Engineering, 2012, 5, 61-64
doi:10.4236/eng.2012.410B016 Published Online October 2012 (http://www.SciRP.org/journal/eng)
Validation of an In Vitr o Model to Study Human Cartilage
Responses to Compression
Natalia de Isla1, Céline Huselstein1, Didier Mainard1,3, Jean-Franç oi s Stoltz1,2
1CNRS UMR 7561, Biopôle, Faculté de Médecine, Université de Lorraine, Vandoeuvre, France
2UTCT, CHU de Nancy, Vandoeuvre, France
3COT, CHU de Nancy, Vandoeuvre, France
Received 2012
ABSTRACT
The aim of this work was to develop an in vitro model to study mechanical compression effects on cartilage. A pressure-controlled
compression device was used in this study. Cartilage explants obtained from human knee were compressed at 1MPa/1Hz for 7 hours
(30 min ON, 30 min OFF) under normoxia (5% CO2, 21% O2) or hypoxia (5% CO2, 5% O2). Cell viability was analyzed while nitric
oxide (NO) and glycosaminoglycans (GAG) release was assayed in culture media. Mechanical stimulation increased NO production
and GAG release by human cartilage explants under normoxia and hypoxia culture. In normoxia and hypoxia conditions, mechanical
stimulation alters human OA cartilage metabolism. There is also, an increase in matrix degradation after compression, as shown by
levels of GAG found in culture media. This study put in evidence the importance of mechanical compression in the progression of
the osteoarthritis and present and in vitro model for mechanobiological and pharmacological studies.
Keywords: Cartilage; Compression; In Vitro Model; Nitric Oxide; GAG
1. Introduction
Cartilage is an avascular tissue submitted in vivo to mechanical
stimuli. These mechanical stimuli result from a complex com-
bination of tension, shearing and compression forces, the latter
being the most important within the cartilage [1]. The compres-
sive forces exerted on the surface of the articular cartilages are
variable according to the weight of the individual, his muscular
tension and its physical-activity. Thus for example the average
pressure being exerted on the hip is of 0.7 MPa but during
physical exercises, it can reach 5 to 10 MPa. These mechanical
forces affect the extracellular matrix as well as the chondrocyte
metabolism [2-9]. During immobilization, the capacities of
synthesis of chondrocytes as well as the thickness of the carti-
lage decrease. In the same way, in the zones subjected to
maximal forces, the balance between the anabolism and the
catabolism of the cartilage are disturbed.
It is know that physiological loading of articular cartilage is
necessary to maintain normal joint function. Articular chon-
drocytes transform mechanical signals into biochemical ones to
maintain the integrity of their extracellular matrix [10-12].
Several studies investigated the effect of mechanical stimula-
tion on chondrocyte metabolism. In general, static compression
decreases biosynthetic activity compared to unloaded tissue
while dynamic compression has been found to stimulate, inhibit
or have no effect on biosynthetic activity depending on the
loading frequency and amplitude [13-15]. Other studies have
shown that cyclic tensile strains of low magnitude (3–8%
equibiaxial strain) and physiological levels of cyclic compres-
sive forces (15% compression) elicit an anabolic response [16,
17], while strains of high magnitude (10–15% equibiaxial strain)
initiate cartilage damage.
Most of the studies investigated the effects of mechanical
compression on osteoarthritis (OA) development and use
healthy cartilage principally from animal origin. In this work
we aimed to investigate the response of human cartilage from
osteoarthritic patients to dynamic unconfined compression. The
aim of this work was to develop an in vitro model to study me-
chanical compression effects on cartilage and to define experi-
mental protocols to be used in cartilage mechanobiology. This
model could be used in pathophysiological or pharmacological
studies of cartilage.
2. Materials and Methods
2.1. Carti lage Expl a nts
Articular cartilage was obtained from preserved areas of femo-
ral condyles of patients undergoing arthroplasty for OA at the
Department of Orthopaedic Surgery, CHU Nancy, France.
Samples from 10 patients (6 women and 4 men, 64  7 years)
were used. Cartilage was separated from the subchondral bone
using a scalpel. Cylindrical explants (5 mm in diameter) were
harvested using a sterile biopsy punch (Stiffel, France) and
immediately incubated in culture medium (DMEM-F12) sup-
plemented with 10% heat inactivated fetal bovine serum, 1% of
antibiotics/antimycotic solution, and 2mM Glutamine at 37°C,
5% CO2. Test and control explants were paired at harvest and
originated from adjacent sites on the joint surface. All com-
pression experiments were performed after allowing explants to
equilibrate in culture for 72 hours after harvest.
2.2. Mechanic al S tim ul a tion
A FX-4000CTM Flexcercell® Compression Plus TM System
(Flexcell International, Hillsborough, NC) was used to apply
dynamic compression. Explants were placed in one well of a

N. de ISLA ET AL.
62
Biopress culture plates (Flexcell International) and mounted
within the apparatus. The plates consist of a 6 well plates con-
taining a flexible silicone rubber membrane at the bottom. The
explant is putted on the plastic disc, into the Foam Sample
Holder and the piston of the Stationary Platen is moved until it
become in contact with the explants. A calibrated air pressure
was applied to the membrane to obtain a compressive stress (σ);
determined from the applied force (F) and the initial crosssec-
tional area (A) of the explant using the equation σ = F/A (Fig-
ure 1). Two millilitres of culture medium were introduced into
each well. Explants were subject to unconfined compression at
compressive stress amplitude of 1 MPa at 1 Hz for 7 hours in
an intermittent manner (30 min ON, 30 min OFF) in a humified
incubator at 37°C, 5% CO2. Unloaded (control) explants were
incubated under the same conditions. Tests were performed in
order to calculate the pressure applied to the sample by using a
force sensor (XFL 205 R, FGP Sensors & Instrumentation)
instead the sample and to be sure that all culture wells on each
of the 4 compression plates of the device were subject to the
same strain (Figure 1).
To analyze the effect of oxygen tension in NO production
and GAG release, experiments were performed in a humidified
incubator at 37°C, 5%CO2, 5%O2. Explants were left overnight
in hypoxia before each experiment in order to let cells to adapt
their metabolism.
2.3. Viability Assay
Cell viability was determined in cartilage explants using the
fluorescent probes Propidium Iodure (PI) and SYTO16 (both
from Molecular Probes). The membrane-permeable SYTO16
labels live and dead cells to yield cytoplasmic and nuclei green
fluorescence, whereas the membrane-impermeable propidium
iodide labels nucleic acids of membrane-compromised cells
with red fluorescence. After each experiment, the loaded and
Flexercell system calibration
y = 0 . 04 24x - 0. 024 4
R
2
= 0.99 92
y = 0. 8333 x - 0. 47 93
R
2
= 0. 9992
0
2
4
6
8
10
12
14
16
18
20
0510 1520 25
Or der pressure(kPa)
Measured strength (N)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Measured pressure ( M Pa)
Figure 1. Calibration of the compression device using a force sen-
unloa
sor.
ded (control) explants were washed in DMEM W/O phe-
the
2.5. Determination of NO
y estimating the stable NO me-
response to compression. Fluorescence staining
compression. The
m
nol red for 5 min and then sectioned perpendicular to the ar-
ticular surface into 1-mm thick slices using parallel razor blades.
The tissue sections were incubated in the SYTO16+PI solution
for 5 min in a dark environment and then washed twice (5 min
each) in DMEM W/O phenol red to remove free dye from the
tissue matrix. The chondrocytes within the cartilage matrix
were then viewed using a fluorescence confocal microscope
(LEICA) in sequential mode (excitation/emission: 488 nm/
520nm (1); 545/633 (2)) to simultaneously observe green and
red fluorescence. Green cells are viable and yellow cells are
dead. The percentage of dead cells was calculated by counting
the total number of cells and the number of yellow cells in five
random, non-contiguous fields.
2.4. Me a s ur ement of GAG Release
GAG levels in the culture medium were determined by
amount of polyanionic material reacting with DMMB (Poly-
sciences, USA). Explants supernatants were removed and 125
μl were combined with 200 µl of DMMB solution. Samples
were examined spectrophotometrically at 525 nm. For this as-
say, standards prepared with control media and chondroitin
sulphate C (Sigma, France) were used. Results are reported as
µg GAG per mg of wet weight of tissue.
NO production was measured b
tabolite, nitrite, in conditioned medium using a spectropho-
tometric method based on the Griess reaction (Griess Reagent
Kit for Nitrite Determination, Molecular Probes). Following
culture of the cartilage explants for the times indicated, 150 µl
of the culture supernatants or sodium nitrite standard dilutions
were mixed with 20 µl Griess reagent (1% sulfanilamide, 0.1%
naphthyl ethylenediamine dihydrochloride, and 5% H3PO4)
and incubated for 30 min at room temperature. Nitrite concen-
trations were determined by measuring absorbance at 570 nm in
a Microplate reader (BioRad, U.S.A.). Values are expressed as
µM nitrite released per mg of wet weight of tissue.
3. Results
Cell viability in
indicated that cell death was confined to the cut edge and to the
superficial zone in uncompressed control samples. In mechani-
cally loaded explants, cell death was evident also in the inter-
mediate region of the explants. When compared the superficial
zone in unloaded and compressed samples, the percentage of
dead cells is higher (p < 0.05) in compressed explants (59%)
than in unloaded ones (23%). Similarly, the percentage of dead
cells in the deep zone is higher (p < 0.05) in compressed sam-
ples (39%) than in unloaded samples (20%).
NO production and GAG release after
echanical compression used in this in vitro study affects the
NO production and GAG release from human cartilage explants.
As shown in Figure 2, NO increases in culture medium of
compressed explants after 7 hours of intermittent compression
when compared to uncompressed explants. In parallel, condi-
tioned medium from compressed and uncompressed cartilage

N. de ISLA ET AL. 63
explants was analyzed for sulphated GAG content after 7 hours.
The results presented in Figure 3 showed that mechanical
stimulation increase GAG release in culture medium of com-
pressed explants when compared to uncompressed ones. We
next investigated the effect of oxygen tension in the level of
NO production and GAG release in response to mechanical
compression. Under static conditions, NO production increased
under hypoxia when compared to normoxia conditions (in-
crease of 202  51 %, p0,05). Moreover, mechanical com-
pression significantly increased NO production (Figure 2) un-
der hypoxia. Furthermore, under static conditions GAG release
under hypoxia did not change when compared to normoxia
conditions. Mechanical compression significantly increased
GAG release (Figure 3) under hypoxia although the increase is
less important than under normoxia.
4. Discussion
ffect of unconfined compression on cartilage
l conditions, large forces which are the
re
In this study, the e
explants from human osteoarthritic knee was studies. Results
showed that in this in vitro model, mechanical compression
increased NO production and GAG release under normoxia and
hypoxia conditions.
Under physiologica
sult of normal joint movements are applied to articular carti-
lage. Mechanical load has been demonstrated in many in vivo
and in vitro investigations to be an important factor affecting
the health of articular cartilage and consequently the function of
Figure 2. Production of Nitrite (M/mg wet weight) by articular
cartilage explants compressed at 1 MPa,1 Hz for 7h (30 min on, 30
min off) under normoxia conditions (21% O2) or under hypoxia
conditions (5% O2). Data are presented as mean ± S.D. of 3 inde-
pendent experiments with n = 3/group/experiment, p <0.05 : w/o
compression vs compression.
Figure 3. GAG released (g/mg wet weight) in culture mediu
pression).
REFERENCES
[1] Wong M and Cilage functional histo-
morphology and rch perspective. Bone
pression modulates matrix biosynthesis in chon-
chondrocyte metabolism in agarose constructs sub-
age to mechanical stimuli in vitro. Osteoarthritis
Biomechanical signals exert
luence the metabolism of chondrocytes seeded in
arwal S. Cyclic Tensile
n vitro mechanical injury of eld-
m oS f
articular cartilage explants compressed at 1 MPa, 1 Hz for 7h (30
min on, 30 min off) under normoxia (21% O2) or hypoxia (5% O2).
Data are presented as mean ± S.D. of 3 independent experiments
with n = 3/group/experiment, p <0.05 (w/o compression vs com-
the diarthrodial joint and the progression of joint degeneration
[18,19]. The effect of the stimulation is strongly dependent on
magnitude and frequency of the applied load. In the in vitro
model used in this study, unconfined compression was used.
Under this condition, the construct is free to expand laterally
and is subject to both compressive strains (along the axial di-
rection) and tensile strains (along the radial and circumferential
directions). This loading regime represents a physiologic load-
ing environment and produces a more uniform mechanical sig-
nal throughout the thickness of a cylindrical cartilage sample
than that of confined compression (where radial expansion is
prevented).
In cartilage, chondrocytes exhibit a predominantly anaerobic
metabolism because they are in hypoxia under physiological
conditions. The hypoxic conditions are further enhanced during
OA because the oxygen consumption of the synovium from OA
patients is elevated and the synovial fluid O2 tension is de-
creased compared with that of normal synovial fluid. In conclu-
sion, the standard cell culture conditions with 21% oxygen
tension do not mimic the physiological situation within carti-
lage. In this study, mechanical compression increased NO pro-
duction and GAG release when experiences were performed at
5% O2 but we noted that the increase is lower than at 21% O2.
In conclusion, this study shows that high levels of compres-
sion increase NO production and GAG release by human OA
under normoxia and hypoxia. In an in vitro model we observed
that OA cartilage is sensitive to mechanical stimulation and
show the importance of mechanical stimulation in the progres-
sion of the osteoarthritis.
arter DR. Articular cart
mechanobiology: a resea
2003; 33:1-13.
[2] Buschmann MD, Gluzband YA, Grodzinsky AJ, Hunziker EB.
Mechanical com
drocyte/agarose culture Journal of Cell Science 1995; 108:
1497-1508.
[3] Chowdhury TT, Bader DL, Shelton JC, Lee DA. Temporal
regulation of
jected to dynamic compression. Arch Biochem Biophys 2003;
417: 105-11.
[4] Plumb MS and Aspden RM. The response of elderly human
articular cartil
and Cartilage 2005;13:1084-1091.
[5] Madhavan S, Anghelina M, Rath-Deschner B, Wypasek E, John
A, Deschner J, Piesco N, Agarwal S.
sustained attenuation of proinflammatory gene induction in ar-
ticular chondrocytes. OsteoArthritis and Cartilage 2006; 14:
1023-1032.
[6] Lee DA, Bader DL. Compressive strains at physiological fre-
quencies inf
agarose. J Orthop Res 1997;15:181-188.
[7] Gassner R, Buckley MJ, Georgescu H, Studer R, Stefano-
vich-Racic M, Piesco NP, Evans CH, Ag
tress Exerts Antiinflammatory Actions on Chondrocytes by In-
hibiting Inducible Nitric Oxide Synthase. The Journal of Immu-
nology 1999;163: 2187–2192.
[8] Démarteau O, Pillet L, Inaebnit A, Borens , Quinn TM. Biome-
chanical characterization and i

N. de ISLA ET AL.
64
, Lark MW, Grodzinsky AJ. Proteoglycan
uman hip joint measur
rtilage engineering: the under
r Functional Tissue Engineering of Articular Carti-
oxide and matrix pro-
S, Misukonis MA, Banes AJ,
F, Levy A, Pigenet A, Thirion S, Saffar
, Piesco NP, Buckley MJ. Cyclic
er DL. Dynamic mechanical
B, Kraus VB, Stabler TV, Guilak F. The
AJ, Schunke M.
erly human femoral head cartilage: comparison to adult bovine
humeral head cartilage. OsteoArthritis and Cartilage
2006 ;14:589-596.
[9] Patwari P, Cook MN, DiMicco MA, Blake SM, James IE,
Kumar S, Cole AA
Guil
degradation after injurious compression of bovine and human ar-
ticular cartilage in vitro: interaction with exogenous cytokines.
Arthritis Rheum 2003;48:1292-301.
[10] Hodge WA, Fujan RS, Carlson KL, Burgess RG, Harris WH,
Mann RW. Contact pressures in the hed in
-
Tensile Strain Suppresses Catabolic Effects of Interleukin-1b in
Fibrochondrocytes From the Temporomandibular Joint Arthritis
& Rheumatism 2001;44:608–617.
[17] Lee DA, Frean SP, Lees P, Bad
vivo. PNAS 1986;83:2879-2883.
[11] Stoltz JF, de Isla N, Huselstein C, Bensoussan D, Muller S,
Decot V. Mechanobiology and ca
lying pathophysiological phenomena. Biorheology 2006;
43:171-80.
[12] Hung CT, Mauck RL., Wang CC, Lima EG, Ateshian GA. A
Paradigm fo
lage via Applied Physiologic Deformational Loading. Annals of
Biomedical Engineering 2004;32:35–49.
[13] Lee MS, Trindade MC, Ikenoue T, Schurman DJ, Goodman SB,
Smith RL. Effects of shear stress on nitric
tein gene expression in human osteoarthritic chondrocytes in vi-
tro. J Orthop Res 2002;20:556-61.
[14] Fermor B, Weinberg JB, Pisetsky D
ak F. The effects of static and intermittent compression on
nitric oxide production in articular cartilage explants. J Orthop
Res 2001;19:729-37.
[15] Gosset M, Berenbaum
JL, Jacques C Prostaglandin E2 synthesis in cartilage explants
under compression: mPGES-1 is a mechanosensitive gene. Ar-
thritis Res Ther. 2006;8:R135.
[16] Agarwal S, Long P, Gassner R
compression influences nitric oxide production by articular
chondrocytes seeded in agarose. Biochem Biophys Res Commun.
1998;251:580-585.
[18] Piscoya JL, Fermor
influence of mechanical compression on the induction of os-
teoarthritis-related biomarkers in articular cartilage explants. Os-
teoArthritis and Cartilage 2005;13:1092-1099.
[19] Kurz B, Lemke AK, Fay J, Pufe T, Grodzinsky
Pathomechanisms of cartilage destruction by mechanical injury.
Annals of Anatomy 2005;187:473-485.