bs_bs_banner

Copyright © 2014 International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.

Tissue-Engineered Tracheal Reconstruction Using Three-Dimensionally Printed Artificial Tracheal Graft: Preliminary Report *Jae Won Chang, †Su A. Park, *‡Ju-Kyeong Park, ‡Jae Won Choi, *Yoo-Suk Kim, *Yoo Seob Shin, and *‡Chul-Ho Kim Departments of *Otolaryngology and ‡Molecular Science & Technology, School of Medicine, Ajou University, Suwon; †Nature-Inspired Mechanical System Team, Nano Convergence & Manufacturing Systems Research Division, Korea Institute of Machinery and Materials, Daejeon, Korea

Abstract: Three-dimensional printing has come into the spotlight in the realm of tissue engineering. We intended to evaluate the plausibility of 3D-printed (3DP) scaffold coated with mesenchymal stem cells (MSCs) seeded in fibrin for the repair of partial tracheal defects. MSCs from rabbit bone marrow were expanded and cultured. A halfpipe-shaped 3DP polycaprolactone scaffold was coated with the MSCs seeded in fibrin. The half-pipe tracheal graft was implanted on a 10 × 10-mm artificial tracheal defect in four rabbits. Four and eight weeks after the operation, the reconstructed sites were evaluated bronchoscopically, radiologically, histologically, and functionally. None of the four rabbits showed any sign of respiratory distress. Endoscopic examination and computed tomography showed successful reconstruction of trachea without any collapse or blockage. The replaced tracheas were completely

covered with regenerated respiratory mucosa. Histologic analysis showed that the implanted 3DP tracheal grafts were successfully integrated with the adjacent trachea without disruption or granulation tissue formation. Neocartilage formation inside the implanted graft was sufficient to maintain the patency of the reconstructed trachea. Scanning electron microscope examination confirmed the regeneration of the cilia, and beating frequency of regenerated cilia was not different from those of the normal adjacent mucosa. The shape and function of reconstructed trachea using 3DP scaffold coated with MSCs seeded in fibrin were restored successfully without any graft rejection. Key Words: Three-dimensional printing— Tracheal regeneration—Fibrin—Tissue engineering— Mesenchymal stem cell—Ciliary function.

Tracheal reconstruction is a very challenging procedure because the trachea is not a simple cylinder, but a complex, multilayered structure (1,2). The trachea is composed of 15–20 C-shaped cartilages, covered with ciliated epithelium on the inside and connective tissue, including smooth muscles and supporting blood supply, on the outside (3). The cylin-

drical shape is supported by the tracheal cartilages, while mucociliary clearance is the essential role of the respiratory epithelium. The connective tissues surrounding the cartilages allows the trachea to bend, expand, or contract (4). As of yet, there is no method to reconstruct this multilayered trachea and all its functions at once. A promising strategy suggested is the tissue-engineered technique (5–7). A scaffold for tracheal reconstruction should provide not only an appropriate shape and strength for sustaining the luminal contour of the trachea, but also a favorable environment for respiratory epithelium and blood vessel restoration. Natural source-derived scaffolds, such as fibrin, alginate, or xenogenic extracellular matrix (ECM), have the advantage in biocompatibility, biodegradability, and neo-vascularization, but

doi:10.1111/aor.12310 Received December 2013; revised February 2014. Address correspondence and reprint requests to: Profs. Yoo Seob Shin and Chul-Ho Kim, Department of Otolaryngology, Molecular Science & Technology, School of Medicine, Ajou University, 164 Worldcup Street, Yeongtong-Gu, Suwon 442–749, Korea. E-mail: [email protected]; [email protected] Yoo Seob Shin and Chul-Ho Kim contributed equally to the article as corresponding authors.

Artificial Organs 2014, 38(6):E95–E105

E96

J. W. CHANG ET AL.

their downfall is their weak mechanical properties (8–10). Synthetic scaffolds are the other way around. Thus, a necessity for an alternative scaffold, which is not only biocompatible but also strong enough to form a neo-trachea, arises. Among various biodegradable synthetic materials, including polyglycolic acid (PGA), polylactic acid/ polyglycolic acid (PLA/PGA), poly(lactic-co-glycolic acid) (PLGA), polyester urethane, polycaprolactone (PCL) has superior mechanical strength and durability with comparable biocompatibility (11). Additionally, in contrast to most synthetic biodegradable polymers such as PLGA, PLA, or PGA, PCL is water soluble, and thus does not require the use of organic solvents that are incompatible with the current threedimensional printing (3DP) systems. The tailorable properties of PCL due to a very low glass transition temperature and melting point enable PCL to be widely used in 3DP devices (12,13). Furthermore, fibrin, a human-derived, fibrous protein involved in the clotting of blood, possesses favorable features including high biocompatibility and biodegradability, so it is well accepted as a useful cell delivery matrix for cartilage tissue engineering (14). We hypothesized that 3DP scaffold which is composed of both naturally derived fibrin and synthetic PCL could be applicable for tissue-engineered tracheal reconstruction. In this pilot investigation, we first determined the feasibility of employing a fibrin/mesenchymal stem cell (MSC)-coated 3DP PCL scaffold in the reconstruction of a partial tracheal defect in an animal model. We evaluated the endoscopic, radiologic, histologic, and functional characterizations of a hybrid tracheal implant composed of fibrin/MSCs plus 3DP PCL scaffold in tracheal reconstruction. In this study, we report preliminary results regarding the plausibility of 3DP scaffold with MSCs in tracheal reconstruction.

MATERIALS AND METHODS Isolation of MSCs Bone marrow-derived MSCs were isolated from rabbit (New Zealand white rabbits, male, 3 months of age) femur and tibia, as previously described (Fig. 1A) (15,16). Briefly, under anesthesia by tiletamine (4.0 mg/kg) (Virbac Ltd, Carros, France) and zolazepam (4.0 mg/kg) (Virbac), the bone marrow was aspirated from the femur and tibia using an 18-G needle on a heparinized 10 mL syringe. Aspirated material was diluted in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/mL penicillin G (Gibco BRL, Grand Island, NY, USA), and 100 μg/mL streptomycin (Gibco BRL). Nonadherent cells were removed from cultures by washing in phosphate-buffered saline and subsequent medium changes. Culture medium was changed every 3 days, and at approximately 50% confluence, the cells were trypsinized and subcultured for further experiments. The MSCs (Fig. 1B) were passaged twice before the experiments. This study was approved by the Committee for Ethics in Animal Experiments, Ajou University School of Medicine. Preparation of 3D printed tracheal graft 3DP scaffold fabrication was done with PCL (Sigma-Aldrich, St. Louis, MO, USA, Catalog no. 440744), as previously described (13). Briefly, polymer pellets were melted at 100–130°C in a heating cylinder, and 3DP scaffolds were made using the Bioplotter System (EnvisionTEC GmbH, Gladbeck, Germany). PCL was ejected through a heated nozzle, and the strand of PCL was plotted as layerby-layer deposition on a stage. The nozzle size and distance between strands were 200 (nozzle size)/300 (distance between strands) with 27 G. A 3DP halfpipe shape scaffold (10 × 10 mm, diameter 10 mm)

FIG. 1. Isolation and culture of mesenchymal stem cells. (A) Bone marrow was aspirated from rabbit femur and tibia. (B) The cells were then plated at a density of 1.5 × 105 cells/cm2 as a monolayer culture. The mesenchymal stem cells were passaged twice before experiments.

Artif Organs, Vol. 38, No. 6, 2014

TRACHEAL RECONSTRUCTION WITH 3D-PRINTED GRAFT

E97

FIG. 2. Preparation of 3D-printed scaffold with MSCs seeded in fibrin is shown. (A,B) The half-pipe-shaped 3D-printed scaffold made with PCL coated with fibrin containing MSCs. The 3DP scaffold was 10 mm × 10 mm, and diameter was 10 mm. The thickness of scaffold was 2 mm, and the grid size which was made by horizontal and vertical strands of PCL was 0.5 mm × 0.5 mm. (C,D) SEM images of PCL 3DP scaffolds before fibrin coating. PCL strands formed grid 3D structure. (E,F) Pores of PCL scaffold were filled with the fibrin-containing MSCs. (G) PCL strands were coated with MSC-containing fibrin as thin as possible. (H) After seeding of MSCs and fibrin hydrogel mixture. Implanted MSCs were safely settled on the 3DP scaffold.

was plotted layer by layer through the extrusion of melted PCL on a cylinder. The thickness of the PCL scaffold was 2 mm, which is comparable with the thickness of a normal rabbit trachea (17,18), and the grid size which was made by horizontal and vertical strands of PCL was 0.5 × 0.5 mm. The MSCs were pelleted by centrifugation, and then resuspended in a solution containing humanderived fibrinogen (9–18 mg/mL) (Mokam Research Center, Suwon, South Korea). MSCs suspension of 5 × 106 cells/mL was then mixed homogeneously with 110 KIU/mL aprotinin (Mokam Research Center), 60 U/mL thrombin (1000 U/mg protein) (Sigma, St. Louis, MO, USA), 50 U fibrin stabilizing factor XIII, and 50 mM CaCl2. The 3DP scaffolds were coated with the fibrin-containing MSCs. The 3DP scaffolds, both pores and PCL strands, were completely coated with the fibrin-containing MSCs in as thin a layer as possible so as not to narrow the implant inner lumen (Fig. 2A–C). The fibrin/MSC-coated 3DP PCL scaffold was observed by scanning electron microscopy (SEM) to investigate the pores and presence of seeded cells (Fig. 2D). The scaffold was cut in cubes (5 × 3 × 1 mm) and fixed to the sample holder. After platinum coating using a model SC 500 K plasma sputter (Emscope, West Sussex, UK), each sample was examined using a model S-4800 scanning electron microscope (Hitachi, Tokyo, Japan) operating at 10 kV.

Assessment of in vitro mesenchymal stem cells survival The survival of implanted MSCs on the 3DP scaffold was evaluated using LIVE/DEAD Viability/ Cytotoxicity Kit, and calcein acetomethoxy-(AM), ethidium homodimer-1 (EthD-1) (Invitrogen, Carlsbad, CA, USA), as previously described (19,20). Briefly, cultured MSCs implanted on the 3DP scaffold were incubated in 2 μM calcein-AM and 1 μM ethidium bromide homodimer-1 at 37°C for 30 min, then mounted on slides containing FluorSave (EMD Biosciences, Gibbston, NJ, USA). Fluorescence image was captured on a Microphot FXA digital fluorescent microscope (Nikon, Melville, NY, USA). Cell survival was quantitated as amount of living cells (green) divided by the total amount of cells, living and dead (green + red in five randomly selected slides). Measurements and calculations were performed with the Image J program (National Institutes of Health, Bethesda, MD, USA). Animals and surgical procedures The rabbit was placed in a supine position with the neck slightly extended. A vertical skin incision was made and the strap muscles were divided at the midline. After the cervical trachea was appropriately exposed, a 10 × 10-mm-sized, half-pipe-shaped tracheal defect was made with a scalpel and an electrocoagulator (Fig. 3A). The defect was covered Artif Organs, Vol. 38, No. 6, 2014

E98

J. W. CHANG ET AL.

FIG. 3. Insertion of tissue-engineered tracheal implant using 3D-printed scaffold. (A) Half-pipe-shaped tracheal defect, approximately 10 mm wide by 10 mm long, was made on a New Zealand white rabbit. (B) The tracheal implant was inserted into the tracheal defect. It was sutured with Vicryl 5-0 and sealed with Greenplast. (C) A schematic drawing of the technique described in this study.

with the 3DP scaffold containing the MSCs, and it was sutured with Vicryl 5-0 (Ethicon, Inc., Somerville, NJ, USA) and sealed with Greenplast fibrin glue (Greencross, Suwon, Korea) (Fig. 3B). Bronchoscopic and radiologic assessment Bronchoscopic examinations were performed 4 and 8 weeks postoperatively with a 4-mm rigid endoscope (Karl Storz, Tuttlingen, Germany) and a camera (Coolpix 3500, Nikon Co., Tokyo, Japan). Computed tomography (CT) was also performed using a CT scanner system (Brilliance 64, Philips, Eindhoven, The Netherlands) at postoperative 8

weeks. The axial images were obtained with 1-mm cut, and 3D image reconstruction was performed with the axial images. The area of the tracheal lumen on the axial images was measured with Image J program (Fig 4D). Histology and SEM After the radiologic examination, the animals were sacrificed. The trachea including the implantation site was resected and prepared for light microscopic examination. After fixation in 10% neutral buffered formalin for 24 h, the samples were embedded in paraffin and cut into 4 μm-thick sections. After that,

FIG. 4. The surgical specimen and CT findings 8 weeks after the surgery in one representative rabbit. (A) Surgical specimen of the trachea and implant showed successful integration of 3DP tracheal implant without disruption or granulation. (B) Axial CT and (C) 3D reconstructed images of the operated sites (B,C; red arrow) revealed a fine luminal portrayal of the repaired trachea. (D) The area of reconstructed trachea (yellow lines) was measured with Image J program. The average area of the reconstructed trachea on the CT axial images was not significantly decreased compared with adjacent normal trachea (3012.00 ± 85.45 pixels vs. 3052.75 ± 128.55 pixels, P = 0.716). Artif Organs, Vol. 38, No. 6, 2014

TRACHEAL RECONSTRUCTION WITH 3D-PRINTED GRAFT the sections were deparaffinized and rehydrated. Next, the safranin-O (S-O)/fast green was stained for proteoglycan of ECM, picric Sirius red was stained with collagen fiber. Finally, the stained sections were counterstained with Harris hematoxylin and eosin (H&E). Samples were examined under the light microscope. The inner surface morphology of the implant and regenerated was examined using SEM (Hitachi). The samples were fixed with 0.4% glutaraldehyde for 24 h and then coated with gold/palladium for 20 min before SEM observation.

Measurement of tracheal ciliary beat frequency Functional assessment of regenerated respiratory epithelium was measured with tracheal ciliary beat frequency (CBF), as previously described (9,10). Briefly, a piece of regenerated tracheal epithelium (1–2 mm thick) was harvested and washed with DMEM, containing 10 000 U/mL penicillin, 10 000 μg/mL streptomycin, and kept in the medium. The ciliary beating video was captured with an inverted microscope (Axiovert 40CLF; Carl Zeiss, Inc., Thornwood, NY, USA) and high-speed digital camera (Moticam 2000; Motic Ltd, Causeway Bay, Hong Kong). The entirety of the captured images was analyzed with a software (Ammons Engineering, Mt. Morris, MI, USA). Amplitudes and frequencies of the cilia were sampled in at least six separate fields and statistically analyzed. All values were expressed as mean ± standard deviation, and statistical analysis was performed by oneway analysis of variance (SPSS, version 19, Chicago, IL, USA). A P value