Hypoxic resistance to articular chondrocyte
apoptosis – a possible mechanism of maintaining
homeostasis of normal articular cartilage
J W. Seol
1
, H B. Lee
1
, Y J. Lee
1
, Y H. Lee
2
, H s. Kang
1
, I s. Kim
1
, N S. Kim
1
and S Y. Park
1
1 Center for Healthcare Technology Development, Bio-Safety Research Institute, College of Veterinary Medicine, Chonbuk National
University, Jeonju, Jeonbuk, South Korea
2 Institute of Oral Bioscience, School of Dentistry, Chonbuk National University, Jeonju, Jeonbuk, South Korea
Keywords
chondrocytes; hypoxia; proteasome; reactive
oxygen species; tumour necrosis
factor-related apoptosis-inducing ligand
(TRAIL)
Correspondence
S Y. Park, College of Veterinary Medicine,
Chonbuk National University, Jeonju,
Jeonbuk 561-756, South Korea

significantly inhibited under hypoxic conditions. These findings suggest that
proteasome inhibition and TRAIL may be possible mechanisms in cartilage
degradation and joint-related diseases. Furthermore, the maintenance
of hypoxic conditions or therapy with hypoxia-related genes in the joint
may be successful for the treatment of joint-related diseases, including
osteoarthritis.
Abbreviations
ALLN, N-acetyl-Leu-Leu-Norleu-al; DCFH
2
-DA, 2¢,7¢-dichlorodihydrofluorescein diacetate; DR-5, death receptor-5; JC-1, 5,5¢,6,6¢-tetrachloro-
1,1¢3,3¢-tetraethylbenzimidazol-carbocyanine iodide; JNK, c-Jun N-terminal kinase; JNK-SAPK, c-Jun N-terminal kinase-stress-activated protein
kinase; MTP, mitochondrial transmembrane potential; OA, osteoarthritis; ROS, reactive oxygen species; Suc-LLVY-AMC, Suc-Leu-Leu-Val-
Tyr-7-amino-4-methylcoumarin; TRAIL, tumour necrosis factor-related apoptosis-inducing ligand.
FEBS Journal 276 (2009) 7375–7385 ª 2009 The Authors Journal compilation ª 2009 FEBS 7375
Introduction
A hallmark of osteoarthritis (OA) is a decrease in the
number of chondrocytes, as they are the only resident
cells in articular cartilage. Chondrocytes regulate the
enzymatic breakdown of the extracellular matrix,
thereby maintaining the equilibrium between synthetic
and degradative processes in the cartilage [1]. There-
fore, the metabolic and structural changes of chondro-
cytes in the articular cartilage play a significant role in
the initiation and progression of the disease. Several
studies have examined cell death in human articular
cartilage affected by OA or in experimental models of
OA [2,3].
Tumour necrosis factor-related apoptosis-inducing
ligands (TRAILs) are type II transmembrane mole-
cules that trigger the apoptotic signal cascade by bind-

required to fully characterize the types of cell death in
aging and arthritic cartilage, together with their respec-
tive frequencies.
Articular cartilage is an avascular tissue that func-
tions at an oxygen tension lower than that of most
other tissues, and derives both its nutrition and oxygen
supply by diffusion from the synovial fluid and
subchondral bone [19,20]. It has been estimated that
articular chondrocytes in the deepest layers may have
access to no more than 1–6% O
2
[21,22]. Oxygen can
be processed to generate reactive oxygen species
(ROS), which play an important role in intracellular
signalling and thus in cell physiology and cellular
destruction. ROS are known to induce a wide range of
responses, depending on cell type and levels of ROS
within the cell [23,24].
The aims of this study were to investigate the major
signalling pathways and effects of hypoxic conditions
in experimental osteoarthritic cartilage degeneration
and cell death of primary cultured chondrocytes. In
particular, we focused on the role of proteasome
inhibition and TRAIL in osteoarthritic disease and
chondrocyte apoptosis, and the hypoxic inhibition of
cartilage and chondrocyte degeneration.
Results
Macroscopic and radiographic examination of the
articular cartilage after experimentally induced
OA

ous inflammatory cells, such as lymphocytes, and other
studies support this [6,8].
Combined treatment with proteasome inhibitor
and TRAIL markedly enhanced apoptosis in
cultured canine chondrocytes
To investigate the proteasome inhibition effect on
TRAIL-induced apoptosis in canine chondrocytes,
N-acetyl-Leu-Leu-Norleu-al (ALLN) was used as a
proteasome inhibitor. Canine chondrocytes were
exposed to ALLN (10 lm) for 12 h and then treated
with recombinant TRAIL protein for an additional
12 h. TRAIL and ALLN alone did not induce apopto-
sis, but combined treatment markedly induced apopto-
sis to 60% in canine chondrocytes (Fig. 3A). The
examination of cell morphology also supported the
enhancing effect of combined ALLN and TRAIL
treatment in canine chondrocytes (Fig. 3B).
To determine whether treatment with the protea-
some inhibitor ALLN affected proteasome-mediated
degradation in canine chondrocytes, proteasome activ-
ity was assayed in cell lysates as a measure of the
hydrolysis of the fluorogenic substrate Suc-LLVY-
AMC. Proteasome activity was significantly inhibited
by treatment with ALLN only and cotreatment with
ALLN and TRAIL (Fig. 3C). Western blot analysis
was also used to investigate whether ALLN induced
proteasome inhibition in canine chondrocytes. In the
absence of ALLN, smears of ubiquitinated proteins
were not observed in control and TRAIL-treated
canine chondrocytes. In the presence of ALLN,

Student’s t-test. OA, osteoarthritis sample; Ubi, ubiquitin.
J W. Seol et al. Hypoxic condition inhibits articular chondrocyte death
FEBS Journal 276 (2009) 7375–7385 ª 2009 The Authors Journal compilation ª 2009 FEBS 7377
expression in cells treated with ALLN and ⁄ or TRAIL
protein. Western blot analysis showed that caspase-8
was slightly activated in control, ALLN-treated and
TRAIL-treated chondrocytes. However, under protea-
some inhibition induced by pretreatment with ALLN,
TRAIL treatment unexpectedly increased the activa-
tion of caspase-8. In addition, the phosphorylation of
JNK protein was markedly increased by combined
treatment with ALLN and TRAIL when compared
with other groups. The expression of DR-5 and
TRAIL protein was also investigated, but these were
not altered in cells with proteasome inhibition (Fig. 4).
Proteasome inhibition and the expression of
TRAIL protein induced the dissipation of the
mitochondrial transmembrane potential (MTP)
and ROS generation
MTP was investigated in order to address the possible
mechanism by which proteasome inhibition enhances
TRAIL-induced apoptosis in canine chondrocytes.
MTP evaluation is based on the ability of a fluorescent
probe to enter the mitochondria selectively and
reversibly change its colour from green to red as the
mitochondrial potential increases. 5,5¢,6,6¢-Tetrachloro-
1,1¢3,3¢-tetraethylbenzimidazol-carbocyanine iodide
(JC-1; Molecular Probes, Eugene, OR, USA) exists as
a monomer at low MTP values and shows green fluo-
rescence, whereas it forms an aggregate at high MTP

7378 FEBS Journal 276 (2009) 7375–7385 ª 2009 The Authors Journal compilation ª 2009 FEBS
To investigate the effects of MTP in cartilage by the
induction of OA, chondrocytes were isolated from
experimentally induced osteoarthritic cartilage. The
photomicrographs and fluorescence values indicated
that these chondrocytes showed a decrease in MTP
compared with normal cartilage (Fig. 5C). In addition,
the chondrocytes isolated from osteoarthritic joints
demonstrated significantly greater ROS generation
than did the controls (Fig. 5D).
Hypoxia inhibited the apoptosis of primary
cultured canine chondrocytes induced by
proteasome inhibition and TRAIL treatment
In order to examine the functional role of ALLN and
TRAIL in apoptotic cell death under hypoxic condi-
tions, canine chondrocytes were exposed to hypoxia
and ALLN (10 lm) for 12 h, and were then treated
with recombinant TRAIL protein for an additional
12 h under normoxic and hypoxic conditions. Hypoxic
conditions inhibited significantly the apoptosis of
chondrocytes induced by cotreatment of the cells with
the proteasome inhibitor and TRAIL (Fig. 6A). Chon-
drocyte survival under hypoxic conditions was
enhanced by 25% compared with the survival of cells
Fig. 4. Proteasome inhibition and TRAIL treatment significantly
increased caspase-8 activation and JNK phosphorylation. Whole-cell
lysates were prepared and total protein (40 lgÆmL
)1
) was electro-
phoretically resolved on SDS gel. Apoptotic proteins were detected

phosphorylation were inhibited under hypoxic condi-
tions (Fig. 6B). MTP and ROS were investigated in
order to address the inhibitory mechanism exerted by
hypoxic conditions. Canine chondrocytes were
exposed to ALLN (10 lm) for 12 h and were then
treated with recombinant TRAIL protein for an addi-
tional 12 h under normoxic and hypoxic conditions.
The fluoroscopic results presented in Fig. 6C show
that the cells fluoresce green after cotreatment with
ALLN and TRAIL, indicating lower MTP under
normoxic conditions. However, the green fluorescence
indicating lower MTP declined under hypoxic condi-
tions (Fig. 6C). Cotreatment of cells with ALLN and
TRAIL induced ROS generation significantly under
normoxic conditions, but hypoxia prevented ROS
generation after cotreatment with ALLN and TRAIL
(Fig. 6D).
A
C
B D
Fig. 6. Hypoxia inhibited chondrocyte death and apoptosis-related signals induced by proteasome inhibition and TRAIL. (A) Cell viability was
determined using the crystal violet staining method. The control cell viability was set at 100%; viability relative to the control is presented.
The experiments were performed in triplicate at least twice independently. The cell morphology was photographed (·200). (B) Whole-cell
lysates were prepared and total protein (40 lgÆmL
)1
) was electrophoretically resolved on SDS gel and then tested for apoptotic proteins by
western blotting analysis. b-Actin was used to normalize equal protein loading; Blot images represent one of three independent experiments.
(C) MTP was determined using a JC-1 probe. The cells were photographed using a fluoroscope. The green fluorescence intensity was mea-
sured under the conditions described in Materials and methods. (D) The ROS level was measured using DCFH-DA. The fluorescence value
for control cells was set at 100%; fluorescence values relative to the control are presented. The experiments were performed in triplicate at

between 40 and 85 mmHg, corresponding to an oxy-
gen concentration of approximately 6–11% [27]. It has
been estimated that articular chondrocytes in the deep-
est layers have access to no more than 1–6% O
2
[20,22]. Moreover, mitochondria are sparse in the
articular chondrocytes, occupying only 1–2% of the
intracellular volume [28], compared with 15–20% in
other typical animal cells (for example, the liver).
Marcus [29] and Otte [30] observed that chondrocytes
produced ATP mostly through substrate-level phos-
phorylation during glycolysis. However, oxygen
tensions below 1% inhibit both glucose uptake and
lactate production, as well as cellular RNA synthesis
[29,30]. This indicates that chondrocytes need at least
some oxygen for their basal metabolic activity. There-
fore, hypoxia is considered to be a key factor in the
growth and survival of chondrocytes.
Hypoxia is known to regulate the expression of
many genes, but little is known about its role in either
apoptosis or anti-apoptosis, especially in canine chon-
drocytes. In this article, we investigated the possible
effects of proteasome inhibition on TRAIL-induced
apoptosis under normoxic and hypoxic conditions.
We found that TRAIL and ALLN alone did not
induce apoptosis, but combined treatment of the cells
with ALLN and TRAIL increased apoptosis markedly
to 60% in canine chondrocytes. However, ALLN ⁄
TRAIL cotreatment-induced apoptosis of canine
chondrocytes was inhibited significantly under hypoxic

chain complexes II and III and the mitochondrial
membrane potential are significantly reduced in cul-
tured human chondrocytes from osteoarthritic donors
when compared with normal donors [35]. In this study,
we demonstrated that hypoxic conditions prevented
ROS generation and restored the loss of MTP seen
after cotreatment with ALLN and TRAIL. These find-
ings suggest that the mitochondrial respiratory chain
complexes are probable sites of ROS production, and
that the inhibition of depolarization of MTP during
hypoxia probably induces a decrease in ROS levels in
canine chondrocytes.
In conclusion, the present study has demonstrated
that proteasome activity, ubiquitinated protein,
TRAIL and ROS are altered significantly in synovial
fluid acquired from experimentally induced osteoar-
thritic joints. At the cellular level, proteasome inhibi-
tion markedly enhances TRAIL-induced apoptosis
through the activation of caspase-8, the phosphoryla-
tion of JNK protein, a decrease in MTP and the gener-
J W. Seol et al. Hypoxic condition inhibits articular chondrocyte death
FEBS Journal 276 (2009) 7375–7385 ª 2009 The Authors Journal compilation ª 2009 FEBS 7381
ation of ROS in primary cultured canine chondrocytes.
However, the enhanced apoptosis of chondrocytes
induced by this combined treatment is inhibited under
hypoxic conditions. All these findings suggest that pro-
teasome inhibition and TRAIL play a pivotal role in
canine chondrocyte death and cartilage degradation.
These findings indicate that the maintenance of
hypoxic conditions in cartilage inhibits articular chon-

and oxygen. During surgery, the jaw reflex, ocular reflex,
heart rate (using electrocardiogram) and respiratory rate
(using capnography) were monitored. Based on these data,
we changed the vaporizer settings if the experimental ani-
mals were in deep or light anaesthesia. After surgery, post-
operative treatment was given with butophanol (Butopan
InjÒ; Hana Pharm. Co. Ltd.), 10 mgÆkg
)1
intramuscularly,
every 12 h for 7 days for pain relief. After 7 days, no anal-
gesic drug was given as the progress of OA was graded
using a clinical scoring system, such as lameness, joint
mobility and weight bearing. All procedures employed in
the animal experiments were approved by the Standard
Operation Procedure of the Institutional Animal Care and
Use Committee, Jeonju, South Korea.
Evaluation of OA
Experimental animals were sacrificed at 12 weeks to evalu-
ate the severity of OA after surgery. Levels of macroscopic
synovial inflammation and cartilage damage were evaluated
with digital high-resolution photographs. The severity of
synovial inflammation was graded on the basis of colour,
angiogenesis and fibrillation: grade 0, no inflammation;
grade 1, slight inflammation; grade 2, strong inflammation.
The cartilage damage severity of the femoral condyles
and tibial plateau was graded from 0 to 4: grade 0,
smooth surface; grade 1, slight fibrillation; grade 2, fibrilla-
tion with shallow grooves; grade 3, deep and sharp
grooves; grade 4, deep and sharp grooves with surrounding
damage.

modified Eagle’s medium (Invitrogen-Gibco, Grand Island,
NY, USA) supplemented with 10% (v ⁄ v) fetal bovine
serum (Invitrogen-Gibco) and antibiotics (100 lgÆmL
)1
gentamycin and 100 lgÆmL
)1
penicillin–streptomycin). Cells
were filtered through a 70 lm cell strainer (Falcon, Frank-
lin Lakes, NJ, USA), washed twice with NaCl ⁄ P
i
and then
seeded into tissue culture flasks. The total cell number was
calculated using a haemocytometer.
Cell viability test
Canine chondrocytes were adjusted to 1.0 · 10
6
cells per
well in 12-well plates, pretreated with ALLN (Sigma,
St Louis, MO, USA) for 12 h, and then further incubated
with recombinant TRAIL protein for 12 h under normoxic
(21% O
2
) and hypoxic (1% O
2
) conditions at the indicated
doses. Cellular morphology was photographed under light
Hypoxic condition inhibits articular chondrocyte death J W. Seol et al.
7382 FEBS Journal 276 (2009) 7375–7385 ª 2009 The Authors Journal compilation ª 2009 FEBS
microscopy (Nikon, Tokyo, Japan), and cell viability was
determined using the crystal violet staining method, as

N-terminal kinase–stress-activated protein kinase (JNK)
and the phosphorylated form (p-JNK) (Upstate Biotechnol-
ogy, Lake Placid, NY, USA).
Proteasome activity test
Proteasome activity was measured as described previously
[38]. The cells were collected by centrifugation and the
synovial fluid was diluted 10 times with NaCl ⁄ P
i
. Protein
concentrations of synovial fluid and cytoplasm were deter-
mined using the Bradford protein assay kit (Bio-Rad, Her-
cules, CA, USA). Two hundred micrograms of synovial
fluid protein and cytoplasm protein were added to the assay
buffer (20 mm Tris ⁄ HCl, pH 8.0, 1 mm ATP, 2 mm MgCl
2
)
in the presence of the synthetic fluorogenic substrate Suc-
LLVY-AMC to a final concentration of 60 lm (Sigma-
Aldrich) suspended in a final volume of 1 mL. The tubes
were incubated at 30 °C for 30 min, after which the reac-
tion was terminated through the addition of 1 mL of cold
ethanol. The lysate was spun at 12 000 g for 10 min at
4 °C. Fluorescence was measured at 380 nm excitation and
440 nm emission using a fluorescence plate reader (Spectra-
Max fluorometer with the softmax program; Molecular
Probes, Eugene, OR, USA).
Evaluation of MTP
The level of MTP was determined using a lipophilic cation,
JC-1 (Molecular Probes). Briefly, chondrocytes were iso-
lated from cartilage obtained from osteoarthritic joints. The

in sodium-containing medium for 10 min to allow de-esteri-
fication to occur. The cells were then collected, and the flu-
orescence signals corresponding to intracellular ROS were
monitored at 490 nm excitation and 525 nm emission using
a fluorescence plate reader (SpectraMax fluorometer with
the softmax program, Molecular Probes).
Hypoxic conditions
A sealed chamber was used to culture the chondrocytes at
low oxygen tension (1%). A gas mixture of 1% O
2
,5%
CO
2
and 94% N
2
was added to the sealed chamber, and
ambient air was evacuated through an outlet tube. The
oxygen flow was allowed to stream through the chamber
for 2–3 min to maintain the desired oxygen tension inside
the chamber. Culture plates were incubated in sealed cham-
bers containing 1% O
2
at 37 °C. For the normoxic condi-
tion (21% O
2
tension), the chondrocytes were incubated at
37 °C in a 95% humidified atmosphere with 5% CO
2
.
There were two controls (normoxia and hypoxia) in this

Lotz M (1998) Chondrocyte apoptosis and nitric oxide
production during experimentally induced osteoarthritis.
Arthritis Rheum 41, 1266–1274.
4 Hymowitz SG, Christinger HW, Fuh G, Ultsch M,
O’Connell M, Kelley RF, Ashkenazi A & de Vos AM
(1999) Triggering cell death: the crystal structure of
Apo2L ⁄ TRAIL in a complex with death receptor 5.
Mol Cell 4, 563–571.
5 Cha SS, Kim MS, Choi YH, Sung BJ, Shin NK, Shin
HC, Sung YC & Oh BH (1999) 2.8 A
˚
resolution crystal
structure of human TRAIL, a cytokine with selective
antitumor activity. Immunity 11, 253–261.
6 Collison A, Foster PS & Mattes J (2009) The emerging
role of TRAIL as key regulator of inflammatory
responses. Clin Exp Pharmacol Physiol, doi:10.1111/j.
1440-1681.2009.05258.x.
7 Brincks EL, Katewa A, Kucaba TA, Griffith TS &
Legge KL (2008) CD8 T cells utilize TRAIL to
control influenza virus infection. J Immunol 181,
4918–4925.
8 Lamhamedi-Cherradi SE, Zheng SJ, Maguschak KA,
Peschon J & Chen YH (2003) Defective thymocyte
apoptosis and accelerated autoimmune diseases in
TRAIL– ⁄ – mice. Nat Immunol 4, 255–260.
9 Robertson NM, Zangrilli JG, Steplewski A, Hastie A,
Lindemeyer RG, Planeta MA, Smith MK, Innocent N,
Musani A, Pascual R et al. (2002) Differential expres-
sion of TRAIL and TRAIL receptors in allergic asth-

mal function in chondrocytes down-regulates growth
plate chondrogenesis and longitudinal bone growth.
Endocrinology 147, 3761–3768.
16 Kuhn K, Shikhman AR & Lotz M (2003) Role of nitric
oxide, reactive oxygen species, and p38 MAP kinase in
the regulation of human chondrocyte apoptosis. J Cell
Physiol 197, 379–387.
17 An WG, Hwang SG, Trepel JB & Blagosklonny MV
(2000) Protease inhibitor-induced apoptosis: accumula-
tion of wt p53, p21WAF1 ⁄ CIP1, and induction of
apoptosis are independent markers of proteasome inhi-
bition. Leukemia 14, 1276–1283.
18 Kim OH, Lim JH, Woo KJ, Kim YH, Jin IN, Han ST,
Park JW & Kwon TK (2004) Influence of p53 and
p21Waf1 expression on G2 ⁄ M phase arrest of colorectal
carcinoma HCT116 cells to proteasome inhibitors. Int J
Oncol 24, 935–941.
19 Falchuk KH, Goetzl EJ & Kulka JP (1970) Respiratory
gases of synovial fluids. An approach to synovial tissue
circulatory–metabolic imbalance in rheumatoid arthritis.
Am J Med 49, 223–231.
20 Lund-Olesen K (1970) Oxygen tension in synovial flu-
ids. Arthritis Rheum 13, 769–776.
21 Silver IA (1975) Measurement of pH and ionic compo-
sition of pericellular sites. Philos Trans R Soc London
B: Biol Sci 271, 261–272.
22 Treuhaft PS & McCarty DJ (1971) Synovial fluid
pH, lactate, oxygen and carbon dioxide partial
pressure in various joint diseases. Arthritis Rheum 14,
475–484.

31 Kuhn K, D’Lima DD, Hashimoto S & Lotz M (2004)
Cell death in cartilage. Osteoarthritis Cartilage 12, 1–16.
32 Sharif M, Whitehouse A, Sharman P, Perry M &
Adams M (2004) Increased apoptosis in human osteoar-
thritic cartilage corresponds to reduced cell density and
expression of caspase-3. Arthritis Rheum 50, 507–515.
33 Ghafourifar P, Bringold U, Klein SD & Richter C
(2001) Mitochondrial nitric oxide synthase, oxidative
stress and apoptosis. Biol Signals Recept 10, 57–65.
34 Salvioli S, Bonafe M, Capri M, Monti D & Franceschi
C (2001) Mitochondria, aging and longevity – a new
perspective. FEBS Lett 492, 9–13.
35 Maneiro E, Martin MA, de Andres MC, Lopez-
Armada MJ, Fernandez-Sueiro JL, del Hoyo P, Galdo
F, Arenas J & Blanco FJ (2003) Mitochondrial respira-
tory activity is altered in osteoarthritic human articular
chondrocytes. Arthritis Rheum 48, 700–708.
36 Chaudhari AA, Seol JW, Kim SJ, Lee YJ, Kang HS,
Kim IS, Kim NS & Park SY (2007) Reactive oxygen
species regulate Bax translocation and mitochondrial
transmembrane potential, a possible mechanism for
enhanced TRAIL-induced apoptosis by CCCP. Oncol
Rep 18, 71–76.
37 Park SY, Billiar TR & Seol DW (2002) Hypoxia inhibi-
tion of apoptosis induced by tumor necrosis factor-
related apoptosis-inducing ligand (TRAIL). Biochem
Biophys Res Commun 291, 150–153.
38 Jamaluddin M, Casola A, Garofalo RP, Han Y, Elliott
T, Ogra PL & Brasier AR (1998) The major component
of IkappaBalpha proteolysis occurs independently of the