Clinical Science (2001) 101, 593–599 (Printed in Great Britain)
Differential roles of nitric oxide and oxygen
radicals in chondrocytes affected by
osteoarthritis and rheumatoid arthritis
Ilaria MAZZETTI*, Brunella GRIGOLO*, Lia PULSATELLI*, Paolo DOLZANI*,
Tania SILVESTRI*, Livia ROSETI*, Riccardo MELICONI† and Andrea FACCHINI*‡
*Laboratorio di Immunologia e Genetica, Istituti Ortopedici Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italy, †Dipartimento
di Medicina Interna, Cardioangiologia, Epatologia, Universita' degli Studi di Bologna, Via Massarenti 9, 40138 Bologna, Italy,
and ‡Dipartimento di Medicina Interna e Gastroenterologia, Universita' degli Studi di Bologna, Via Massarenti 9,
40138 Bologna, Italy
A
B
S
T
R
A
C
T
Osteoarthritis and rheumatoid arthritis are characterized by focal loss of cartilage due to an upregulation of catabolic pathways, induced mainly by pro-inflammatory cytokines, such as
interleukin-1 (IL-1) and tumour necrosis factor α (TNFα). Since reactive oxygen species are also
involved in this extracellular-matrix-degrading activity, we aimed to compare the chondrocyte
oxidative status responsible for cartilage damage occurring in primarily degenerative (osteoarthritis) and inflammatory (rheumatoid arthritis) joint diseases. Human articular
chondrocytes were isolated from patients with osteoarthritis or rheumatoid arthritis, or from
multi-organ donors, and stimulated with IL-1β and/or TNFα. We evaluated the oxidative stress
related to reactive nitrogen and oxygen intermediates, measuring NO−2 as a stable end-product
of nitric oxide generation and superoxide dismutase as an antioxidant enzyme induced by radical
oxygen species. We found that cells from patients with osteoarthritis produced higher levels of
NO−2 than those from patients with rheumatoid arthritis. In addition, IL-1β was more potent than
TNFα in inducing nitric oxide in both arthritides, and TNFα alone was almost ineffective in cells
from rheumatoid arthritis patients. We also observed that the intracellular content of
copper/zinc superoxide dismutase (Cu/ZnSOD) was always lower in rheumatoid arthritis
chondrocytes than in those from multi-organ donors, whereas no differences were found in
intracellular manganese SOD (MnSOD) or in supernatant Cu/ZnSOD and MnSOD levels.
Moreover, intracellular MnSOD was up-regulated by cytokines in osteoarthritis chondrocytes.
In conclusion, our results suggest that nitric oxide may play a major role in altering chondrocyte
functions in osteoarthritis, whereas the harmful effects of radical oxygen species are more
evident in chondrocytes from patients with rheumatoid arthritis, due to an oxidant/antioxidant
imbalance.
INTRODUCTION
The progressive deterioration and loss of articular cartilage leading to an irreversible impairment of joint
motion are the final pathogenic events common to
osteoarthritis (OA) and rheumatoid arthritis (RA). In
both arthritides the loss of integrity of the cartilage
extracellular matrix depends on an imbalance between
anabolic and catabolic pathways [1,2]. It is clearly
established that pro-inflammatory cytokines, mainly
Key words : chondrocytes, nitric oxide, osteoarthritis, oxygen radicals, rheumatoid arthritis, superoxide dismutases.
Abbreviations: IL-1, interleukin-1 ; MD cells, cells from multi-organ donors ; OA, osteoarthritis ; RA, rheumatoid arthritis ; RNI,
reactive nitrogen intermediates ; ROI, reactive oxygen intermediates ; SOD, superoxide dismutase ; TNFα, tumour necrosis factor α.
Correspondence: Dr Andrea Facchini, Laboratorio di Immunologia e Genetica, I. O. R. (e-mail facchini!alma.unibo.it).
# 2001 The Biochemical Society and the Medical Research Society
593
594
I. Mazzetti and others
interleukin-1 (IL-1) and tumour necrosis factor α
(TNFα), can activate chondrocytes to enter a catabolic
condition and prime a matrix-degrading activity [3],
including protease secretion [4], radical species production [5,6], down-regulation of matrix and protease
inhibitor synthesis [7,8], inhibition of chondrocyte proliferation [9] and cell death [10].
Radical species with oxidative activity, which include
reactive nitrogen intermediates (RNI) and reactive oxygen intermediates (ROI), play an important role in this
chondrocyte catabolic activity, since they represent the
mediators and effectors of cartilage damage [11,12]. RNI
are derived from the oxidation of the guanido nitrogen of
L-arginine, with production of a nitrogen-centred radical,
nitric oxide (NO). NO production depends on a large
family of NO synthases, which includes constitutively
expressed isoforms (neuronal and endothelial NO
synthases) and an inducible isoform [13]. Evidence is
accumulating that there is an overproduction of inducible
NO synthase and NO in OA and RA cartilage compared
with normal tissue [13,14]. It has been demonstrated that
high local concentrations of NO may exert detrimental
effects on chondrocyte functions, including inhibition of
collagen and proteoglycan synthesis [7], activation of
metalloproteinases [15], decreased expression of IL-1
receptor antagonist [16], inhibition of chondrocyte proliferation [9] and induction of apoptotic death [10]. On
the other hand, it has been suggested that the ROI
produced inside the joints may also contribute significantly to the pathogenesis of arthritis, since these
inorganic oxidants are able to degrade matrix components
by direct action or by indirect activation of latent
collagenases [17]. Moreover, it has been suggested that
ROI act as intracellular second messengers mediating
IL-1-induced collagenase gene expression [18]. To prevent ROI being toxic, cells possess a co-ordinated antioxidant enzyme system. Superoxide dismutase (SOD),
being the first enzyme in this radical-scavenging system
and catalysing the dismutation of superoxide anion into
hydrogen peroxide (H O ), plays a critical and limiting
# #
role. Indeed, if the levels of some antioxidant enzymes are
lower or higher than others, H O can be stored as an
# #
intermediate of scavenging reactions and may exert
detrimental effects [19]. It has been demonstrated that
cytokines, besides inducing ROI production [6], may in
parallel modulate the expression and activity of oxygen
radical scavengers [20,21] ; in particular, IL-1 and TNFα
are able to induce SOD expression [22,23].
In the present study, we wanted to compare the
chondrocyte oxidative status responsible for cartilage
damage occurring in primarily degenerative or inflammatory joint diseases, in order to clarify its role in the
pathogenic mechanisms of the two different arthritic
disorders, OA and RA. It is generally accepted [13,19]
that a reliable estimation of RNI and ROI production can
be obtained by evaluating the level either of radical
# 2001 The Biochemical Society and the Medical Research Society
byproducts or of the enzymes induced by these species,
whereas the precise measurement of free radical levels is
difficult because of their extremely short half-lives.
Therefore we evaluated the RNI- and ROI-related
oxidative stress indirectly, by measuring the production
of NO− and SOD respectively by human OA- and RA#
affected chondrocytes, stimulated in vitro with the two
pro-inflammatory cytokines IL-1β and\or TNFα.
METHODS
Specimens
Articular cartilage fragments were obtained from the
tissues of 14 patients with OA (mean age 59 years, range
28–77 years ; mean disease duration 4 years, range 3–6
years) and 14 patients with RA (mean age 62 years, range
43–75 years ; mean disease duration 17.5 years, range 1–40
years) undergoing hip and knee replacement. The diagnosis of OA was based on clinical, laboratory and
radiological evaluations [24,25]. The diagnosis of RA was
made according to American College of Rheumatology
classification criteria [26]. Cartilage samples were also
obtained from three multi-organ donors with no history
of joint diseases at the time of explantation (MD cells)
(mean age 60 years, range 56–64 years). Informed consent
was obtained from all patients, and the Institutional
Review Board approved the study. All OA patients were
receiving non-steroidal anti-inflammatory drugs and all
RA patients were on immunosuppressive treatment.
Chondrocyte isolation
Chondrocytes were isolated from articular cartilage by
sequential enzymic digestion : 30 min with 0.1 % hyaluronidase (Sigma, St. Louis, MO, U.S.A.), 1 h with 0.5 %
Pronase (Sigma), and 1 h with 0.2 % collagenase (Sigma)
at 37 mC in Dulbecco’s modified Eagle’s medium
(GIBCO BRL, Grand Island, NY, U.S.A.) containing
25 mM Hepes (Sigma), 100 units\ml penicillin (Biological Industries, Kibbutz Beit Haemek, Israel), 100 µg\ml
streptomycin (Biological Industries), 50 µg\ml gentamicin (Flow Laboratories, Biaggio, Switzerland) and
2.5 µg\ml amphotericin (Biological Industries). The isolated chondrocytes were filtered using 100 µm and 70 µm
nylon meshes, washed and centrifuged. The pellet was
seeded at high density (2i10& cells\cm#) and cultured in
Dulbecco’s modified Eagle’s medium supplemented with
10 % (v\v) fetal calf serum at 37 mC in a humidified
atmosphere of 5 % CO . For all experiments, only
#
primary cultures were used, to ensure the stability of the
chondrocyte phenotype.
Chondrocyte stimulation
At 24 h after seeding, the cell monolayers were incubated
for 72 h with or without recombinant human IL-1β
Oxidative stress in arthritides
(100 units\ml ; specific activity 5i10( units\mg ;
Boehringer Mannheim, Mannheim, Germany) and\or
recombinant human TNFα (100 units\ml ; specific activity 1.0i10) units\mg ; Boehringer Mannheim). The
cytokine concentrations and incubation time were
standardized based on results obtained in previous dosedependence and kinetic experiments. After 72 h the
supernatants were collected, divided into aliquots, and
stored frozen at k80 mC until analysis. Cell monolayers
were lysed directly in the culture plate by adding 0.250 ml
of 0.5 % Triton in PBS per cm#, incubating for 30 min on
ice and sonicating in iced water (5i20 s) with an
Ultrasonic 300 sonicator equipped with a sonotrode
(Branson Scientific Instruments, NJ, U.S.A.). The suspensions were then harvested and stored frozen at
k80 mC until analysis.
NO production
NO production was measured by estimating the stable
NO metabolite, nitrite, in conditioned medium using a
microplate adaptation of the Griess assay. Briefly, 100 µl
of each supernatant was incubated with 100 µl of Griess
reagent (Molecular Probes, Eugene, OR, U.S.A.) for
10 min at room temperature and the absorbance at
540 nm was measured with an ELISA plate reader
(Multiskan Ex ; Labsystems, Helsinki, Finland). Concentrations of nitrite were calculated using sodium nitrite in
fresh culture medium as a standard. The results were
expressed as nmol\10' cells.
Cu/ZnSOD and MnSOD determination
The cell lysates were thawed and centrifuged for 15 min
at 4 mC at 13000 g, giving clear supernatants. The cell
lysates and cell culture supernatants were then examined
for concentrations of copper\zinc-dependent (Cu\Zn)
and manganese-dependent (Mn) SOD by means of two
specific enzyme immunoassays (Cu\ZnSOD ELISA :
Bender MedSystems, Vienna, Austria ; MnSOD ELISA :
Sceti, Tokyo, Japan). The results were expressed as ng\10'
cells.
RESULTS
NO production
Unstimulated chondrocytes released NO− spon#
taneously, and NO− levels from OA-derived cells were
#
higher than those from RA-derived cells (P 0.05)
(Figure 1). OA chondrocytes also generated more NO−
#
than RA cells in response to IL-1β or TNFα (P 0.05)
(Figure 1). However, NO− release induced by the
#
combination of the two cytokines was not significantly
different between the two diseases.
We also evaluated the efficacy of the different stimuli
in NO− induction. In OA chondrocytes both IL-1β and
#
TNFα increased NO− synthesis significantly compared
#
with the basal condition (P 0.005) ; IL-1β was more
potent than TNFα (P l 0.002). In RA cells, by contrast,
NO− production was increased only slightly in response
#
to TNFα, but IL-1β had a significant effect (P 0.01).
Finally, the combined use of IL-1β and TNFα induced a
remarkable release of NO− in both arthritides (P 0.05) ;
#
this effect was probably due primarily to the activity of
IL-1β, since a statistically significant difference was only
found between the combined treatment with the two
pro-inflammatory cytokines and the treatment with
TNFα alone (P 0.05).
Cu/ZnSOD and MnSOD content and
modulation
Under basal conditions, different amounts of intracellular Cu\ZnSOD were found in chondrocytes from the
various groups of subjects (Table 1). RA chondrocytes
had the lowest levels of the enzyme, whereas MD cells
had the highest levels (P 0.05). Treatment with cytokines failed to cause a statistically significant increase in
the amount of intracellular Cu\ZnSOD, and the difference among groups paralleled the results obtained in
absence of stimuli (RA compared with MD : P 0.05)
Statistical analysis
Non-parametric ANOVA tests for multiple comparisons
of unpaired and paired data were used (distribution-free
Kruskall–Wallis test and Friedman ANOVA), followed
by Dunn’s test or the Wilcoxon test respectively for
unpaired and paired data. Comparisons between only
two independent groups were performed using the
Mann–Whitney U-test. Statistical computations were
performed using CSS Statistica statistical software
(Statsoft Inc., Tulsa, OK, U.S.A.).
Figure 1
Effects of IL-1β and TNFα on NO production
Cells from patients with OA (n l 14) or RA (n l 14) were cultured for 72 h with
culture medium alone or in the presence of IL-1β (100 units/ml), TNFα
(100 units/ml) or a combination of the two cytokines. Values are meanspS.E.M.
Significance of differences (Mann–Whitney U-test) : *P 0.05. UN, unstimulated
cells.
# 2001 The Biochemical Society and the Medical Research Society
595
596
I. Mazzetti and others
Table 1 Intracellular Cu/ZnSOD levels in chondrocytes from
OA (n l 5) and RA (n l 5) patients and MD (n l 3) subjects
Table 3 CuZnSOD release into the supernatant by OA (n l
14), RA (n l 14) and MD (n l 3) chondrocytes
Results are expressed as absolute amounts of intracellular enzyme (ng/106 cells)
after 72 h of culture in the basal condition or after stimulation with IL-1β and/or
TNFα. Values are meanspS.E.M. Significance of differences (ANOVA Kruskal–Wallis
and Dunn’s test) : *P 0.05 for OA compared with RA ; †P 0.05 for RA
compared with MD. Statistical analysis showed no significant differences between
intracellular Cu/ZnSOD levels in cytokine-stimulated and untreated cells.
Results are expressed as absolute amounts of enzyme released into the supernatant
(ng/106 cells) after 72 h of culture in the basal condition or after stimulation with
IL-1β and/or TNFα. Values are meanspS.E.M. Friedman ANOVA (comparison
between cytokine-treated cells and unstimulated cells) and ANOVA Kruskal–Wallis
(comparison among different disease-affected cells and normal cells) showed no
significant differences.
Intracellular Cu/ZnSOD (ng/106 cells)
CuZnSOD release (ng/106 cells)
Treatment
OA
RA
MD
Treatment
OA
RA
MD
Unstimulated
IL-1β
TNFα
IL-1βjTNFα
191.2p45.7
249.1p22.2
285.4p46.2*
253.0p30.0*
107.8p23.9†
130.7p25.8†
103.2p26.3†
84.6p12.2†
376.3p35.4
419.5p62.9
357.3p56.7
380.2p81.9
Unstimulated
IL-1β
TNFα
IL-1βjTNFα
135.1p21.8
143.7p25.0
132.9p18.2
117.3p21.7
128.8p25.0
129.2p27.6
120.8p23.9
128.0p33.8
200.2p43.7
171.7p46.1
178.1p45.2
184.3p60.1
Table 2 Intracellular MnSOD levels in chondrocytes from OA
(n l 5) and RA (n l 5) patients and MD (n l 3) subjects
Results are expressed as absolute amounts of intracellular enzyme (ng/106 cells)
after 72 h of culture in the basal condition or after stimulation with IL-1β and/or
TNFα. Values are meanspS.E.M. Significance of differences (Friedman ANOVA and
Wilcoxon matched pairs test) : *P 0.05 for cytokine-treated cells compared
with unstimulated cells. Statistical analysis showed no significant differences in
intracellular MnSOD levels among OA, RA and MD chondrocytes under equivalent
conditions.
Intracellular MnSOD (ng/106 cells)
Treatment
OA
RA
MD
Unstimulated
IL-1β
TNFα
IL-1βjTNFα
1070.6p431.1
1906.2p422.4*
2533.3p867.3*
2166.3p840.5*
786.3p459.1
1293.0p502.9
1136.9p442.1
1082.1p330.7
901.9p137.1
1859.1p635.3
2158.5p938.0
1610.1p463.2
(Table 1). Finally, OA chondrocytes possessed always
more Cu\ZnSOD than RA cells, but this difference
became significant only after addition of TNFα (P
0.05) (Table 1).
With regard to intracellular MnSOD, we found that the
content of MnSOD was not significantly different among
OA, RA and MD cells, both in the basal condition and
upon activation with cytokines (Table 2). On comparing
cytokine-treated cells with unstimulated cells within each
disease condition, the up-regulation of MnSOD by
cytokines was statistically significant only in OA
chondrocytes (P 0.05) (Table 2). Finally, when we
grouped all the chondrocyte samples, a higher degree of
statistical difference compared with the basal condition
was obtained with treatment with either IL-1β or TNFα
alone (P 0.005 for IL-1β or TNFα alone ; P 0.05 for
the combined treatment).
# 2001 The Biochemical Society and the Medical Research Society
Table 4 MnSOD release into the supernatant by OA (n l 4),
RA (n l 4) and MD (n l 3) chondrocytes
Results are expressed as absolute amounts of enzyme released into the supernatant
(ng/106 cells) after 72 h of culture in the basal condition or after stimulation with
IL-1β and/or TNFα. Values are meanspS.E.M. Friedman ANOVA (comparison
between cytokine-treated cells and unstimulated cells) and ANOVA Kruskal–Wallis
(comparison among different disease-affected cells and normal cells) showed no
significant differences.
MnSOD release (ng/106 cells)
Treatment
OA
RA
MD
Unstimulated
IL-1β
TNFα
IL-1βjTNFα
65.6p11.6
83.8p16.5
77.8p18.2
90.7p22.0
56.9p32.4
33.8p13.5
34.0p15.8
25.0p19.8
170.5p73.9
198.5p95.7
207.7p103.1
197.0p94.4
Cu\ZnSOD release into the supernatant was abundant
under basal conditions for all samples studied (Table 3).
MnSOD was also released into the culture medium of
unstimulated chondrocytes, but was less abundant (Table
4). Neither Cu\ZnSOD nor MnSOD levels in the
supernatant were influenced significantly by stimulation
with pro-inflammatory cytokines, and there were no
differences between chondrocytes from OA, RA and
MD subjects (Tables 3 and 4).
DISCUSSION
The breakdown of articular cartilage in RA and OA is
likely to be related to the synthesis and release of catabolic
factors in the microenvironment of the joint. We demonstrated previously [27] that pro-inflammatory cytokines
and inducible NO synthase were highly expressed in
synovial membranes from patients with inflammatory
arthritides, but rarely expressed by synovial cells in OA
Oxidative stress in arthritides
and traumatic arthritis. In contrast, these mediators were
frequently expressed by OA-affected chondrocytes, and
were almost absent from cartilage obtained from patients
with inflammatory joint disease. In the present study, we
showed that the production of NO by isolated chondrocytes cultured in vitro was significantly higher in OA
than in RA chondrocytes both under basal conditions
and after stimulation with IL-1β or TNFα. These findings
further support the hypothesis that OA chondrocytes are
a more important source of NO than RA cartilage. The
spontaneous overproduction of NO in OA compared
with RA may indicate the presence of inflammatory
factors such as autocrine cytokines or growth factors in
osteoarthritic cartilage [13], or it may be due to a
constitutively up-regulated NO synthase in OA [28].
OA chondrocytes also released higher levels of NO−
#
after cytokine treatment. It has been shown that the
number of type I IL-1 receptors is significantly increased
in OA chondrocytes [29], and it is well known that the
increased release of NO by IL-1β-stimulated OA
chondrocytes inhibits synthesis of the IL-1 receptor
antagonist. Furthermore, TNFα receptor expression is
significantly increased on OA chondrocytes [30]. On the
other hand, pro-inflammatory cytokines and their
specific receptors seem to be more highly expressed in
synovial tissue and at the cartilage\pannus junction in
RA than in OA [31,32]. On comparing the efficacy of the
different stimuli in the induction of NO release in the
two diseases, we confirmed an enhanced susceptibility of
OA cells to cytokine-induced NO production compared
with RA chondrocytes. It is noteworthy that IL-1β
induced the most prominent effects and was more potent
than TNFα, which was ineffective when added alone to
RA-affected chondrocytes, further underlining the lower
responsiveness of these cells. It has been reported
previously that normal cartilage releases lower levels of
NO than arthritic cartilage [13,16], but no information
was available concerning the comparison between OA
and RA, which was the aim of the first part of our study.
On the other hand, to our knowledge there has been no
report dealing with the ability of chondrocytes to
produce the superoxide-scavenging enzymes (Cu\ZnSOD and MnSOD) in response to cytokine-induced
ROI generation. Therefore we performed this second
part of the study on both cell lysates and supernatants
from normal and arthritic chondrocytes. We found that
the intracellular Cu\ZnSOD content was always significantly lower in RA chondrocytes than in MD cells.
Although OA chondrocytes also possessed higher levels
of Cu\ZnSOD than RA cells, we observed a significant
difference only when the cells were treated with TNFα or
with IL-1β plus TNFα. These results could lend themselves to different interpretations. We might hypothesize
that RA cartilage is more susceptible to ROI attack
because of a pre-existing disease-related deficit in antioxidant enzyme levels compared with normal chon-
drocytes. Indeed, since pro-inflammatory cytokines
failed to modulate the amount of intracellular Cu\ZnSOD, the basal inadequacy of the antioxidant enzyme
persists during stimulation with IL-1β and TNFα, while
ROI production probably rises, thus worsening the
oxidant\antioxidant imbalance. On the other hand, we
did not observe any difference in intracellular MnSOD
levels between arthritic cells and normal chondrocytes
either in the basal condition or after stimulation with
cytokines. Therefore, since cytoplasmic Cu\ZnSOD, in
contrast with mitochondrial MnSOD, can be released
upon cell activation and degranulation [33], we might
alternatively postulate that RA chondrocytes have been
depleted in vivo of Cu\ZnSOD by the persistent
presence of pro-inflammatory cytokines inducing cell
degranulation and release into the synovial fluid. Indeed,
SOD activity in knee joint fluids from patients with RA
has been found to be significantly higher than in those
from control subjects [34,35], although conflicting data
are presented in the literature [36]. Moreover, our group
has reported elevated Cu\ZnSOD levels in RA sera [37],
probably resulting from a high SOD concentration in
synovial fluid. It has been hypothesized that a similar
increase in Cu\ZnSOD could be inadequate to exert
effective antioxidant protection, but could in turn increase the oxidative burden via overproduction of H O
# #
and generation of the highly reactive OHd radical during
the Fenton [38] or Haber–Weiss [39] reaction. Our
findings, however, independent of their interpretation,
provide evidence that the harmful effects of oxygen
radicals seem to be involved primarily in articular
destruction in RA compared with OA.
It is interesting to note that stimulation with IL-1β and
TNFα increased the intracellular MnSOD content significantly in OA chondrocytes. The difference between
cytokine-treated cells and unstimulated cells became
more highly significant when grouping data from all
chondrocyte samples, indicating that chondrocytes react
to the increased oxidant stress by inducing compensatory
changes in the levels of this antioxidant enzyme. This
observation is fully in keeping with the concept that
MnSOD is more inducible by cytokines than Cu\
ZnSOD [20]. In contrast, investigation of the release of
Cu\ZnSOD and MnSOD into the cell culture supernatant did not reveal any modulation by cytokines or any
significant difference between arthritic and normal
chondrocytes, suggesting that cells counteract the effect
of ROI, which are indeed generated endogenously during
intracellular aerobic reactions, mainly by increasing their
intracellular antioxidant activities. These results underline the importance of our study performed directly on
chondrocytes in order to assay SOD as a marker of local
oxidative damage. Until now, the amount and activity of
this antioxidant enzyme have been evaluated mainly in
serum [35–37], synovial fluid [34–36] or erythrocytes
[40] from patients with RA, with conflicting findings that
# 2001 The Biochemical Society and the Medical Research Society
597
598
I. Mazzetti and others
probably reflect the existence of complex systemic
mechanisms or represent the contributions of different
types of cells.
Taken together, our data may identify the types of
molecules mainly responsible for local oxidant-induced
joint injury in OA and RA. We can postulate that, in
OA-affected chondrocytes, the harmful effects of free
radicals are due mainly to more abundant NO production, whereas ROI attack is more prominent in RA
chondrocytes due to an oxidant\antioxidant imbalance.
Since it has been demonstrated that, in cytokine-activated
chondrocytes, increased NO levels are associated with
apoptosis only in the presence of oxygen radical
scavengers, and increased levels of ROI cause necrosis
only under conditions in which the simultaneous production of NO is lowered [10], we can also suggest that
cell death in OA-affected chondrocytes is principally
NO-mediated apoptosis [41,42], in contrast with minor
apoptotic chondrocyte death [43], but more frequent
ROI-mediated necrosis, in RA chondrocytes.
6
7
8
9
10
11
12
13
14
ACKNOWLEDGMENTS
We thank Dr Franco Piras (I. O. R.) for his help with
statistical analysis, and Mr Keith Smith for language
editing. We are grateful to Dr Rosa Maria Borzı' and Dr
Samantha Paoletti for their suggestions and encouragement. We acknowledge Dr Giorgia Magagnoli for her
assistance with graphics. We also thank Mrs Graziella
Salmi for her assistance in the preparation of the
manuscript, and Mr Luciano Pizzi for his technical
assistance. This work was supported by grants from
IRCCS ‘‘ Istituti Ortopedici Rizzoli ’’ Bologna, MURST
Universita' di Bologna and Progetto Finalizzato
Ministero della Sanita' .
15
16
17
18
19
20
21
REFERENCES
1
Wilder, R. L. (1993) Rheumatoid Arthritis. A.
Epidemiology, pathology and pathogenesis. In Primer on
the Rheumatic Diseases (Schumacher, H. R., Klippel,
J. H. and Koopman, W. J., eds), pp. 86–89, Arthritis
Foundation, Atlanta
2 Brandt, K. D. and Slemenda, C. W. (1993) Osteoarthritis.
A. Epidemiology, pathology and pathophysiology. In
Primer on the Rheumatic Diseases (Schumacher, H. R.,
Klippel, J. H. and Koopman, W. J., eds), pp. 184–188,
Arthritis Foundation, Atlanta
3 Lotz, M., Blanco, F. J., von Kempis, J. et al. (1995)
Cytokine regulation of chondrocyte functions. J.
Rheumatol. 22, 104–108
4 Okada, Y., Shinmei, M., Tanaka, O. et al. (1992)
Localization of matrix metalloproteinase 3 (stromelysin)
in osteoarthritic cartilage and synovium. Lab. Invest. 66,
680–690
5 Olee, T., Hashimoto, S., Quach, J. and Lotz, M. (1999)
IL-18 is produced by articular chondrocytes and induces
proinflammatory and catabolic responses. J. Immunol.
162, 1096–1100
# 2001 The Biochemical Society and the Medical Research Society
22
23
24
25
Tiku, M. L., Liesch, J. B. and Robertson, F. M. (1990)
Production of hydrogen peroxide by rabbit articular
chondrocytes. Enhancement by cytokines. J. Immunol.
145, 690–696
Taskiran, D., Stefanovic-Racic, M., Georgescu, H. and
Evans, C. (1994) Nitric oxide mediates suppression of
cartilage proteoglycan synthesis by interleukin-1.
Biochem. Biophys. Res. Commun. 200, 142–148
Martel-Pelletier, J., McCollum, R., Fujimoto, N., Obata,
K., Cloutier, J. M. and Pelletier, J. P. (1994) Excess of
metalloproteases over tissue inhibitor of metalloprotease
may contribute to cartilage degradation in osteoarthritis
and rheumatoid arthritis. Lab. Invest. 70, 807–815
Blanco, F. J. and Lotz, M. (1995) IL-1-induced nitric
oxide inhibits chondrocyte proliferation via PGE2. Exp.
Cell Res. 218, 319–325
Blanco, F. J., Ochs, R. L., Schwarz, H. and Lotz, M.
(1995) Chondrocyte apoptosis induced by nitric oxide.
Am. J. Pathol. 146, 75–85
Lotz, M. (1999) The role of nitric oxide in articular
cartilage damage. Rheum. Dis. Clin. North Am. 25,
269–282
Tiku, M. L., Gupta, S. and Deshmukh, D. R. (1999)
Aggrecan degradation in chondrocytes is mediated by
reactive oxygen species and protected by antioxidants.
Free Radical Res. 30, 395–405
Clancy, R. M., Amin, A. R. and Abramson, S. B. (1998)
The role of nitric oxide in inflammation and immunity.
Arthritis Rheum. 41, 1141–1151
St Claire, E. W. (1998) Nitric oxide – friend or foe in
arthritis? J. Rheumatol. 25, 1451–1453
Murrel, G. A., Jang, D. and Williams, R. J. (1995) Nitric
oxide activates metalloprotease enzymes in articular
cartilage. Biochem. Biophys. Res. Commun. 206, 15–21
Pelletier, J. P., Mineau, F., Ranger, P., Tardif, G. and
Martel-Pellettier, J. (1996) The increased synthesis of
inducible nitric oxide inhibits IL-1ra synthesis by human
articular chondrocytes : possible role in osteoarthritic
cartilage degradation. Osteoarthritis Cartilage 4, 77–84
Shingu, M., Isayama, T., Yasutake, C. et al. (1994) Role of
oxygen radicals and IL-6 in IL-1-dependent cartilage
matrix degradation. Inflammation 18, 613–623
Lo, Y. Y., Conquer, J. A., Grinstein, S. and Cruz, T. F.
(1998) Interleukin-1 beta induction of c-fos and
collagenase expression in articular chondrocytes :
involvement of reactive oxygen species. J. Clin. Biochem.
69, 19–29
Harris, E. D. (1992) Regulation of antioxidant enzymes.
FASEB J. 6, 2675–2683
Flanders, K. C., Bhandiwad, A. R. and Winokur, T. S.
(1997) Transforming growth factor-betas block cytokine
induction of catalase and xanthine oxidase mRNA levels
in cultured rat cardiac cells. J. Mol. Cell. Cardiol. 29,
273–280
Niwa, Y., Ozaki, Y., Kanoh, T., Akamatsu, H. and
Kurisaka, M. (1996) Role of cytokines, tyrosine kinase,
and protein kinase C on production of superoxide and
induction of scavenging enzymes in human leukocytes.
Clin. Immunol. Immunopathol. 79, 303–313
Sugino, N., Telleria, C. M. and Gibori, G. (1998)
Differential regulation of copper-zinc superoxide
dismutase and manganese superoxide dismutase in the rat
corpus luteum : induction of manganese superoxide
dismutase messenger ribonucleic acid by inflammatory
cytokines. Biol. Reprod. 59, 208–215
Tannahill, C. L., Stevenot, S. A., Eaker, E. Y., Sallustio,
J. E., Nick, H. S. and Valentine, J. F. (1997) Regulation of
superoxide dismutase in primary cultures of rat colonic
smooth muscle cells. Am. J. Physiol. 272, G1230–G1235
Altman, R., Alarcon, G., Appelrouth, D. et al. (1991) The
American College of Rheumatology criteria for the
classification and reporting of osteoarthritis of the hip.
Arthritis Rheum. 34, 505–514
Altman, R., Asch, E., Bloch, D. et al. (1986) Development
of criteria for the classification and reporting of
osteoarthritis. Classification of osteoarthritis of the knee.
Diagnostic and therapeutic criteria committee of the
American Rheumatism Association. Arthritis Rheum. 29,
1039–1049
Oxidative stress in arthritides
26
27
28
29
30
31
32
33
Arnett, F. C., Edworthy, S. M., Bloch, D. A. et al. (1988)
The Arthritis and Rheumatism Association 1987 revised
criteria for the classification of rheumatoid arthritis.
Arthritis Rheum. 31, 315–324
Melchiorri, C., Meliconi, R., Frizziero, L. et al. (1998)
Enhanced and coordinated in vivo expression of
inflammation cytokines and nitric oxide synthase by
chondrocytes from patients with osteoarthritis. Arthritis
Rheum. 41, 2165–2174
Amin, A. R., Di Cesare, P. E., Vyas, P. et al. (1995) The
expression and regulation of nitric oxide synthase in
human osteoarthritis-affected chondrocytes : evidence for
up-regulated neuronal nitric oxide synthase. J. Exp. Med.
182, 2097–2102
Martel-Pelletier, J., McCollum, R., DiBattista, J. et al.
(1992) The interleukin-1 receptor in normal and
osteoarthritic human articular chondrocytes.
Identification as the type I receptor and analysis of
binding kinetics and biologic function. Arthritis Rheum.
35, 530–540
Westacott, C. I., Atkins, R. M., Dieppe, P. A. and Elson,
C. J. (1994) Tumor necrosis factor-α receptor expression
on chondrocytes isolated from human articular cartilage.
J. Rheumatol. 21, 1710–1715
Deleuran, B. W., Chu, C. Q., Field, M. et al. (1992)
Localization of tumor necrosis factor receptors in the
synovial tissue and cartilage-pannus junction in patients
with rheumatoid arthritis. Implications for local actions
of tumor necrosis factor α. Arthritis Rheum. 35,
1170–1178
Deleuran, B. W., Chu, C. Q., Field, M., Brennan, F. M.,
Katsikis, P. and Maini, R. N. (1992) Localization of
interleukin-1 alpha, type 1 interleukin-1 receptor and
interleukin-1 receptor antagonist in the synovial
membrane and cartilage\pannus junction in rheumatoid
arthritis. Br. J. Rheumatol. 31, 801–809
Pham Huu, T., Marquetty, C., Amit, N. and Hakim, J.
(1986) Effect of degranulation on superoxide dismutase
activity in human neutrophils. Free Radical Biol. Med. 2,
213–217
34
35
36
37
38
39
40
41
42
43
Sumii, H., Inoue, H., Onoue, J., Mori, A., Oda, T. and
Tsubokura, T. (1996) Superoxide dismutase activity in
arthropathy : its role and measurement in the joint.
Hiroshima J. Med. Sci. 45, 51–55
Igari, T., Kaneda, H., Horiuchi, S. and Ono, S. (1982) A
remarkable increase of superoxide dismutase activity in
synovial fluid of patients with rheumatoid arthritis. Clin.
Orthop. 162, 262–287
Biemond, P., Swaak, A. J. G. and Koster, J. F. (1984)
Protective factors against oxygen free radicals and
hydrogen peroxide in rheumatoid arthritis synovial fluid.
Arthritis Rheum. 27, 760–765
Mazzetti, I., Grigolo, B., Borzı' , R. M., Meliconi, R. and
Facchini, A. (1996) Serum copper\zinc superoxide
dismutase levels in patients with rheumatoid arthritis. Int.
J. Clin. Lab. Res. 26, 245–249
Mao, G. D., Thomas, P. D., Lopaschuk, G. D. and
Poznansky, M. J. (1993) Superoxide dismutase (SOD)catalase conjugates. Role of hydrogen peroxide and the
Fenton reaction in SOD toxicity. J. Biol. Chem. 268,
416–420
Tiku, M. L., Yan, Y. P. and Chen, K. Y. (1998) Hydroxyl
radical formation in chondrocytes and cartilage as
detected by electron paramagnetic resonance
spectroscopy using spin trapping reagents. Free Radical
Res. 29, 177–187
Imadaya, A., Terasawa, K., Tosa, H., Okamoto, M. and
Toriizuka, K. (1988) Erythrocyte antioxidant enzymes are
reduced in patients with rheumatoid arthritis. J.
Rheumatol. 15, 1628–1631
Blanco, F. J., Guitian, R., Vazquez-Martul, E., De Toro,
F. J. and Galdo, F. (1998) Osteoarthritis chondrocytes die
by apoptosis. A possible pathway for osteoarthritis
pathology. Arthritis Rheum. 41, 284–289
Hashimoto, S., Ochs, R. L., Komiya, S. and Lotz, M.
(1998) Linkage of chondrocyte apoptosis and cartilage
degradation in human osteoarthritis. Arthritis Rheum. 41,
1632–1638
Kim, H. A. and Song, Y. W. (1999) Apoptotic
chondrocyte death in rheumatoid arthritis. Arthritis
Rheum. 42, 1528–1537
Received 22 January 2001/27 April 2001; accepted 30 July 2001
# 2001 The Biochemical Society and the Medical Research Society
599