SEARCH

SEARCH BY CITATION

Keywords:

  • corneal cells;
  • decellularization;
  • porcine cornea;
  • tissue engineering;
  • xenografting

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Purpose:  To evaluate the potential use of decellularized porcine corneas (DPCs) as a carrier matrix for cultivating human corneal cells in tissue engineering.

Methods:  Corneal cells were isolated from human corneoscleral rims. Porcine corneas were decellularized using hypotonic tris buffer, ethylene diamine tetra-acetic acid (EDTA, 0.1%), aprotinin (10 KIU/ml) and 0.3% sodium dodecyl sulphate. Haematoxylin–eosin (HE) and 4,6-diamidino-2-phenylindole (DAPI) staining were performed to confirm removal of the corneal cells. Quantitative analysis was performed to determine levels of desoxyribonucleic acid (DNA) using DNA Purification Kit (Fermentas, St. Leon-Rot, Germany). Alcian blue staining was carried out to analyse the structure of the extracellular matrix (ECM). Corneal stromal cells were injected into the DPCs; limbal corneal epithelial cells and corneal endothelial cells were seeded onto the anterior and posterior surfaces of the DPCs, respectively. Evaluation was undertaken at days 14 and 30. The phenotypical properties of the cultivated corneal cells were investigated using Immunolocalization of type I collagen, keratocan, lumican, cytokeratin 3 (AE5) and type VIII collagen.

Results:  Haematoxylin–eosin and DAPI staining showed efficient elimination of porcine corneal cells, whereas alcian blue confirmed gross preservation of the ECM. The quantitative analysis of the DNA content showed a significant reduction (mean before decellularization: 75.45 ± 13.71 ng/mg; mean after decellularization: 9.87 ± 2.04 ng/mg, p < 0.001). All three types of corneal cells were efficiently cultured and expanded on the DPCs.

Conclusions:  Decellularized porcine corneas might serve as a potential scaffold for tissue engineering of the cornea, possibly providing xenogenic substrate for corneal transplantation.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The cornea may be injured by various types of diseases leading to profound visual impairment or blindness (Wilson 1975; Kadar et al. 2009). So far, the only treatment for irreversible corneal damage is corneal transplantation. Since the first human corneal transplantation in 1905, perforating keratoplasty has been one of the most successful forms of tissue transplantation (Moffatt et al. 2005; Hanada et al. 2008). Despite its benefits for the patient, a severe shortage of donor corneas exists worldwide (Golchet et al. 2000).

Hence, alternatives for tissue procurement ameliorate the supply of corneal material. At the same time complications, especially rejection and transplant failure, should be addressed (Thompson et al. 2003; Panda et al. 2007).

This lead to the concept of a tissue-engineered cornea using autologous cells cultured in various extracellular 3-D matrices (Griffith et al. 1999; Doillon et al. 2003; Alaminos et al. 2006; Liu et al. 2006; Mimura et al. 2008; Vrana et al. 2008; Fagerholm et al. 2010). Although different types of scaffolds have been evaluated, problems remain including cell culture disadvantages, primary rejection, biotoxicity and biocompatibility (Mimura et al. 2004; Shimmura et al. 2005). A new strategy for preparing a scaffold, besides the use of biomaterials and synthetic polymers, is the use of decellularized xenogenic tissue. Donor cells and antigens are completely removed, thereby diminishing the host immune reaction, and a repopulation of the matrix with recipient cells performed (Wilshaw et al. 2006; Ingram et al. 2007; Mirsadraee et al. 2007; Ott et al. 2008). This technique enables us to utilize xenogenic tissue as a natural scaffold for tissue engineering.

The aim of this work was to evaluate the potential use of decellularized porcine corneas (DPCs) as a scaffold to expand and transplant human corneal cells.

Materials and Methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Isolation of human corneal cells and culture conditions

Human corneoscleral rims from six donors were obtained from the Tuebingen Eye Bank. The rims were to serve as a source of human corneal cells to be transferred to the porcine corneal scaffolds. For the use of human corneascleral rims, a written approval was obtained from the legal representatives of the donors according to the protocols established by the German federal law and the guidelines of the University of Tuebingen. In addition, two whole human corneas, which were not suitable for transplantation, were obtained from our eye bank after written consent for clinical and research purposes. These corneas served as control material in subsequent histological analyses. The mean age of the donors was 72.0 ± 6.3 years. The mean time between death and enucleation of the eye globes and the mean time in organ culture I (Biochrom AG, Berlin, Germany) was 10.3 ± 3.4 h and 16.2 ± 5.1 days, respectively.

Limbal corneal epithelial cells (LCECs) were obtained from discoid samples (1.2 mm diameter) taken under a microscope from the limbus. Each sample was incubated at 37 °C in 3 mg/ml dispase II dissolved in Dulbecco’s Modified Eagle’s Medium (DMEM/F12; Gibco, Invitrogen, Carlsbad, CA, USA) for 1 h. Samples were incubated with 0.25% Trypsin/ethylene diamine tetra-acetic acid (EDTA) (Gibco, Invitrogen) for 15 min to produce single-cell suspensions and were seeded in 24-well plates (Costar; Corning Incorporated, NY, USA). For preparation of corneal stromal cells (CSCs), the stroma was cut into pieces and incubated at 37 °C for 30 min in DMEM/F12 containing 1 mg/ml of collagenase A (Gibco, Invitrogen). After incubation, collagenase A was removed, and the digested stromal segments were incubated in a second aliquot of collagenase A for 30 min. The digested tissue was centrifuged at 215 g for 5 min. The solution was decanted, and the CSCs-containing pellet was re-suspended in 24-well plates. Small explants from the endothelial layer, including the Descemet membrane, were removed with sterile surgical forceps under a stereoscopic dissection microscope and trypsinized for 10 min and seeded in 24-well plates. Corneal cells were maintained in DMEM/F12 containing 5% foetal calf serum, 100 U/ml penicillin G, 100 μg/ml streptomycin sulphate, 1.25 g/ml amphotericin B, 0.1 ng/ml epidermal growth factor, 1.0 ng/ml basic fibroblast growth factor, and 1.0 μg/ml hydrocortisone in a 5% carbon dioxide humidified environment. The medium was changed every 2–3 days, and the isolation of the corneal cells was performed by using 0.25% Trypsin/EDTA solution at 37 °C for 10 min.

Decellularization of porcine corneas

Fresh porcine eyes were obtained from a slaughterhouse and were sterilized for 10 min in a 10% iodine solution (Braunol; Braun, Melsungen, Germany) before decellularization. Porcine eyes were washed three times with phosphate-buffered saline (PBS; Gibco, Invitrogen) and a central corneal button was trephinated (9.0 mm). Corneas were incubated in 10 mm hypotonic tris buffer (Sigma-Aldrich, Steinheim, Germany) and incubated with protease inhibitors EDTA (0.1% w/v) and aprotinin (10 KIU/ml, pH 7.6; Sigma-Aldrich) at 4°°C for 12 h. Corneas were transferred to 0.3% solution of sodium dodecyl sulphate (SDS; Sigma-Aldrich) in tris-buffered saline (TBS) containing EDTA (0.1% w/v) and aprotinin (10 KIU/ml; pH 7.6), and incubated for 20 h at room temperature. After treatment with SDS, corneas were washed three times in TBS (pH 7.6) containing no protease inhibitors. The tissue was then incubated with 50 U/ml Deoxyribonuclease I (DNase; Sigma-Aldrich) and 1 U/ml Ribonuclease A (RNase; Sigma-Aldrich) from bovine pancreas in a reaction buffer (50 mm tris–hydrochloric acid, 10 mm magnesium chloride and 50 μg/ml bovine serum albumin, pH 7.5) for 3 h at 37 °C, with bland agitation. Thereafter, the tissue was terminally washed using TBS three times and embedded in PBS.

Repopulation of decellularized porcine corneas

First, the isolated human CSCs were injected at ten locations in the decellularized corneal stroma of six eyes with a 20-gauge syringe (103 cells/ml). The DPCs with injected cell suspensions were transferred to 24-well plates and were incubated at aforementioned culture conditions. After 30 days, porcine corneas were fixed in 4% paraformaldehyde and 3% sucrose in PBS (pH 7.4) for 15 min at room temperature. Human LCECs were seeded on the anterior surface of six of the DPCs. A special ring construction enabled us to reproduce a multilayer sheet. Sections were analysed after 14 days of incubation. The same procedure was carried out for human corneal endothelial cells (CECs) for the posterior surface in six eyes. Correct orientation was assured by markings on the anterior surface.

Histological examination and immunohistochemistry

Before starting with the staining procedure, the sections were dewaxed and incubated for 3 min with 3% acetic acid and 30 min with an alcian blue solution (1% in 3% acetic acid). For haematoxylin–eosin (HE) staining, the sections were incubated for 10 min with Harris Haematoxylin solution (Sigma-Aldrich). Thereafter, the sections were stained with eosin and observed under a microscope (Axiovert 135; Zeiss, Oberkochen, Germany). Porcine corneas (control, decellularized and recellularized) and human corneas (control) were fixed in 4% paraformaldehyde and 3% sucrose in PBS (pH 7.4) for 15 min at room temperature. Primary antibodies against keratocan (1:150; Sigma-Aldrich), lumican (1:150; Sigma-Aldrich), type I collagen (1:100; Sigma-Aldrich), cytokeratin 3 (AE 5) and type VIII collagen (1:100; Zymed Laboratories, South San Francisco, CA, USA) were used. Slides were incubated with the primary antibody overnight at 4 °C and washed three times in PBS. Monolayers were then incubated for 1 h at room temperature with the secondary antibody alkaline phosphatase, rabbit/mouse (Dako, Glostrup, Denmark) or fluorescent secondary antibody Alexa 488 goat anti-mouse (1:400; Gibco, Invitrogen). Slides were examined under a microscope (Axiovert 135; Zeiss). Negative controls were incubated without a primary antibody.

4,6-diamidino-2-phenylindole (DAPI) staining, DNA extraction and quantification

The sections were dewaxed and incubated with TBS for 10 min and for 1 min with DAPI (2.0 μg/ml in TBS). The sections were washed twice with deionized water and covered with Fluorsafe (Calbiochem, San Diego, CA, USA) and were examined under a fluorescence microscope. 4,6-diamidino-2-phenylindole positive cells were counted per five fields at 100-fold magnification. DNA was extracted from 15 to 50 mg of fresh and DPCs. First the samples were digested for 24 h at 40 °C with 200 μl papain solution (Merck, Darmstadt, Germany). The further DNA extraction was performed with the DNA Purification Kit (Fermentas GmbH, St. Leon-Rot, Germany) according to the manufacturer`s protocol. Briefly, 200 μl of the sample were mixed with 600 μl lysis solution and incubated for 5 min at 65 °C. Afterwards, 600 μl chloroform was added and the sample was emulsified by inverting the tube. After centrifugation for 2 min at 215 g, the upper aqueous phase was transferred to a new tube and 800 μl of precipitation solution was added. The sample was mixed and centrifuged for 2 min at 215 g. The supernatant was removed, and the DNA was dissolved in 100 μl of 1.2 m NaCl. Afterwards the DNA was precipitated with 300 μl of ice-cold ethanol for 10 min at −20 °C and spun down for 4 min at 215 g. The ethanol was removed and the pellet was washed once with ice-cold 70% ethanol, left to air dry and dissolved in 100 μl of sterile deionized water. The DNA concentration was quantified photometrically at a wavelength of 260 nm.

Expression of results and statistics

All data were expressed as the mean ± SD. Student’s t-test was used to compare mean values from two groups. p < 0.05 was considered as statistically significant and marked with an asterisk. All analyses were performed with commercial software (spss version 17.0; SPSS Inc., Chicago, IL, USA).

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Corneal cells and repopulation of porcine corneas

The isolated human corneal cells maintained their phenotypical properties under culture conditions. The isolated LCECs were immunopositive for AE5, the CSCs for type 1 collagen, keratocan and lumican and the CECs for type VIII collagen (Fig. 1). After isolation of CSCs and injection into the DPCs, the human CSCs grew rapidly and formed a three-dimensional network (Fig. 2). After 30 days of incubation, CSCs maintained a dendritic morphology and formed widespread networks in the DPC. Cultivated LCECs proliferated on the denuded DPCs anterior surface and formed a confluent sheet of epithelial cells within 5–8 days. After 2 weeks, the cultivated LCECs consisted of two to three layers of cells, mostly in direct contact with neighbouring cells (Fig. 3).

image

Figure 1.  Upper row showing the expression patterns of the human cornea with positive immunostaining for keratocan (B), lumican (C), type I collagen (D). (A) negative control. The isolated corneal stromal cells maintain their phenotypical properties and express the same antigens under culture conditions (F–H). (E) negative control. Magnification ×100.

Download figure to PowerPoint

image

Figure 2.  Light-microscope photograph of a comprehensive network of human corneal fibroblasts in the decellularized porcine corneas after 30 days at the injected area. Magnification ×100.

Download figure to PowerPoint

image

Figure 3.  Haematoxylin–eosin staining of porcine acellular cornea matrix after decellularization (A) and 2 weeks after recellularization with human corneal limbal epithelial cells (B). Two layers of human corneal epithelial cells were observed on the surface of the matrix, mostly in direct contact with neighbouring cells. Magnification ×100.

Download figure to PowerPoint

The isolated human CECs attached and spread on the posterior surface of the denuded DPCs. The CECs grew rapidly to reach confluence after 2 weeks of cultivation (Fig. 4).

image

Figure 4.  The isolated human corneal endothelial cells (CECs) attached and spread on the posterior surface of denuded decellularized porcine corneas. After decellularization, no CECs were visible (A). The cultivated CECs grew rapidly to reach confluence after 2 week cultivation (B). Magnification ×100.

Download figure to PowerPoint

Transparency, histological examination and immunohistochemistry

The decellularization process left the DPCs with low transparency. However, treatment with 100% glycerol for 6 h restored transparency close to the native state (Fig. 5). After decellularization, HE staining showed that the majority of the immunogenic porcine corneal cells were removed. Alcian blue staining confirmed an intact extracellular matrix (ECM) with gross preservation of the corneal stroma and intact Bowmann and Descemet′s membrane (Fig. 6). After recellularization with human CSCs, nuclei were again detectable on the stroma of DPCs, indicating stromal repopulation.

image

Figure 5.  Semi-transparent to cloudy porcine cornea immediately after complete decellularization (B), almost complete restoration of transparency after deswelling with 100% glycerol for 6 h (C). Image A presents a normal porcine cornea before decellularization (A).

Download figure to PowerPoint

image

Figure 6.  (A) Haematoxylin–eosin (HE) staining of porcine corneas with corneal stromal cells (CSCs). After decellularization, the HE staining displayed only sparse nuclei in the corneal stroma (B). In the recellularized porcine cornea, human CSCs were visible again (C). Alcian blue staining confirmed an intact extra cellular matrix with gross preservation of the corneal stroma and intact Bowmann and Descemet′s membrane both in the control (D) and the decellularized (E) and recellularized porcine corneas (F). Magnification ×100.

Download figure to PowerPoint

Furthermore, human CSCs maintain their phenotypical properties in the porcine matrix and express lumican, keratocan and type I collagen. The expression patterns of tissue-specific keratins were similar in human corneas. AE 5 was expressed in the cultivated epithelial cell layer on the anterior phase. Immunohistochemical staining of type VIII collagen of the cultivated CECs indicated a well-differentiated endothelial cell layer on the posterior surface of the DPCs. In general, a lower intensity of the immunohistochemical markers was noted compared with the control (Fig. 7).

image

Figure 7.  (A) Represents a native human cornea with immunohistochemical staining for cytokeratin 3 and (C) A native human cornea with immunohistochemical staining for type VIII collagen. (B and C) Recellularized porcine corneas stained with AE5 cytokeratin 3 and type VIII collagen, respectively (B, D). The phenotypical properties are expressed on the recellularized corneas similar to native corneas. Magnification ×100.

Download figure to PowerPoint

DAPI staining and DNA quantification

4,6-diamidino-2-phenylindole staining showed cellular nuclei in the corneal stroma and the endothelial and epithelial layers. 62.0 ± 3.9 cells were counted per five fields at 100-fold magnification. After decellularization, only 3.2% of the cellular nuclei were present in the porcine corneal stroma, indicating a highly significant reduction (p < 0.001). After repopulation with human corneal cells 57.6 ± 3.6 cells were counted per X fields at 100-fold magnification, meeting our expectations. The DNA content of porcine corneas prior to decellularization was 75.45 ± 13.71 ng/mg and afterwards 9.87 ± 2.04 ng/mg, indicating a highly significant drop. Data represent mean value of five samples (Fig. 8).

image

Figure 8.  The DNA content of porcine corneas before treatment (75.45 ± 13.71 ng/mg) has significantly dropped to 9.87 ± 2.04 ng/mg after decellularization (p < 0.001). Data represent mean value of five samples.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The idea of tissue engineering represents a paradigm shift in transplantation surgery and has the theoretical potential to overcome the present donor shortage worldwide (Moffatt et al. 2005). To date, several corneal tissue models have been produced in vitro (Schneider et al. 1999; Orwin et al. 2003; Panda et al. 2007; Ruberti & Zieske 2008).

However, many difficulties still exist for tissue engineering of the cornea and its clinical use. In general, the properties of native soft tissues cannot easily be duplicated by synthetic materials or biomaterials (Porter et al. 1998; Tegtmeyer et al. 2001). The most appropriate matrix for a tissue-engineered cornea is a cornea itself. While human material is limited, xenogenic corneal material is abundant.

Porcine material especially has been investigated extensively as a candidate for xenografting. Previously, it has been shown that the α-gal epitope, which induces hyperacute rejection, is almost absent in the porcine cornea except for several keratocytes in the most anterior part, suggesting that the porcine cornea has clinical potential. Although this may limit the risk of graft rejection, cross-species transmission of porcine pathogens is still a major concern (Amano et al. 2003).

Decellularization is a suitable way to obtain non-immunogenic and functional xenografts for human transplantation (Wilshaw et al. 2006; Ingram et al. 2007; Mirsadraee et al. 2007; Ott et al. 2008; Oh et al. 2009; Du et al. 2011). While decellularization of these rather simple biological structures has shown the general feasibility of this technique, it is a greater challenge to achieve a proper function in complex organs. Recently, Ott et al. (2008) described a method for generating a complete rat heart decellularized by coronary perfusion with detergents.

Our results confirm that the decellularization method used for other organs may also be suitable for the porcine cornea. The three different corneal cell types were successfully incorporated into the decellularized scaffold. Although an incomplete reduction of porcine DNA with 0.3% SDS was noted. A threshold for an immunogenic reaction needs still to be determined. Moreover, SDS seems to lower the corneal transparency most evidently immediately after its application. An acceptable transparency level was restored by 100% glycerol for 6 h. Alcian blue staining of the ECM confirmed a partial disorganization of the fibres possibly explaining the reduced transparency. Still, no gross histological disruption was observed. The intensity of the immunohistochemical images was lower in the porcine tissue compared with the human controls. An explanation of this phenomenon could have been the usage of passage 1 of cultivated cells or an altered surface of the DPCs.

Gonzalez-Andrades et al. compared two decellularization processes. It was shown that both NaCl and SDS achieved a sufficient decellularization level for all three corneal cell types (Gonzalez-Andrades et al. 2011). The authors concluded that the application of a 1.5 m NaCl solution on porcine corneas produces an acellular corneal stroma with adequate histological and optical properties. Human keratocytes were able to penetrate, spread and differentiate within this scaffold. In conclusion, the authors suggest that the decellularization of animal corneas with 1.5 m NaCl represents a useful method for the development of human bioengineered corneas with therapeutic potential. In comparison with our study, the authors used 0.1% SDS and no endothelial or epithelial cells were investigated. However, Du et al. optimised a protocol to produce an acellular porcine corneal scaffold and investigated its mechanical integrity and biocompatibility. They concluded that only SDS resulted in an almost complete decellularization after 24 h. Histological analysis of the acellular matrix showed that the CSCs had been removed substantially, collagen fibres were kept arranged in an orderly fashion, and Bowman’s layer and Descemet’s membrane were both still intact (Du et al. 2011). Sasaki et al. evaluated stability and biocompatibility of an artificial corneal stroma which was decellularized by ultra-high, hydrostatic pressure. They concluded that completely decellularized porcine corneal stroma has an extremely high biocompatibility and has the potential to serve as a scaffold for an artificial cornea (Sasaki et al. 2009).

Zhou et al. supported the results that SDS is able to produce a porcine corneal acellular matrix. Although water absorption and light transmission of the acellular matrix decreased, it showed similar biocompatibility and biomechanical properties as the natural equivalent. After xenotransplantation into rabbit corneal stromal layers, and observation for 4 weeks, complete transparency was noted. For almost 1 year postoperatively, the corneas remained transparent with regular stromal structures and appeared stable in situ without an immunological rejection. They concluded that these acellular matrices, similar to natural corneas in structure, strength, and transparency, have tremendous potential for tissue engineering and corneal transplantation (Zhou et al. 2011).

Fu et al. used Triton X-100 (1%) plus a freeze-drying process for decellularization of porcine corneas and measured the in vivo biocompatibility in a rabbit model with a follow-up of 3 months. The authors observed no sign of rejection and that the acellular matrix gradually integrated into the rabbit cornea, indicating a good biocompatibility. They concluded that the corneal acellular matrix can be used as a scaffold for a tissue-engineered cornea and that a biological corneal equivalent can be reconstructed in a dynamic culturing system (Fu et al. 2010).

Further in vivo studies are necessary to determine whether the xenocorneas can maintain stromal dehydration, long-term transparency and in-vivo biocompatibility. Also, it has to be clarified whether a significant reduction of immunogenic epitopes was achieved to guarantee a long-term immunological tolerance. Although promising studies on other decellularized organs give a genuine hope for decellularized xenocorneas, the above stated questions have to be meticulously addressed before defining its future potential as a realistic treatment alternative. In conclusion, our results demonstrate that by decellularization of porcine corneas and recellularization with human corneal cells, it is possible to obtain viable xenografts for corneal transplantation. In principle, this technique could establish a new dimension of tissue engineering of the cornea and has the theoretical potential to overcome the immanent problem of corneal transplant shortage.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References