Airway reconstruction using decellularized aortic xenografts in a dog model

ABSTRACT

Tracheal reconstruction after extensive resection remains a challenge in thoracic surgery. Aortic allograft has been proposed to be a potential tracheal substitute. However, clinically, its application is limited for the shortage of autologous aortic segment. Whether xenogeneic aortic biosheets can be used as tracheal substitutes remains unknown. In the present study, we investigated the possibility in dog model. The results show that all dogs were survived without airway symptoms at 6 months after tracheal reconstruction with gently decellularized bovine carotid arteries. In the interior of engrafted areas, grafted patch integrated tightly with the residual native tracheal tissues and tracheal defects in the lumen were repaired smoothly without obvious inflammation, granulation, anastomotic leakage, or stenosis. In addition, histological and scanning electron microscopy examination showed that grafted patches were covered with ciliated columnar epithelium similar to epithelium in native trachea, which indicated successfully re-epithelialization of decellularized bovine carotid arteries in dogs. These findings provide preclinical investigation of xenogeneic aortic biosheets in serving as tracheal substitute in a dog model, which proposes that decellularized biosheets of bovine carotid may be a potential material for bioartificial tracheal graft.

Introduction

Tracheal resection is a major treatment for the patients with significant tracheal injury caused by multiple etiologies including trauma, congenital malformation, neoplastic disease, or stenotic inflammatory lesions.^1,2 Although primary end-to-end anastomosis can be achieved if the affected segment does not exceed half the entire tracheal length in adults or one-third in children, the reconstruction of extensive tracheal defects following long-segment resections remains challenging in tracheal surgery because of the structural complexity of trachea.³ In the last few years, multiple creative techniques such as transplantation, tracheal substitutes, and tissue engineering have been developed in attempts to repair such defects. However, success is still limited because of restrictions inherent to them, such as inability to mimic the properties of native tissues, as well as the occurrence of extrusion, immune rejection, incomplete host–tissue integration, and infection.^4–6 Therefore, finding safe and effective tracheal substitute is still an urgent target in tracheal surgery.

The ideal tracheal substitute should be laterally rigid yet longitudinally flexible with an airtight lumen and an internal lining of ciliated respiratory epithelium. In addition, it must also be biocompatible, nontoxic, nonimmunogenic, noncarcinogenic, and resist dislocation, erosion, and stenosis over time.⁷ Recently, aortic allograft has been proposed to be a potential tracheal substitute for the specific advantages such as its similar diameter, solidity, and elasticity.^8,9 Animal studies have shown that the aortic allograft progressively transformed into tracheal tissue comprising respiratory epithelium and cartilage.^10–12 Long-term evaluation has shown that autologous aortic grafts remain functional for periods up to 3 y.¹³ Besides, clinically, aortic allograft reportedly develops respiratory epithelium in patients 1 y after transplantation, as observed in animal models.¹⁴ These studies indicate that aortic segment might be a valuable tracheal substitute. However, this material has major disadvantages including a suitable donor and donor site complications. Whether xenogeneic aortic segment can be used as tracheal substitutes, which may resolve the problem of the pressing requirement or shortage of autologous aortic segment remains unknown.

In this study, we aimed to investigate the possibility of xenogeneic aortic segment as tracheal substitutes. First, bovine aortic segments were decellularized to remove their endothelial cell layer. Second, histological changes of decellularized bovine aortic segments were examined. Then, in dog model, tracheal reconstruction was performed with decellularized bovine carotid arterial graft.

Materials and methods

All procedures for the present experiments were approved by the animal care committee of Shanghai Sixth People’s Hospital. All animal use procedures were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals as described by the US National Institutes of Health, and all animal experiments were designed to minimize both the number of animals used and their suffering.

Vessel harvest and decellularization

Bovine common carotid arteries were obtained from adult cattle in a local abattoir. Animals were killed by exsanguination through a puncture wound in the neck. Within 30 minutes after death, carotid arteries were dissected from a ventral midline neck incision from 15 to 20 cm proximal to the carotid bifurcation to the bifurcation using a careful, sharp dissection technique. Vessels were immediately rinsed in ice-cold Hanks’ balanced salt solution (HBSS, Gibco, Grand Island, NY) to remove any residual blood clots and then dissected into pieces of 5 cm×10 cm and stored in HBSS on ice until returned to the laboratory.

Vessels were decellularized using a complex process recently developed by our team involving hypotonic breakdown, multiple enzymatic digestions and detergent treatments, and mechanical means.¹⁵ Briefly, the harvested bovine carotid arteries were placed in hypotonic and hypertonic Tris (hydroxymethyl) aminomethane hydrochloride (Tris–HCl, Sigma, St. Louis, MO) with 0.02% ethylenediamine tetraacetic acid (EDTA, Gibco, Grand Island, NY) and 100kIU/ml aprotinin (Sigma, St. Louis, MO) at 4°C for 24 hours. Afterward, the carotid arteries were agitated in Tris–HCl with 1% octylphenoxypolyethanol (Triton X-100, Sigma, St. Louis, MO) and 0.02% EDTA at 4°C for 24 hours. These decellularized carotid arteries were washed with Hank’s (Gibco, Grand Island, NY) four times for periods of 15 minutes each. These procedures were followed by incubation with 2 mg/mL ribonuclease A (RNase A, Boehringer, Mannheim, Germany) and 10 mg/mL deoxyribonuclease I (DNase I, Boehringer, Mannheim, Germany) at 37°C for 1–2 hours. The carotid arteries were subsequently placed in Tris–HCl with 1% Triton X-100 at 48°C for 24 hours. Then, the carotid arteries were agitated in Tris–HCl with 1% Triton X-100 and 0.02% EDTA at 4°C for 24 hours. These decellularized carotid arteries were rinsed in Hank’s (pH7.4) at 4°C for 48 hours to remove residual substances. Finally, the decellularized bovine carotid arteries were sealed and irradiated for sterilization. The detailed protocol is presented in Table 1.

Table 1.

Decellularization protocol.

1. Aortic allograft harvest

2. Wash with Hank’s (pH7.4) at 4°C

3. Rinse in hypotonic Tris-NaCl (0.05 mol/L NaCl, 0.02% EDTA, 0.05 mol/L Tris-Cl, 100kIU/ml aprotinin, pH8.0) at 4°C for 24 hours

4. Rinse in hypertonic Tris-NaCl (1.5 mol/L NaCl, 0.02% EDTA, 0.05 mol/L Tris-Cl, 100kIU/ml aprotinin, pH8.0) at 4°C for 24 hours

5. Wash with Hank’s (pH7.4) at 4°C, 15 minutes for 4 times

6. Rinse in Tris- Triton X − 100 (1%Triton X-100, 0.02% EDTA, 0.05 mol/L Tris-Cl, 100kIU/ml aprotinin, pH7.4) at 4°C for 24 hours

7. Wash with Hank’s (pH7.4 without Mg2+, Ca2+) at 4°C

8. Enzyme extraction: Rinse in Tris- Enzyme (10 mg/L DNAase I, 1 mg/L RNAaseA, 3 mmol/L Mg2+, 1 mmol/L Ca2+, 10 mmol/L Tris-Cl) at 37°C for 1–2 hours

9. Repeat step 6

10. Rinse in Hank’s (pH7.4) at 4°C for 48 hours

Preparation of tracheoplasty

Preparation of tracheoplasty was aseptically performed with 10 adult beagle dogs weighing 16–25 kg. Briefly, all dogs fasted for 12 hours before being anesthetized by intramuscular injection of ketamine (10 mg/kg; Shanghai Reagent Company, Shanghai, China), diazepam (1 mg/kg; Shanghai Reagent Company, Shanghai, China), and atropine (0.05 mg/kg; Shanghai Reagent Company, Shanghai, China).

Tracheoplasty

The dog was placed in the supine position. A longitudinal incision was made in the anterior cervical region of the dog. Then, paratracheal tissues were dissociated carefully to fully expose the trachea. Three C-shaped tracheal cartilage rings below the thyroid cartilage were excised, and the posterior tracheal membrane was preserved. Decellularized bovine carotid arterial graft described above was tailored to the same size as the defect with 4–0 Prolene suture and covered with carotid muscles. Then, the incision was closed (Figure 1a,b).

Figure 1.

Transplantation of decellularized bovine carotid arterial graft (a) The dog trachea was exposed and a midline longitudinal tracheotomy was performed. (b) Decellularized bovine carotid arterial graft was implanted into a defect created in the midventral portion of the cervical trachea.

Postoperative care

All dogs received postoperative antibiotics by intramuscular injection of ampicillin (1 g per day) for 7 d, and bronchoscope examinations were performed at 1 and 2 months post operation to determine the healing state of the operation area. The dogs were sacrificed at 6 months after operation, and grafts with adjacent tracheal tissues were obtained for histological and Scanning electron microscopy (SEM) examinations.

Morphological and histologic examination

The harvested specimens were stained with H&E to confirm histology. Briefly, tissue blocks were paraffin embedded and 5 µm thick sections were cut and stained with hematoxylin-eosin. All the results were read simultaneously by two pathologists who were blind to the study.

Scanning electron microscopy (SEM)

For SEM examination, pieces of harvested specimens were fixed by immersion with 2.5% glutaraldehyde in 0.1 mol/L cacodylate buffer, pH 7.4 for 24 hours. All specimens were postfixed with 2% osmium tetroxide in the same buffer for 2 hours. Next, the specimens were dehydrated in graded alcohol and critical point dried over CO₂. Then, the specimens were made conductive by progressive osmium impregnation in a vacuum evaporator. Finally, the specimens were coated with gold-palladium and examined in a Phillips 505 microscope (XL29, 25 kV, Kassel, Germany). All the results were read simultaneously by two pathologists who were blind to the study.

Transmission electron microscopy (TEM)

For TEM examination, pieces of harvested specimens were fixed overnight in a mixture of cold 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH7.2) and 2% paraformaldehyde in 0.1 M phosphate or cacodylate buffer (pH 7.2) and embedded with epoxy resin. The epoxy resin-mixed samples were loaded into capsules and polymerized at 38°C for 12 hours and 60°C for 48 hours. Thin sections were sliced on an ultramicrotome (RMC MT-XL; RMC Products, Tucson, AZ, USA) and collected on a copper grid. Appropriate areas for thin sectioning were cut at 65 nm and stained with saturated 4% uranyl acetate and 4% lead citrate before the examination on a transmission electron microscope (JEM-1400; JEOL, Tokyo, Japan) at 80 kV.

DNA quantification

To determine the DNA concentration, decellularized carotid arterial grafts and bovine native carotid arteries were cut into small cross-sectional pieces of 25 mg each. Total DNA was extracted using the EZ1 DNA tissue kit (Qiagen, Valencia, CA, USA) following the supplier’s instructions. Then, DNA concentration (ng/mL) was determined using a microplate reader (Epoch, BioTek, Luzern, Switzerland).

Total collagen analysis

Total collagen content in each sample was measured using total collagen assay kit (MyBioSource, San Diego, CA, USA) following the supplier’s instructions.

Statistical analysis

All data were expressed as means ± SEM and statistical analysis was performed using GraphPad Prism5. The data were analyzed by Student’s t-test for unpaired comparisons. Differences with p values of less than 0.05 were considered significant.

Results

Characteristics of decellularized carotid arteries

Over 30 bovine common carotid arteries were decellularized. Macroscopic view showed that the decellularized carotid arteries were white, thin flexible sheets (Figure 2a). Histological examination of H&E staining showed that, in the endothelial cell layer, blue-stained nuclei, which were clearly visible in the native carotid artery, were not observed any signs in the decellularization carotid artery. In the smooth muscle cell layer, blue-stained nuclei were visible both in native and decellularization carotid arteries (Figure 2b).

Figure 2.

Decellularized bovine carotid artery (a) Macroscopic view of decellularized bovine carotid artery. (b) Representative images show the H&E staining of native bovine carotid artery (Native) and decellularized bovine carotid artery (Decellularized), yellow arrowheads showed the blue-stained nuclei in the endothelial cell layer. (c) Representative SEM images of native bovine carotid artery (Native) and decellularized bovine carotid artery (Decellularized), scale bar = 10 μm. The covered area of SEM image of the each sample is about 500 μm².

Furthermore, SEM was performed to investigate microstructural changes in the carotid arteries after the process of decellularization. The results showed that the inner surface of decellularization carotid artery was rougher for the complete removal of the endothelial layers compare to native carotid artery in which the inner surface was smooth and endothelial layer was connectively intact. While the basic extracellular matrix (ECM) microstructure remained intact in decellularization carotid artery (Figure 2c). These results indicated that endothelial cells in the carotid arteries of bovine were successfully decellularized.

In addition, the DNA concentration and total collagen level were also investigated to determine the biochemical change of the decellularized carotid arteries. The results showed that no obvious differences were observed in the DNA concentration and total collagen level between decellularized bovine carotid arteries (n = 10) and bovine native carotid arteries (n = 10), respectively (Figure 3a,b). The results indicated that the decellularization process in our study had less change in the DNA concentration and total collagen level of the artery.

Figure 3.

Analysis of DNA concentration and collagen level of bovine carotid artery after the process of decellularization. (a) Statistic shows DNA concentration in native bovine carotid artery (Native) and decellularized bovine carotid artery (Decellularized). (b) Statistic shows relative collagen level in decellularized bovine carotid artery (Decellularized) compared to native bovine carotid artery (Native). Native: n = 10; Decellularized: n = 10. Data are expressed as mean±SEM.

Morphological and histological characteristics of the decellularized carotid arterial graft after implantation

All dogs which were implanted with decellularized carotid arterial graft survived and no dogs displayed airway symptoms during the observation period of 6 months after implantation, specimens were collected from the central parts of the transplants. Macroscopic view showed that trachea was restored well without inflammation, granulation, anastomotic leakage, or stenosis in the operation area (Figure 4a). Interior of engrafted area showed that the graft integrated tightly with the residual native tracheal tissues and artificial defects in the lumen were repaired smoothly (Figure 4b). Histological examination of H&E staining revealed that connective epithelium lining was clearly visible in engrafted area. Newly formed columnar epitheliums were mature epitheliums which characterization was pseudostratified and ciliated. In addition, ciliated epithelium was very similar to that seen in the native trachea (Figure 4c).

Figure 4.

Morphological and histological characteristic of decellularized bovine carotid arterial graft after implantation (a) Macroscopic view of exteriorized recipient trachea engrafted with decellularized bovine carotid arterial graft. (b) Macroscopic view of interior of engrafted areas of decellularized bovine carotid arterial graft. (c) Representative images show the H&E staining of dog native trachea (Native) and decellularized bovine carotid arterial graft (Graft), black arrowheads show the lining of ciliated pseudostratified columnar epithelium. Scale bar = 100 μm.

In addition, the morphology of cilia in the regenerated respiratory epithelium of the grafted patch was evaluated with SEM images. The results revealed that regenerated cilia covered on the reconstructed luminal surface and the length of cilia were very similar to that in normal trachea. In addition, morphology of basal body of multiciliated cells was very similar to that seen in the native trachea (Figure 5a,b). Furthermore, to further confirm the newly formed columnar epitheliums in decellularized carotid arterial graft, we performed TEM examination. The result showed the presence of cilia on the columnar epitheliums in graft. Morphology of cilia was very similar to that seen in the native trachea (Figure 5c). These results indicated that decellularized bovine carotid arterial graft successfully re-epithelialization after implantation.

Figure 5.

Representative image of SEM of patch of decellularized bovine carotid artery after implantation. (a) Representative 2000× SEM image of inner surface of native trachea of dog (Native) and patch of decellularized bovine carotid artery (Patch), scale bar = 50 μm. (b) Representative 20000× SEM image of inner surface of native trachea of dog (Native) and patch of decellularized bovine carotid artery (Patch), scale bar = 5 μm. (c) Representative TEM image of inner surface of native trachea of dog (Native) and patch of decellularized bovine carotid artery (Patch), black arrowheads showed the cilia in the columnar epithelium, scale bar = 2 μm.

Discussion

Clinically, tracheal resection is the preferred treatment to both relieve airway obstruction and cure the disease, which results in the long-term survival of patients.¹⁶ However, tracheal reconstruction is one of the greatest challenges in thoracic surgery when tracheal defect is extensive and direct end-to-end anastomosis is impossible or after this procedure has failed. The lack of ideal tracheal substitutes is the major cause for this challenge. Recent studies show that aortic allografts are valuable tracheal substitutes for their histological transformation to tracheal tissue few months after transplantation and keeping function for periods up to 3 y.¹³ Despite clinically successful grafting of aortic allografts has been reported following replacement of segments of trachea, an autologous tracheal substitute is limited by limited donor site availability and the lack of a dominant vascular pedicle for microsurgical reconstruction. Whether aortic grafts can be used as a reliable and widespread practical xenograft remains unknown. The preclinical research is explored in the present study.

The leading reason for the failure of xenograft is graft-specific host immune rejection response.¹⁷ Decellularization, a process removing the xenogeneic cell components and retaining its ECM is considered an effective method to minimize or eliminate the incidence of the host response related to immune rejection.¹⁸ Endothelial cells, one of the main components of artery, are reportedly important initiators of graft rejection for its expression of class I MHC molecules, character of antigen-presenting cells, and ability to activate resting T cells.^19,20 In addition, the presence of the cell membrane antigens, such as the oligosaccharide α-Gal (Galα1,3-Galβ1–4GlcNAc-R) (Gal epitope) on the surface of the vascular endothelium is the primary cause of rejection of xenogeneic organ transplants.²¹ Accordingly, in our study, the endothelial cells of bovine carotid arteries were decellularized with chemical detergent and enzymatic methods. The results of H&E staining and SEM showed that no obvious inflammation or immune response were observed 6 months after the decellularized bovine aortic arterial grafts were implanted into tracheal defect of beagle dogs. These results indicated that removal of endothelial cells in bovine artery might effectively minimize host response related to immune rejection. Our result further confirmed that vascular endothelium of the graft plays a significant role in host immune responses to xenografts.

Another leading reason for the failure of segmental tracheal defect reconstruction is mucous impaction and tracheostoma dependency which were mainly caused by failure to regenerate inner lining epithelium and ciliated columnar epithelium.^22,23 Studies on tracheal replacement by autogenous aorta show ciliated columnar epithelium coverage of grafted aorta after implantation, which indicates that arterial graft may be a better material for regeneration of ciliated columnar epithelium.^13,24 In line with these studies, our results showed that ciliated columnar epithelium and regenerated cilia on the reconstructed luminal surface. These results indicated that aortic xenografts might also be re-epithelialization after implantation.

Experimental evidence indicates that a decellularized biologic material composed of ECM matrix provide a popular scaffold for functional recipient cells.²⁵ However, the method of decellularization closely affects the biochemical composition, tissue ultrastructure, and mechanical behavior of the remaining ECM scaffold, which is important for the subsequent regeneration and adhesion of host cells on the graft.^15,26 Decellularized aortic graft by sodium dodecylsulfate (SDS) treatment reportedly leads to a complete loss of cellular structures from the three layers of the arterial wall and only conservation of the main component of the ECM. However, that decellularized aortic graft failed to transform into a new tracheal epithelium.²⁷ Study has shown that fibroblasts are important in activating epithelial cell proliferation and migration when co-cultured tracheal epithelial cells with fibroblasts.²⁸ In our study, aortic grafts were decellularized with method including detergent and enzyme extraction which only removed the endothelial cell of the arterial wall and better conserved the majority of arterial wall including fibroblasts, muscle cell layer, and ECM. The results of H&E staining and SEM revealed regenerated epithelium and cilia on the reconstructed luminal surface at 6 months after implantation. Whether re-epithelialization after the implantation of decellularized bovine aortic graft is caused by conserved fibroblasts is needed further investigation.

Conclusion

In conclusion, the aim of the present study is to provide the evidence of tracheal reconstruction using decellularized aortic xenograft in a dog model. The study offers a new strategy for the development of the reproducible and widespread applicable artificial tracheal substitution. However, this study is the first step in a multistep project. Additional studies are required to explore in humans if these methods prove to be successful and reliable.

Funding Statement

This study was supported by Shanghai Municipal Health Commission (201540267)

Abbreviations

ECM

Extracellular matrix

MHC

Major histocompatibility complex

SDS

Sodium dodecylsulfate

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.