Construction Of Bioengineered Corneal Posterior Lamella With Acellular Corneal Matrix And Endothelial Cell-like Cells

Corneal disease is the second most common cause for blindness worldwide, and so as it is in China. Corneal transplantation is the most effective treatment for irreversible corneal blindness at present, but extreme shortage of donor corneas is a big obstacle for its broad application. Bioengineered corneas have affluent sources, which can effectively relieve the burden and will be promising substitutes for natural donor corneas.Bioengineered cornea is constructed by seeding in vitro cultured cells (including epithelial, stromal and endothelial cells) on/in a scaffold and culturing them in a3-dimensional system. Scaffolds and seed cells are two key elements, which are also the nodi and hot topics in bioengineering research.Due to their deficiency in biocompatibility, transparency or mechanical strength, most scaffold materials are only restricted to basic researches such as histology, biophysics, pathology, toxicology, pharmacology and corneal injury repair. Acellular porcine corneal matrix (APCM) was created in our previous studies and proved to be a kind of favorable scaffold for bioengineered corneal construction. It has been successfully applied in the construction of rabbit corneal anterior lamella, but if it can support the growth of endothelial cells and form functional bioengineered corneal endothelium is still unknown. The construction of bioengineered corneal endothelium will not only provide models for ex vivo studies of corneal endothelium, but also will establish technical foundation for full-thickness bioengineered corneal construction and also will be a potential source of implant for corneal endothelium transplantation.On the other hand, the construction of bioengineered corneal endothelium requires large amount of endothelial cells. Since natural human corneal endothelial cells have limited proliferative ability and are hard to be amplificated in vitro, looking for new sources is an urgent issue at the moment. Now most bioengineered corneal endothelia are not suitable for clinical use, because they are constructed with animal cells or immortalized human cells, which have relatively stronger proliferative ability. Neural crest stem cell (NCSC) is the natural precursor of corneal endothelial cell (CEC) and has affluent source (induced for human embryonic stem cell), it can be an ideal source of endothelial cells if induced to differentiate successfully in vitro.Concerned with all the issues mentioned above, this study firstly tested the biocompatibility of APCM with human endothelial cell line by constructing a bioengineered human corneal endothelium equivalent, which is similar to their natural counterpart in structure and function. Secondly, this study investigated the feasibility of inducing rat neural crest stem cells to differentiate into corneal endothelial cells and we acquired functional corneal endothelial cell-like cells. At last, this study made preliminary research on construction of epithelium-less bioengineered corneas with APCM, stromal cells and induced corneal endothelial cell-like cells, and the bioengineered corneas were subjected to animal transplantation; the results indicated that there should be some improvements in implant construction, choice of experimental animals and surgery skills.Part I A Human Corneal Endothelium Equivalent Constructed with Posterior Lamella APCM and B4G12Cells[Purpose] This study aimed to demonstrate the feasibility of constructing human corneal endothelium equivalents using posterior lamella APCM scaffolds and human corneal endothelial cells.[Methods]1. Preparation of posterior lamella APCM scaffolds:Posterior lamella with1/2thickness of fresh porcine corneas were thoroughly washed with phosphate buffered saline containing1%penicillin-streptomycin. Then they were decellularized with0.5%sodium dodecyl sulfate (SDS) solution at4℃on a shaker for24hours. After decellularization was finished, tissues were thoroughly washed with sterile phosphate buffered saline8times in16hours and put in-20℃for12hours to be freeze-dried, then transferred to cabinet for air-drying for3hours and stored at-20℃.2. Morphology and toxicity detection of posterior lamella APCM; APCM scaffolds were subjected to HE staining to evaluate the effect of decellularization. Leaching liquid was extracted from APCM and MTT was used to test its toxicity for B4G12cells.3. Culture of Human corneal endothelial cells B4G12:B4G12cells were cultured with human endothelium serum free medium containing basic fibroblast growth factor in flasks coated with laminin and chondroitin-6sulphate under normal conditions. Cells were digested with0.05%trypsin-0.02%EDTA and passaged according to routine procedure.4. Construction of bioengineered corneal endothelium equivalents:Posterior lamella APCM in lmm thickness and8mm diameter were cut under anatomical microscope and soaked in B4G12medium for24hours at37℃before cell seeding. B4G12cells were seeded on the Descemet's membrane at a density of2000/mm2and cultured for2weeks. Corneal endothelium equivalents were analyzed using trypan blue and alizarin red S co-staining, HE staining, immunofluorescence for zonula occluden-1(ZO-1) and Na+/K+ATPase and corneal swelling assay.[Results] HE staining showed that porcine cells were completely removed from posterior lamella APCM scaffolds and structure of collagen was well preserved. Leaching liquid from APCM had no obvious toxic impact on proliferative curve of B4G12cells (P>0.05). Trypan blue and alizarin red S co-staining showed that B4G12were alive after2weeks in organ culture and formed a monolayer covering the surface of APCM with a density of2061±344/mm2. HE and DAPI staining confirmed that a monolayer of cells covered the Descemet's membrane of APCM. Proper localizations of ZO-1and pump protein Na+/K+ATPase were detected by immunofluorescence assay. Functional experiments were conducted to show that Na+/K+ATPase inhibitor ouabain could block the pump function of this protein, leading to persistent swelling by51.7%compared with the control which did not contain oubain and went through reversible corneal swelling; these results indirectly proved that bioengineered corneal endothelium equivalents had a pump function. [Conclusions] Corneal endothelium equivalents had properties similar to those of native corneal endothelium and could serve as good models for in vitro study of human corneal endothelium. Part II Derivation of Corneal Endothelial Cell-like Cells from Rat Neural Crest Stem Cells in vitro[Purpose] The aim of this study was to investigate the feasibility of inducing rat NCSC to differentiate toward functional CEC-like cells in vitro.[Methods]1. Primary culture and identification of rat NCSC:Rat embryos were removed at embryonic day9and were placed in a Petri dish; the neural tubes were dissected at the midbrain level (first10somites) using tungsten needles. Then the neural tubes were transferred to a new Petri dish coated with fibronectin and allowed to adhere before flooded with enough NCSC culture medium. Forty-eight hours later, the neural tubes were removed and NCSC that migrated out of the explants were cultured to reach confluence and harvested for the following experiments; NCSC markers P75and HNK-1were detected by immunofluorescence.2. Primary culture of rat CEC:Rat corneas were incubated in growth medium at37℃overnight to stabilize the cells. After the corneas were centrifuged the next morning, they were incubated in0.02%EDTA for1hour, and the loosened endothelial cells were detached from Descemet's membrane by pipetting the corneas several times through a pipette. Endothelial cells were then centrifuged, resuspended in fresh medium and cultured under normal conditions.3. Preparation of conditioned medium:Collecting the supernatant from cultured rat CEC when they reached70-90%confluence every12hours. The supernatant was filtered and stored at-80℃to preserve its biological activity. Then the supernatant was mixed with NCSC medium at different ratios of1:3,1:1and3:1to acquire25%,50%and75%conditioned medium.4. NCSC induction and detection of CEC differentiation markers: NCSC were respectively induced by conditioned medium and cell co-culture protocols. Growth and morphological change of NCSC were observed everyday and successful induction protocol was initially screened out, and then the induced cells were further analyzed for CEC differentiation markers detection, the specific experimental grouping was shown as below:1) Induction of NCSC with conditioned medium:A:Extracellular matrix coating group:The wells of culture plates were coated with laminin and chondroitin-6sulphate and NCSC were induced with25%,50%and75%conditioned medium respectively;B:Non-coating control group:The wells of culture plates were not coated, and NCSC were induced with25%,50%and75%conditioned medium respectively;2) Induction of NCSC by cell co-culture:A:Extracellular matrix coating group:The upper chamber sides of polycarbonate membranes of Transwell were coated with laminin and chondroitin-6sulphate, rat NCSC and CEC were seeded respectively in the upper and lower chamber of Transwell at a ratios of1:1,1:5,1:10and1:20;B:Non-coating control group:The polycarbonate membranes of Transwell were not coated with laminin and chondroitin-6sulphate, rat NCSC and CEC were seeded respectively in the upper and lower chamber of Transwell at a ratios of1:1,1:5,1:10and1:20;The induced cells that had positive morphological change were collected and subjected to immunofluorescence for CEC differentiation markers N-cadherin, ZO-1and Na+/K+ATPase detection, flow cytometric analysis for confirming the proportion of N-cadherin positive cells, and RT-PCR for CEC differentiation-related genes FoxCl and Pitx2detection.5. Transplantation of CEC-like cells:Rat corneal endothelium deficiency models were created by nitrogen-freezing protocol. Fluorescent dye-labeled CEC-like cells were injected to the anterior chamber of rat corneal endothelium deficiency models with a cell density of3000/mm2, and an eye-down position was maintained for24hours to allow the cell attachment, blank rat models were used as controls. The animals were observed by slit lamp and confocal microscopy through focusing after surgery, as well as histological examination to evaluate the function and bio-safety of CEC-like cells in vivo.[Results] The cells that migrated from the neural tube were both P75and HNK-1positive, indicating their NCSC origin. Spindle-like NCSC turned to polygonal CEC-like morphology1week after being induced by75%conditioned medium+10%fetal bovine serum in the coating group. About40-50%NCSC changed their morphology after2weeks and the proportion kept stable after2weeks. The induced cells expressed N-Cadherin, FoxC1, Pitx2, ZO-1and sodium-potassium pump Na+/K+ATPase. The corneas of experimental group were much clearer than those of control group and the mean corneal thickness in experimental group was back to normal1month after surgery. The corneas in the control group were continuously swelling and hazy during the observation. Confocal microscopy though focusing and histological analysis confirmed green fluorescence-positive CEC-like cells formed a monolayer covering the Descemet's membrane in experimental group with a cell density of2872.6±172.7/mm2. And there were no cells on the Descemet's membrane in the control group.[Conclusion] NCSC could be induced to differentiate to functional CEC-like cells with conditioned medium. Part Ⅲ Construction of Epithelium-less Bioengineered Cornea and Animal Transplantation[Purpose] To explore the feasibility of constructing epithelium-less bioengineered cornea with posterior lamella APCM, CEC-like cells and stromal cells and to evaluate the in vivo performance of the construct.[Methods]1. Posterior lamella APCM scaffolds and CEC-like cells were prepared according to previous methods.2. Culture of primary rabbit corneal stromal cells:Fresh rabbit corneas were deprived of epithelium and endothelium, and the stroma was cut into small pieces and allowed to adhere to the culture dish until stromal cells migrated out. The pieces were removed and stromal cells were cultured under normal conditions.3. Construction of epithelium-less bioengineered corneas and penetrating corneal transplantation:Stromal cells were injected into posterior lamella APCM scaffolds and gradually trained to adapte to the culture medium of CEC-like cells, then CEC-like cells were seeded on the Descemet's membrane and the constructs were cultured for2weeks. New Zealand rabbit penetrating corneal transplantation was performed using epithelium-less bioengineered corneas as implants, and amniotic membranes were used to cover the nude epithelial side of the implants, posterior lamella APCM alone was used as controls. TobraDex eye drop were used4times a day and TobraDex ointment were used once at night. Animals were subjected to slit lamp observation at specific time points and HE-staining of the implants was also conducted.[Results] Conjunctival congestion and secretion were obvious1week after the surgery. Amniotic membranes had melted and fell off the implants naturally. Bioengineered cornea and blank posterior lamella APCM implants were both edematous, and fluorescein staining showed that about70%surface of the bioengineered cornea implant was covered with regenerative epithelium, while the epithelium healed at a much lower speed for blank posterior lamella APCM implant. The inflammation reaction regressed a bit, and bioengineered cornea implant was less edematous2weeks after the surgery, its epithelium was almost completely repaired. Blank posterior lamella APCM implant kept edematous2weeks after the surgery, and its epithelial surface was not smooth enough. HE-staining showed that there were more inflammatory cell-like cells in bioengineered cornea implant than in blank posterior lamella APCM implant, but the collagen arranged much better in bioengineered cornea implant and the Descemet's membrane was covered by a monolayer of cells.[Conclusion] The rejection was severe after the surgery. There should be further improvements in aspects of bioengineered corneal construction, experimental animal preparation and surgery manipulation.