Original Article

Engineering Ear Constructs with a Composite Scaffoldto Maintain Dimensions

Libin Zhou, D.M.D.,1,2 Irina Pomerantseva, M.D., Ph.D.,1,3 Erik K. Bassett, M.S.,1 Chris M. Bowley, B.S.,4

Xing Zhao, M.D.,3,5 David A. Bichara, M.D.,3,5 Katherine M. Kulig, B.S.,1 Joseph P. Vacanti, M.D.,1,3

Mark A. Randolph, M.A.S.,3,5 and Cathryn A. Sundback, Sc.D.1,3

Engineered cartilage composed of a patient’s own cells can become a feasible option for auricular reconstruction.However, distortion and shrinkage of ear-shaped constructs during scaffold degradation and neocartilagematuration in vivo have hindered the field. Scaffolds made of synthetic polymers often generate degradationproducts that cause an inflammatory reaction and negatively affect neocartilage formation in vivo. Porouscollagen, a natural material, is a promising candidate; however, it cannot withstand the contractile forces exertedby skin and surrounding tissue during normal wound healing. We hypothesised that a permanent support in theform of a coiled wire embedded into a porous collagen scaffold will maintain the construct’s size and ear-specificshape. Half-sized human adult ear-shaped fibrous collagen scaffolds with and without embedded coiled tita-nium wire were seeded with sheep auricular chondrocytes, cultured in vitro for up to 2 weeks, and implantedsubcutaneously on the backs of nude mice. After 6 weeks, the dimensional changes in all implants with wiresupport were minimal (2.0% in length and 4.1% in width), whereas significant reduction in size occurred in theconstructs without embedded wire (14.4% in length and 16.5% in width). No gross distortion occurred over thein vivo study period. There were no adverse effects on neocartilage formation from the embedded wire. His-tologically, mature neocartilage extracellular matrix was observed throughout all implants. The amount of DNA,glycosaminoglycan, and hydroxyproline in the engineered cartilage were similar to that of native sheep earcartilage. The embedded wire support was essential for avoiding shrinkage of the ear-shaped porous collagenconstructs.

Introduction

Current approaches for complete auricular reconstruc-tion, including carved autologous rib cartilage and al-

loplastic implants, are prone to complications and oftenresult in suboptimal aesthetic outcomes.1 The shortcomingsof available auricular implant options stimulated a search foralternative strategies. The ideal auricular implant wouldhave low extrusion rates and be able to grow, remodel, andwithstand secondary trauma. Tissue-engineered cartilage,derived from autologous cells combined with biodegradablescaffold material, has the potential to meet these require-ments; cartilage engineered from chondrocytes suspended inhydrogels or seeded onto resorbable scaffolds has beendemonstrated in vitro and in vivo.2,3

During the past two decades, attempts have been reportedby our research groups4–7 and others8–13 that have uncov-

ered difficulties related to engineering three-dimensionalhuman ear-shaped cartilage with its complex architectureand largely unsupported, protruding, three-dimensionalstructure. The biggest challenge, however, remains to dem-onstrate specific shape retention of the auricle in longer termin vivo studies. Shape changes inevitably occurred upondegradation of the internal supporting polymer scaffold,6,11

or removal of external stents, which were preserving auricleshape.5,7,9

The success of engineering auricular cartilage largely de-pends on the ability of the scaffold to support cartilage for-mation, withstand contractile healing forces, and degradewithout inducing a deleterious effect on the newly formedtissue. To maintain implant structural integrity in thepresence of immature developing tissue, the supportingscaffold must withstand the aggressive healing forces en-countered upon subcutaneous implantation, especially in an

1Departments of Surgery and Pediatric Surgery, Massachusetts General Hospital, Boston, Massachusetts.2Department of Oral and Maxillofacial Surgery, School of Stomatology, Fourth Military Medical University, Shaanxi, P.R. China.3Harvard Medical School, Boston, Massachusetts.4Kensey Nash Corporation, Exton, Pennsylvania.5Plastic Surgery Research Laboratories, Massachusetts General Hospital, Boston, Massachusetts.

TISSUE ENGINEERING: Part AVolume XX, Number XX, 2011ª Mary Ann Liebert, Inc.DOI: 10.1089/ten.tea.2010.0627

1

immunocompetent animal. The degradation rate of thescaffold material must match the rate of the new tissue for-mation; premature degradation leads to scaffold collapse andloss of implant shape.

For human ear-shaped constructs, scaffolds have beenoften manufactured from a combination of polyglycolic acid,poly-L-lactic acid, and polycaprolactone.2,4–6,8,10,11,13,14 Me-chanical properties and degradation rates of synthetic ma-terials can be modified, and polymers can be combined invarious ratios to meet the mechanical and degradation re-quirements discussed above. Indeed, the ear-shaped con-structs containing polymers with slower degradation rates,such as polycaprolactone, were better preserved at the end ofthe studies because the shape of the auricle was maintainedmostly by the still present scaffold material.6,8,10,11 However,the degradation products of the synthetic materials oftencause chronic inflammation that can negatively affect neo-cartilage formation.15–18

Natural materials, such as collagen, are promising candi-dates for cartilage engineering; being part of extracellularmatrix (ECM), natural materials are abundant and biocom-patible and their use eliminates the negative impact of thedegrading synthetic polymers on neocartilage.19,20 Althoughsignificant immune response can be mounted to collagen-based products, advances in collagen purification and pro-cessing have rendered them biocompatible.21 Scaffolds madeof collagen originating from diverse animal tissues arecommercially available and have been actively used in re-search and clinical applications. Employing collagen scaf-folds, several types of tissues, including meniscal cartilage,have been successfully regenerated22 and osteochondral de-fects have been repaired in patients.23,24

Cartilage formation from sheep auricular chondrocytescultured in vitro on fibrous collagen (type I collagen origi-nating from bovine dermis; Kensey Nash Corporation, Ex-ton, PA) scaffolds using various methods and for varyingtimes before implantation has been evaluated in the concur-rent studies by our laboratories in nude mice and in sheep.Robust neocartilage formation was demonstrated in all studygroups after 6 weeks in vivo. However, fibrous collagen is softand lacks the strength to withstand contraction forces exertedby skin and surrounding tissue during healing. Therefore, apermanent coiled titanium wire framework was embedded

within the collagen scaffold to maintain the size and ear-likeshape of the construct during neocartilage formation, scaffoldremodelling, and exposure to wound healing forces. The re-sults of this proof-of-concept study in immunocompromisedmice are presented in this article.

Materials and Methods

Ear-shaped scaffold design and manufacture

Human ear-shaped scaffolds were fabricated for implan-tation on the back of a mouse. A single half size human adultear master (28.2 mm · 18.4 mm) was carved by hand in clayand used to create polydimethylsiloxane molds. Metalframeworks bent to mimic the shape of the human ear weremade of 0.25–mm-diameter coiled titanium wire (Small Parts,Inc., Logansport, IN). Composite metal and collagen (fibrousbovine dermis-derived type I collagen) ear-shaped scaffoldswere manufactured by Kensey Nash Corporation (Fig. 1).Metal frameworks were embedded in half of the collagenscaffolds; the remaining collagen scaffolds were manufacturedwithout internal wire support. Scaffolds were sterilized withcold ethylene oxide gas before seeding with chondrocytes.

Chondrocyte isolation and culture

Chondrocytes were isolated from auricular cartilage of 11-month-old Polypay sheep. Ear skin, subcutaneous tissues,and perichondrium were removed and discarded. Cartilagewas minced into 1 mm3 fragments and digested with 0.1%collagenase type II (Worthington Biochemical Corporation,Lakewood, NJ) at 37�C for 16 h. Isolated chondrocytes werewashed twice with phosphate-buffered saline; cells werecounted using trypan blue and a hemacytometer and platedinto roller bottles (Corning, Inc., Acton, MA) at 3 · 103 cells/cm2. Chondrocytes were cultured for *10 days in the culturemedium, which consisted of Ham’s F12 medium (Invitrogen,Grand Island, NY) supplemented with 10% FBS (Sigma-Aldrich, St. Louis, MO), 100 U/mL penicillin, 100 mg/mLstreptomycin, 292 mg/mL L-glutamine (Sigma-Aldrich),0.1 mM nonessential amino acids (Invitrogen), and 50mg/mLascorbic acid (Sigma-Aldrich). Upon reaching confluency,the chondrocytes were trypsinized with 0.05% trypsin–ethylenediaminetetraacetic acid and used for this study.

FIG. 1. A half size human adult ear master was carved by hand in clay (A) and used to create polydimethylsiloxane molds(B). Titanium wire frameworks (C) were bent to simulate the ridges of human auricle. Porous human ear-shaped scaffoldswith or without metal frameworks were manufactured from bovine fibrous collagen (D). Color images available online atwww.liebertonline.com/tea.

2 ZHOU ET AL.

Cell seeding and construct culture

Chondrocytes were suspended in the culture mediumat a concentration of 50 · 106 cells/mL. One milliliter of cellsuspension was pipetted onto each scaffold and the cellswere allowed to adhere for 3 h with the scaffolds flippedupside down every 20 min to facilitate more uniform distri-bution of cells. Constructs were cultured in six-well plates in4 mL of the culture medium on the platform of an orbitalRotoMix mixer (Krackeler Scientific, Inc., Albany, NY),which was rotating at 55 rpm25 in standard incubator con-ditions (37�C and 5% CO2) for 2 or 14 days. The culturemedium was changed twice a week.

Construct implantation

All procedures were approved by the Institutional AnimalCare and Use Committee of the Massachusetts GeneralHospital and performed according to the National Institutesof Health Guidelines for the Care and Use of LaboratoryAnimals. Sixteen ear-shaped constructs, eight with wiresupport and eight without, were implanted subcutaneouslyon the backs of 6–8-week-old female athymic nude mice(Cox-7 Laboratories, Massachusetts General Hospital, Bos-ton, MA), one construct per mouse. General anesthesia wasachieved with intraperitoneal injection of 300–500 mg/kgtribromoethanol. Under aseptic conditions, a horizontal in-cision was performed 1.5 cm proximally from the base of thetail, and a subcutaneous pocket was created through bluntdissection. After insertion of the ear-shaped construct, theskin was closed with nonresorbable monofilament suturethat was removed after 7 days. Additionally, in separatemice, four 5–mm-diameter discs were implanted to serve asacellular controls; these discs were punched out of a 2-mm-thick sheet of fibrous collagen identical to the ear-shapedscaffold material (Kensey Nash Corporation).

Gross evaluation and histology

The length and width of all constructs were measuredwith a sterilized digital calliper by three blinded observers atfour time points: before seeding, after in vitro culture on days2 or 14, and after 6 weeks in vivo.

The implants were harvested at 6 weeks and carefully dis-sected from the surrounding mouse tissue. For histologicalevaluation, full-thickness 5-mm-diameter biopsies were pun-ched at three areas of constructs with wire; complete crosssections were obtained at the similar levels from the constructswithout wire. Three full-thickness, 5-mm-diameter biopsies forbiochemical testing were obtained from similar locations inboth types of constructs. Samples for histology were fixed in10% buffered formalin. Specimens for biochemical testing weresnap-frozen and stored at - 80�C until analyzed. To assesscartilage formation within wire coils, wires were carefully re-moved from the fixed tissue before paraffin embedding.

Paraffin-embedded specimens were sectioned at 8 mm.Sections were stained with hematoxylin and eosin; cartilageECM formation was evaluated with safranin O, toluidineblue, and Verhoeff’s elastic stains.

Immunohistochemistry

Tissue sections were pretreated with 1 mg/mL pepsin inTris HCl (pH 2.0) for 15 min at room temperature, followed

by peroxidase block and serum block from M.O.M. kit(Vector Laboratories, Inc., Burlingame, CA). Sections wereincubated with mouse anti-human collagen type I antibody(Accurate Chemical & Scientific Corporation, Westbury, NY)or mouse anti-human collagen type II antibody (Develop-mental Studies Hybridoma Bank, Iowa City, IA) for 30 min.EnVision + System Peroxidase kit (Dako, Carpinteria, CA)was used to identify the antigens; sections were counter-stained with hematoxylin.

Quantitative DNA and ECM analyses

Frozen samples were weighed, minced, and digested with10% proteinase K from tritirachium album (Sigma-Aldrich)at 56�C overnight; the DNA was extracted and purified witha Qiagen DNeasy kit (Qiagen, Inc., Valencia, CA) accordingto the manufacturer’s instructions. Total DNA content wasdetermined using a PicoGreen dsDNA assay.26

For biochemical analysis, engineered constructs and na-tive sheep ear cartilage specimens were minced and ly-ophilized for 24 h. The dehydrated specimens were weighedand digested with papain solution (125 mg/mL papaintype III, 100 mM phosphate, 10 mM l-cysteine, and 10 mMethylenediaminetetraacetic acid, pH 6.3) at 60�C for 16 h.Aliquots of these digests were assayed for glycosami-noglycan (GAG) and hydroxyproline (OH-proline) con-tent. GAG content was measured spectrophotometricallyusing dimethylmethylene blue dye from the Blyscan Gly-cosaminoglycan Assay kit (Biocolor Ltd., Carrickfergus,United Kingdom) with chondroitin sulfate as a standard.27

OH-proline content was measured in the aliquots of thesame papain digests using Stegemann’s hydroxyprolineassay.28 All samples and standards were analyzed induplicate.

Statistical analysis

Construct size and biochemical analyses values are ex-pressed as mean – standard deviation. Statistical analyseswere performed using SPSS 11.0 (SPSS, Chicago, IL). Com-parison of means was assessed by a one-way analysis ofvariance and the Tukey multiple comparison test ( p < 0.05was considered significant).

Results

Gross evaluation and histological analyses

All animals survived until the predetermined endpoints;no extrusion of constructs or wire supports was observedduring the study period. All implants maintained theiroriginal shape and resembled a human ear (Fig. 2).

During the 2-week in vitro culture, the constructs withwire support maintained their size while the constructswithout wire support decreased in both length and width(Fig. 3). There was no considerable change in dimensions ofeither construct type during the 6 weeks in vivo. Constructsize changes during the experiment are presented in Figure4. Significantly less dimensional changes were observed inear-shaped constructs with wire support than in constructswithout wire support ( p < 0.05). Little changes in length andwidth were found in constructs with internal wire support(2.0% length and 4.1% width). Constructs without wiresupport initially swelled at 2 days in vitro but both dimensions

COMPOSITE ENGINEERED EAR SCAFFOLD MAINTAINS DIMENSIONS 3

decreased and remained smaller at the 6-week in vivo timepoint (14.4% length and 16.5% width).

At 6 weeks postimplantation, all constructs were sur-rounded by a thin, fibrous capsule that could be easilyremoved. Grossly, the tissue resembled cartilage, and all ear-shaped constructs, both with and without internal wiresupport, were flexible (Fig. 5).

The morphology of neocartilage was similar in ear-shapedconstructs with and without internal wire support (Fig. 6).The chondrocytes in the newly formed tissue demonstratedsimilar morphologic characteristics to those seen in nativesheep auricular cartilage and were located within evenlydistributed ovoid lacunae. Collagen fibers of the scaffoldwere seen throughout the neocartilage ECM (Fig. 6A, E, I).Similarly, the neocartilage ECM, like native cartilage ECM,stained intensely with safranin O and toluidine blue, indi-cating the presence of abundant sulfated GAG (Fig. 6B, C, F,G, J, K). Weak positive staining for elastin was detected inthe engineered cartilage in both types of constructs at the 6-week time point (Fig. 6D, H).

A composite image of the cross section of the ear-shapedconstruct without wire (Fig. 7A) demonstrates cartilage ECM

formation throughout the construct, as evidenced by safraninO staining. Small areas in the middle of the construct did notstain positively for cartilage ECM; some of those areas ap-peared to have densely packed scaffold fibers and low cel-lularity, some areas contained no scaffold fibers and werefilled with loose connective tissue. In the cross section of theear-shaped construct with wire, neocartilage formation wasobserved within the rings of the titanium coil of the internalwire support (Fig. 7B).

Immunohistochemistry

Cartilage-specific ECM was demonstrated in constructs,with and without wire support, as collagen type II wasexpressed in the ECM surrounding the chondrocyte ovoidlacunae (Fig. 8A, B). Collagen type II staining was moreintense at the periphery and was less intense near the cen-ter of the constructs. Collagen type I was expressed atthe surface of the both construct types (Fig. 8D, E). Thesecollagen staining patterns are similar to that observed innative sheep ear cartilage (Fig. 8C, F). Collagen fibers of thescaffold (bovine collagen type I) did not stain for collagen

FIG. 2. Human ear-shapedconstructs with (A) andwithout (B) internal wiresupport on the backs of nudemice retained characteristicear shape at 6 weeks. Colorimages available online atwww.liebertonline.com/tea.

FIG. 3. Gross images of ear-shaped constructs. (A, E) Be-fore seeding, (B, F) after 2weeks of in vitro preculture,(C, G) after 6 weeks in vivo,and (D, H) comparison of si-zes at different stages. (A–D)Constructs with wires and(E–H) without wires. Theconstruct size was main-tained in scaffolds with wires(D) but was not maintainedin scaffolds without wires (H)after 2 weeks in vitro culture.Color images available onlineat www.liebertonline.com/tea.

4 ZHOU ET AL.

type I, indicating the lack of species cross reactivity for thisantibody.

Quantitative DNA and ECM analyses

The DNA content (Fig. 9) of engineered, ear-shaped con-structs was similar in constructs with and without wiresupport (245.4 – 95.7 and 226.1 – 67.4Zg/mg wet weight re-spectively) and similar to that of native sheep ear cartilage(147.2 – 18.6Zg/mg, p > 0.1). The DNA content of the controlacellular scaffolds was low (79.1 – 9.2Zg/mg, p < 0.05) andcould be attributed to the migration of mouse cells into thescaffold upon implantation. The amount of GAG in theconstructs was similar: 127.4 – 61.5 and 122.1 – 43.6 mg ofGAG/mg dry weight in constructs with and without wiresupport, respectively ( p > 0.1). The GAG content of en-gineered cartilage was similar to that of native sheep earcartilage (133.4 – 18.1 mg/mg, p > 0.1) and no GAG could bedetected in the control acellular scaffolds. The amount ofOH-proline was 89.5 – 12.7 and 93.3 – 19.8 mg/mg dry weightin the constructs with and without wire support, respectively( p > 0.1). OH-proline content of ear-shaped engineered car-tilage, both with and without wire support, was higher thanthat of the native sheep ear cartilage (50.2 – 3.4 mg/mg,p < 0.05) and the acellular scaffold control (60.7 – 3.7 mg/mg,p < 0.05). The higher amount of OH-proline in the engineered

cartilage, as compared to native sheep auricular cartilage,can be attributed in part to the collagen material of scaffold,which has been digested along with the collagen of en-gineered cartilage ECM and possibly contributed to theoverall OH-proline content. In the acellular scaffold control,OH-proline content is attributed mostly to the collagen of thescaffold material. After 6 weeks in vivo, however, no differ-ence was detected between DNA, GAG, and OH-prolinecontent of the constructs cultured in vitro for 2 days versusthose cultured for 2 weeks before implantation (results notshown).

Discussion

Multiple efforts to engineer human ear-shaped cartilagehave been hindered by the inability to retain the size andshape of the construct for the duration of in vivo stud-ies.6,8,9,11,29–31 To preserve the specific shape of a humanauricle, many approaches have been investigated includingreinforcement of a scaffold with an additional syntheticpolymer poly-L-lactic acid,4–6,8,10,12,13,32 use of temporaryexternal stents,5,7,9 acrylic sheet,12 and implantable externalperforated mold.30 In this study, we proposed a new strat-egy: reinforce the ear-shaped porous collagen scaffold withan internal titanium wire skeleton. A coiled titanium wire,bent to simulate the ridges of a human auricle, was embed-

FIG. 4. Construct size chan-ges during the experiment.The length and width of thescaffolds containing wire sup-ports (A) did not change.Scaffolds without wire sup-ports (B) decreased in sizeafter initial swelling at 2 daysand remained smaller after 2weeks in vitro culture and after6 weeks in vivo.

FIG. 5. Demonstration offlexibility of ear-shaped con-structs with internal wiresupport (A, C) and withoutwire support (B, D) after 6weeks in vivo. Color imagesavailable online at www.liebertonline.com/tea.

COMPOSITE ENGINEERED EAR SCAFFOLD MAINTAINS DIMENSIONS 5

ded into porous collagen, thereby combining the advantagesof the biological nature of collagen material and the me-chanical properties of the titanium wire. Titanium has beendemonstrated to be a biocompatible material and is usedroutinely in medical implants for numerous applications,including auricular replacement.33

Our results demonstrate that the size and ear-like shapewere preserved throughout the experiment in all implantswith internal wire support. After the initial swelling, signif-icant reduction in size occurred in constructs made of porouscollagen alone; however, the human ear-like shape of theconstructs was grossly preserved. The reduction in size oc-

curred after 2 weeks of in vitro culture without any furtherreduction during subsequent 6 weeks in vivo. The shrinkageis possibly caused by the beginning of ECM formation. Thisfinding corroborates the assessment of the Kensey Nashmultiphasic composite scaffold for osteochondral defect re-pair; the authors observed slight contraction of the cell-seeded collagen layer after 3 weeks of in vitro culture.34 Thelack of further reduction in size during the in vivo period maybe attributed to rather loose subcutaneous connective tissuein rodents and the reduced inflammatory response in im-munocompromised nude mice as evidenced by the forma-tion of a thin fibrous capsule. In a large animal model,

FIG. 6. Histological appearance of engineered cartilage after 6 weeks in vivo was similar to that of native sheep ear cartilage.(A–D) Constructs with wire supports, (E–H) constructs without wire supports, and (I–L) native sheep ear cartilage. (A, E, I)Hematoxylin and eosin, (B, F, J) safranin O, (C, G, K) toluidine blue, and (D, H, L) Verhoeff’s elastin stain. Residual collagenfibers of the scaffold stained red on hematoxylin and eosin and elastin-stained slides and green on safranin O-stained slides.Scale bar: 100mm. Color images available online at www.liebertonline.com/tea.

FIG. 7. Composite image of the cross section of the ear-shaped construct without wire demonstrates neocartilage formationthroughout the construct after 6 weeks in vivo (A). The small areas in the middle of the construct (arrowhead) did not stainpositively for cartilage extracellular matrix possibly due to scaffold production artifacts such as uneven distribution ofcollagen fibers within the scaffold or air bubbles. Neocartilage formation was observed within the rings of the coils of thetitanium wire (*) in the scaffolds with internal wire support (B). Safranin O staining; scale bars: (A) 1 mm, (B) 200mm. Colorimages available online at www.liebertonline.com/tea.

6 ZHOU ET AL.

stronger contraction forces are expected to be exerted by skinand surrounding tissue during healing, approximating con-ditions in humans.

Our results suggest that the internal wire frameworkwas essential for the preservation of the dimensions of theengineered ear during in vitro culture and after implantationinto the animal model. Titanium wire was well incorporatedinto the neocartilage, without any adverse effects on chon-drocyte viability, adhesion to scaffold material, and cartilageECM formation, suggesting low possibility of extrusion inthe future. Histologically, we found no difference betweenneocartilage that formed in the constructs with and with outinternal wire support. Weak elastin expression was observedin both types of constructs after 6 weeks in vivo, suggestingthat elastic cartilage started to form at this early time point.

We were unable to identify any differences between thecartilage that formed in nude mice after 2 days and 2 weeksof in vitro culture before implantation. However, in a largeanimal model, in vitro preculture is important to achieveautologous cartilage formation before implantation. Such anapproach may help reduce the inflammatory and foreignbody response that can be induced by a scaffold made fromcollagen originating from a different species and by antigen-presenting chondrocyte surface in an immunologically activesubcutaneous environment.35

Staining of the cross sections obtained from the ear-shapedimplants without wires with safranin O demonstrated a fewareas in the center of the constructs that did not show thepresence of cartilage-specific GAG (Fig. 7A). Some of thesestaining defects appeared to have lower cellularity and maybe due to scaffold production artifacts such as uneven dis-tribution of collagen fibers within the scaffold or air bubbles.On the other hand, the thickness of the constructs often ex-ceeded 3 mm, which might have negatively affected chon-drocyte survival in the central part of the constructs due tolimited nutrient and gas diffusion. We are currently modi-fying the design of the ear-shaped scaffold to reduce thethickness of engineered cartilage so that it more closely re-sembles human auricular cartilage; this thickness reductionshould eliminate this central defect if it is related to constructthickness.

In this study, auricular cartilage was engineered and thesize of the human ear-like construct was retained with thehelp of the internal titanium wire framework. The embeddedwire support was essential for avoiding shrinkage of the ear-shaped porous collagen constructs. Engineering human ear-shaped cartilage with preserved dimensions represents animportant milestone in our translational program to developa replacement living auricle for patients with congenital andacquired external ear defects. The improved composite

FIG. 8. Immunohistochemical staining for collagen type II (A–C) and type I (D–F) of engineered cartilage in the constructswith (A, D) and without wire support (B, E), and of native sheep auricular cartilage (C, F). Positive collagen type II staining(brown) was observed throughout the engineered cartilage, similar to native cartilage, whereas positive collagen type Istaining was seen at the surface in both types of constructs and in the perichondrium of native cartilage. Scale bar: 200mm.Color images available online at www.liebertonline.com/tea.

FIG. 9. DNA, glycosaminoglycan (GAG), and hydro-xyproline (OH-proline) content. Data are presented asmean – standard deviation. DNA is presented as Zg/mgconstruct wet weight, and GAG and OH-proline as mg/mgconstruct dry weight.

COMPOSITE ENGINEERED EAR SCAFFOLD MAINTAINS DIMENSIONS 7

scaffold will become a viable asset for future clinical appli-cations and serve as an initial step in the development of afully resorbable ear-shaped scaffold.

Acknowledgments

This research was sponsored by the Armed Forces Instituteof Regenerative Medicine award number W81XWH-08-2-0034. The U.S. Army Medical Research Acquisition Activity,820 Chandler Street, Fort Detrick MD 21702-5014, is theawarding and administering acquisition office. The contentof the article does not necessarily reflect the position orthe policy of the Government, and no official endorsementshould be inferred. Dr. Libin Zhou was sponsored by ChinaScholarship Council for his study in Massachusetts GeneralHospital.

Disclosure Statement

No competing financial interests exist.

References

1. Bauer, B.S. Reconstruction of microtia. Plast Reconstr Surg124, 14e, 2009.

2. Kim, W.S., Vacanti, J.P., Cima, L., Mooney, D., Upton, J.,Puelacher, W.C., and Vacanti, C.A. Cartilage engineered inpredetermined shapes employing cell transplantation onsynthetic biodegradable polymers. Plast Reconstr Surg 94,

233, 1994.3. Chung, C., Mesa, J., Randolph, M.A., Yaremchuk, M., and

Burdick, J.A. Influence of gel properties on neocartilageformation by auricular chondrocytes photoencapsulated inhyaluronic acid networks. J Biomed Mater Res A 77, 518,2006.

4. Vacanti, C.A., Cima, L.G., Ratkowski, D., Upton, J., andVacanti, J.P. Tissue engineered growth of new cartilage inthe shape of a human ear using synthetic polymers seededwith chondrocytes. Mater Res Soc Symp Proc 252, 367, 1992.

5. Cao, Y., Vacanti, J.P., Paige, K.T., Upton, J., and Vacanti,C.A. Transplantation of chondrocytes utilizing a polymer-cell construct to produce tissue-engineered cartilage in theshape of a human ear. Plast Reconstr Surg 100, 297, 1997.

6. Shieh, S.J., Terada, S., and Vacanti, J.P. Tissue engineeringauricular reconstruction: in vitro and in vivo studies. Bio-materials 25, 1545, 2004.

7. Xu, J.W., Johnson, T.S., Motarjem, P.M., Peretti, G.M., Ran-dolph, M.A., and Yaremchuk, M.J. Tissue-engineered flexi-ble ear-shaped cartilage. Plast Reconstr Surg 115, 1633, 2005.

8. Kusuhara, H., Isogai, N., Enjo, M., Otani, H., Ikada, Y.,Jacquet, R., Lowder, E., and Landis, W.J. Tissue engineeringa model for the human ear: assessment of size, shape,morphology, and gene expression following seeding of dif-ferent chondrocytes. Wound Repair Regen 17, 136, 2009.

9. Neumeister, M.W., Wu, T., and Chambers, C. Vascularizedtissue-engineered ears. Plast Reconstr Surg 117, 116, 2006.

10. Isogai, N., Morotomi, T., Hayakawa, S., Munakata, H., Ta-bata, Y., Ikada, Y., and Kamiishi, H. Combined chondrocyte-copolymer implantation with slow release of basic fibroblastgrowth factor for tissue engineering an auricular cartilageconstruct. J Biomed Mater Res A 74, 408, 2005.

11. Isogai, N., Asamura, S., Higashi, T., Ikada, Y., Morita, S.,Hillyer, J., Jacquet, R., and Landis, W.J. Tissue engineering ofan auricular cartilage model utilizing cultured chondrocyte-

poly(L-lactide-epsilon-caprolactone) scaffolds. Tissue Eng10, 673, 2004.

12. Kamil, S.H., Kojima, K., Vacanti, M.P., Bonassar, L.J., Va-canti, C.A., and Eavey, R.D. In vitro tissue engineering togenerate a human-sized auricle and nasal tip. Laryngoscope113, 90, 2003.

13. Haisch, A., Klaring, S., Groger, A., Gebert, C., and Sittinger,M. A tissue-engineering model for the manufacture of au-ricular-shaped cartilage implants. Eur Arch Otorhinolar-yngol 259, 316, 2002.

14. Liu, Y., Zhang, L., Zhou, G., Li, Q., Liu, W., Yu, Z., Luo, X.,Jiang, T., Zhang, W., and Cao, Y. In vitro engineering ofhuman ear-shaped cartilage assisted with CAD/CAMtechnology. Biomaterials 31, 2176, 2010.

15. Fujihara, Y., Takato, T., and Hoshi, K. Immunological re-sponse to tissue-engineered cartilage derived from auricularchondrocytes and a PLLA scaffold in transgenic mice. Bio-materials 31,1227, 2010.

16. Hsu, S.H., Chang, S.H., Yen, H.J., Whu, S.W., Tsai, C.L., andChen, D.C. Evaluation of biodegradable polyesters modifiedby type II collagen and Arg-Gly-Asp as tissue engineeringscaffolding materials for cartilage regeneration. Artif Organs30, 42, 2006.

17. Lotz, A.S., Havla, J.B., Richter, E., Frolich, K., Staudenmaier,R., Hagen, R., and Kleinsasser, N.H. Cytotoxic and geno-toxic effects of matrices for cartilage tissue engineering.Toxicol Lett 190, 128, 2009.

18. Rotter, N., Ung, F., Roy, A.K., Vacanti, M., Eavey, R.D.,Vacanti, C.A., and Bonassar, L.J. Role for interleukin 1alphain the inhibition of chondrogenesis in autologous implantsusing polyglycolic acid-polylactic acid scaffolds. Tissue Eng11, 192, 2005.

19. Parenteau-Bareil, R., Gauvin, R., and Berthod, F. Collagen-based biomaterials for tissue engineering applications. Ma-terials 3, 1863, 2010.

20. Glowacki, J., and Mizuno, S. Collagen scaffolds for tissueengineering. Biopolymers 89, 338, 2008.

21. Lynn, A.K., Yannas, I.V., and Bonfield, W. Antigenicity andimmunogenicity of collagen. J Biomed Mater Res B ApplBiomater 71, 343, 2004.

22. Stone, K.R., Steadman, J.R., Rodkey, W.G., Li ST. Re-generation of meniscal cartilage with use of a collagenscaffold. Analysis of preliminary data. J Bone Joint Surg Am79, 1770, 1997.

23. Steinwachs, M., and Kreuz, P.C. Autologous chondrocyteimplantation in chondral defects of the knee with a type I/IIIcollagen membrane: a prospective study with a 3-year fol-low-up. Arthroscopy 23, 381, 2007.

24. Kon, E., Verdonk, P., Condello, V., Delcogliano, M,,Dhollander, A., Filardo, G., Pignotti, E., and Marcacci, M.Matrix-assisted autologous chondrocyte transplantation forthe repair of cartilage defects of the knee: systematic clinicaldata review and study quality analysis. Am J Sports Med 37,

156S, 2009.25. Vunjak-Novakovic, G., Freed, L.E., Biron, R.J., and Langer,

R. Effects of mixing on the composition and morphologyof tissue-engineered cartilage. J Am Inst Chem Eng 42, 850,1996.

26. Singer, V.L., Jones, L.J., Yue, S.T., and Haugland, R.P.Characterization of PicoGreen reagent and development of afluorescence-based solution assay for double-stranded DNAquantitation. Anal Biochem 249, 228, 1997.

27. Enobakhare, B.O., Bader, D.L., and Lee, D.A. Quantificationof sulfated glycosaminoglycans in chondrocyte/alginate

8 ZHOU ET AL.

cultures, by use of 1,9-dimethylmethylene blue. Anal Bio-chem 243, 189, 1996.

28. Stegemann, H., and Stalder, K. Determination of hydro-xyproline. Clin Chim Acta 18, 267, 1967.

29. Ting, V., Sims, C.D., Brecht, L.E., McCarthy, J.G., Kasabian,A.K., Connelly, P.R., Elisseeff, J., Gittes, G.K., and Longaker,M.T. In vitro prefabrication of human cartilage shapes usingfibrin glue and human chondrocytes. Ann Plast Surg 40, 413,1998.

30. Kamil, S.H., Vacanti, M.P., Aminuddin, B.S., Jackson, M.J.,Vacanti, C.A., and Eavey, R.D. Tissue engineering of a hu-man sized and shaped auricle using a mold. Laryngoscope114, 867, 2004.

31. Chung, C., and Burdick, J.A. Engineering cartilage tissue.Adv Drug Deliv Rev 60, 243, 2008.

32. Kawazoe, N., Inoue, C., Tateishi, T., and Chen, G. A cellleakproof PLGA-collagen hybrid scaffold for cartilage tissueengineering. Biotechnol Prog 26, 819, 2010.

33. Tjellstrom, A. Osseointegrated implants for replacement ofabsent or defective ears. Clin Plast Surg 17, 355, 1990.

34. Heymer, A., Bradica, G., Eulert, J., and Noth, U. Multiphasiccollagen fibre-PLA composites seeded with human mesen-chymal stem cells for osteochondral defect repair: an in vitrostudy. J Tissue Eng Regen Med 3, 389, 2009.

35. Haisch, A. Ear reconstruction through tissue engineering.Adv Otorhinolaryngol 68, 108, 2010.

Address correspondence to:Joseph P. Vacanti, M.D.

Departments of Surgery and Pediatric SurgeryMassachusetts General Hospital

55 Fruit St., Warren 1151Boston, MA 02114

E-mail: jvacanti@partners.org

Received: October 29, 2010Accepted: February 1, 2011

Online Publication Date: March 14, 2011

COMPOSITE ENGINEERED EAR SCAFFOLD MAINTAINS DIMENSIONS 9