Connective Tissue Research, 51:216–223, 2010
c Informa UK Ltd.
Copyright ISSN: 0300-8207 print / 1607-8438 online
DOI: 10.3109/03008200903281683
Young Adult Chondrocytes Proliferate Rapidly
and Produce a Cartilaginous Tissue at the Gel-Media
Interface in Agarose Cultures
Nicolas Tran-Khanh, Anik Chevrier, Viorica Lascau-Coman,
Caroline D. Hoemann, and Michael D. Buschmann
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
Institute of Biomedical Engineering and Department of Chemical Engineering, Ecole Polytechnique,
Montreal, Quebec, Canada
Primary chondrocytes cultured in agarose can escape the gel,
accumulate at the interface between agarose and the culture
medium, and form an outgrowing tissue. These outgrowths can
appear as voluminous cartilage-like nodules that have never been
previously investigated. In the present study, bovine articular
chondrocytes from three age groups (fetal, young adult, aged)
were seeded and cultured in agarose to test the hypothesis that
hyaline-like cartilage outgrowths develop at the interface by
appositional growth, in an age-dependant manner. Macroscopic
appearance, cell content, cell division, cytoskeletal morphology,
and extracellular matrix (ECM) composition were analyzed. Fetal
chondrocytes produced a fibrous interfacial tissue while aged
chondrocytes produced ECM-poor cell clusters. In contrast young
adult chondrocytes produced large cartilaginous outgrowths, rich
in proteoglycan and collagen II, where cells in the central region
displayed a chondrocyte morphology. Cell proliferation was confined to the peripheral edge of these outgrowths, where elongated
cell morphology, cell-cell contacts, and cell extensions toward the
culture medium were seen. Thus these voluminous cartilaginous
outgrowths formed in an appositional growth process and only
for donor chondrocytes from young adult animals. This system
offers an interesting ability to proliferate chondrocytes in a manner
that results in a chondrocyte morphology and a cartilaginous
ECM in central regions of the outgrowing tissue. It also provides
an in vitro model system to study neocartilage appositional
growth.
Keywords
Chondrocytes; Cartilage; Tissue Engineering; Agarose
Culture; Phenotype
Received 3 May 2009; Revised 3 August 2009; Accepted 21 August
2009.
Address correspondence to Michael D. Buschmann, PhD, Chemical and Biomedical Engineering, Ecole Polytechnique, PO 6079
Station Centre-ville, Montreal Qc, Canada H3C 3A7. E-mail:
[email protected]
INTRODUCTION
Culture in agarose gel preserves chondrocyte phenotype
[1]. However, fibroblast-like cells are frequently observed at
the surface of the hydrogel when chondrocytes from 1- to
2-week-old calves are cultured following their encapsulation
[2–4]. The presence of these elongated outgrowing cells is
usually considered undesirable since they produce a fibrous
tissue and confound analyses of bulk properties of the constructs.
In contrast to these fibrous outgrowths, cartilage-like outgrowths
have been noted at the agarose-media interface in cultures of
chondrocytes from older 4- to 6-month-old animals [5]. These
cartilagenous nodules have never been characterized in terms
of cell morphology, composition of the extracellular matrix
(ECM), or nature of their growth processes. In addition, this
interesting interfacial culture environment has not been fully
appreciated for its practical importance, since it may provide
an alternative to existing culture systems rather than just being
viewed as a culture artifact.
In the present study, based on the above observations, we
initially examined the age-dependence of the interfacial outgrowths, using primary bovine articular chondrocytes seeded in
agarose. Then, we tested the hypothesis that the large cartilagelike outgrowths contain cells with chondrocyte morphology and
a cartilagenous ECM. Our final hypothesis was that chondrocyte
growth at the agarose-media interface occurs in an appositional
manner rather than interstitially, i.e., proliferation preferentially
occurs at the outgrowth periphery rather than at more central
locations.
MATERIALS AND METHODS
Reagents were obtained from Sigma-Aldrich (St. Louis, MO,
USA) unless otherwise indicated.
Chondrocytes Cultured in Agarose
Articular cartilage was removed from humeral heads of Fetal
(last trimester), Young (1 to 2-year-old), and Aged (5–7 year-old)
216
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
CHONDROCYTES IN AGAROSE CULTURES
bovine shoulders (3–4 joints per age group) and finely chopped
using a scalpel. As previously described [6], chondrocytes were
released by sequential digestions with protease type XIV and
collagenase CLS2 (Worthington Biochemical, Lakewood, NJ,
USA), filtered, seeded at ∼1 × 107 cells/mL in 2% (w/v) low
melting temperature agarose at 37◦ C, and poured into a 100-mm
Petri dish. Agarose concentration was chosen in accordance
with previous studies [2–5]. After gelling at room temperature
for 5 min and at 4◦ C for 1 hr, disks (6 mm diameter, 1.8 to
2.0 mm thick) were cored from the solidified slab gel using
a stainless steel biopsy punch and cultured in 48-well culture
plates (Corning, Acton, MA, USA) for up to 1 month, on top of
1-mm thick nylon mesh (Spectrum, Rancho Dominguez, CA,
USA).
Each day, samples were transferred to a new plate filled
with complete media: high glucose DMEM (Gibco), 3.7 g/L
sodium bicarbonate, 1×penicillin-streptomycin solution, 10%
fetal bovine serum, nonessential amino acids (1×NAA), 2 mM
L-glutamine, and 0.4 mM L-proline. Media were supplemented
with 30 µg/mL fresh sodium L-ascorbate beginning on day 2 of
culture.
Macroscopic Appearance of Outgrowths
On day 26, samples were fixed in fresh 4% (w/v)
paraformaldehyde in 0.1 M sodium cacodylate (Fisher, Ottawa,
Canada), pH 7.3. For samples to be stained for proteoglycan
(PG), 2.5% (w/v) cetylpyridinium chloride (CPC) was added to
promote PG retention. Low-magnification images of the fixed
disks were acquired with a Stemi 2000-C stereomicroscope
(Zeiss Canada, Toronto, ON, Canada).
ECM Staining
Fixed samples were embedded in paraffin and sectioned
(6 µm thick). Sections were stained for PGs with 0.2% (w/v)
Safranin-O and counterstained with 0.04% (w/v) Fast Green
FCF and Weigert iron hematoxylin. For collagen immunodetection, sections were subjected to antigen retrieval by heating
at 60◦ C in 10 mM Tris pH 10. Enzymatic treatments were
as follow: 0.1% (w/v) pronase (EC 3.4.24.31) in phosphatebuffered saline (PBS) for 12 min at RT (for collagen II) or
0.25% (w/v) trypsin (EC 3.4.21.4) in Tris-buffered saline (TBS)
for 30 min at RT (for collagen I), then 2.5% (w/v) hyaluronidase
from bovine testes (EC 3.2.1.35) in PBS for 30 min at 37◦ C.
Sections were immersed for 1 h at RT in a blocking solution of
20% (v/v) goat serum (GS) and 0.1% (v/v) Triton-X100 (TX) in
PBS. Immunodetection was performed as previously described
[7]. Antibodies were diluted with 10% (v/v) GS/0.1% (v/v) TX
in PBS: 10× for anticollagen II (II-II6B3; DSHB, University of
Iowa, IA, USA), 50× for anticollagen I (I-8H5; VWR, Montreal,
QC, Canada), and 200× for biotinylated goat antimouse IgG
(Fab specific). Detection was performed with the Vectastain
ABC-AP system and the Alkaline phosphatase red substrate
kit added with Levamisole solution (Vector Laboratories,
217
Burlington, Ontario, Canada). Sections were counterstained
with Weigert Iron Hematoxylin.
Analysis of Cell Number Inside Agarose and in
Outgrowths
Cell number inside the agarose and in outgrowths was
estimated via DNA content as previously described [8]. Disks
were removed from culture on days 1, 13, and 26; weighed;
and stored frozen (−80◦ C). Papain digestion was performed
(125 µg/mL papain type III and 5 mM L-cysteine in 100 mM
phosphate buffer with 10 mM EDTA, pH 6.5, at 60◦ C) during
16 hr, leading to a complete digestion of the outgrowths (which
are exterior to the agarose disks) and release of DNA originating
from the outgrowths and the digestion of the pericellular matrix,
while leaving the DNA entrapped inside the undigested agarose
disk, for the cells located in the gel. The agarose disks were
transferred to an empty tube and crushed with a pestle. The
agarose was then melted in phosphate buffer with EDTA (PBE)
at 70◦ C for 10 min to free the DNA from digested cells that
were entrapped inside agarose.
Aliquots from each extract (the papain solution containing
the DNA from the outgrowths, the melted agarose containing the
DNA from the disks) were separately diluted in PBE buffer to
quantify DNA content by the fluorometric Hoechst 33258 assay.
Aliquots from identically processed acellular disks were added
to standard curves to account for any interference from agarose.
Cell number was extrapolated from DNA content assuming
7.7 pg of DNA per cell [9].
Cell Expansion Ratios
Cell ratios (R) that approximate cell expansion ratio at
day 26 were evaluated for the constructs seeded with young
adult chondrocytes. We calculated the ratio, Rin , of cells inside
agarose to the initial cell number at seeding (Rin = Cellsin
(day 26)/Cellsseeded (day 0)) and the ratio, Rout , of cells in the
outgrowth to the initial cell number at seeding (Rout = Cellsout
(day 26)/Cellsseeded (day 0)). Rin would correspond to a cell
expansion ratio if all seeded cells stayed in the agarose, while
Rout would correspond to a cell expansion ratio if all seeded
cells went to the outgrowth.
Given that only a limited number of cells that are near
the surface of agarose can actually reach the interface and
contribute to the outgrowths, we also calculated a cell ratio
representing an effective cell expansion ratio for the outgrowths:
Reff = Cellsout /Cellsshell . Cellsshell were the number of seeded
cells located in an outer shell of the agarose that has thickness
δ (see Figure 4A), which was varied between 10 and 100 µm
to represent a range of physically reasonable values; our
histological observations showed cells accessing the peripheral
surface from several tens of microns inside (as in Figure 2A).
Cellsshell was calculated as Cellsshell = Cellsseeded ×
(Vshell /Vtot ) Vtot being the total volume of the disk, and Vshell
218
N. TRAN-KHANH ET AL
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
the volume of the outer shell. With h and r the height and radius
of the disk, Vtot = hπ r2 and Vshell = hπ r2 − (h-2δ)π (r-δ)2 .
Cytoskeleton Imaging
Triple labeling was used, based on a previously described
method [10]. The buffer was a modified Hank’s balanced salt
solution (mHBSS) [10], unless otherwise mentioned. After 30
days in culture, live samples were fixed for 40 min in 0.6%
(v/v) glutaraldehyde and 5% (v/v) TX, at 37◦ C. To block
autofluorescence, samples were immersed twice in 5 mg/mL
NaBH4 for 10 min on ice. Outgrowths were transversally cut
through the center to increase antibody accessibility to cells.
Outgrowth halves were then digested at 37◦ C for 16 hr with
200 mU/mL chondroitinase ABC and 400 mU/mL keratanase
(MJS Biolynx, Brockville, ON, Canada) in digestion buffer of
100 mM NaCl, 0.01% (m/v) BSA, and 100 mM Tris, pH 7.4.
The following steps were then performed at RT. Samples
were kept for 1 hr in blocking solution of 10% (v/v) GS/0.1%
(v/v) Tween-20. Actin and vimentin were labeled for 4 hr with
Alexa Fluor 488 phalloidin (Invitrogen) and Cy3-conjugated
antivimentin monoclonal antibody (Clone V9), diluted 40× and
600×, respectively in 1% (w/v) BSA/0.05% (v/v) Tween-20.
RNA was digested by 1 mg/mL RNase A (Qiagen, Mississauga,
Ontario, Canada) solubilized in a solution of 1% (w/v) BSA
in PBS. Nuclear DNA was then stained with 3 µM TOTO-3
(Invitrogen) in PBS for 2 hr. Samples were finally immersed
for 30 min in glucose oxidase/catalase antifading reagent and
mounted with Mowiol. Images were acquired with a LSM 510
META Axioplan 2 confocal laser scanning microscope (Carl
Zeiss Canada).
Cell Proliferation Analyses via Ki-67
Immunohistochemistry and BrdU Incorporation
For Ki-67 immunohistochemistry, samples after 22 and
26 days in culture were rinsed with PBS, fixed in 4% w/v
paraformaldehyde in 0.1 M sodium cacodylate pH 7.3, cryoprotected in graded sucrose followed by graded sucrose-OCT
solutions, and cryosectioned to generate 6 to 8 µm sections.
Sections were digested with 0.25 U/mL chondroitinase (C-2905,
Sigma-Aldrich) in PBS for 1 hr at 37◦ C. Antigen retrieval was
performed with a microwave oven by heating twice for 1 min,
at a temperature below boiling, sections immersed in 10 mM
citrate buffer, pH 6.1, followed by cooling for 20 min at RT.
Sections were permeabilized with 0.4% (v/v) TX in PBS, for
20 min at RT, incubated for 20 min in 3% H2 O2 in PBS to
block endogenous peroxidase, and blocked for 1 hr at RT with
5% (w/v) milk powder and 3% (w/v) BSA in PBS. Mouse
anti-Ki-67 IgG1 (MAB4062, Chemicon, Temecula, CA, USA)
was diluted 50× in 1% (w/v) BSA in PBS and applied to sections
overnight at 4◦ C. Sections were incubated for 1 hr at RT in
biotinylated goat antimouse IgG (Fab specific), diluted 200× in
1% (m/v) BSA in PBS, followed by detection with the Vectastain
Elite ABC kit (Vector Laboratories) and the diaminobenzidine
enhanced liquid substrate system. Finally, sections were rinsed
with deionized water, dehydrated, and mounted with Permount
mounting medium (Fisher).
BrdU incorporation was performed by adding 10 µM to
culture media and incubating for 48 hr on days 19 and 20
and then chased in fresh media on day 21 for 24 hr. On day
22, samples were fixed, cryosectioned, and stained for nuclear
BrdU by immunohistochemistry as described above for Ki-67,
except that antigen retrieval in citrate buffer was replaced by
DNA denaturation in 2 N HCl, by immersion for 30 min at
37◦ C. Endogenous peroxidase was blocked with a solution of
4 parts of methanol to 1 part 3% H2 O2 solution. The primary
mouse anti-BrdU antibody (clone BU-33) was diluted 2000×.
FIG. 1. Macroscopic appearance of chondrocyte/agarose disks after 26 days
in culture following initial seeding with bovine articular chondrocytes from aged
(A and E), fetal (B and F), and young adult (D and H) animals. Young adult
chondrocytes were also cultured without ascorbic acid as controls (C and G).
Disks were fixed without CPC (A–D) or with CPC (E–H). Significant nodular
tissue outgrowths (arrowheads) were observed with young adult chondrocytes
only (D, H). Scale bars = 1 mm.
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
CHONDROCYTES IN AGAROSE CULTURES
219
FIG. 2. Paraffin sections of agarose disks seeded with young adult (A, B) and fetal (C–G) bovine chondrocytes after 26 days in culture. (A) Safranin-O/Fast-Green
staining and (B) collagen type II immunostaining show impressive cartilaginous outgrowths (OUT) developed outside the agarose gel (AG) by young adult
chondrocytes. Cells in contact with both the agarose gel and the outgrowth (arrowheads in insets) were observed at the interface. Fetal chondrocytes were
fibroblast-like at the disk surface as shown by the Safranin-O/Fast-Green staining (arrows in C and D) and could produce a fibrous tissue as shown by the collagen
type I immunostaining (E). Fetal chondrocytes also produced sparsely distributed small outgrowths (stars), which (F) always stained positive for collagen type I
and (G) sometimes for collagen type II. Scale bars = 100 µm.
Negative controls had omission of the primary antibody or the
complete omission of BrdU in culture media.
Statistical Analysis
Statistical analysis was performed with Statistica 6.1 (StatSoft, Tulsa, OK, USA) where significant differences were
determined by the Tukey HSD post-hoc test, following ANOVA.
RESULTS
Outgrowths Characteristics Depend on Age of Donor
Animal
Outgrowths from aged cells were barely visible (Figure 1A,
1E) while fetal cells produced flattened fibrous outgrowths giving a slightly dimpled appearance to the disk surface (Figure 1B,
1F). The outgrowths produced by young adult cells were the
most voluminous and had the appearance of dome-shaped
cartilaginous nodules (Figure 1D, 1H, arrowheads).
All outgrowths produced by young adult animals were
GAG-rich (Figure 2A, OUT) and stained positive for collagen
II (Figure 2B) while only in rare cases could a faint signal for
collagen type I be detected (not shown). Cell clusters in contact
with the agarose gel on one side and with the outgrowing tissue
on the other side were observed at the interface in these cultures
(arrowheads, insets of Figure 2A, 2B). At the interface, fetal cells
mostly accumulated as layers of elongated cells (Figure 2C,
2D) sometimes accompanied by a fibrous tissue (Figure 2E),
or produced outgrowths that showed a positive staining for
collagen I (Figure 2F) with an occasional staining for collagen
type II (Figure 2G). Collagen II and GAGs were detected
pericellularly (Figure 2G, 2C, respectively) inside the agarose,
while collagen type I was not (Figure 2E, 2F), indicating that
the age-dependent phenotype modulation was specific to the
outgrowth region. Aged cells produced microscopic aggregates
that were poor in ECM (not shown).
Cell number inside the agarose just after cell seeding
(Figure 3A, day 1) showed no significant difference between age
groups. At the end of the culture, cell number inside the agarose
was the highest for fetal cells and the lowest for aged cells
(Figure 3A, day 26). DNA was undetectable in the papain digest
at the beginning of the culture (Figure 3B, day 1), confirming the
absence of cells outside the agarose disk just after seeding. At
the end of the culture, the number of cells in outgrowths external
to agarose was highest for young adult cells, next highest for
fetal cells, and lowest for aged cells (Figure 3B, day 26).
Cell ratios that approximate expansion ratios inside agarose
and in the outgrowths were evaluated as described in Material
and Methods section. Rout was higher than Rin for young adult
samples (Figure 4B), but not for the other age groups (not
shown). Both Rout and Rin were relatively low for young adult
chondrocytes (3.0 and 1.8, respectively). However, the effective
expansion ratio for chondrocytes contributing to the outgrowth,
Reff , was well above 10 (Figure 4B, Reff ) suggesting an
impressive expansion capability for young adult chondrocytes at
the interface of the agarose construct. This ratio is calculated by
220
N. TRAN-KHANH ET AL
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
finding since it was mainly detected in this superficial layer
(Figure 6D, arrowheads), although some cells in deeper more
central zones displayed BrdU staining (Figure 6D, arrows). The
higher number of positive cells for BrdU comes from the fact that
it represents all the cells that divided during the incorporation
period (48 hr), whereas Ki-67 is only present in dividing cells
at the time of fixation.
FIG. 3. Cell number (mean ± SD, n = 4) inside agarose (A) and in the
outgrowth (B) for the three different age groups. † indicates significantly higher
(p < 0.05) than the same age group on day 1. ‡ indicates significantly higher (p
< 0.05) than the same age group on day 13. ∗ indicates a significant difference
(p < 0.05) between two age groups, on the same day. ND: nondetectable.
DISCUSSION
As cells divide and accumulate ECM inside agarose, a portion
of them that are close to the sample surface reach the interface in
contact with culture media. Access to this interface is more likely
facilitated by the disruption of the agarose structure provoked
by cell and ECM growth rather than by cell migration through
the agarose gel, since the pore size in 2% agarose is ∼100 nm
[13]. Curiously, cell accumulation outside the agarose did not
show the same age-dependence as within the gel, since cells
from young adult donors had the greatest proliferation in the
outgrowth (Figure 3B) while those from fetal donors had the
greatest proliferation inside agarose (Figure 3A). This result
concurred with macroscopic and histological observations of
the outgrowths where chondrocytes from young adult animals
(1–2 years old) produced the largest outgrowths (Figure 1),
which were also rich in GAG and collagen II (Figure 2), and
macroscopically resembled those seen previously [5] using
assuming that only the cells within a distance δ from the interface
could contribute to the cells in the outgrowth (Figure 4A,
dark grey zone of thickness δ). We did not determined this
distance exactly, but histological observations (Figure 2) suggest
that δ is unlikely to be higher than several tens of micrometers.
Spatial Patterns of Cell Morphology and Proliferation in
Cartilaginous Outgrowths
In central regions of large outgrowths, cells were polygonal or
spherically shaped and separated from one another (Figure 5A).
Actin in this central region did not form stress fibers but was
dense, punctuate, and cortically located within cells (Figure 5A,
5B, in green) similar to its structure in chondrocytes of cartilage
explants observed previously [11, 12]. Vimentin staining in this
central region displayed a cytoplasmic mesh spanning from the
plasma membrane to the nucleus (Figure 5A, 5B, in yellow).
In contrast, cells in the superficial and most peripheral layer
of large outgrowths were elongated and in contact with each
other (Figure 5C). Vimentin (in yellow) was either distributed in
the cytoplasm (Figure 5D) or concentrated around the nucleus
(Figure 5F), and actin staining (in green) sometimes showed
stress fibers along cell bodies (Figure 5D, arrowhead), as well
as large processes reaching outside the tissue (Figure 5E, 5F,
arrows).
The proliferation marker Ki-67 was only detected in cells
at the superficial and most peripheral surface, of both nascent
(Figure 6A, arrowheads) and more developed (Figure 6B, C,
arrowheads) outgrowths. Incorporated BrdU corroborated this
FIG. 4. (A) Schematic representation of a vertical section from the center
of the cylindrical construct with diameter 2r and height h. An outer shell of
thickness δ from which encapsulated cells can escape to reach the interface is
shown (B) Cell ratios show Rout > Rin , demonstrating greater cell proliferation
at the interface than inside the gel. An effective cell ratio (Reff ) of the cell number
in the outgrowth to the cell number at seeding in a shell of thickness δ of the
agarose disk was calculated to approximate the cell expnsion ratio for cells that
have contributed to the outgrowth.
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
CHONDROCYTES IN AGAROSE CULTURES
221
FIG. 5. Confocal imaging of actin (green, color online), vimentin (yellow), and the nucleus (red) of an outgrowth exterior to an agarose gel seeded with
chondrocytes from a young adult animal and cultured for 30 days. Chondrocytes from central regions of the outgrowth bear chondrocyte cytoskeletal characteristics
(A and B) while those in more peripheral regions (C–F) of the outgrowth display cell-cell contacts, actin stress fibres (arrowhead), and actin-rich cell extensions
(arrows). S = surface of the outgrowth. Scale bars = 15 µm.
chondrocytes from 4- to 6-month-old steers. The superficial
layer of proliferating cells in these large outgrowths bore
similarity to the proliferating zone in epiphyseal growth plates,
where chondrocytes divide at the highest rate in this tissue [14].
Such a structure is also present in the superficial zone of articular
cartilage during development and furnishes the cellular pool of
proliferating chondrocytes implicated in the appositional growth
at the articular surface [15, 16].
Fetal chondrocytes produced more fibrous outgrowths,
resembling observations made previously [2–4] when using
chondrocytes from young 1- to 2-week-old calves. The exact
mechanisms by which the age-related differences in outgrowth
phenotype arise remain to be elucidated. Some possibilities
include an age-dependence in the response to exogenous factors
present in the culture medium, in the expression levels of
endogenous factors from donor chondrocytes [17], and in
the synthesis and accumulation of ECM molecules [18–20].
Chondrocytes from aged animals (5–7 years old) produced
very small and ECM-poor outgrowths, reflecting low levels
of proliferation and reduced ECM synthesis that agree with
previous in vitro findings [6, 19, 21–25].
Preliminary results suggest that hypertrophy did not take
place within the outgrowths (no collagen X was detected by
immunohistochemistry, data not shown). However, it would be
interesting to investigate chondrocyte maturation by culturing
the outgrowths in presence of factors known to induce articular
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
222
N. TRAN-KHANH ET AL
FIG. 6. Immunocytochemical detection of cell proliferation markers (Ki-67 and BrdU) in outgrowths (OUT) exterior to the agarose gel (AG), which was seeded
with chondrocytes from young adult animals and cultured for 22 (A) or 26 (B) days. Ki-67 indicated proliferation in peripheral zones only, for both nascent (A)
and more developed outgrowths (B). (C) Higher magnification of a Ki-67-positive cell located in the rectangle of (B). (D) BrdU (10 µM) incorporation over 48 hr
indicates a greater number and region of proliferating cells, also primarily located at the periphery. Arrowheads indicate positive cells located at the periphery of
outgrowths and arrows in deeper regions. Scale bars = 30 µm.
chondrocytes hypertrophy in vitro, like thyroid hormones
[26]. It also would be interesting to determine by detection
of specific phenotypic markers whether the outgrowing fetal
cells are dedifferentiated or more like prechondrogenic cells,
and to verify if they can express a chondrocyte phenotype
when reseeded in agarose or allowed to mature further in
outgrowths.
It is generally accepted that the culture of isolated primary
chondrocytes in three dimensional environments promotes
chondrocyte phenotype and accumulation of an ECM rich in
GAGs and collagen II. However, cell expansion in these systems
is very low and far below the level required for chondrocyte
production in cell therapy [27] or to study growth processes, as
discussed elsewhere [28, 29]. Multistep procedures are generally
required to increase cell number more than 10-fold, often
leading to the use of monolayer cultures which unfortunately
induces dedifferentiation [1] and irreversible collagen type IX
downregulation [30]. An alternative is successive cycles of
encapsulation/culture/recovery using alginate beads [29].
In the present study, we observed a typical small degree
of cell expansion inside agarose (Figure 4B, Rin ). However,
when considering that only the chondrocytes seeded near
the construct surface contribute cells to the outgrowth, we
calculated a relatively high cell expansion ratio (Reff >10
for young adult samples, Figure 4B) at the agarose/media
interface. Thus, a particular environment seems to exist at this
surface, which activates cell division concomitantly with the
production of a cartilaginous ECM (Figure 2A, 2B) at least
for chondrocytes from young adult donors. Another study also
suggests that interface culture may aid phenotypically stable cell
expansion for chondrocytes [28]. In this latter study, primary
porcine chondrocytes from 8- to 12-month-old donors, at low
concentration, are brought into porous alginate sponges by
capillary action, attach on the internal cavity walls, and grow in
cell clusters able to accumulate PG and collagen II [28]. It thus
appears that interface culture systems, in which the initial seeded
chondrocytes are individually anchored in a stable phenotypic
state and the daughter cells are free to proliferate with minimal
space confinement, may provide an effective means to both
multiply chondrocytes and grow neocartilaginous tissue constructs. Nonetheless, it is important to point out that the hyaline
cartilaginous tissue observed at the agarose/media interface and
in the alginate sponges [28] were produced by bovine or porcine
chondrocytes from young adult animals. Equivalent results
using chondrocytes from aged human patients have not been
obtained to date, and thus the potential of interfacial systems to
generate an engineered tissue for autologous transplantation in
adult humans is yet to be demonstrated.
Taken together, our results indicate that chondrocyte culture
at a gel/media interface bears several interesting features,
including high levels of cell proliferation combined with the
generation of a hyaline cartilagenous extracellular matrix. This
system can be further exploited to study biological, chemical,
and physical factors that influence chondrocyte growth and
CHONDROCYTES IN AGAROSE CULTURES
neocartilage formation during development and tissue repair
processes.
Connect Tissue Res Downloaded from informahealthcare.com by Ecole Polytechnique-Crea on 08/16/10
For personal use only.
ACKNOWLEDGMENTS
This study was supported by the Canadian Institutes of Health
Research (CIHR) and N. Tran-Khanh had fellowship support
from the Fonds pour la formation de chercheurs et l’aide à
la recherche du Québec and from the CIHR strategic training
program in skeletal health research.
Declaration of Interest
The authors report no conflicts of interest. The authors alone
are responsible for the content and writing of the paper.
REFERENCES
1. Benya, P.D., and Shaffer, J.D. (1982). Dedifferentiated chondrocytes
reexpress the differentiated collagen phenotype when cultured in agarose
gels. Cell, 30(1), 215–224.
2. Buschmann, M.D., Gluzband, Y.A., Grodzinsky, A.J., Kimura, J.H.,
and Hunziker, E.B. (1992). Chondrocytes in agarose culture synthesize
a mechanically functional extracellular matrix. J. Orthop. Res., 10(6),
745–758.
3. Kisiday, J.D., Kurz, B., DiMicco, M.A., and Grodzinsky, A.J. (2005).
Evaluation of medium supplemented with insulin-transferrin-selenium for
culture of primary bovine calf chondrocytes in three-dimensional hydrogel
scaffolds. Tissue Eng., 11(1–2), 141–151.
4. Hoemann, C.D., Sun, J., Legare, A., McKee, M.D., and Buschmann, M.D.
(2005). Tissue engineering of cartilage using an injectable and adhesive
chitosan-based cell-delivery vehicle. Osteoarthr. Cart., 13(4), 318–329.
5. Mauck, R.L., Wang, C.C., Oswald, E.S., Ateshian, G.A., and Hung, C.T.
(2003). The role of cell seeding density and nutrient supply for articular
cartilage tissue engineering with deformational loading. Osteoarthri. Cart.,
11(12), 879–890.
6. Tran-Khanh, N., Hoemann, C.D., McKee, M.D., Henderson, J.E., and
Buschmann, M.D. (2005). Aged bovine chondrocytes display a diminished
capacity to produce a collagen-rich, mechanically functional cartilage
extracellular matrix. J. Orthop. Res., 23(6), 1354–1362.
7. Chevrier, A., Rossomacha, E., Buschmann, M.D., and Hoemann, C.D.
(2005). Optimization of histoprocessing methods to detect glycosaminoglycan, collagen type II, and collagen type I in decalcified rabbit osteochondral
sections. J. Histotechnol., 28(3), 165–175.
8. Hoemann, C.D., Sun, J., Chrzanowski, V., and Buschmann, M.D. (2002). A
multivalent assay to detect glycosaminoglycan, protein, collagen, RNA, and
DNA content in milligram samples of cartilage or hydrogel-based repair
cartilage. Anal. Biochem., 300(1), 1–10.
9. Kim, Y.J., Sah, R.L., Doong, J.Y., and Grodzinsky, A.J. (1988). Fluorometric assay of DNA in cartilage explants using Hoechst 33258. Anal.
Biochem., 174(1), 168–176.
10. Blanc, A., Tran-Khanh, N., Filion, D., and Buschmann, M.D. (2005).
Optimal processing method to obtain four-color confocal fluorescent images
of the cytoskeleton and nucleus in three-dimensional chondrocyte cultures.
J. Histochem. Cytochem., 53(9), 1171–1175.
11. Langelier, E., Suetterlin, R., Hoemann, C.D., Aebi, U., and Buschmann,
M.D. (2000). The chondrocyte cytoskeleton in mature articular cartilage:
structure and distribution of actin, tubulin, and vimentin filaments. J.
Histochem. Cytochem., 48(10), 1307–1320.
12. Durrant, L.A., Archer, C.W., Benjamin, M., and Ralphs, J.R. (1999).
Organisation of the chondrocyte cytoskeleton and its response to changing
mechanical conditions in organ culture. J. Anat., 194, 343–353.
223
13. Narayanan, J., Xiong, J.Y., and Lui, X.Y. (2006). Determination of agarose
gel pore size: absorbance measurements vis a vis other techniques. J. Phys.
Conf. Ser., 28, 83–86.
14. Hunziker, E.B. (1988). Growth plate structure and function. Pathol.
Immunopathol. Res., 7(1–2), 9–13.
15. Hayes, A.J., MacPherson, S., Morrison, H., Dowthwaite, G., and Archer,
C.W. (2001). The development of articular cartilage: evidence for an
appositional growth mechanism. Anat. Embryol. (Berl.), 203(6), 469–479.
16. Hunziker, E.B., Kapfinger, E., and Geiss, J. (2007). The structural
architecture of adult mammalian articular cartilage evolves by a synchronized process of tissue resorption and neoformation during postnatal
development. Osteoarthr. Cart., 15(4), 403–413.
17. Chen, A.L., Fang, C., Liu, C., Leslie, M.P., Chang, E., and Di Cesare, P.E.
(2004). Expression of bone morphogenetic proteins, receptors, and tissue
inhibitors in human fetal, adult, and osteoarthritic articular cartilage. J.
Orthop. Res., 22(6), 1188–1192.
18. Bolton, M.C., Dudhia, J., and Bayliss, M.T. (1999). Age-related changes
in the synthesis of link protein and aggrecan in human articular cartilage:
implications for aggregate stability. Biochem. J., 337(pt 1), (77–82.
19. Kamada, H., Masuda, K., D’Souza, A.L., Lenz, M.E., Pietryla, D., Otten,
L., and Thonar, E.J. (2002). Age-related differences in the accumulation
and size of hyaluronan in alginate culture. Arch. Biochem. Biophys., 408(2),
192–199.
20. Roughley, P.J., Melching, L.I., and Recklies, A.D. (1994). Changes in the
expression of decorin and biglycan in human articular cartilage with age
and regulation by TGF-beta. Matrix Biol., 14(1), 51–59.
21. Hauselmann, H.J., Aydelotte, M.B., Schumacher, B.L., Kuettner, K.E.,
Gitelis, S.H., and Thonar, E.J. (1992). Synthesis and turnover of proteoglycans by human and bovine adult articular chondrocytes cultured in alginate
beads. Matrix., 12(2), 116–129.
22. Martin, J.A., Ellerbroek, S.M., and Buckwalter, J.A. (1997). Age-related
decline in chondrocyte response to insulin-like growth factor-I: the role of
growth factor binding proteins. J. Orthop. Res., 15(4), 491–498.
23. Plaas, A.H., and Sandy, J.D. (1984). Age-related decrease in the linkstability of proteoglycan aggregates formed by articular chondrocytes.
Biochem. J., 220(1), 337–340.
24. Ribault, D., Habib, M., Abdel-Majid, K., Barbara, A., and Mitrovic,
D. (1998). Age-related decrease in the responsiveness of rat articular
chondrocytes to EGF is associated with diminished number and affinity
for the ligand of cell surface binding sites. Mech. Ageing. Dev., 100(1),
25–40.
25. Verbruggen, G., Cornelissen, M., Almqvist, K.F., Wang, L., Elewaut, D.,
Broddelez, C., de Ridder, L., and Veys, E.M. (2000). Influence of aging on
the synthesis and morphology of the aggrecans synthesized by differentiated
human articular chondrocytes. Osteoarthr. Carti., 8(3), 170–179.
26. Rosenthal, A.K., and Henry, L.A. (1999). Thyroid hormones induce features
of the hypertrophic phenotype and stimulate correlates of CPPD crystal
formation in articular chondrocytes. J. Rheumatol., 26(2), 395–401.
27. Brittberg, M., Lindahl, A., Nilsson, A., Ohlsson, C., Isaksson, O., and
Peterson, L. (1994). Treatment of deep cartilage defects in the knee
with autologous chondrocyte transplantation. N. Engl. J. Med., 331(14),
889–895.
28. Yen, C.N., Lin, Y.R., Chang, M.D., Tien, C.W., Wu, Y.C., Liao, C.J., and
Hu, Y.C. (2008). Use of porous alginate sponges for substantial chondrocyte
expansion and matrix production: effects of seeding density. Biotechnol.
Prog., 24(2), 452–457.
29. Lee, D.A., Reisler, T., and Bader, D.L. (2003). Expansion of chondrocytes
for tissue engineering in alginate beads enhances chondrocytic phenotype
compared to conventional monolayer techniques. Acta Orthop. Scand.,
74(1), 6–15.
30. Zaucke, F., Dinser, R., Maurer, P., and Paulsson, M. (2001). Cartilage
oligomeric matrix protein (COMP) and collagen IX are sensitive markers
for the differentiation state of articular primary chondrocytes. Biochem. J.,
358, 17–24.