cells. An example of this is presented in Figure 5. Explants from sheep and human corneas were placed onto Attachment Factor-coated glass coverslips in 24-well plates and cultured for 4 weeks. The explants were removed and then the cultures were analyzed using immunofluorescence for the presence of ZO-1, a tight junction protein, and N-cadherin, an adherents junction protein, according to our published protocol9. The same anti-ZO-1 and anti-N-cadherin antibodies were used for both sheep and human cells, and the results showed that both proteins were detected in the plasma membranes of cells from both species. ZO-1 is normally present as a distinct band at the cell border but becomes weak or absent in cells undergoing EMT. Therefore, the robust ZO-1 expression in these cultures indicated that the cells had not undergone EMT.
Our custom-made micro-Boyden chambers are designed to suspend a biomaterial membrane within the well of a 6-well tissue culture plate (Figure 6). The procedure for mounting a biomaterial membrane into a micro-Boyden chamber is shown in Figure 7. The design of the micro-Boyden chamber allows both sides of the membrane suspended within it to be used as cell culture surfaces simultaneously. To demonstrate this, sheep stromal cells derived from corneal stromal tissue were seeded at a density of 100,000 cells/cm2 onto one side of a collagen type I membrane and then 24 h later the chamber was flipped over and sheep corneal endothelial cells were seeded onto the other side of the membrane at a density of 400,000 cells/cm2. The tissue-engineered cell layers were cultured for 4 weeks, then fixed with 10% neutral buffered formalin and stained using rhodamine phalloidin and Hoechst nuclear dye 33342. They were then examined and photographed using a confocal microscope (Figure 8). A cross-section view of the tissue-engineered cell construct revealed a single layer of corneal endothelial cells on one surface of the collagen membrane and a multi-layered culture of corneal stromal cells on the other surface.
Figure 1: Technique for obtaining explants of endothelium/Descemet's membrane from fresh corneas. (A) The cornea is placed endothelium-side up in a Petri dish under a dissecting microscope. (B) Close up view of the area indicated by a red rectangle in (A). Watchmaker forceps are used to gently peel away Descemet's membrane from the underlying stroma. Please click here to view a larger version of this figure.
Figure 2: Procedure for establishing and expanding cultures of corneal endothelial cells from endothelium/Descemet's membrane explants. Please click here to view a larger version of this figure.
Figure 3: Representative phase contrast images of endothelial cell cultures during initial establishment from explants of sheep corneal endothelium/Descemet's membrane. (A) A sheep corneal endothelium/Descemet's membrane explant after 3 days in culture. Corneal endothelial cells have begun to migrate onto the plate. (B) A sheep explant culture after 1 week. The explant is surrounded by a confluent sheet of cells. (C) A sheep explant culture after 2 weeks. Small cells surround the explant while larger cells are located further away. (D) A sheep explant culture after 6 weeks. A region of small, tightly packed cells is seen next to a region of larger cells. Please click here to view a larger version of this figure.
Figure 4: Isolation of small, tightly packed human corneal endothelial cells for subcultures. (A) A human corneal endothelium/Descemet's membrane explant culture after 7 weeks. Many small, tightly packed cells are present next to the explant. (B) Regions of small, tightly packed cells develop in human explant cultures in a similar manner to that observed in sheep explant cultures. (C) A human explant culture after removal of the explant and after 20 min exposure to TrypLE. Small, tightly packed cells have retained their attachment to the plate while the larger cells have floated away. (D) A human corneal endothelial cell culture after 1 h in Dispase II. Most cells have detached from the plate as free-floating clumps. (E) A human corneal endothelial cell subculture after 1 day. Cells have migrated outwards from cell clumps that were isolated from the original explant culture. (F) A human corneal endothelial cell subculture after 12 days. The cells have formed a confluent monolayer. Please click here to view a larger version of this figure.
Figure 5: Localization of ZO-1 and N-cadherin proteins in the membranes of sheep and human corneal endothelial cells by dual-labelling immunofluorescence. Sheep and human corneal endothelium/Descemet's membrane explant cultures were established on Attachment Factor-coated glass coverslips and analyzed after 4 weeks. Both ZO-1 (green stain) and N-cadherin (red stain) were detected in the membranes of sheep (A and B) and human (D and E) corneal endothelial cells. Analysis of the merged images revealed that the two proteins were highly co-localized within the cultures (C and F). Please click here to view a larger version of this figure.
Figure 6: Diagram of a micro-Boyden chamber shown in cross section. Our custom-made micro-Boyden chamber consists of an upper chamber, a lower chamber and an O-ring. It can be used to suspend any type of membranous material within a tissue culture well. Please click here to view a larger version of this figure.
Figure 7: The procedure for mounting a biomaterial membrane in a micro-Boyden chamber for tissue culture. (A) The equipment required for this procedure includes a polytetrafluoroethylene cutting board, a pair of forceps, a trephine of 18 mm in diameter, a custom-made micro-Boyden chamber and a biomaterial membrane. (B) Use the trephine to punch out a disc from the biomaterial membrane. (C) Place the O-ring into the upper chamber of the mounting device and then lay the biomaterial disc over it. (D) Screw the lower chamber onto the upper chamber of the mounting device. (E) The assembled micro-Boyden chamber is ready to be sterilized with 70% ethanol. (F) Immerse the sterilized micro-Boyden chamber in tissue culture medium in the well of a 6-well tissue culture plate. Please click here to view a larger version of this figure.
Figure 8: Sheep corneal endothelial and stromal cells on opposing sides of a collagen type I membrane. The cells were stained with phalloidin rhodamine to visualize actin (red) and Hoechst to visualize nuclei (blue). The collagen type I membrane was not stained and is therefore not visible in these images that were collected using confocal microscopy. (A) A low magnification, cross section view of the tissue-engineered construct. A thin layer of actin representing a corneal endothelial cell culture is visible on the upper surface of the membrane, and a thicker layer of actin representing a stromal cell culture is present on the lower surface of the membrane. Blue nuclei are not shown in this image. (B) En face view of the corneal endothelial cell layer showing both actin and nuclei staining. (C) En face view of the corneal stromal cell layer showing both actin and nuclei staining. Please click here to view a larger version of this figure.
Attachment factor | Gibco | S006100 | A 1X sterile solution containing gelatin that is used to coat tissue culture surfaces. Store at 4 °C. |
Bovine pituitary extract | Gibco | 13028014 | A single vial contains 25 mg. Freeze in aliquots. |
Calcium chloride | Merck | C5670 | Dissolve in HBSS to make a 1 mM stock solution. Filter sterilise. |
Centrifuge tube, 50 ml | Labtek | 650.550.050 | |
Chondroitin sulphate | LKT Laboratories | C2960 | This is bovine chondroitin sulphate. Dissolve in HBSS to make a 0.08 g/mL stock solution. Filter sterilise and freeze in aliquots. |
Dispase II | Gibco | 17105-041 | Dissolve in DPBS to make a 2 mg/mL stock solution. Filter sterilise and freeze in aliquots. |
Ethanol | Labtek | EA043 | 100% undenatured ethanol should be diluted to 70% in deionised water for sterilising instruments and surfaces. |
Foetal bovine serum | GE Healthcare Australia Pty Ltd | SH30084.03 | This is a HyClone brand of foetal bovine serum. |
Coverglass No. 1, Ø 13 mm | Proscitech | G401-13 | Place sterilised cover slips into 24-well plates for tissue culture. |
HBSS | Gibco | 14025-092 | Hank's balanced salt solution, 1X, containing calcium chloride and magnesium chloride. |
L-ascorbic acid 2-phosphate | Merck | A8960 | Dissolve in HBSS to make a 150 mM stock solution. Filter sterilise. |
Micro-Boyden chamber | CNC Components Pty. Ltd. | Upper ring: QUT-0002-0006, Base ring: QUT-0002-0007 | Both components are made from polytetrafluoroethelyne (PTFE). |
O-ring for micro-Boyden chamber | Ludowici Sealing Solutions | RSB012 | Composed of silicon rubber. |
Opti-MEM 1 (1X) + GlutaMAX-1 | Gibco | 51985-034 | A reduced serum medium containing glutamine. |
DPBS | Gibco | 14190-144 | Dulbecco's phosphate buffered saline, 1X, without calcium chloride and magnesium chloride. |
Pen Strep | Gibco | 15140-122 | A 100X antibiotic solution containing 10,000 Units/mL penicillin and 10,000 µg/mL streptomycin. |
Petri dish | Sarstedt | 82.14473.001 | Sterile Petri dish, 92 X 16 mm, for tissue dissections. |
Tissue culture plate, 24 well | Corning Incorporated | Costar 3524 | A plate containing 24 wells, each with a surface area of 2 cm2. |
Tissue culture plate, 6 well | Corning Incorporated | Costar 3516 | A plate containing 6 wells, each with a surface area of 9 cm2. |
TrypLE Select | Gibco | 12563-011 | A 1X enzyme solution for dissociating cells. |
Versene | Gibco | 15040-066 | A 1X EDTA solution for dissociating cells. |
Watchmaker forceps | Labtek | BWMF4 | Number 4 watchmaker forceps work well for removing strips of endothelium/Descemet's membrane from corneas. |
Corneal endothelial cell cultures have a tendency to undergo epithelial-to-mesenchymal transition (EMT) after loss of cell-to-cell contact. EMT is deleterious for the cells as it reduces their ability to form a mature and functional layer. Here, we present a method for establishing and subculturing human and sheep corneal endothelial cell cultures that minimizes the loss of cell-to-cell contact. Explants of corneal endothelium/Descemet's membrane are taken from donor corneas and placed into tissue culture under conditions that allow the cells to collectively migrate onto the culture surface. Once a culture has been established, the explants are transferred to fresh plates to initiate new cultures. Dispase II is used to gently lift clumps of cells off tissue culture plates for subculturing. Corneal endothelial cell cultures that have been established using this protocol are suitable for transferring to biomaterial membranes to produce tissue-engineered cell layers for transplantation in animal trials. A custom-made device for supporting biomaterial membranes during tissue culture is described and an example of a tissue-engineered graft composed of a layer of corneal endothelial cells and a layer of corneal stromal cells on either side of a collagen type I membrane is presented.
Corneal endothelial cell cultures have a tendency to undergo epithelial-to-mesenchymal transition (EMT) after loss of cell-to-cell contact. EMT is deleterious for the cells as it reduces their ability to form a mature and functional layer. Here, we present a method for establishing and subculturing human and sheep corneal endothelial cell cultures that minimizes the loss of cell-to-cell contact. Explants of corneal endothelium/Descemet's membrane are taken from donor corneas and placed into tissue culture under conditions that allow the cells to collectively migrate onto the culture surface. Once a culture has been established, the explants are transferred to fresh plates to initiate new cultures. Dispase II is used to gently lift clumps of cells off tissue culture plates for subculturing. Corneal endothelial cell cultures that have been established using this protocol are suitable for transferring to biomaterial membranes to produce tissue-engineered cell layers for transplantation in animal trials. A custom-made device for supporting biomaterial membranes during tissue culture is described and an example of a tissue-engineered graft composed of a layer of corneal endothelial cells and a layer of corneal stromal cells on either side of a collagen type I membrane is presented.
Corneal endothelial cell cultures have a tendency to undergo epithelial-to-mesenchymal transition (EMT) after loss of cell-to-cell contact. EMT is deleterious for the cells as it reduces their ability to form a mature and functional layer. Here, we present a method for establishing and subculturing human and sheep corneal endothelial cell cultures that minimizes the loss of cell-to-cell contact. Explants of corneal endothelium/Descemet's membrane are taken from donor corneas and placed into tissue culture under conditions that allow the cells to collectively migrate onto the culture surface. Once a culture has been established, the explants are transferred to fresh plates to initiate new cultures. Dispase II is used to gently lift clumps of cells off tissue culture plates for subculturing. Corneal endothelial cell cultures that have been established using this protocol are suitable for transferring to biomaterial membranes to produce tissue-engineered cell layers for transplantation in animal trials. A custom-made device for supporting biomaterial membranes during tissue culture is described and an example of a tissue-engineered graft composed of a layer of corneal endothelial cells and a layer of corneal stromal cells on either side of a collagen type I membrane is presented.