Tissue Engineered Trachea Using Decellularized Aorta

Experimental and clinical tracheal, bronchial repair or anastomosis began in the late 19th century [1]. Since then the technique of tracheal reconstruction by end-to-endanostomosis has been used in a variety of tracheal diseases, such as stenosis,tracheomalacia, tuberculosis, traumatic injury, and malignancies [2]. However, this established method is not appropriate for the repair of extensive lesions. Therefore, for these types of lesions other replacement methods, such as autografts [2,3], allografts [4,5], prosthetics materials [6-8] and tissue engineering [9-12] have been studied and developed. Some partial successes have been announced with some of these methods, butso far not one of them has archived a completely satisfactory long-term tracheal replacement for extensive lesions of the trachea.


Introduction
Experimental and clinical tracheal, bronchial repair or anastomosis began in the late 19 th century [1]. Since then the technique of tracheal reconstruction by end-to-endanostomosis has been used in a variety of tracheal diseases, such as stenosis,tracheomalacia, tuberculosis, traumatic injury, and malignancies [2]. However, this established method is not appropriate for the repair of extensive lesions. Therefore, for these types of lesions other replacement methods, such as autografts [2,3], allografts [4,5], prosthetics materials [6][7][8] and tissue engineering [9][10][11][12] have been studied and developed. Some partial successes have been announced with some of these methods, butso far not one of them has archived a completely satisfactory long-term tracheal replacement for extensive lesions of the trachea.
Tissue engineering is an interdisciplinary field that applies the principles of engineering and life science toward the development of biological substitutes that restore, maintain, orimprove tissue function [13]. The last decade of the 20 th century saw an influx of newresearch concepts in terms of tissue engineering, which has had a revolutionary impact on the research of tracheal prosthesis [14]. In 1994 two of the first reports on the effort to engineer a trachea were made by Osada and colleagues [9] and Vacanti and colleagues [10]. With different approaches, both groups were trying to recreate the cylindrical shape of the trachea, combining material and cells to make a more biocompatible tracheal replacement. Our group believes that the requirements for an ideal TET are: (1) easy,less-invasive harvest procedure that yields sufficient cells in a relatively short time, (2) abiocompatible and biodegradable scaffolding that avoids inflammatory responses, (3) proper flexibility and stiffness, (4) air tightness, (5) neovascularization, (6) epithelial regeneration and (7) growth on the insert, which is especially important in the treatment of pediatric patients [15].
Scaffolds derived from xenogenic extracellular matrix have been used in numerous tissue-engineering applications [16]. These types of scaffolds are obtained by removing the cellular components of the tissues through different methods. The most effective method for the decellularization of tissues includes a combination of physical, chemical and enzymatic approaches [17]. In an optimal decellularization process, cells have to be completely removed from the tissue while preserving the components of the extracellular matrix [17,18]. Porcine extracellular matrix from different tissues, such as valves [18,19], urinary bladder [16] and small intestinal submucosa SIS [20,21], have yielded positive results when used as a scaffold for tissue repair.
Previously, our group demonstrated the feasibility to engineer a trachea using tissue engineered cartilage and epithelium from different cells sources and with various biomaterials, such as PGA or Pluronic [11,22,23]. The present study is considered to have improved the design and functionality of the tissue engineered trachea by using aporcine decellularized aorta as a scaffold providing structural support, so as to create a tracheal tube with defined C-shape cartilage rings.

Cell seeding and implantation
After trypsinization, suspensions of approximately 100 x 106 cell/ml was obtained. Each suspension was seeded onto 60 x50 x 2 mm matrix of non-woven PGA fiber meshpolymer. The cell-polymer constructs were incubated in-vitro for 1 week. Subsequently,6 pieces of 3x4 cm were cut from a porcine decellularized aorta ( Figure 1A) and four 0.5x3cm pieces were excised from one side in a comb like fashion from each piece ( Figure 1B). A 4 cm silicon tube stent was inserted into each of the decellularized aortas ( Figure 1C) and the spaces were filled with chondrocyte-seeded PGA to create the tracheal rings ( Figure 1D). Three-dimensional cell-polymer-aorta constructs were then implanted into a subcutaneous pocket of 6 nude rats for 4 weeks.

Porcine aorta decellularization
Porcine aortas were decellularized using a unique device for tissue-decellularization ( Figure 2A). The device was developed by Dr. Iwasaki and his colleagues at the WasedaInstitute of Advanced Study, in Tokyo, Japan [19]. The device consists of a microwave irradiation unit and pulsatile circulation system. A porcine aorta with a length of 5 cm and a diameter of 16 mm was connected to a supporting chamber at both ends and incorporated into a pulsatile circulation system. Deoxycholic acid detergent with a concentration of 1wt% was circulated. In parallel, a microwave at the frequency of 2.45GHz was irradiated. In order to prevent degeneration of extracellular matrix by heat generated by the microwaves, a cooling system for the working Regenerative Medicine fluid was equipped in the system. The temperature of the circulating detergent was monitored and regulated below 37°C using a feedback control mechanism throughout the period of decellularization. Porcine aortic tissue was aseptically harvested. The six aorta pieces were treated with the microwaves for a total of 24 hours in the pulsatile pressure and flow environment ( Figure 2B).

Histological examination
Samples of TET and native trachea were fixed in 10% phosphatebuffered formalin,embedded in paraffin and sectioned. Sections were stained with H&E, Safranin-O and anti-Collagen Type II antibody.

Fluorescent immunohistochemistry or Immunofluorescence of collagen Type II antibody
Paraffin sections of TET and native trachea were heated at 60ºC for 10 min and rehydrated with Xylene and decreasing concentrations of Ethanol in water. Subsequently,tissue was permeabilized in 1mg/ml of Pepsin (Sigma, St Louis, MO) in 10mM Tris HCL for 1.5 hours at 37°C and blocked with 10% rabbit serum (Jackson Immunoresearch,West Grove, PA). Incubation with the first antibody, anti-collagen type II (1/50,Chemicon, Temecula, CA), was carried overnight. Control slides were not exposed to the first antibody; instead they were incubated overnight with just the blocking solution. The visualization of the first antibody was made with anti-mouse IgG Texas Red (1/100,Jackson Immunoresearch, West Grove, PA). Cell nucleus was counter stained with 4', 6-Diamidino-2-phenyindole, dilactate, DAPI (Sigma, St Louis, MO). Finally, tissue was dehydrated with different concentrations of Ethanol and Xylene and covered with Xylene based media.

Biochemical assay
Due to the fact that proteoglycans and collagen type II are the main components of thecartilage extracellular matrix (ECM), it is important that the TET and the native tracheal rings have similar amounts of these components. The amount of proteoglycans was calculated measuring the amount of S-GAG, which form chain(s) that are attached through covalent bonding to a core protein to form proteoglycans. Additionally, the amount of collagen type II was calculated measuring the amount of Hydroxyproline, which is a major component of the collagen type II protein.
Biochemical analyses were performed on harvested tissue engineered (n=6) and native trachea cartilage (n=6) to quantify and compare the level of cartilage specific ECM components. First, around 30 mg of each sample was measured and placed to dry out for 24 hours in a Vacuum Desiccator using Drierite (W.A. HAMMOND DRIERITE CO.LTD., Xenia, OH) as a drying agent. Then, samples were weighted again and were digested for 16 hours at 60°C in 1ml of Papain solution, which contained 100 mM sodium phosphate (Na2HPO4), 10 mM sodium EDTA (Na2EDTA), 10 mM cysteine hydrochloride (Sigma, St Louis, MO), 5 mM EDTA, and 125 mg/mL papain (Sigma, StLouis, MO). Finally, the digested samples were stored at -20°C. S-GAG content of the digested samples was quantified with the Blycan Assay kit (Biocolor Ltd., Belfast, North Ireland). Briefly, 1 ml of Blycan Dye Reagent, composed of 1,9-dymethyl-methylene blue in an inorganic buffer, was added to each digested sample. After 30 min the Blycan Dye Reagent bonded to S-GAG. Samples were centrifuged at > 10000 g for 10 min and the supernatant was removed. 1 ml of Blycan Dissociation Reagent, which dissociates the S-GAGdye complex, was added to each sample. Absorbance was measured at 656 nmwith a spectrophotometer. The S-GAG content of each sample was determined by reading the concentration in the standard curve. Hydroxyproline content of the digested samples was determined by the procedure of Stegemann and Stadler [24]. Briefly, the digested samples were hydrolyzed with equal volumes of 6N HCL at 115°C for 16 hours.Chloramine T and p-dimethyl amino-benzaldehyde were added to the hydrolyzed samples; absorbance was measured at 560 nm with a spectrophotometer immediately after addition of the dye. The hydroxyproline content of each sample was determined by reading the concentration in the standard curve. Means and standard deviations   were calculated for each assay. Statistical significance was determined with a Welch's t-test.

Gross morphology
After 4 weeks, the three-dimensional cell-polymer-aorta constructs formed a TET with defined C-shape cartilage rings, which visibly recreated the structure of a normal trachea (Figure 3 A,B). The decellularized aorta appeared intact and well integrated to the de novocartilage, thus the entire construct felt firm and stable to the touch, and seemed to have adequate stiffness (Figure 3 C,D). Finally, the decellularized aorta appeared to provide appropriate support to the tracheal cartilage.

Histology assay
Histological examination of the specimens showed evidence of the formation of mature cartilage in the TET cartilage, and the decellularized aorta connected well with the denovo cartilage. Hematoxyline and Eosin (H&E) staining revealed a resemblance of themicroscopic morphology in TET cartilage to that of normal hyaline cartilage, and showed the decellularized aorta without infiltration of any type of cells. Safranin-O staining demonstrated the present of proteoglycans in the ECM of both TET and native trachealcartilage ( Figure 4A). Finally, immunostaining for Collagen type II antibody in TET and native tracheal cartilage was predominant in both ECMs and also showed the de novocartilage did not invade the decellularized aorta ( Figure 4B).

Discussion
Our group has previously demonstrated the feasibility of engineering a trachea using PGA, a thermo reversible hydrogel (pluronic F127) [22] and a thermo reversible gelation polymer (TGP) [26] alone. These past results showed mature and adequate engineered cartilage in a tracheal shape in an animal model with a T cell deficiency. When we moved on to creating an autologous model in vivo in an immunocompetent host, the chondrocytes seeded on PGA, Pluronic F127 or TGP were sufficient for creating healthy cartilage. However, the tissue did not hold a proper tracheal shape due to the lack of sufficient glycosaminoglycan and collagen content. This was likely due to the development of a strong inflammatory response or perhaps because the gelatinous consistency of Pluronic F127 and TGP is poor at maintaining a proper shape. To attempt to circumvent the effects of inflammation, we experimented with the use of an anti-inflammatory drug during the implantation period, but the results showed no significant differences to those without the use of anti-inflammatory agents. Moreover, the tissue generated in the autologous immunocompetent model did not demonstrate sufficient engineered tracheal cartilage using either, PGA, Pluronioc F127 and TGP. Thus, we hypothesized that using an alternative material that led to a reduced or absent inflammatory response would generate improved tracheal cartilage. This belief guided our group to focus on using aorta because of its flexibility, and similar size and tubular shape to the human trachea. However, as shown by Tsukada and colleagues [26], using an aortic allograft alone for tracheal replacement leads to necrosis with calcifications or connective tissue replacement of the allograft. Furthermore, the allograft showed no signs of angiogenesis or even evidence of tracheal cartilage regeneration. Finally, the 8 cm aortic allografts shortened up to 87.5% when used for tracheal replacement.
Therefore, based on our previous results, our group concluded that a combination of cell and polymer with a decellularized aortic scaffold would be able to prevent the generation of a strong inflammatory response, avoid necrosis, calcification and contraction, improve overall tracheal cartilage regeneration and structure, and enhance the angiogenic response.
Current decellularization protocols that have become standard are based on 0.1% sodiumdodecyle sulfate (SDS) or 3% Triton X-100 (tert-octylphenylpolyoxyethylen). However,this detergent-based methods sacrifice the mechanical integrity of the underlying ECM. It has been reported that ECM damage is due to collagen fibers becoming more loosely packed following decellularization due to the loss of glycosaminoglycans, thus there populated decellularized material does not have the same mechanical strength in vivo [28][29][30].
Our new decellularization method consists of a microwave irradiation unit along with apulsatile circulation system. It treats aortas with microwave irradiation enhancing the immersion of detergent into the thick aortic tissue by its resonance effect on water molecules. This then leads to the cell membranes being dissolved, causing less damage to collagen fibers. As a result, the unique decellularization method used in this study has advantages in preserving mechanical integrity while dissolving cell membranes. A previous study on this decellularization method demonstrated that it allows the preservation of the ultimate strength and elastic modulus of decellularized aortic valve leaflets and aorta, both results comparable to untreated controls [19].
Based on the success of this novel decellularization method, our group conceived the idea of using a decellularized aortic graft as a flexible support structure for our engineered cartilage created from a PGA scaffold. Our results from this study show that we engineered mature rigid cartilage tissue in between the decellularized aortic graft, generating a well-connected TET with no obstructions. Thus, by creating a cell-polymer decellularized aortic graft construct we were able of engineering a construct withenhanced structural support.
Of note, in this initial study in a nude rat model we were unable to reconstitute the pseudostratified columnar, ciliated epithelium native to the trachea. However, our group has demonstrated that wrapping of an injured deepithialized trachea with an external epithelial/ endothelial cell sheet provided the tissue with similar or enhanced benefits as a native luminal epithelial layer [31]. Thus, when moving to an autologous immunocompetent model, we plan to wrap the engineered tissue and decellularized graft construct with an epithelial and/or endothelial cell sheet to provide the construct with a means to control injury and fibrous response, and preserve tissue perfusion, leading to an enhanced overall reparative response.
In conclusion, our study has demonstrated the feasibility of creating a structurally soundtissue engineered tracheal replacement by combining chondrocytes seeded on a PGA scaffold with a decellularized aortic scaffold. By using the aorta as a backbone, our group was able to improve the structural stability of the regenerated cartilage, and generate tissue strikingly similar to native cartilage. The next step is to repeat the study in an autologous immunocompetent animal model, and examine the tissue histologically aswell as biomechanically. The ultimate goal is to optimize our novel technique with the hopes of providing a new therapy for patients with extensive tracheal disease.