TISSUE ENGINEERING: Part A Volume 00, Number 00, 2013 Mary Ann Liebert, Inc. DOI: 10.1089/ten.tea.2013.

0150

Original Article

Successful Creation of Tissue-Engineered Autologous Auricular Cartilage in an Immunocompetent Large Animal Model
1,2 David A. Bichara, MD, Irina Pomerantseva, MD, PhD,2,3 Xing Zhao, MD,1,2 Libin Zhou, DMD,3,4 Katherine M. Kulig, BS,3 Alan Tseng, BS,3 Anya M. Kimura, BA,3 Matthew A. Johnson, BS,1 Joseph P. Vacanti, MD,2,3 Mark A. Randolph, MAS,1,2 and Cathryn A. Sundback, ScD 2,3

Tissue-engineered cartilage has historically been an attractive alternative treatment option for auricular reconstruction. However, the ability to reliably generate autologous auricular neocartilage in an immunocompetent preclinical model should rst be established. The objectives of this study were to demonstrate engineered autologous auricular cartilage in the immunologically aggressive subcutaneous environment of an immunocompetent animal model, and to determine the impact of in vitro culture duration of chondrocyte-seeded constructs on the quality of neocartilage maturation in vivo. Auricular cartilage was harvested from eight adult sheep; chondrocytes were isolated, expanded in vitro, and seeded onto brous collagen scaffolds. Constructs were cultured in vitro for 2, 6, and 12 weeks, and then implanted autologously in sheep and in control nude mice for 6 and 12 weeks. Explanted tissue was stained with hematoxylin and eosin, safranin O, toluidine blue, collagen type II, and elastin. DNA and glycosaminoglycans (GAGs) were quantied. The quality of cartilage engineered in sheep decreased with prolonged in vitro culture time. Superior cartilage formation was demonstrated after 2 weeks of in vitro culture; the neocartilage quality improved with increased implantation time. In nude mice, neocartilage resembled native sheep auricular cartilage regardless of the in vitro culture length, with the exception of elastin expression. The DNA quantication was similar in all engineered and native cartilage ( p > 0.1). All cartilage engineered in sheep had signicantly less GAG than native cartilage ( p < 0.02); signicantly more GAG was observed with increased implantation time ( p < 0.02). In mice, the GAG content was similar to that of native cartilage and became signicantly higher with increased in vitro or in vivo durations ( p < 0.02). Autologous auricular cartilage was successfully engineered in the subcutaneous environment of an ovine model using expanded chondrocytes seeded on a brous collagen scaffold after a 2-week in vitro culture period.

Introduction

limited number of treatment options exist for patients who require reconstruction of the external auricular framework. Depending on the clinical scenario, patients receive either an ear-shaped framework manually carved from their costal cartilage,1 or a porous polyethylene implant (Medpor) available in predetermined shapes and sizes.2 Although acceptable aesthetic outcomes can be attained, costal cartilage harvest donor site morbidity and extrusion of rigid polyethylene frameworks are complications3 that have incited the search of alternate reconstructive techniques.

Autologous cartilage tissue engineering has historically been an attractive alternative option for auricular reconstruction that could improve postoperative outcomes. Considerable success has been accomplished in engineering ear-shaped elastic cartilage utilizing isolated chondrocytes and diverse scaffolds in immunocompromised animal models.412 However, in immunocompetent models, inconsistent results have been reported, hindering the translation of tissue engineering methodologies into clinical practice.13 To date, the only successful clinical report of auricular reconstruction with autologous engineered tissue utilized a two-step implantation approach in which

1 2

Plastic Surgery Research Laboratory, Massachusetts General Hospital, Boston, Massachusetts. Harvard Medical School, Boston, Massachusetts. 3 Department of Surgery, Massachusetts General Hospital, Boston, Massachusetts. 4 Department of Stomatology, the 513rd Hospital of PLA, Lanzhou, Gansu, P.R. China.

2 chondrocytes were injected subcutaneously in the lower abdomen; after a 6-month implantation period, the engineered cartilage was carved into an ear-shaped framework and re-implanted in the nal position.14 Engineering auricular cartilage in immunocompetent models is a function of scaffold material selection, extent of in vitro chondrocyte culture, and subsequent handling of cellseeded constructs before implantation. Currently, auricular neocartilage has not been reliably demonstrated in a subcutaneous environment of an immunocompetent preclinical animal model. Only a limited number of studies have reported the generation of autologous auricular tissue with histological, biochemical, and mechanical properties similar to those of native cartilage.1517 However, auricular cartilage engineered in immunocompetent animal models was difcult to reproduce15,18 or was not homogenous16; scaffolds with suboptimal mechanical properties were utilized that could not be readily translated to form complex 3D cartilage constructs.17 In addition, elastin bersa key feature of auricular cartilagewere only identied in a few instances and reported after 6,17 8,19 and 16 weeks16 in a swine model. Other studies failed to identify elastin bers in tissue generated in an ovine model after 30 weeks.15 In lapine20 and canine21 models, the complete resorption of autologously engineered auricular cartilage, accompanied by severe inammatory response, was described. In other cases, the engineered auricular cartilage was found to be immature and nonhomogeneous containing a brovascular core and prominent vascular and brous tissue ingrowth; focal calcications accompanied by extensive inammatory and foreign body response have also been reported.18,2226 Inammation negatively affects chondrogenesis and contributes to autologous neocartilage resorption.26 Inammatory and immune responses are thought to be initiated by scaffold materials, their degradation products, or unprotected antigen-presenting chondrocyte surfaces.2630 In vitro culture of immature neocartilage has been known to improve the quality of tissue. Such conditions improved the characteristics of ectopic articular neocartilage in a nude mouse model.31,32 In other instances, inammation was reduced, and the quality of tracheal33 and articular neocartilage in lapine34 and ovine35 models was enhanced. However, a severe inammatory response was described to an autologous ear-shaped implant after prolonged in vitro culture in a lapine model.20 Our research group has previously effectively engineered an ear-shaped construct utilizing brous collagen in a nude mouse model.10 Before engineering a human-shaped, adultsized auricle, the objective of this study was to reproducibly demonstrate high-quality, stable engineered autologous auricular neocartilage in the subcutaneous environment of an ovine model. In addition, we investigated the role of in vitro culture duration of chondrocyte-seeded constructs in the generation of auricular neocartilage. Materials and Methods Cartilage harvest, chondrocyte isolation and expansion The Institutional Animal Care and Use Committee of the Massachusetts General Hospital approved all actions outlined next. After induction of endotracheal anesthesia, a fullthickness biopsy of auricular cartilage, free of surrounding

BICHARA ET AL. tissue and perichondrium and measuring *2 5 cm, was harvested under aseptic conditions from adult sheep (n = 8). The incision was closed using absorbable sutures. Analgesia was achieved using buprenorphine (0.3 mg/kg intramuscular) for 72 h postoperatively. The resected tissue was taken to the laboratory for chondrocyte isolation. Any remaining perichondrium was removed sterilely using a periosteal elevator. Cartilage was washed twice in phosphate-buffered saline and nely minced into 1 mm3 pieces using razor blades. The minced tissue was digested for 16 h at 37C in a solution containing Hams F12 medium (Invitrogen, Grand Island, NY), 0.1% collagenase type II (Worthington Biochemical Corp., Lakewood, NJ), 100 U/mL penicillin, and 100 mg/mL streptomycin (Sigma-Aldrich, St. Louis, MO). After the digestion period, any remnants of undigested cartilage tissue were removed using a 100 mm nylon cell strainer. Trypan blue exclusion test determined cell viability and quantity. The chondrocytes were plated at 3 103 cells/cm2 into roller bottles (Corning Life Sciences, Inc., Lowell, MA) and maintained in Hams F12 medium with L-glutamine, supplemented with 10% fetal bovine serum (FBS; SigmaAldrich), 100 U/mL penicillin, 100 mg/mL streptomycin (Sigma-Aldrich), and 0.1 mM nonessential amino acids (Invitrogen). Roller bottles were rotated at 1 rpm on a roller apparatus (Wheaton Scientic Products, Millville, NJ) in an incubator (Forma Scientic, Marietta, OH) at 37C and 5% CO2. Medium was changed twice a week until cells were subconuent. Enzymatic digestion with 0.05% trypsinEDTA was used to collect passage 1 (P1) cells, which were seeded onto scaffolds. Scaffold preparation and cell seeding Sheets of brous bovine collagen measuring 2 mm in thickness were provided by Kensey Nash Corp. (Exton, PA). Ten-mm-diameter scaffolds were created using a dermal biopsy punch (Acuderm, Inc., Ft. Lauderdale, FL). The scaffolds were sterilized with cold ethylene oxide gas before cell seeding. The collected chondrocytes were suspended in cell culture medium at 50 106 cells/mL, and 120 mL of this suspension (6 106 cells) was pipetted onto each side of each scaffold. Scaffolds were ipped upside down every 20 min for 3 h to facilitate homogenous cell distribution. After that, constructs were placed in dynamic in vitro culture. In vitro culture For in vitro culture, each cell-seeded construct was placed in 4 mL of cell culture medium that was additionally supplemented with 50 mg/mL ascorbic acid in a well of a sixwell plate (Corning Life Sciences, Inc.). The plates were rotated at 40 rpm36 on the orbital shaker (RotoMix; Krackeler Scientic, Inc., Albany, NY) inside a 37C and 5% CO2 incubator for 2, 6, or 12 weeks. Cell culture medium was changed twice a week. Five to 7 days before implantation, FBS in the culture medium was replaced with autologous sheep serum. To obtain autologous serum, 100 mL of blood were drawn during ear cartilage harvest and allowed to clot at 37C for 3060 min; serum was decanted into a new centrifuge tube and centrifuged at 5000 rpm for 10 min. The supernatant was collected and stored at - 20C until use. At

SUCCESSFUL CREATION OF TISSUE-ENGINEERED AUTOLOGOUS AURICULAR CARTILAGE the end of in vitro culture period, the constructs were evaluated grossly, photographed, and either further analyzed or implanted autologously in sheep and in control nude mice. In vivo construct implantation After 2, 6, or 12 weeks of in vitro culture, autologous constructs were implanted subcutaneously in the dorsolateral aspect of the neck in sheep (n = 8) and on the back of control nude mice (n = 8) for 6 and 12 weeks. Constructs from 4 sheep, cultured for 2 weeks in vitro, were harvested at 6 weeks (two sheep) or at 12 weeks (two sheep) post implantation. Constructs, cultured in vitro for 6 weeks (two sheep) and 12 weeks (two sheep), were harvested after 6 weeks of implantation. After induction of endotracheal anesthesia, a 10 cm incision was made starting at the condyle of the mandible and extending posteriorly. Individual subcutaneous pockets were created through blunt dissection, and one construct was implanted into each pocket along with a 12-mm diameter titanium (Ti) ring. The Ti rings were used to facilitate identication of the constructs at the time of explantation. Biocompatibility of the Ti rings was tested in immunocompetent BL6/C57 mice, and no inammatory or foreign body reaction to this metal has been found (unpublished data). The incisions were closed using 30 absorbable sutures, and postoperative observations and pain management were performed as described earlier. Control constructs were implanted subcutaneously on the back of nude mice (Cox 7 Laboratories, Massachusetts General Hospital, Boston, MA). Briey, female nude mice were anesthetized with intraperitoneal tribromoethanol (400 mg/ kg), and a 1 cm midline incision was made on the dorsum under sterile conditions. Two subcutaneous pockets were created by blunt dissection, and constructs were implanted. The incision was closed with stainless steel Autoclip staples (Becton Dickenson, Sparks, MD) that were removed after 14 days. For each sheep, n = 12 scaffolds were seeded: n = 3 were analyzed after in vitro culture, n = 6 were implanted back into the sheep, and n = 3 were implanted into nude mice as controls. All constructs were subjected to both histological and biochemical analyses as described next. Histological and immunohistochemical analyses After gross evaluation, all constructs targeted for histological analysis were bisected; one half of each construct was stored in - 80C, and one half was xed in 10% buffered formalin and dehydrated in increasing concentrations of ethanol before embedding in parafn. The embedded tissue was cut into 5 mm sections and stained with hematoxylin and eosin (H&E), toluidine blue, and safranin O following standard protocols. Collagen type I and type II were detected by means of immunohistochemistry. Briey, tissue sections were digested with 1 mg/mL pepsin (Sigma-Aldrich) in TrisHCl (pH 2.0) for 15 min at room temperature, and peroxidase block was applied for 5 min. The sections were incubated with antibodies against collagen type I (1:100; Accurate Chemical & Scientic Corp., Westbury, NY) and type II (1:100; Developmental Studies Hybridoma Bank, Iowa City, IA) for 1 h at room temperature. The secondary antibody from the EnVision + System Peroxidase kit (Dako,

Carpinteria, CA) was applied for 30 min. For antigen visualization, 3,3-diaminobenzidine was used and cell nuclei were counterstained with hematoxylin. Detection of elastin was performed using a Verhoeffs elastic stain kit (American Master Tech Scientic, Lodi, CA) by following the manufacturers instructions. Quantitative DNA and extracellular matrix analyses Frozen samples were weighed, minced, and digested with 10% proteinase K (Sigma-Aldrich) at 56C overnight; the DNA was extracted and puried with a QiagenDNeasy kit (Qiagen, Inc., Valencia, CA) according to the manufacturers instructions. Total DNA content was determined using a PicoGreen dsDNA assay and expressed in Zg/mg wet weight. To determine the glycosaminoglycan (GAG) content, engineered constructs and native sheep ear cartilage specimens were minced and lyophilized for 24 h. The dehydrated specimens were weighed and digested with papain solution (125 mg/mL papain type III, 100 mM disodium phosphate, 10 mM L-cysteine, and 10 mM EDTA, pH 6.3) at 60C for 16 h. GAG content was measured spectrophotometrically using dimethylmethylene blue dye from the Blyscan Glycosaminoglycan Assay kit (Biocolor, Ltd., Carrickfergus, United Kingdom) with chondroitin sulfate as a standard and expressed in mg/mg dry weight. At least n = 3 samples were analyzed per experimental condition, and all assays were performed in duplicate as per the manufacturers instructions. Statistical analysis Values were expressed as mean standard deviation. Statistical analyses were performed using Students t-test; p < 0.05 was considered signicant. Results Chondrocyte yield and expansion An average of 10.6 4.2 106 cells was isolated per gram of harvested sheep auricular cartilage, and the cell viability was 96.7% 0.5% after cartilage digestion. Despite uniform cell attachment after plating, proliferation of cells in roller bottles was uneven with areas of dense and sparse cell growth. Cell morphology changed from rounded, polygonal to more elongated during proliferation. The average cell expansion factor was 6.4 2.5 after 14.6 0.9 days when the cells were harvested from the roller bottles and seeded onto the scaffolds. In vitro culture The results of the gross and histological evaluation after in vitro culture are presented in Figure 1. Grossly, the constructs appeared translucent after 2 weeks in vitro, while their circular shape and size were mostly preserved; the constructs were soft. After 6 weeks in vitro, the constructs were more opaque, their shape was slightly elongated, and minor warping occurred; some shrinkage was also noted. The constructs became more rm. By 12 weeks, the constructs had glossy, white surfaces and warping increased; the constructs were rm, slippery, exible, and grossly

BICHARA ET AL.

FIG. 1. Gross and histological images of constructs after 2, 6, and 12 weeks in vitro culture. Gradual maturation of engineered cartilage is demonstrated by gross appearance and increasing staining intensity for cartilage-specic glycosaminoglycans (GAG) (safranin O [red] and toluidine blue [blue]) and collagen type II (brown). Scaffold bers stained red with hematoxylin and eosin (H&E). Scale bars: 500 mm. Color images available online at www.liebertpub.com/tea

resembled native cartilage. After 2 weeks in vitro, cells were evenly distributed throughout the constructs; safranin O and toluidine blue staining were negative, indicating that no cartilage-specic GAG was present in the extracellular matrix (ECM). No collagen type II could be identied. At 6 weeks in vitro, cartilage matrix was forming as indicated by weak positive safranin O, toluidine blue, and collagen II staining. At 12 weeks in vitro, neocartilage matrix staining became more intense, chondrocytes were rounded, and typical chondrocyte lacunae were forming. No elastin was detected at any time point after in vitro culture (not shown). Quantitative analysis revealed that the DNA content was similar to that of native cartilage, after 2 weeks in vitro (Fig.

2A).The DNA content slightly increased with prolonged in vitro culture, but the differences were not statistically signicant at all time points ( p > 0.1). The GAG content (Fig. 2B) was low after 2 weeks (6.6 1.0 mg/mg dry weight) but signicantly increased to 15.3 2.5 and 21.8 0.4 mg/mg dry weight after 6 and 12 weeks in vitro, respectively ( p < 0.02). However, the GAG content remained signicantly lower than that of native cartilage during 12 weeks in vitro ( p < 0.02). In vivo construct implantation After implantation, no construct extrusion, edema, or erythema of surrounding tissue was present in both sheep and control nude mice. All explants had white glistening surfaces, were exible grossly, and resembled native cartilage regardless of the experimental conditions. In vivo histological and immunohistochemical analyses Histologically, homogenous neocartilage was found throughout the cross sections of the constructs explanted from both sheep (Fig. 3) and control nude mice (Fig. 4). Chondrocytes within the small ovoid-shaped lacunae were evenly distributed throughout the entire scaffold, and neocartilage matrix stained intensely with hematoxylin in all experimental conditions at all time points, similar to that in native sheep auricular cartilage. Scaffold collagen bers stained intensely with eosin and were seen throughout the contiguous neocartilage ECM in all experimental conditions for all time points. In sheep samples, mild macrophage inltration was seen on the periphery of engineered cartilage (not shown). Neocartilage formation was conrmed by staining with toluidine blue and safranin O, identifying deposition of negatively charged sulfated GAGs characteristic of cartilage ECM. In sheep, the most intense and homogeneous staining identifying the GAGs was observed in samples cultured in vitro for 2 weeks and implanted for 6 and 12 weeks (Fig. 3). The staining intensity and pattern in these samples resembled those of native sheep auricular cartilage. In samples cultured in vitro for 6 and 12 weeks and implanted for 6

FIG. 2. The DNA and GAG contents of neocartilage engineered in vitro. (A) No difference in the DNA content was observed among all experimental groups ( p > 0.1). (B) The GAG content signicantly increased at each time point during in vitro culture ( p < 0.02); however, it remained signicantly lower than in native sheep auricular cartilage ( p < 0.02). An * indicates statistical signicance.

SUCCESSFUL CREATION OF TISSUE-ENGINEERED AUTOLOGOUS AURICULAR CARTILAGE

FIG. 3. Histological images of native sheep auricular cartilage and neocartilage engineered in sheep at 6 and 12 weeks after various in vitro culture periods. Neocartilage was demonstrated in all experimental conditions by positive staining for cartilage-specic GAGs and collagen type II. Neocartilage cultured for 2 weeks and implanted for 6 and 12 weeks closely resembled native cartilage. Elastin bers started appearing after 12 weeks in vivo in samples cultured for 2 weeks before implantation. Collagen scaffold bers stained red on H&E sections. Scale bars: 200 mm, except 50 mm for elastin. Color images available online at www.liebertpub.com/tea weeks, the staining intensity was less homogeneous. Cartilage-specic collagen type II immunohistochemical staining was positive in all experimental conditions; staining intensity and pattern were similar to those described for toluidine blue and safranin O staining. Collagen type I was seen in the surrounding tissue but not within the engineered cartilage in all samples, similar to native sheep cartilage; in samples cultured in vitro for 6 and 12 weeks and implanted for 6 weeks, some positive collagen type I staining could be observed within the constructs (not shown). Elastin staining was negative in all experimental samples except the ones cultured in vitro for 2 weeks and implanted in sheep for 12 weeks. In these samples, elastin bers stained with less intensity than in native sheep cartilage; elastic bers were not as well organized and their distribution within the constructs was uneven. In mice, neocartilage was similar to native sheep auricular cartilage, based on the safranin O and toluidine blue staining, regardless of the experimental condition (Fig. 4). However, staining for collagen type II appeared to be more intense in samples cultured for 2 weeks in vitro and implanted for 12 weeks than in samples with longer in vitro culture durations or shorter implantation. Elastin bers were identied histologically only in samples cultured for 2 weeks in vitro. After 6 weeks in vivo, the distribution of elastic bers was uneven, and staining was not intense; after 12 weeks in vivo, elastin expression resembled native cartilage. Quantitative DNA and ECM analyses The DNA content of all experimental sheep samples was not signicantly different from that of native sheep ear cartilage (79.9 18.4 Zg/mg wet weight) (Fig. 5A, p > 0.1) and no statistically signicant difference was found in the DNA content among the experimental groups. Similar ndings were observed in the corresponding control nude mice groups. The GAG content was signicantly lower (Fig. 5B, p < 0.02) in 6 weeks in vivo sheep samples cultured in vitro for 2, 6, and 12 weeks (38.6 7.0, 31.1 6.6, and 24.6 7.0 mg/mg dry weight, respectively) as compared with the corresponding control mouse samples (91.4 9.4, 99.6 12.3, and 123.8 13.2 mg/mg dry weight, respectively) and with native cartilage (88.4 14.6 mg/mg). In sheep, increasing in vivo implantation time to 12 weeks resulted in signicantly higher GAG content (53.1 8.4 mg/mg) as compared with all sheep experimental groups with the 6-week in vivo implantation

BICHARA ET AL.

FIG. 4. Histological images of neocartilage engineered in control nude mice at 6 and 12 weeks after various in vitro culture periods. Neocartilage was demonstrated in all experimental conditions by positive staining for cartilage-specic GAGs and collagen type II. Elastin bers started appearing after 6 weeks of implantation only in samples cultured for 2 weeks before implantation. After 12 weeks in vivo, elastin expression resembled native cartilage. Collagen scaffold bers stained red on H&E sections. Scale bars: 200 mm, except 50 mm for elastin. Color images available online at www.liebertpub.com/tea

FIG. 5. The DNA and GAG content of neocartilage engineered in vitro, in sheep, and in control nude mice. (A) No difference in the DNA content was observed among all experimental groups and in comparison with native sheep cartilage ( p > 0.1). (B) The GAG content in sheep samples was signicantly lower than in native cartilage in all experimental groups ( p < 0.02). No difference in GAG content was found among the experimental groups except after 12 weeks of implantation in sheep, where more GAG was quantied when compared with other experimental groups ( p < 0.02). In control mice, the GAG content was similar to that of native sheep cartilage and became signicantly higher with increased in vitro or in vivo times ( p < 0.02). An * indicates statistical signicance. Color images available online at www.liebertpub.com/tea

SUCCESSFUL CREATION OF TISSUE-ENGINEERED AUTOLOGOUS AURICULAR CARTILAGE times, regardless of the length of the in vitro culture period ( p < 0.02). In control mice, the GAG content also increased from 91.4 9.4 to 121.0 15.9 mg/mg with increased implantation time; however, the difference was not statistically signicant ( p < 0.1). In sheep, the GAG content remained signicantly lower than native after 12 weeks in vivo ( p < 0.02) but in control mice, it exceeded the native value ( p < 0.02). Within the group of samples implanted into sheep for 6 weeks, no statistically signicant differences were found between the GAG content of samples cultured for 2 and 6 weeks ( p > 0.1); increasing in vitro time to 12 weeks resulted in lower GAG content ( p < 0.02) as compared with samples cultured for 2 weeks. However, in control mice, increasing in vitro time to 12 weeks resulted in higher GAG content (123.8 13.2 mg/mg) as compared with samples cultured for 2 weeks ( p < 0.05) after implantation for 6 weeks. Discussion Several limitations exist that should be addressed in order to advance auricular cartilage tissue-engineering methodologies from the benchtop to the bedside. A critical issue is the engineering of stable auricular cartilage in immunologically aggressive immunocompetent environments. In this study, we demonstrated engineered autologous auricular cartilage in the subcutaneous environment of an ovine model. The ovine craniofacial anatomy and dermal thickness were found to more closely match those of humans, and was therefore chosen as a model for this study. Sheep auricular chondrocytes were expanded in culture, seeded onto brous collagen disk-shaped scaffolds, and cultured in vitro for 2, 6, or 12 weeks before autologous subcutaneous implantation in sheep. Fibrous collagen was utilized, as we had previously observed high-quality neocartilage formation in this scaffold material in nude mice.10 In addition, the dimensions and shape of human ear-shaped cartilage constructs with brous collagen scaffolds were precisely maintained during neocartilage formation and maturation in immunocompromised mice. We utilized cellular approaches that could be readily translated into clinical practice if successful. For example, in vitro culture was performed without additional growth factors other than those present in FBS and autologous serum; the use of growth factors, such as transforming growth factor, insulin-like growth factor, and broblast growth factor, has been studied in detail and is known to signicantly improve the engineered cartilage qualities.31,3739 However, the biologic and regulatory implications are currently unclear for cytokine supplementation in engineered constructs for clinical use. In this study, we engineered auricular cartilage from expanded auricular chondrocytes. Freshly isolated chondrocytes harvested from an unrestricted amount of donor cartilage of experimental animals are most often used in autologous auricular cartilage engineering.1517,22,23,25,26 However, in patients, the cartilage biopsy size is limited, and, therefore, expansion of harvested chondrocytes is necessary to obtain a sufcient cell population to engineer a replacement auricle. Chondrocytes are known to dedifferentiate, losing their cartilage-forming ability during extensive traditional two-dimensional in vitro proliferation.40,41 We successfully expanded chondrocytes without the loss of chondrogenic potential over a sixfold expansion. This ex-

pansion factor is insufcient for clinical needs; in vitro expansion conditions need to be further optimized. The subcutaneous environment of immunocompetent animals is characterized by active inammatory and immunological responses. Inammatory cytokines such as interleukin 1 (IL-1) have been implicated in the inhibition of chondrogenesis both in vitro4244 and in autologous engineered implant resorption.26 It has been suggested that implantation of more mature neocartilage may provide superior engineered cartilage survival because of its resistance to IL-1 exposure.45 A more mature neocartilage matrix may dampen the inammatory and immune responses that can be generated by scaffold materials, their degradation products, or by unprotected antigen-presenting chondrocyte surfaces.2630,46 In our experiments, neocartilage quality was assessed as a function of in vitro culture time after 6 weeks of implantation in sheep and control nude mice. Our studies determined that cartilage with superior morphological and biochemical characteristics had developed at 6 weeks of implantation in sheep after 2 weeks of in vitro culture, as compared with more prolonged in vitro periods (Figs. 3 and 5). Morphologically, neocartilage engineered in sheep after 2 weeks of in vitro culture looked similar to native sheep auricular cartilage. In vitro culture period increases to 6 and 12 weeks did not improve the properties of engineered cartilage in our immunocompetent sheep model. In samples cultured for 6 and 12 weeks in vitro and implanted for 6 weeks, the cartilage-specic safranin O, toluidine blue, and collagen type II staining was less intense and less homogeneous especially after 12 weeks of in vitro culture. Shorter in vitro culture times (03 days) were tested during our initial studies and resulted in a signicant inammatory response and absence of neocartilage formation in sheep (unpublished data). In the control nude mice, all samples resembled native sheep cartilage and morphological differences were not apparent in neocartilage engineered at 6 weeks in mice after 2, 6, or 12 weeks in vitro. Elastin ber formation presented an exception and was only noted in samples cultured in vitro for 2 weeks, but not for 6 or 12 weeks (Fig. 4). The GAG content of neocartilage implanted in sheep for 6 weeks was low as compared with native sheep auricular cartilage and remained low regardless of in vitro culture time (Fig. 5B). Signicant GAG content increase after 12 weeks of in vitro culture (Fig. 2B) was not accompanied by a further increase after implantation in sheep (21.8 0.04 and 24.6 7.0 mg/mg GAG after 12 weeks in vitro and after additional 6 weeks in vivo respectively, p > 0.1). At 6 weeks of in vivo implantation in control mice, the GAG content approximated or exceeded native sheep auricular cartilage values regardless of the in vitro culture period length and increased from 91.8 9.4 mg/mg after a 2 week in vitro culture to 123.8 13.2 mg/mg after 12 week in vitro ( p < 0.05). In mice, the GAG content approximated or exceeded native sheep auricular cartilage values in all engineered cartilage. This nding corroborates our earlier studies involving auricular cartilage engineering in a nude mouse model.10,20 Several groups investigated the impact of in vitro culture before implantation on the engineered articular cartilage properties in an ectopic nude mouse model31,32 and in immunocompetent species, such as rabbits47 and goats,34,35 as

8 well as tracheal cartilage in a rabbit model.33 All researchers acknowledged the benets of in vitro culture before implantation for improvement of neocartilage properties and alleviation of post-implantation inammation; however, the ideal in vitro culture time greatly varied from 1 to 4 weeks. These diverse results suggest that optimal times and culture conditions depend on scaffold material, implantation site, and chondrocyte differentiation state.29,30 Longer in vitro culture and more mature cartilage before implantation do not necessarily lead to better neocartilage in vivo.32,35 Although not clearly understood, prolonged in vitro culture appears to have a negative effect on chondrocyte viability and phenotype stability, which is not reversed on implantation. To the best of our knowledge, no study investigated the role of in vitro culture in generating engineered auricular cartilage in the subcutaneous environment of an immunocompetent model. Our study suggests that prolonged in vitro culture negatively affects neocartilage production, as evident from the results of neocartilage matrix staining and determination of GAG content in constructs cultured for prolonged periods of time and implanted in sheep (Figs. 3 and 5B). It was suggested that prolonged in vitro culture negatively affected cell viability and activity, causing matrix degeneration that was unable to be rescued in vivo.32 In our study, superior cartilage was produced in sheep after 2 weeks in vitro culture; engineered auricular cartilage was stable and signicant improvement in neocartilage quality was demonstrated with increasing implantation time from 6 to 12 weeks. Neocartilage was morphologically similar to native auricular cartilage; the GAG content signicantly increased, and elastin expression was observed after 12 weeks of implantation in sheep. To our knowledge, this is the only demonstration of elastin expression in engineered auricular cartilage in an ovine model. This trend was not observed in nude mice because of the forgiving immunocompromised environment. Superior GAG production was demonstrated in mice as compared with sheep. However, the quality of cartilage engineered in sheep signicantly improved with prolonged implantation time. These encouraging results warrant further studies to determine the impact of longer implantation times on autologous neocartilage stability in a subcutaneous environment of an immunocompetent animal. Conclusions and Future Directions Auricular cartilage was successfully engineered autologously in the subcutaneous environment of an ovine model using moderately expanded (P1) chondrocytes and a brous collagen scaffold after a 2-week period of in vitro culture. Prolonged in vitro culture resulted in suboptimal cartilage quality after implantation in sheep. Cartilage engineered in sheep that was cultured in vitro for 2 weeks before implantation had superior qualities and signicantly improved with prolonged implantation time. Utilizing this approach, engineering of human ear-shaped and sized auricular cartilage is under way in a subcutaneous sheep model. This work is critical to the eld of auricular cartilage engineering, as it represents an essential step toward advancement of replacement engineered auricular cartilage into clinical practice. Acknowledgments

BICHARA ET AL.

This research was sponsored by the Armed Forces Institute of Regenerative Medicine award number W81XWH08-2-0034. The U.S. Army Medical Research Acquisition Activity, 820 Chandler St., Fort Detrick MD 21702-5014 is the awarding and administering acquisition ofce. The content of this article does not necessarily reect the position or the policy of the Government, and no ofcial endorsement should be inferred. Libin Zhou was sponsored by the China Scholarship Council for his study at the Massachusetts General Hospital. The brous collagen scaffolds were generously provided by the Kensey Nash Corporation. Disclosure Statement No competing nancial interests exist. References
1. Walton, R.L., and Beahm, E.K. Auricular reconstruction for microtia: Part II. Surgical techniques. Plast Reconstr Surg 110, 234 quiz 501, 387, 2002. 2. Braun, T., Gratza, S., Becker, S., Schwentner, I., Stelter, K., Patscheider, M., et al. Auricular reconstruction with porous polyethylene frameworks: outcome and patient benet in 65 children and adults. Plast Reconstr Surg 126, 1201, 2010. 3. Bauer, B.S. Reconstruction of microtia. Plast Reconstr Surg 124, 14e, 2009. 4. Xu, J.W., Johnson, T.S., Motarjem, P.M., Peretti, G.M., Randolph, M.A., and Yaremchuk, M.J. Tissue-engineered exible ear-shaped cartilage. Plast Reconstr Surg 115, 1633, 2005. 5. Isogai, N., Morotomi, T., Hayakawa, S., Munakata, H., Tabata, Y., Ikada, Y., et al. Combined chondrocyte-copolymer implantation with slow release of basic broblast growth factor for tissue engineering an auricular cartilage construct. J Biomed Mater Res A 74, 408, 2005. 6. Haisch, A., Klaring, S., Groger, A., Gebert, C., and Sittinger, M. A tissue-engineering model for the manufacture of auricular-shaped cartilage implants. Eur Arch Otorhinolaryngol 259, 316, 2002. 7. Isogai, N., Asamura, S., Higashi, T., Ikada, Y., Morita, S., Hillyer, J., et al. Tissue engineering of an auricular cartilage model utilizing cultured chondrocyte-poly(L-lactide-epsilon-caprolactone) scaffolds. Tissue Eng 10, 673, 2004. 8. Isogai, N., Nakagawa, Y., Suzuki, K., Yamada, R., Asamura, S., Hayakawa, S., et al. Cytokine-rich autologous serum system for cartilaginous tissue engineering. Ann Plast Surg 60, 703, 2008. 9. Kusuhara, H., Isogai, N., Enjo, M., Otani, H., Ikada, Y., Jacquet, R., et al. Tissue engineering a model for the human ear: assessment of size, shape, morphology, and gene expression following seeding of different chondrocytes. Wound Repair Regen 17, 136, 2009. 10. Zhou, L., Pomerantseva, I., Bassett, E.K., Bowley, C.M., Zhao, X., Bichara, D.A., et al. Engineering ear constructs with a composite scaffold to maintain dimensions. Tissue Eng Part A 17, 1573, 2011. 11. Xue, J., Feng, B., Zheng, R., Lu, Y., Zhou, G., Liu, W., et al. Engineering ear-shaped cartilage using electrospun brous membranes of gelatin/polycaprolactone. Biomaterials 34, 2624, 2013. 12. Reiffel, A.J., Kafka, C., Hernandez, K.A., Popa, S., Perez, J.L., Zhou, S., et al. High-delity tissue engineering of patient-

SUCCESSFUL CREATION OF TISSUE-ENGINEERED AUTOLOGOUS AURICULAR CARTILAGE

specic auricles for reconstruction of pediatric microtia and other auricular deformities. PLoS One 8, e56506, 2013. Bichara, D.A., OSullivan, N.A., Pomerantseva, I., Zhao, X., Sundback, C.A., Vacanti, J.P., et al. The tissue-engineered auricle: past, present, and future. Tissue Eng Part B Rev 18, 51, 2012. Yanaga, H., Imai, K., Fujimoto, T., and Yanaga, K. Generating ears from cultured autologous auricular chondrocytes by using two-stage implantation in treatment of microtia. Plast Reconstr Surg 124, 817, 2009. Chang, S.C., Tobias, G., Roy, A.K., Vacanti, C.A., and Bonassar, L.J. Tissue engineering of autologous cartilage for craniofacial reconstruction by injection molding. Plast Reconstr Surg 112, 793, discussion 8001, 2003. Xia, W., Liu, W., Cui, L., Liu, Y., Zhong, W., Liu, D., et al. Tissue engineering of cartilage with the use of chitosangelatin complex scaffolds. J Biomed Mater Res B Appl Biomater 71, 373, 2004. Cao, Y., Rodriguez, A., Vacanti, M., Ibarra, C., Arevalo, C., and Vacanti, C.A. Comparative study of the use of poly(glycolic acid), calcium alginate and pluronics in the engineering of autologous porcine cartilage. J Biomater Sci Polym Ed 9, 475, 1998. Kamil, S.H., Vacanti, M.P., Aminuddin, B.S., Jackson, M.J., Vacanti, C.A., and Eavey, R.D. Tissue engineering of a human sized and shaped auricle using a mold. Laryngoscope 114, 867, 2004. Arevalo-Silva, C.A., Eavey, R.D., Cao, Y., Vacanti, M., Weng, Y., and Vacanti, C.A. Internal support of tissueengineered cartilage. Arch Otolaryngol Head Neck Surg 126, 1448, 2000. Shieh, S.J., Terada, S., and Vacanti, J.P. Tissue engineering auricular reconstruction: in vitro and in vivo studies. Biomaterials 25, 1545, 2004. Kanazawa, S., Fujihara, Y., Sakamoto, T., Asawa, Y., Komura, M., Nagata, S., et al. Tissue responses against tissueengineered cartilage consisting of chondrocytes encapsulated within non-absorbable hydrogel. J Tissue Eng Regen Med 7, 1, 2013. Saim, A.B., Cao, Y., Weng, Y., Chang, C.N., Vacanti, M.A., Vacanti, C.A., et al. Engineering autogenous cartilage in the shape of a helix using an injectable hydrogel scaffold. Laryngoscope 110, 1694, 2000. Monroy, A., Kojima, K., Ghanem, M.A., Paz, A.C., Kamil, S., Vacanti, C.A., et al. Tissue engineered cartilage bioshell protective layer for subcutaneous implants. Int J Pediatr Otorhinolaryngol 71, 547, 2007. Christophel, J.J., Chang, J.S., and Park, S.S. Transplanted tissue-engineered cartilage. Arch Facial Plast Surg 8, 117, 2006. Fussenegger, M., Meinhart, J., Hobling, W., Kullich, W., Funk, S., and Bernatzky, G. Stabilized autologous brinchondrocyte constructs for cartilage repair in vivo. Ann Plast Surg 51, 493, 2003. Rotter, N., Ung, F., Roy, A.K., Vacanti, M., Eavey, R.D., Vacanti, C.A., et al. Role for interleukin 1alpha in the inhibition of chondrogenesis in autologous implants using polyglycolic acid-polylactic acid scaffolds. Tissue Eng 11, 192, 2005. Fujihara, Y., Asawa, Y., Takato, T., and Hoshi, K. Tissue reactions to engineered cartilage based on poly-L-lactic acid scaffolds. Tissue Eng Part A 15, 1565, 2009. Fujihara, Y., Takato, T., and Hoshi, K. Immunological response to tissue-engineered cartilage derived from auricular

13.

29. 30.

14.

31.

15.

32.

16.

33.

17.

34.

18.

35.

19.

36.

20.

37.

21.

38.

22.

39.

23.

40.

24.

25.

41.

42.

26.

43.

27.

28.

44.

chondrocytes and a PLLA scaffold in transgenic mice. Biomaterials 31, 1227, 2010. Haisch, A. Ear reconstruction through tissue engineering. Adv Otorhinolaryngol 68, 108, 2010. Liu, W., and Cao, Y. Application of scaffold materials in tissue reconstruction in immunocompetent mammals: our experience and future requirements. Biomaterials 28, 5078, 2007. Moretti, M., Wendt, D., Dickinson, S.C., Sims, T.J., Hollander, A.P., Kelly, D.J., et al. Effects of in vitro preculture on in vivo development of human engineered cartilage in an ectopic model. Tissue Eng 11, 1421, 2005. Deponti, D., Di Giancamillo, A., Mangiavini, L., Pozzi, A., Fraschini, G., Sosio, C., et al. Fibrin-based model for cartilage regeneration: tissue maturation from in vitro to in vivo. Tissue Eng Part A 18, 1109, 2012. Luo, X., Zhou, G., Liu, W., Zhang, W.J., Cen, L., Cui, L., et al. In vitro precultivation alleviates post-implantation inammation and enhances development of tissue-engineered tubular cartilage. Biomed Mater 4, 025006, 2009. Jin, C.Z., Cho, J.H., Choi, B.H., Wang, L.M., Kim, M.S., Park, S.R., et al. The maturity of tissue-engineered cartilage in vitro affects the repairability for osteochondral defect. Tissue Eng Part A 17, 3057, 2011. Miot, S., Brehm, W., Dickinson, S., Sims, T., Wixmerten, A., Longinotti, C., et al. Inuence of in vitro maturation of engineered cartilage on the outcome of osteochondral repair in a goat model. Eur Cell Mater 23, 222, 2012. Schaefer, D., Martin, I., Jundt, G., Seidel, J., Heberer, M., Grodzinsky, A., et al. Tissue-engineered composites for the repair of large osteochondral defects. Arthritis Rheum 46, 2524, 2002. Blunk, T., Sieminski, A.L., Gooch, K.J., Courter, D.L., Hollander, A.P., Nahir, A.M., et al. Differential effects of growth factors on tissue-engineered cartilage. Tissue Eng 8, 73, 2002. Pei, M., Seidel, J., Vunjak-Novakovic, G., and Freed, L.E. Growth factors for sequential cellular de- and re-differentiation in tissue engineering. Biochem Biophys Res Commun 294, 149, 2002. Morotomi, T., Wada, M., Uehara, M., Enjo, M., and Isogai, N. Effect of local environment, brin, and basic broblast growth factor incorporation on a canine autologous model of bioengineered cartilage tissue. Cells Tissues Organs 196, 398, 2012. Schulze-Tanzil, G., Mobasheri, A., de Souza, P., John, T., and Shakibaei, M. Loss of chondrogenic potential in dedifferentiated chondrocytes correlates with decient Shc-Erk interaction and apoptosis. Osteoarthritis Cartilage 12, 448, 2004. Bobick, B.E., Chen, F.H., Le, A.M., and Tuan, R.S. Regulation of the chondrogenic phenotype in culture. Birth Defects Res C Embryo Today 87, 351, 2009. Beekman, B., Verzijl, N., de Roos, J.A., and TeKoppele, J.M. Matrix degradation by chondrocytes cultured in alginate: IL1 beta induces proteoglycan degradation and proMMP synthesis but does not result in collagen degradation. Osteoarthritis Cartilage 6, 330, 1998. Lima, E.G., Tan, A.R., Tai, T., Bian, L., Stoker, A.M., Ateshian, G.A., et al. Differences in interleukin-1 response between engineered and native cartilage. Tissue Eng Part A 14, 1721, 2008. Scotti, C., Osmokrovic, A., Wolf, F., Miot, S., Peretti, G.M., Barbero, A., et al. Response of human engineered cartilage

10
based on articular or nasal chondrocytes to interleukin-1beta and low oxygen. Tissue Eng Part A 18, 362, 2012. 45. Francioli, S., Cavallo, C., Grigolo, B., Martin, I., and Barbero, A. Engineered cartilage maturation regulates cytokine production and interleukin-1beta response. Clin Orthop Relat Res 469, 2773, 2011. 46. Lotz, A.S., Havla, J.B., Richter, E., Frolich, K., Staudenmaier, R., Hagen, R., et al. Cytotoxic and genotoxic effects of matrices for cartilage tissue engineering. Toxicol Lett 190, 128, 2009. 47. Sterodimas, A., and de Faria, J. Human auricular tissue engineering in an immunocompetent animal model. Aesthet Surg J 33, 283, 2013.

BICHARA ET AL. Address correspondence to: Joseph P. Vacanti, MD Plastic Surgery Research Laboratory Massachusetts General Hospital 55 Fruit St. Warren 1151 Boston, MA 02114 E-mail: jvacanti@partners.org Received: March 15, 2013 Accepted: August 7, 2013 Online Publication Date: October 4, 2013