TISSUE ENGINEERING: Part A

Volume 20, Numbers 5 and 6, 2014

ª Mary Ann Liebert, Inc.
DOI: 10.1089/ten.tea.2013.0159

Development of Scaffold-Free Elastic Cartilaginous

Constructs with Structural Similarities to Auricular Cartilage

Renata Giardini-Rosa, PhD,1,2 Paulo P. Joazeiro, PhD,2 Kathryn Thomas, BASc,3 Kristina Collavino, BASc,1,3
Joanna Weber, MSc,1,4 and Stephen D. Waldman, PhD, PEng1,3,4

External ear reconstruction with autologous cartilage still remains one of the most difficult problems in the
fields of plastic and reconstructive surgery. As the absence of tissue vascularization limits the ability to
stimulate new tissue growth, relatively few surgical approaches are currently available (alloplastic implants or
sculpted autologous cartilage grafts) to repair or reconstruct the auricle (or pinna) as a result of traumatic loss or
congenital absence (e.g., microtia). Alternatively, tissue engineering can offer the potential to grow autogenous
cartilage suitable for implantation. While tissue-engineered auricle cartilage constructs can be created, a sub-
stantial number of cells are required to generate sufficient quantities of tissue for reconstruction. Similarly, as
routine cell expansion can elicit negative effects on chondrocyte function, we have developed an approach to
generate large-sized engineered auricle constructs ( ‡ 3 cm2) directly from a small population of donor cells
(20,000–40,000 cells/construct). Using rabbit donor cells, the developed bioreactor-cultivated constructs
adopted structural-like characteristics similar to native auricular cartilage, including the development of distinct
cartilaginous and perichondrium-like regions. Both alterations in media composition and seeding density had
profound effects on the formation of engineered elastic tissue constructs in terms of cellularity, extracellular
matrix accumulation, and tissue structure. Higher seeding densities and media containing sodium bicarbonate
produced tissue constructs that were closer to the native tissue in terms of structure and composition. Future
studies will be aimed at improving the accumulation of specific tissue constituents and determining the clinical
effectiveness of this approach using a reconstructive animal model.

Introduction neering can offer the potential to overcome the limitations

with these conventional surgical approaches as autologous
cartilaginous tissue can be grown in vitro.2,6 In addition,
R econstruction of the external ear (total or par-
tial) remains one of the most difficult challenges in
reconstructive and plastic surgery.1–4 However, there are a
several studies have demonstrated that tissue-engineered
cartilage constructs can precisely be developed in the pre-
few approaches currently available for external ear recon- determined shape of the external ear using cell seeded-
struction, such as alloplastic implants and sculpted autolo- polymer constructs.2,7 Despite this potential, recapitulation
gous cartilage grafts—both of which have their advantages of the native structure of auricle cartilage and cell sourcing
and disadvantages. While the availability of alloplastic im- remain as technical hurdles toward the development of
plants in the shape of the ear is typically not an issue, these functional tissue-engineered implants.1,6
implants are susceptible to both infection and immunogenic Auricle cartilage is an elastic cartilaginous tissue with a
issues and, as such, their long-term durability is uncertain.3,5 distinct structure. The central cartilaginous region is similar
Sculpted autologous cartilage grafts, alternatively, have to hyaline cartilage as it is rich in collagen type II fibers and
good long-term durability as well as the potential to grow large aggregating proteoglycans while also possessing
with the patient.2,3,5 In contrast to alloplastic implants, abundant elastic fibers (cross-linked elastin and associated
sculpted autologous cartilage grafts require shaping at the microfibrils). The sparsely populated chondrocytes that re-
time of surgery, can result in donor-site morbidity,2,5 and side in this region are quite large in size (compared to
patients must have a sufficient supply of donor cartilage to chondrocytes in hyaline cartilage) and are responsible for
be candidate for this approach.2,3 Alternatively, tissue engi- the maintenance and remodeling of this tissue.8,9 This

1
Human Mobility Research Centre, Kingston General Hospital and Queen’s University, Kingston, Canada.
2
Department of Histology and Embryology, Institute of Biology, University of Campinas (UNICAMP), Campinas, São Paulo, Brazil.
Departments of 3Chemical Engineering and 4Mechanical & Materials Engineering, Queen’s University, Kingston, Canada.

1012
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1013

central cartilaginous region is surrounded by two layers of were isolated by centrifugation (700 g for 7 min) with the
dense connective tissue on both sides and referred to as the resultant cell pellet washed three times with Ham’s F12
perichondium.10 The perichondrium, believed to be essential media. Viable cells, determined by Trypan Blue dye (Sigma-
for the growth and maintenance of elastic cartilage,2,9–11 is Aldrich) exclusion,30 were then seeded in low-density
rich in collagen type I fibers, small nonaggregating proteo- monolayers in a continuous flow bioreactor in Ham’s F12
glycans, and contains numerous fibroblastic-like cells. media (Hyclone) supplemented with 25 mM HEPES (4-2
However, the cells in the inner layer of the perichondrium (2-hydroxyethyl) piperazine-1-ethanesulfonic acid) (Sigma-
(adjacent to the cartilaginous region), while resembling fi- Aldrich) and an antibiotic solution containing 100 U/mL
broblasts, are chondroblasts and can easily differentiate into penicillin, 100 mg/mL streptomycin, and 0.25 mg/mL am-
chondrocytes.10–15 photericin B (Sigma-Aldrich).21,22,31 To minimize inter-
Although the use of autologous cells in tissue engineering animal variability, tissue was obtained from several ears (up
applications is preferable,3,16 only a reasonable number of to three per experiment) and pooled together.
cells can be harvested from an individual. To overcome this
limitation, a common approach has been to include a cell Continuous flow bioreactor
expansion phase before seeding the cells in, or on, a suitable
scaffolding material.11,17–19 Although successful in producing A continuous flow bioreactor system was used to main-
large quantities of cells, routine cell expansion of chon- tain a constant medium supply to the developing construct in
drocytes tends to elicit negative side effects, such as dedif- a low-shear environment.28,32,33 Briefly, the reactor con-
ferentiation and loss of potency,20–22 thereby requiring a vast sisted of multichannel vented polypropylene chambers to
number of cells to synthesize sufficient quantities of en- house single constructs (3 cm2 containing a maximum media
gineered tissue. Alternatively, other sources of chondrocytes volume of 4 mL). A constant flow of fresh media (from
(e.g., costal) have been explored for auricular cartilage tissue aerated reservoirs) at 10 mL/min was provided by a peri-
engineering3,23–27; however, the cytological differences be- staltic pump (Ismatec; Cole Parmer Canada, Anjou, Canada)
tween these cells and elastic chondrocytes may make it dif- with waste media collected in a vented reservoir. The bio-
ficult to recapitulate a functional elastic cartilaginous reactor was housed in an incubator maintained at 37C with
extracellular matrix (ECM). Choosing the appropriate in vitro 95% relative humidity and 5% CO2.
model and creating a microenvironment that sustains elastin
production may be important to tissue engineer elastic carti- Effect of different seeding densities
lage, but the prerequisites for elastin formation are still lar- and media compositions
gely unknown.10 To overcome these limitations, we have To determine the most effective method to synthesize large
recently developed an approach to generate hyaline carti- elastic cartilaginous constructs in reactor culture ( ‡ 3 cm2),
laginous tissue directly from limited population of donor cells two different seeding densities and media formulations were
without a separate expansion phase.28 Long-term bioreactor investigated. Cultures were seeded at either 6500 or 13,000
cultivation of articular chondrocytes resulted in the formation isolated cells/cm2 (i.e., 20,000 or 40,000 cells/well) directly
of large, scaffold-free cartilaginous tissue constructs (as little on the bottom surface of the reactor wells. These seeding
as 20,000 cells to create 3 cm2 of engineered tissue). Upon densities reflect multiples of a routine cell passaging density
implantation, the constructs were able to repair critical-sized (6500 cells/cm2). Similarly, constructs (from both seeding
defects and also adopted the structure of native articular densities) were cultured in the media (Ham’s F12 supple-
cartilage.29 Due to these promising results, the purpose of this mented with 20 mM HEPES, 20% FBS, 100 mg/mL ascorbic
study was to investigate whether this approach could be used acid, and antibiotics) in the presence or absence of 14 mM of
to create scaffold-free auricular cartilaginous constructs that sodium bicarbonate (NaHCO3) to stimulate synthesis and
possess the distinct structure of elastic cartilage from a lim- accumulation of cartilaginous ECM.28 After seeding, all
ited population of extracted cells. preparations were maintained under no-flow conditions for
48 h. Preparations were then cultured under a constant media
Materials and Methods flow rate of 10 mL/min for a period of 4 weeks. Media res-
Rabbit auricular chondrocyte harvest and isolation ervoirs were changed every 2–3 days and supplemented with
fresh ascorbic acid and antibiotics. After the 4-week culture
This study was performed with the approval from the period, developed constructs were harvested, weighed (wet
University Animal Care Committee (UACC) at the Queen’s weight), and then divided into three parts and processed for
University. Full thickness auricular cartilage was harvested histological/immunohistochemical evaluation, biochemical
from the ear of adolescent female New Zealand white rab- analyses, and transmission electron microscopy (TEM), re-
bits (2 – 0.5 kg; *12 weeks old) (Charles River Labora- spectively. Four separate trials of this study were conducted
tories, Wilmington, MA) to reflect the general age of (n = 8 samples/group).
patients undergoing auricular reconstruction. Ears were
completely dissected free of skin with the inner chondro-
Histological evaluation
genic layer and the outer perichondrial layer were kept en-
tirely intact. The tissue was cut in small fragments and Engineered constructs as well as native auricular cartilage
digested with 0.5% protease (w/v) (Sigma-Aldrich, Oak- tissue samples were fixed in 4% paraformaldehyde (in 0.1 M
ville, Canada) followed by 0.15% collagenase A (w/v) phosphate buffer, pH 7.2) for 24 h, dehydrated in graded
(Roche Diagnostics Canada, Laval, QC, Canada) in Ham’s ethanol solutions, and embedded in paraffin at 65C. Thin
F12 media (Hyclone, Logan, UT) overnight at 37C with (5 mm thick) sections were cut and mounted on Superfrost
95% relative humidity and 5% CO2. Auricular chondrocytes slides (Fisher Scientific, Mississauga, Canada) and dried for
1014 GIARDINI-ROSA ET AL.

24 h at 37C. Sections were stained with either hematoxylin

and eosin (H&E, a general connective tissue stain), safranin-
O (proteoglycan stain), and Weigert’s resorcin-fuchsin
(elastin/elastic fiber stain).34 Stained sections were exam-
ined by light microscopy using a Zeiss Axio-Image M1
microscope (Göttingen, Germany).

Tissue thickness measurements

and mechanical testing
The thickness of the tissue constructs and native tissue
samples were determined using the bright-field images of
the histology sections. Bright-field images (at 20 · ) were
digitally captured and calibrated using a calibration slide
from which the thicknesses of the tissue samples could be
determined using ImageJ software (Open Source; National
Institutes of Health). Thickness measurements were then
FIG. 1. Macroscopic appearance of the tissue generated
from the monolayer cell preparations after 4 weeks of bio- taken at 10 random locations throughout each section and
reactor culture. Scale bar: 1 cm. Color images available the average value recorded. Mechanical testing of native
online at www.liebertpub.com/tea auricular cartilage and engineered cartilage constructs was

FIG. 2. General architec-

tural organization of the tis-
sue generated from the
monolayer cell preparations
after 4 weeks in bioreactor
culture compared to native
auricular cartilage. (A) Na-
tive auricular cartilage. (B)
Engineered construct gener-
ated from 6500 cells/cm2
without sodium bicarbonate
(NaHCO3). (C) Engineered
construct generated from
6500 cells/cm2 with 14 mM
NaHCO3. (D) Engineered
construct generated from
13,000 cells/cm2 without
NaHCO3. (E) Engineered
construct generated from
13,000 cells/cm2 with 14 mM
NaHCO3. Hematoxylin and
eosin stain; scale bar: 50 mm.
Color images available online
at www.liebertpub.com/tea
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1015

performed using a Mach-1 Micromechanical Testing system Briefly, after deparaffinization and dehydration, sections
(Biomomentum, Laval, QC, Canada) equipped with a 1 kg were enzymatically treated with 0.25 units/mL chon-
load cell (0.05 g resolution). Tissue mechanical properties droitinase ABC (Sigma-Aldrich) in Tris-acetate buffer
(elastic modulus and Poisson’s ratio at 37C in Ham’s F12 (40 mM Tris acetate with 1 mM EDTA), pH 8.5, for 1 h at
media) were determined using a double compressive in- 37C followed by 0.25 units/mL Keratinase (Sigma-
dentation method35,36 using two plane-ended indentors (2 Aldrich) Tris-acetate buffer (40 mM Tris acetate with 1 mM
and 6 mm diameter). Compressive indentations were con- EDTA), pH 8.5, for 30 min at 37C to facilitate antibody
ducted at a ramp rate of 10% strain/s to a maximum of 10% binding. To reduce nonspecific protein binding, the sections
strain. Samples were then allowed to equilibrate in media were blocked with phosphate-buffered saline (PBS) with 1%
for *20 min between indentations. bovine serum albumin (BSA) for 30 min at room tempera-
ture. Sections were then incubated with mouse monoclonal
antibodies against elastin (BA4 at 1:100 dilution; Abcam,
Immunochemical localization of elastin
Cambridge, MA) or with rabbit polyclonal antibodies
and collagen types
against collagen type X (ab58632 at 1:150 dilution; Abcam),
Immunofluorescence localization of elastin and collagen all diluted in phosphate buffer (pH 7.4) containing 1% BSA
X in the engineered constructs and native auricular cartilage overnight at 4C. Following primary antibody incubation,
tissues samples was performed as previously described.28 sections were rinsed in PBS (pH 7.4) and incubated with

FIG. 3. Proteoglycan orga-

nization of the tissue gener-
ated from the monolayer cell
preparations after 4 weeks in
bioreactor culture compared
to native auricular cartilage.
(A) Native auricular carti-
lage. (B) Engineered con-
struct generated from 6500
cells/cm2 without NaHCO3.
(C) Engineered construct
generated from 6500 cells/
cm2 with 14 mM NaHCO3.
(D) Engineered construct
generated from 13,000 cells/
cm2 without NaHCO3. (E)
Engineered construct gener-
ated from 13,000 cells/cm2
with 14 mM NaHCO3.
Safranin-O stain; scale bar:
50 mm. Color images avail-
able online at www.liebertpub
.com/tea
1016 GIARDINI-ROSA ET AL.

Texas Red (elastin) or FITC-labeled (collagen type X) anti- 187 mg/mL; Developmental Studies Hybridoma Bank,
mouse or anti-rabbit secondary antibodies (1:200 dilution; Iowa, IA), all diluted in phosphate buffer (pH 7.4) con-
Abcam) for 2 h at room temperature. The sections were taining 1% BSA overnight at 4C. Following primary an-
counterstained with DAPI (Vector Laboratories, Inc., Bur- tibody incubation, sections were rinsed in PBS (pH 7.4)
lingame, CA) and examined by fluorescent microscopy us- and incubated with biotinylated anti-mouse secondary an-
ing a Zeiss Axio-Image M1 microscope with Axiovision tibodies (Vector Laboratories) using the Vectastain Elite
software (Carl Zeiss, Oberkochen, Germany). ABC kit (Vector Laboratories) for 2 h at room temperature,
Immunohistochemical localization of collagen types I followed by incubation with diaminobenzidine for 6 min
and II in the engineered constructs and native auricular at room temperature. The sections were counterstained
cartilage tissues samples was performed as previously de- with Harris’ hematoxylin and mounted in the permanent
scribed.37 Briefly, after deparaffinization and dehydration, mounting medium. Stained sections were examined by
sections were enzymatically treated with 0.05% of trypsin light microscopy using a Zeiss Axio-Image M1 microscope
(pH 7.8) for 30 min at 37C to facilitate antibody binding. (Göttingen, Germany).
Endogenous peroxidase activity was blocked with 1% For all immunohistochemical studies, nonspecific staining
H2O2 and 1% BSA in phosphate buffer for 30 min. Sec- was assessed by the replacement of the primary antibody
tions were then incubated with mouse monoclonal anti- with nonimmune serum (data not shown). All the experi-
bodies against collagen type I (I-8H5 at 40 mg/mL; Daiichi ments were completed at least three times with no positive
Fine Chemicals Co. Ltd.) or collagen type II (II-II6B3 at staining detected in the negative controls.

FIG. 4. Elastic fiber orga-

nization of the tissue gener-
ated from the monolayer cell
preparations after 4 weeks in
bioreactor culture compared
to native auricular cartilage.
(A) Native auricular carti-
lage. (B) Engineered con-
struct generated from 6500
cells/cm2 without NaHCO3.
(C) Engineered construct
generated from 6500 cells/
cm2 with 14 mM NaHCO3.
(D) Engineered construct
generated from 13,000 cells/
cm2 without NaHCO3; (E)
Engineered construct gener-
ated from 13,000 cells/cm2
with 14 mM NaHCO3. Ar-
rows denote the presence of
thin elastic fibers. Resorcin-
fuchsin elastic fibers stain;
scale bars: (A) 50 mm; (B–E)
20 mm. Color images avail-
able online at www.liebertpub
.com/tea
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1017

Transmission electron microscopy well as native auricular cartilage tissue samples was then
Engineered constructs and native auricular cartilage determined after overnight lyophilization. Samples were
tissue samples were fixed in a Karnovsky solution for 2 h at then digested by papain (bioreactor samples: 40 mg/mL;
room temperature. Samples were then postfixed in 1% native tissue samples: 80 mg/mL in 20 mM ammonium ac-
osmium tetroxide for 1 h at 4C, dehydrated in graded etate, 1 mM EDTA, and 2 mM dithiothreitol) for 72 h at
ethanol solutions, and embedded in epoxy resin Epon 812 65C and stored at - 20C until analysis. Aliquots of the
(Electron Microscope Science, Hatfield, PA). Ultrathin digest were assayed separately for DNA, proteoglycan, and
sections (70 nm) were cut and collected on copper collagen content. The DNA assay was estimated using
grids, stained with uranyl acetate and lead citrate, and Hoechst 33258 dye (Sigma-Aldrich)39 with calf thymus
examined using a LEO 906 transmission electron micro- DNA as the standard (Sigma-Aldrich). The proteoglycan
scope (LEO Elektronenmikroskopie GmbH, Oberkochen, content was estimated by quantifying the amount of sulfated
Germany).38 glycosaminoglycans using 1,9-dimethylmethylene blue dye
binding assay 40,41 with chondroitin sulfate sodium salt from
bovine cartilage as the standard (Sigma-Aldrich). Total
Biochemical quantification of matrix constituents
collagen content (combined content of all collagen types)
Engineered construct samples were weighed again to was determined by the determination of hydroxyproline
determine the percent sample mass of the entire developed content. Briefly, aliquots of the papain digest were hydro-
construct. The dry weight of the engineered construct as lyzed in 6 N HCl for 18 h at 110C, and the hydroxyproline

FIG. 5. Immunohisto-
chemical localization of
elastin in the engineered
elastic cartilaginous con-
structs generated from
monolayer preparations af-
ter 4 weeks of bioreactor
culture compared to native
auricular cartilage. (A, B)
Native auricular cartilage.
(C cartilaginous region; C¢
perichondrium-like region)
Engineered construct gener-
ated from 6500 cells/cm2
without NaHCO3. (D carti-
laginous region; D¢ peri-
chondrium-like region)
Engineered construct gener-
ated from 6500 cells/cm2
with 14 mM NaHCO3. (E
cartilaginous region; E¢
perichondrium-like region)
Engineered construct gener-
ated from 13,000 cells/cm2
without NaHCO3. (F carti-
laginous region; F¢ peri-
chondrium-like region)
Engineered construct gener-
ated from 13,000 cells/cm2
with 14 mM NaHCO3. Bi-
nucleated cells were ob-
served in the native tissue
(circled in B) and also on
the constructs (circled in E
and F). Scale bars: (A)
50 mm; (B) 20 mm; (C–F¢)
10 mm. Color images avail-
able online at www.liebertpub
.com/tea
1018 GIARDINI-ROSA ET AL.

content was determined in the hydrolyzate using chlora- Results

mine-T/Ehrlich’s reagent assay.42 Total collagen content Structure of engineered elastic cartilage constructs
was estimated assuming hydroxyproline accounts for 10%
of the total collagen mass in cartilage.43 Using a small population of isolated cells, scaffold-free
elastic cartilage constructs were generated from long-term
bioreactor culture of different monolayer preparations (6500
Statistical analyses
or 13,000 cells/cm2) with or without media supplemented
All results were expressed as the mean – standard error of with 14 mM NaHCO3. Each of the preparations resulted in
the mean. Biochemical quantification data were analyzed the formation of large tissue constructs (*3 cm2) (Fig. 1).
using two-way analysis of variance (ANOVA) to determine Histological evaluation of the engineered constructs re-
the effect of cell seeding density and media formulation on vealed that the generated neotissue possessed a similar
elastic cartilaginous tissue growth. In addition, pair-wise morphology to that of native auricular cartilage with the
comparisons between individual groups was conducted us- appearance of defined cartilaginous and perichondrium-like
ing a one-way ANOVA and Tukey’s post hoc testing. Data regions. The neocartilage was composed of round-to-oval
were checked before performing statistical tests for both lacunae cells homogeneously distributed in a basophilic
normality and equal variance. Statistical tests were con- matrix (Fig. 2) with larger cells arranged in cartilaginous
ducted using statistical software (SPSS version 16; SPSS, region and smaller cells in the perichondrium-like region.
Inc., Chicago, IL), and the significance was associated with The ECM of the cartilaginous region stained positive for
p-values less than 0.05. sulfated proteoglycans (Fig. 3) as well as displaying the

FIG. 6. Immunohisto-
chemical localization of
collagen X in the en-
gineered elastic cartilagi-
nous constructs generated
from monolayer prepara-
tions after 4 weeks of bio-
reactor culture compared to
native auricular cartilage.
(A, B) Native auricular
cartilage. (C: cartilaginous
region; C¢: perichondrium-
like region) Engineered
construct generated from
6500 cells/cm2 without
NaHCO3. (D: cartilaginous
region; D¢: perichondrium-
like region) Engineered
construct generated from
6500 cells/cm2 with 14 mM
NaHCO3. (E: cartilaginous
region; E¢: perichondrium-
like region) Engineered
construct generated from
13,000 cells/cm2 without
NaHCO3. (F: cartilaginous
region; F¢: perichondrium-
like region) Engineered
construct generated from
13,000 cells/cm2 with
14 mM NaHCO3. Binu-
cleated cells were observed
in the native tissue (circled
in B) and also on the con-
structs (circled in panel F).
Scale bars: (A) 50 mm; (B)
20 mm; (C–F¢) 10 mm. Color
images available online at
www.liebertpub.com/tea
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1019

presence of thin elastin fibers (Fig. 4). Immunohisto- constructs, collagen X was only detected intracellularly
chemical assessment confirmed the presence of elastin within the cartilaginous and perichondrium-like regions
(Fig. 5). In the native tissue, elastin was evident in the (Fig. 6C–F). Binucleated cells were also detected in these
matrix as well as inside the cells. For the engineered images in both the native tissue and the engineered con-
constructs, however, elastin was only observed intracellu- structs (Fig. 6B, F). Collagen type I was exclusively lo-
larly. In addition, the presence of binucleated cells was cated in the perichondium-like region (Fig. 7A, C, E), with
observed in both the native tissue and the engineered more intense staining the boundary between this region and
constructs (Fig. 5B, E, F). The perichondrium-like region elastic cartilaginous matrix (Fig. 7C, E). Interestingly,
was predominantly fibrous and displayed little positive collagen type I was not detected in the any of the con-
proteoglycan staining. In the native tissue, the perichon- structs grown in the absence of NaHCO3 (Fig. 7B, D). The
drium was negative for elastin (Fig. 5A, B); however, distribution of collagen type II was more homogenous with
elastin was present in the perichondrium-like region of the staining throughout the construct (elastic cartilaginous
engineered constructs, which appeared to decrease with matrix and perichondrium-like regions), with more intense
higher seeding densities (Fig. 5C–F). staining in the elastic cartilaginous matrix region (Fig. 8).
Immunohistochemical assessment for collagen types Among the different preparations investigated, tissues
showed that collagen X was present in the native tissue on generated from the higher seeding density (13,000 cells/
the cartilaginous region, intra- and extracellularly but not cm2) with NaHCO3 supplemented media appeared to
within the perichondrium (Fig. 6A, B). In the engineered possess a structure more similar to the native tissue in

FIG. 7. Immunohisto-
chemical localization of col-
lagen I in the engineered
elastic cartilaginous con-
structs generated from
monolayer preparations after
4 weeks of bioreactor culture
compared to native auricular
cartilage. (A) Native auricu-
lar cartilage. (B) Engineered
construct generated from
6500 cells/cm2 without
NaHCO3. (C) Engineered
construct generated from
6500 cells/cm2 with 14 mM
NaHCO3. (D) Engineered
construct generated from
13,000 cells/cm2 without
NaHCO3. (E) Engineered
construct generated from
13,000 cells/cm2 with 14 mM
NaHCO3. Arrows indicate
positive detection of collagen
I in the perichondrium-like
regions of the construct.
Scale bar: 50 mm. Color
images available online at
www.liebertpub.com/tea
1020 GIARDINI-ROSA ET AL.

FIG. 8. Immunohisto-
chemical localization of col-
lagen II in the engineered
elastic cartilaginous con-
structs generated from
monolayer preparations after
4 weeks of bioreactor culture
compared to native auricular
cartilage. (A) Native auricu-
lar cartilage. (B) Engineered
construct generated from
6500 cells/cm2 without
NaHCO3. (C) Engineered
construct generated from
6500 cells/cm2 with 14 mM
NaHCO3. (D) Engineered
construct generated from
13,000 cells/cm2 without
NaHCO3. (E) Engineered
construct generated from
13,000 cells/cm2 with 14 mM
NaHCO3. Scale bar: 50 mm.
Color images available online
at www.liebertpub.com/tea

terms of cellular morphology and having a more defined chondroblasts of all cartilage tissues, and it was also ob-
perichondrium-like region. served on the native tissue. Similar morphology was also
observed in the constructs grown in the absence of NaHCO3
(data not shown). All groups possessed the same key ele-
Tissue ultrastructure
ments of elastic tissue in terms of cytoskeleton, lipid drop-
Ultrastructure of cells from the engineered elastic tissue lets, and defined territorial and interterritorial matrices. It
constructs, compared to native elastic cartilage, was exam- was possible to observe the presence of elastic fibers in the
ined by TEM. In the native elastic cartilage, chondrocytes in constructs grown in the presence of NaHCO3. These ap-
the central region were observed with single, large cyto- peared as electron-dense amorphous aggregates delineated
plasmic lipid droplets (Fig. 9A). These droplets were sur- by a thin fibrillary meshwork at the border of interterritorial
rounded by cytofilaments primarily arranged in parallel (Fig. matrix and not detected in the territorial matrix (Fig. 9C, E).
9A, inset). These elements were more readily observed in Slight differences in ultrastructure were observed between
cells that were less elongated (typical chondrocyte mor- the different preparations with the presence of NaHCO3
phology). Similar morphologies were observed in the en- appearing to have a greater effect than seeding density.
gineered tissues after 4 weeks of bioreactor cultivation (Fig. Constructs grown in the absence of NaHCO3 (data not
9B, D). Cells present in the ECM of the engineered con- shown) displayed less preserved cells and contained more
structs were generally oval in shape and presented many thin/delicate fibrilar meshes compared to the construct
indentations and deep recesses. These irregularly shaped grown in the presence of NaHCO3. Similar to the histo-
margins represent a peculiar structural feature of young logical and immunohistochemical evaluations, constructs
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1021

FIG. 9. Transmission elec-

tron micrographs of the
middle of the engineered
elastic cartilaginous con-
structs and native auricular
cartilage. (A) Native auricu-
lar cartilage with inset
showing the detail of the ex-
tracellular matrix. (B) Cel-
lular ultrastructure within the
engineered construct gener-
ated from 6500 cells/cm2
with NaHCO3. (C) Presence
of elastic fiber formation
(circled) within the en-
gineered construct generated
from 6500 cells/cm2 with
14 mM NaHCO3. (D) Cel-
lular ultrastructure within the
engineered construct gener-
ated from 13,000 cells/cm2
with NaHCO3. (E) Presence
of elastic fiber formation
(circled) within the en-
gineered construct generated
from 13,000 cells/cm2 with
14 mM NaHCO3. LD, lipid
droplets; N, nucleus; IM, in-
terterritorial matrix; TM,
territorial matrix. Scale bars:
(A, B) 5 mm; (A inset) 1 mm;
(D) 2 mm; (C, E) 1 mm.

Table 1. Biochemical and Physical Properties of Engineered Elastic Cartilaginous Tissue

Seeding density and NaHCO3 concentration
6500 (cells/cm2) 13,000 (cells/cm2)
0 (mM) 14 (mM) 0 (mM) 14 (mM)
DNA (mg/mg dry wt.) 1.0 – 0.04a 1.1 – 0.1 1.13 – 0.06a,b 1.0 – 0.1b
PG (mg/mg dry wt.) 1.0 – 0.05 1.5 – 0.5 1.3 – 0.1b 2.2 – 0.9b
Collagen (mg/mg dry wt.) 1.0 – 0.1 1.2 – 0.2 1.2 – 0.2 1.4 – 0.1
PG/DNA (mg/mg) 1.0 – 0.1 1.2 – 0.3a 1.0 – 0.1b 2.1 – 0.8a,b
Collagen/DNA (mg/mg) 1.0 – 0.1 1.3 – 0.2a 0.89 – 0.09b 1.6 – 0.2a,b
Collagen/PG (mg/mg) 1.0 – 0.2b 1.6 – 0.6a,b 0.76 – 0.07 0.9 – 0.3a
Thickness (mm) 1.0 – 0.1b 1.5 – 0.3b 1.2 – 0.1b 1.5 – 0.2b
Four separate experiments were performed, data expressed relative to control (6500 cells/cm2; 0 mM NaHCO3), presented as
mean – SEM (n = 6–8 samples/group).
a
Significant effect of seeding density ( p < 0.05).
b
Significant effect of sodium bicarbonate ( p < 0.05).
PG, proteoglycans; SEM, standard error of the mean; NaHCO3, sodium bicarbonate.
1022 GIARDINI-ROSA ET AL.

grown at the higher seeding density and in the presence of

NaHCO3 displayed the greatest similarity to the ultrastruc-
ture of native elastic cartilage.

Physical and biochemical properties

of accumulated ECM
Both the presence of NaHCO3 in the media and seeding
density had a profound effect on the formation of engineered
elastic tissue constructs in terms of tissue thickness, cellu-
larity, and ECM accumulation. Tissue thickness was sig-
nificantly affected by the presence of NaHCO3 (50%
increase; p < 0.01), with no observable effect of seeding
density (Table 1). Tissue cellularity was also only signifi-
cantly affected by the presence of NaHCO3 (72–99% in-
crease in DNA content; p < 0.01) (Fig. 10A). Cartilaginous
ECM accumulation was significantly affected by both
NaHCO3 (83–139% increase in proteoglycan content, 129–
221% increase in collagen content; p < 0.01) and seeding
density (30% increase in proteoglycan content, 40% in-
crease in collagen content; p < 0.01) (Fig. 10B, C). When
normalized to tissue dry weight or cellularity, similar effects
of NaHCO3 and seeding density were observed (Table 1).
However, when compared to native auricular cartilage, the
obtained thickness, ECM accumulation, and cellularity in
the engineered elastic constructs were lower with the ex-
ception of proteoglycan accumulation, which appeared to
approach the values of native tissue (Table 2). Similar to the
previous results, constructs grown at the higher seeding
density and in the presence of NaHCO3 displayed the
greatest similarity to native elastic cartilage in terms of
tissue thickness and biochemical composition (Table 2). In
addition, while the elastic modulus of the engineered elastic
constructs were significantly lower ( p < 0.01) than native
auricular cartilage (13,000 cells/cm2 with NaHCO3:
13 – 4 kPa, n = 8; native auricular cartilage: 100 – 19 kPa,
n = 7); both tissues appeared to possess a similar Poisson’s
ratio (13,000 cells/cm2 with NaHCO3: 0.305 – 0.003, n = 8;
native auricular cartilage: 0.281 – 0.019, n = 7).

Discussion
FIG. 10. Quantification of biochemical constituents in the
As there are relatively few surgical approaches currently engineered elastic cartilaginous constructs generated from
available for total or partial ear reconstruction (alloplastic monolayer preparations after 4 weeks of bioreactor culture.
implants or sculpted autologous cartilage grafts),2,3,6,44 tis- (A) DNA content; (B) proteoglycan content; (C) total col-
sue engineering methods have been explored as a means to lagen content (combined content of all collagen types). Data
represented as mean – SEM (n = 8 samples/group) normal-
grow autologous cartilage suitable for implantation. How- ized to the constructs generated from 6500 cells/cm2 without
ever, as routine cell expansion can elicit negative effects NaHCO3. aStatistical effect of NaHCO3 ( p < 0.01). bStatis-
on chondrocyte function,20–22 we have developed an ap- tical effect of seeding density ( p < 0.01).
proach to generate large-sized,28,45 scaffold-free engineered
auricle constructs without the use of a separate cell ex-
pansion phase. After 4 weeks of bioreactor cultivation, boundary between the cartilaginous and perichondrium-like
the developed scaffold-free tissue constructs approached regions of the developed constructs and there was some ev-
the thickness of native auricular cartilage (38–65% of native idence of both collagen type X and elastin, which appeared to
tissue thickness). Interestingly, the engineered constructs be modulated by changes in seeding density and the presence
appeared to adopt a structural appearance similar to that of of NaHCO3. Interestingly, while the formation of a cartilag-
native auricular cartilage. The top portion of the constructs inous region in the developed constructs were observed
resembled the perichondrium of native auricular cartilage among all the investigated conditions (seeding density and
and contained smaller spindle-shaped cells in a fibrous ECM NaHCO3 supplementation), the cultures supplemented with
that stained positive for the presence of collagen types I and NaHCO3 developed more distinct perichondrium-like re-
II with little staining for sulfated proteoglycans. However, gions. Although the development of perichondrium-like re-
the presence of collagen type I was primarily located at the gions in neocartilage constructs have been observed in
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1023

Table 2. Representative Data of the Biochemical and Physical Properties of Engineered

Elastic Cartilaginous Tissue and Native Auricular Tissue from a Single Experiment
Seeding density and NaHCO3 concentration
6500 (cells/cm2) 13,000 (cells/cm2)
0 (mM) 14 (mM) 0 (mM) 14 (mM) Native tissue
DNA (mg/mg dry wt.) 3.44 – 0.05 3.3 – 0.1 4.1 – 0.7 3.3 – 0.4 2.3 – 0.2
PG (mg/mg dry wt.) 202 – 41 263 – 2 315 – 64 234 – 15 231 – 26
Collagen (mg/mg dry wt.) 130 – 18 185 – 48 144 – 31 152 – 35 496 – 35
PG/DNA (mg/mg) 59 – 11 79 – 2 76 – 2 71 – 5 133 – 14
Collagen/DNA (mg/mg) 38 – 5 26 – 15 35 – 1 45 – 5 230 – 30
Collagen/PG (mg/mg) 0.65 – 0.04 0.3 – 0.1 0.46 – 0.01 0.6 – 0.1 2.2 – 0.2
Thickness (mm) 0.10 – 0.01 0.162 – 0.03 0.13 – 0.01 0.17 – 0.03 0.26 – 0.03
Bioreactor: n = 2–8/group; native tissue: n = 6/group.

relatively few previous studies—most notably by Yanaga and Although the developed cartilage constructs appeared to
coworkers who also showed a spontaneous recapitulation of a adopt structural-like characteristics of native auricular car-
perichondrium-like region after implantation,46 the majority tilage, the accumulation of ECM constituents and resultant
of elastic cartilage tissue engineering approaches tend to only mechanical properties were lower than that of the native
utilize cells isolated from the cartilaginous region.10,11,47–51 tissue. Although the accumulation of proteoglycans in the
The creation of engineered constructs possessing both carti- developed constructs was similar to native auricular carti-
laginous and perichondrium regions, similar to that observed lage, the engineered constructs only accumulated between
in native auricular cartilage, is of benefit toward the devel- 26% and 37% of the collagen content of the native tissue.
opment of functional engineered constructs suitable for im- This effect resulted in lower collagen-to-proteoglycan ratios
plantation.2,10,44 with the constructs achieving similar levels compared to that
The cartilaginous portion of the engineered constructs of native auricular cartilage. Previous studies have also
also appeared to be similar to that of native auricular car- noted reduced ECM accumulation and mechanical proper-
tilage with the presence of elastin, collagen types II and X, ties within engineered elastic cartilage constructs.3,6,9
and sulfated proteoglycans. The presence of elastin ex- However, the biochemical and mechanical properties of the
pression and elastic fiber formation was confirmed by his- developed constructs could potentially be affected by the
tology, immunohistochemistry, and TEM—which appeared culture conditions. Both seeding density and the use of
as electron-dense aggregates delineated by a thin fibrillary medium supplemented with NaHCO3 were sufficient to in-
meshwork located at the border of the interterritorial matrix. crease the accumulation of both proteoglycans (twofold
The reduced elastin expression in the engineered constructs increase) and collagen (threefold increase) relative to the
(with little to no observable extracellular staining) may be control cultures (lower seeding density without NaHCO3
due to mature chondrocytes being less capable of elastin supplementation). Although increasing maturation of the
synthesis or that in vitro culture models do not necessarily engineered constructs was observed through the modulation
support elastin production by these cells.10,48 Thus, the of the culture conditions, the development of constructs with
upregulation of elastin synthesis and elastic fiber formation improved biochemical and mechanical properties will most
in the developed constructs needs to be improved for the likely be required for the development of functional con-
development of functional elastic cartilage constructs. structs that are suitable for implantation. Engineered con-
While collagen type II appeared to be widely expressed, structs also appeared to be more cellular than native
collagen type X was also expressed in the engineered con- auricular cartilage (43–78% increase in tissue cellularity);
structs. Although collagen type X is typically associated as a an effect that has also been observed by others.58 Interest-
marker of hypertrophic and mineralized cartilage,52–55 it is ingly, no significant differences in construct cellularity were
also present in nonmineralizing tissues, including auricular observed with increased seeding densities (as opposed to the
and neoseptal cartilages,55 and on the surface of articular presence of NaHCO3) suggesting that higher seeding den-
cartilage56 and the meniscus.57 In addition, a small population sities could potentially be used to accelerate tissue forma-
of binucleated cells was observed in the cartilaginous region tion within the bioreactor without deleterious effects on the
of the native tissue and bioreactor-cultivated constructs. Ac- cellularity of the developed constructs. Ultrastructural
cording to Moskalewski et al.,47 the presence of binucleated characterization of the constructs confirmed the observa-
cells is one of the hallmark observations in rabbit auricular tions of the similarities between the engineered constructs
cartilage, which appeared to be maintained after bioreactor and native auricular cartilage. Key elements of the native
cultivation. Although smaller populations of binucleated cells tissue could also be observed in the constructs after 4 weeks
were observed in the present study (native tissue and en- of cultivation, including defined territorial and interterrito-
gineered constructs) compared to the observations of Mos- rial matrices, cytoskeleton, and presence of lipid droplets.
kalewski (*25% of the cell population),49 the engineered The cells in the engineered constructs were generally oval in
constructs were capable of maintaining the characteristic bi- shape and had cytoplasmic margins with many indentations
nucleated cell phenotype of elastic cartilage.47,49 and recesses similar to that observed in the native tissue.
1024 GIARDINI-ROSA ET AL.

These irregularly shaped margins represent a peculiar support of Foreign Affairs and International Trade Canada/
structural feature of young chondroblasts of all cartilaginous L’appui d’Affaires étrangères et Commerce international
tissues.59 The intercellular matrix within the constructs did Canada. The authors also declare that there are no conflicts
not appear to differ morphologically from native auricular of interest associated with this work.
cartilage. However, the microfibrillar component of the
elastic fibers was more prominent in the native tissue. In- Disclosure Statement
terestingly, due to the presence of a large cytoplasmic lipid
droplets surrounded by cytofilaments and sparse ECM, some No competing financial interests exist.
authors have suggested that tissue with these features re-
semble more of an adipose-like tissue, referred to as a References
secondary cartilage (a primitive chondroid tissue of lower 1. Langer, R., and Vacanti, J.P. Tissue engineering. Science
vertebrates).60–62 In the present study, while we did observe 260, 920, 1993.
some of these characteristics in both the engineered con- 2. Cao, Y., Vacanti, J.P., Paige, K.T., Upton, J., and Vacanti,
structs and native auricular cartilage samples, further ana- C.A. Transplantation of chondrocytes utilizing a polymer-
lyses are required to determine whether these tissues can be cell construct to produce tissue-engineered cartilage in
classified as secondary cartilages since these lipid droplets the shape of a human ear. Plast Reconstr Surg 100,
were also observed in the chondrogenic perichondrium of 297, 1997.
the native tissue (data not shown). 3. Shieh, S.-J., Terada, S., and Vacanti, J.P. Tissue engi-
In this proof-of-principle study, isolated chondrocytes neering auricular reconstruction: in vitro and in vivo stud-
from adolescent animals were used to reflect the general age ies. Biomaterials 25, 1545, 2004.
of patients undergoing auricular reconstruction for microtia. 4. Yanaga, H., Imai, K., Koga, M., and Yanaga, K. Cell-
However, as surgical reconstruction is also performed in engineered human elastic chondrocytes regenerate natural
cases of ear injury (e.g., burns, bites), older and potentially scaffold in vitro and neocartilage with neoperichondrium in
less synthetically active autologous cells are available, the human body post-transplantation. Tissue Eng Part A 18,
which may affect the amount and quality of the tissue that 2020, 2012.
can be generated through this approach. This is especially 5. Kim, W.S., Vacanti, J.P., Cima, L., Mooney, D., Upton, J.,
relevant in terms of elastogenesis, which is known to be Puelacher, W.C., and Vacanti, C.A. Cartilage engineered in
adversely affected with increasing age.8,10,47,50 This could predetermined shapes employing cell transplantation on
potentially be overcome through the maturation of the synthetic biodegradable polymers. Plast Reconstr Surg 94,
233, 1994.
constructs in vivo44 or the use of hormone or growth factor
6. Bichara, D.A., O’Sullivan, N.A., Pomerantseva, I., Zhao,
supplementation (insulin, IGF-1, or TGF-b1), which have
X., Sundback., C.A., Vacanti, J.P., and Randolph, M.A.
been shown to stimulate the elastogenesis.63–68 Another The tissue-engineered auricle: past, present, and future.
hurdle for the development of functional auricular tissue Tissue Eng Part B Rev 18, 51, 2012.
implants is the need to develop anatomically shaped con- 7. Yotsuyanagi, T., Urushidate, S., Watanabe, M., and Sa-
structs. To address this issue, patient-specific rapid-proto- wada, Y. Reconstruction of a three-dimensional structure
typed tissue culture molds could be used, which have using cartilage regenerated from the perichondrium of
previously been explored for the creation of engineered rabbits. Plast Reconstr Surg 103, 1120, 1999.
articular cartilage constructs tailored to the anatomy of a 8. Moskalewski, S., and Rybicka, E. The influence of the
particular defect.28 Finally, the translation toward the de- degree of maturation of donor tissue on the reconstruction
velopment of human engineered auricular constructs need to of elastic cartilage by isolated chondrocytes. Acta Anat
be undertaken before this approach could used clinically. (Basel) 97, 231, 1977.
In this study, we have described an approach to generate 9. Madsen, K., Moskalewski, S., Mark, von der, K., and Fri-
large-size auricular cartilage constructs from a small popu- berg, U. Synthesis of proteoglycans, collagen, and elastin
lation of isolated cells. The cells were cultivated in a con- by cultures of rabbit auricular chondrocytes-relation to age
tinuous flow bioreactor to stimulate the production of of the donor. Dev Biol 96, 63, 1983.
cartilaginous tissue. Developed scaffold-free tissue con- 10. Hellingman, C.A., Verwiel, E.T., Slagt, I., Koevoet, W.,
structs appeared to adopt structural-like characteristics Poublon, E.M., Nolst-Trenité, G.J., Baatenburg de Jong,
similar to native auricular cartilage with the development of R.J, Jahr, H., and Van Osch, G.J. Differences in cartilage-
distinct cartilaginous and perichondrium-like regions. Al- forming capacity of expanded human chondrocytes from
terations in media composition and seeding density had ear and nose and their gene expression profiles. Cell
Transplant 20, 925, 2011.
profound effects on the formation of engineered auricular
11. Takebe, T., Kobayashi, S., Kan, H., Suzuki, H., Yabuki, Y.,
tissue constructs in terms of cellularity, ECM accumulation,
Mizuno, M., Adegawa, T., Yoshioka, T., Tanaka, J., Mae-
thickness, and tissue structure. Future studies will be aimed gawa, J., and Taniguchi, H. Human elastic cartilage engi-
at improving the accumulation of specific tissue constitu- neering from cartilage progenitor cells using rotating wall
ents, especially elastin and elastic fibers, and determining vessel bioreactor. Transplant Proc 44, 1158, 2012.
the clinical effectiveness of this approach using a recon- 12. Duynstee, M.L.G., Verwoerd-Verhoef, H.L., Verwoerd,
structive animal model. C.D.A., and van Osch, G.J.V.M. The dual role of peri-
chondrium in cartilage wound healing. Plast Reconstr Surg
Acknowledgments 110, 1073, 2002.
13. Skoog, T., Ohlsén, L., and Sohn, S.A. Perichondrial potential
Funding for this work was provided by the Canadian for cartilagenous regeneration. Scand J Plast Reconstr Surg
Institutes of Health Research (CIHR) and also with the 6, 123, 1972.
ENGINEERED AURICULAR CARTILAGE CONSTRUCTS 1025

14. Togo, T., Utani, A., Naitoh, M., Ohta, M., Tsuji, Y., 31. Kuettner, K.E., Pauli, B.U., Gall, G., Memoli, V.A., and
Morikawa, N., Nakamura, M., and Suzuki, S. Identification Schenk, R.K. Synthesis of cartilage matrix by mammalian
of cartilage progenitor cells in the adult ear perichondrium: chondrocytes in vitro. I. Isolation, culture characteristics,
utilization for cartilage reconstruction. Lab Invest 86, 445, and morphology. J Cell Biol 93, 743, 1982.
2006. 32. Khan, A.A., Suits, J.M.T., Kandel, R.A., and Waldman,
15. Van Osch, G.J., Van Der Veen, S.W., Burger, E.H., and S.D. The effect of continuous culture on the growth and
Verwoerd-Verhoef, H.L. Chondrogenic potential of in vitro structure of tissue-engineered cartilage. Biotechnol Prog
multiplied rabbit perichondrium cells cultured in alginate 25, 508, 2009.
beads in defined medium. Tissue Eng 6, 321, 2000. 33. Khan, A.A., and Surrao, D.C. The importance of bicar-
16. Ruszymah, B.H., Lokman, B.S., Asma, A., Munirah, S., bonate and nonbicarbonate buffer systems in batch and
Chua, K., Mazlyzam, A.L., Isa, M.R., Fuzina, N.H., and continuous flow bioreactors for articular cartilage tissue
Aminuddin, B.S. Pediatric auricular chondrocytes gene engineering. Tissue Eng Part C Methods 18, 358, 2012.
expression analysis in monolayer culture and engineered 34. Consonni, S.R., Werneck, C.C., Sobreira, D.R., Kühne, F.,
elastic cartilage. Int J Pediatr Otorhinolaryngol 71, 1225, Moraes, S.G., Alvares, L.E., and Joazeiro, P.P. Elastic fiber
2007. assembly in the adult mouse pubic symphysis during
17. Asawa, Y., Ogasawara, T., Takahashi, T., Yamaoka, H., pregnancy and postpartum. Biol Reprod 86, 151, 2012.
Nishizawa, S., Matsudaira, K., Mori, Y., Takato, T., and 35. Jin, H., and Lewis, J.L. Determination of Poisson’s ratio of
Hoshi, K. Aptitude of auricular and nasoseptal chon- articular cartilage by indentation using different-sized in-
drocytes cultured under a monolayer or three-dimensional denters. J Biomech Eng. 126, 138, 2004.
condition for cartilage tissue engineering. Tissue Eng Part 36. Hayes, W.C., Keer, L.M., Herrmann, G., and Mockros, L.F.
A 15, 1109, 2009. A mathematical analysis for indentation tests of articular
18. Strehl, R., Schumacher, K., de Vries, U., and Minuth, W.W. cartilage. J Biomech. 5, 541, 1972.
Proliferating cells versus differentiated cells in tissue en- 37. Rosa, R.G., Akgul, Y., Joazeiro, P.P., and Mahendroo, M.
gineering. Tissue Eng 8, 37, 2002. Changes of large molecular weight hyaluronan and versi-
19. Terada, S., Fuchs, J.R., Yoshimoto, H., Fauza, D.O., and can in the mouse pubic symphysis through pregnancy. Biol
Vacanti, J.P. In vitro cartilage regeneration from proliferated Reprod 86, 1, 2012.
adult elastic chondrocytes. Ann Plast Surg 55, 196, 2005. 38. Rosa, R.G., Tarsitano, C.A., Hyslop, S., Yamada, A.T.,
20. Schulze-Tanzil, G., Mobasheri, A., de Souza, P., John, T., Toledo, O.M., and Joazeiro P.P. Relaxation of the mouse
and Shakibaei, M. Loss of chondrogenic potential in ded- pubic symphysis during late pregnancy is not accompanied
ifferentiated chondrocytes correlates with deficient Shc-Erk by the influx of granulocytes. Microsc Res Tech 71, 169,
interaction and apoptosis. Osteoarthr Cartil 12, 448, 2004. 2008.
21. Darling, E.M., and Athanasiou, K.A. Rapid phenotypic 39. Kim, Y.J., Sah, R.L., Doong, J.Y., and Grodzinsky, A.J.
changes in passaged articular chondrocyte subpopulations. Fluorometric assay of DNA in cartilage explants using
J Orthop Res 23, 425, 2005. Hoechst 33258. Anal Biochem 174, 168, 1988.
22. Darling, E.M., and Athanasiou, K.A. Retaining zonal 40. Farndale, R.W., Buttle, D.J., and Barrett, A.J. Improved
chondrocyte phenotype by means of novel growth envi- quantitation and discrimination of sulphated glycosamino-
ronments. Tissue Eng 11, 395, 2005. glycans by use of dimethylmethylene blue. Biochim Bio-
23. Walton, R.L., and Beahm, E.K. Auricular reconstruction phys Acta 883, 173, 1986.
for microtia: Part II. Surgical techniques. Plast Reconstr 41. Goldberg, R.L., and Kolibas, L.M. An improved method
Surg 110, 234, 2002. for determining proteoglycans synthesized by chondrocytes
24. Tanzer, R.C. Total reconstruction of the auricle: a 10-year in culture. Connect Tissue Res 24, 265, 1990.
report. Plast Reconstr Surg 40, 547, 1967. 42. Woessner, J.F. The determination of hydroxyproline in
25. Tanzer, R.C. Microtia-a long-term follow-up of 44 re- tissue and protein samples containing small proportions of
constructed auricles. Plast Reconstr Surg 61, 161, 1978. this imino acid. Arch Biochem Biophys 93, 440, 1961.
26. Brent, B. Technical advances in ear reconstruction with 43. Heinegard, D., Bayliss, M., and Lorenzo, P. Biochemistry
autogenous rib cartilage grafts: personal experience with and metabolism of normal and osteoarthritic cartilage. In:
1200 cases. Plast Reconstr Surg 104, 319, 1999. Brandt, K.D., Doherty, M., and Lohmander, L.S., eds.
27. Brent, B. Microtia repair with rib cartilage grafts: a review Osteoarthritis. New York: Oxford University Press, 1998,
of personal experience with 1000 cases. Clin Plast Surg 29, pp. 74–84.
257, 2002. 44. Yanaga, H., Imai, K., and Yanaga, K. Generative surgery of
28. Brenner, J.M., Kunz, M., Tse, M.Y., Winterborn, A., Bar- cultured autologous auricular chondrocytes for nasal aug-
dana, D.D., Pang, S.C., and Waldman, S.D. Development mentation. Aesthetic Plast Surg 33, 795, 2009.
of large engineered cartilage constructs from a small pop- 45. Waldman, S.D., Couto, D.C., Omelon, S.J., and Kandel,
ulation of cells. Biotechnol Prog 29, 213, 2013. R.A. Effect of sodium bicarbonate on extracellular pH,
29. Brenner, J.M., Ventura, N.M., Tse, M.Y., Winterborn, A., matrix accumulation, and morphology of cultured articular
Bardana, D.D., Pang, S.C., Hurtig, M.B., and Waldman, chondrocytes. Tissue Eng 10, 1633, 2004.
S.D. Implantation of scaffold-free engineered cartilage 46. Yanaga, H., Imai, K., Koga, M., and Yanaga, K. Cell-
constructs in a rabbit model for chondral resurfacing. Tis- engineered human elastic chondrocytes regenerate natural
sue Eng Regen Med 2013. DOI: 10.1111/aor.12199. scaffold in vitro and neocartilage with neoperichondrium in
30. Mishell, R.I., Shiigi, J.M., Mishell, B.B., Grabstein, K.H., the human body post-transplantation. Tissue Eng Part A 18,
and Shiigi, S.M. Prevention of the immunosuppressive 2020, 2012.
effects of glucocorticosteroids by cell-free factors from 47. Moskalewski, S., Adamiec, I., and Go1aszewska, A. Ma-
adjuvant-activated accessory cells. Immunopharmacology turation of rabbit auricular chondrocytes grown in vitro in
2, 233, 1980. monolayer culture. Am J Anat 155, 339, 1979.
1026 GIARDINI-ROSA ET AL.

48. Thyberg, J., and Hinek, A. Fine structure of rabbit ear 60. Bradamante, Z., Kostović-Knezevic, L., Levak-Svajger, B.,
chondrocytes in vitro and after autotransplantation. Cell and Svajger, A. Differentiation of the secondary elastic
Tissue Res 180, 341, 1977. cartilage in the external ear of the rat. Int J Dev Biol 35,
49. Moskalewski, S. Elastic fiber formation in monolayer and 311, 1991.
organ cultures of chondrocytes isolated from auricular 61. Luckenbill, L.M., and Cohen, A.S. The association of lipid
cartilage. Am J Anat 146, 443, 1976. droplets with cytoplasmic filaments in avian subsynovial
50. Moskalewski, S. The elastogenetic process in transplants adipose tissue. J Cell Biol 31, 195, 1966.
and cultures of isolated auricular chondrocytes. Connect 62. Wood, E.M. An ordered complex of filaments surrounding
Tissue Res 8, 171, 1981. the lipid droplets in developing adipose cells. Anat Rec
51. Madsen, K., von der Mark, K., van Menxel, M., and Fri- 157, 437, 1967.
berg, U. Analysis of collagen types synthesized by rabbit 63. Foster, J., Rich, C.B., and Florini, J.R. Insulin-like growth
ear cartilage chondrocytes in vivo and in vitro. Biochem J factor I, somatomedin C, induces the synthesis of tropoe-
221, 189, 1984. lastin in aortic tissue. Coll Relat Res 7, 161, 1987.
52. Pelttari, K., Steck, E., and Richter, W. The use of mesen- 64. Rich, C.B., Ewton, D.Z., Martin, B.M., Florini, J.R., Bashir,
chymal stem cells for chondrogenesis. Injury 39 Suppl 1, M., Rosenbloom, J., and Foster, J.A. IGF-I regulation of
S58, 2008. elastogenesis: comparison of aortic and lung cells. Am J
53. Chiang, H., Hsieh, C.H., Lin, Y.H., Lin, S., Tsai-Wu, J.J., Physiol 263, L276, 1992.
and Jiang, C.C. Differences between chondrocytes and 65. Wolfe, B.L., Rich, C.B., Goud, H.D., Terpstra, A.J., Bashir,
bone marrow-derived chondrogenic cells. Tissue Eng Part M., Rosenbloom, J., Sonenshein, G.E., and Foster, J.A.
A 17, 2919, 2011. Insulin-like growth factor-I regulates transcription of the
54. Lammi, P.E., Lammi, M.J., Hyttinen, M.M., Panula, H., elastin gene. J Biol Chem 268, 12418, 1993.
Kiviranta, I., and Helminen, H.J. Site-specific im- 66. Long, J.L., and Tranquillo, R.T. Elastic fiber production in
munostaining for type X collagen in noncalcified articular cardiovascular tissue-equivalents. Matrix Biol 22, 339, 2003.
cartilage of canine stifle knee joint. Bone 31, 690, 2002. 67. Venkataraman, L., and Ramamurthi, A. Induced elastic
55. Naumann, A., Dennis, J.E., Aigner, J., Coticchia, J., Ar- matrix deposition within three-dimensional collagen scaf-
nold, J., Berghaus, A., Kastenbauer, E.R., and Caplan, A.I. folds. Tissue Eng Part A 17, 2879, 2011.
Tissue engineering of autologous cartilage grafts in three- 68. Shi, J., Wang, A., Sen, S., Wang, Y., Kim, H.J., Mitts, T.F.,
dimensional in vitro macroaggregate culture system. Tissue and Hinek, A. Insulin induces production of new elastin in
Eng 10, 1695, 2004. cultures of human aortic smooth muscle cells. Am J Pathol
56. Rucklidge, G.J., Milne, G., and Robins, S.P. Collagen type 180, 715, 2012.
X: a component of the surface of normal human, pig, and
rat articular cartilage. Biochem Biophys Res Commun 224,
Address correspondence to:
297, 1996.
Stephen D. Waldman, PhD, PEng
57. Bluteau, G., Labourdette, L., Ronziere, M., Conrozier, T.,
Mathieu, P., Herbage, D., and Mallein-Gerin, F. Type X Department Mechanical & Materials Engineering
collagen in rabbit and human meniscus. Osteoarth Cartilage Queen’s University
7, 498, 1999. Kingston K7L 3N6
58. Fedewa, M.M., Oegema, T.R., Schwartz, M.H., MacLeod, Ontario
A., and Lewis, J.L. Chondrocytes in culture produce a Canada
mechanically functional tissue. J Orthop Res 16, 227, 1998. E-mail: waldman@me.queensu.ca
59. Quintarelli, G., Starcher, B.C., Vocaturo, A., Di Gianfi-
lippo, F., Gotte L., and Mecham, R.P. Fibrogenesis and Received: March 5, 2013
biosynthesis of elastin in cartilage. Connect Tissue Res 7, 1, Accepted: October 10, 2013
1979. Online Publication Date: January 21, 2014