Archives of Biochemistry and Biophysics 583 (2015) 150e157

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics

journal homepage: www.elsevier.com/locate/yabbi

Quercetin reduced inflammation and increased antioxidant defense in

rat adjuvant arthritis
C. Gardi a, b, *, 1, K. Bauerova c, d, 1, B. Stringa b, V. Kuncirova c, L. Slovak c, S. Ponist c,
F. Drafi c, L. Bezakova d, I. Tedesco a, A. Acquaviva b, S. Bilotto a, G.L. Russo a, **
a
Institute of Food Sciences, National Research Council, 83100 Avellino, Italy
b
Department of Molecular and Developmental Medicine, University of Siena, Via A. Moro, I-53100 Siena, Italy
c
Institute of Experimental Pharmacology and Toxicology, Slovak Academy of Sciences, Dubravska Cesta 9, SK-841 04 Bratislava, Slovak Republic
d
Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic

a r t i c l e i n f o a b s t r a c t

Article history: Novel therapies for rheumatoid arthritis also include the use of naturally occurring compounds pos-
Received 14 July 2015 sessing antioxidant properties. In the present work, the effects of oral administration of quercetin were
Received in revised form investigated in a rat model of adjuvant arthritis. Arthritis was induced by a single intradermal injection of
14 August 2015
heat-inactivated Mycobacterium butyricum in incomplete Freund's adjuvant. The experimental groups
Accepted 17 August 2015
were treated with an oral daily dose of 150 mg/kg b.w. of quercetin for 28 days. Results indicated that
Available online 20 August 2015
quercetin was able to ameliorate all markers of inflammation and oxidative stress measured. Quercetin
lowered levels of interleukin-1b, C-reactive protein, and monocyte chemotactic protein-1 and restored
Keywords:
Quercetin
plasma antioxidant capacity. In addition, quercetin inhibited the enzymatic activity of pro-inflammatory
Adjuvant arthritis 12/15-lipoxygenase in lung and liver and increased the expression of heme oxygenase-1 in joint and lung
NF-kB of arthritic rats. Finally, quercetin inhibited the 2-fold increase of NF-қB activity observed in lung, liver
ERK pathway and joint after induction of arthritis.
Inflammation © 2015 Elsevier Inc. All rights reserved.
Oxidative stress

1. Introduction agreement with findings of other authors who referred to the

important role of oxidative stress in the pathogenesis of RA [5e7].
Rheumatoid arthritis (RA) is a chronic autoimmune disease In the last decade, the potential involvement of flavonoids with
affecting approximately 1% of the whole world population. Patients antioxidant properties in RA has been evaluated [8e10]. In this
with RA have a reduced life quality (joints and bones degeneration, context, the limited side effects of quercetin (QUE) and its well-
muscle weakness, persistent pain) and require long-life therapy. A known pharmacological activities suggested a potential applica-
common effect of long-term therapy is the development of resis- tion as an adjuvant natural drug for the treatment of RA [10]. QUE
tance to treatment and also an increased occurrence of adverse (3,30,40,5,7-pentahydroxyflavone) is the major dietary flavonol
effects. Due to these reasons, a continuous need for new agents in found in fruits, vegetables and beverages, such as tea and red wine
the therapy of RA is envisaged. Primary and dominant processes in [11]. Several epidemiological and experimental studies support the
the etiopathogenesis of RA are immunological mechanisms, closely antioxidant, anti-inflammatory, anti-angiogenic, anti-proliferative
related to redox imbalance in the organism, which may potentiate and pro-apoptotic effects of this molecule [12e14]. In Western
chronic inflammatory processes [1]. Our studies [2e4] are in populations, the estimated daily intake of total flavonols is in the
range of 20e50 mg/day, of which about 15e20 mg correspond to
QUE glycosides [15].
The existence of a functional link between the intake of QUE and
* Corresponding author. Department of Molecular and Developmental Medicine, other flavonoids and RA is supported by circumstantial evidence
University of Siena, Via A. Moro 2, I-53100 Siena, Italy. deriving from pre-clinical studies on primary cells and animal
** Corresponding author. Istituto Scienze dell'Alimentazione, Consiglio Nazionale models, as well as clinical studies. Early in the 1997, it was reported
Delle Ricerche, Via Roma 64, 83100 Avellino, Italy.
that QUE suppressed the increase in the mRNA for interleukin 8 (IL-
E-mail addresses: [email protected] (C. Gardi), [email protected]
(G.L. Russo). 8) and monocyte chemotactic protein-1 (MCP-1) in cultured human
1
Equal contribution. synovial cells stimulated by tumor necrosis factor-alpha (TNF-a) in

http://dx.doi.org/10.1016/j.abb.2015.08.008
0003-9861/© 2015 Elsevier Inc. All rights reserved.
 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157 151

a dose dependent manner. TNF-a is present in synovial fluid and oxidized low density lipoprotein (Ox-LDL) and high sensitivity CRP
induces the expression of pro-inflammatory cytokines in synovial did not show any significant difference between QUE and placebo
cells of patients with RA. The suppression was dose dependent and groups [24]. On the opposite, in an ex vivo study, where neutrophils
probably induced by the inhibition of TNF-a mediated nuclear were isolated from RA patients versus healthy subjects and stim-
factor kappa-light-chain-enhancer of activated B cells (NF-kB) ulated by in vitro prepared immune complex before treatment with
activation [16]. A decade later, the anti-RA capacity of QUE was 4 different flavonols (galangin, kaempferol, QUE, and myricetin),
confirmed in synoviocytes isolated from rabbit where the molecule QUE was the most effective in reducing superoxide anion produc-
inhibited proliferation of 30e40% at very low micromolar concen- tion with an IC50 of 1.71 mM [25]. It is worthwhile to note that the
trations (<10 mM). It must be considered that proliferation of syn- applied concentrations were in the physiological range of those
oviocytes in RA contributes to the establishment of the so-called measured in vivo after supplementation with QUE or other flavo-
“pannus formation”, a lesion accompanied by restriction of joint nols [15,26]. Matsuno et al. [27] performed a study with osteoar-
movement and the generation of pro-inflammatory cytokines [17]. thritic patients and RA patients, in which QUE was administered in
In human rheumatoid synovial fibroblasts activated by interleukin form of glucosamine-chondrotin-QUE glucoside combination. The
1 beta (IL-1b), QUE inhibited proliferation and induced apoptosis patients were treated for 3 months with oral doses of QUE gluco-
starting from 20 mM concentration. The mode of action was double: side (45 mg/day). Significant improvement in pain symptoms, daily
i. inhibition of both the expression of IL-1b-induced mRNA and activities (walking and climbing up and down stairs) and changes
protein of matrix metalloproteases MMP-1, MMP-3, and COX-2 and in the synovial fluid properties were observed in osteoarthritic
PGE2 production; ii. inhibition of extracellular signal-regulated patients. No beneficial effects were observed in RA subjects [27].
kinases (ERK) signal pathways and NF-kB activation both medi- Therefore, in our study we re-investigated the effect of QUE
ated by IL-1b [18]. orally administered in a dose of 150 mg/kg in AA with the aim to
The anti-RA effect of QUE was confirmed in several animal prove its anti-arthritic potential, as well as to study its mechanisms
models of experimentally induced arthritis. In a rat model of gouty of action. We focused on the key two processes in arthritis:
arthritis, QUE treatment (100e400 mg/kg) ameliorated edema by inflammation and oxidative stress. Both processes were evaluated
decreasing histological signs of acute inflammation and attenuating in plasma and in selected tissues as joint, liver and lung
several markers of inflammation [19]. QUE was more effective than homogenates.
hesperidin, but less than rutin (all tested at a dose of 80 mg/kg and
administered intraperitoneally) in inhibiting acute and chronic 2. Materials and methods
inflammation in rats where experimental arthritis was induced
following the method of adjuvant-carrageenan-induced inflam- 2.1. Animals, experimental design and treatments
mation [20]. Considering that rutin differs from QUE for the pres-
ence of rutinose in position 3, it is possible to hypothesize that the Male Lewis rats weighing 160e180 g were obtained from the
2-fold higher efficacy of rutin than QUE in arthrogram scores can be Breeding Farm Dobra Voda (Slovakia). The rats had free access to
attributed to pharmacokinetic factors [20]. In a subsequent work, standard pelleted diet and tap water. The experimental protocol
the same group, comparing the effects of different flavonoids on was approved by the Ethics Committee of the Institute of the
different rats and mice models, confirmed that rutin was the only Experimental Pharmacology and Toxicology, by the Slovak State
effective against chronic-like arthritis, principally in adjuvant Veterinary and Food Administration in accordance with the Euro-
arthritis (AA), but QUE resulted the most active in reducing the paw pean Convention for the Protection of Vertebrate Animals Used for
edema induced by carrageenan [21]. In AA induced in female Lewis Experimental and Other Scientific Purposes, and by Slovak legis-
rats by subcutaneous injection of inactivated Mycobacterium lation. AA was induced by a single intradermal injection of heat-
butyricum, oral administration of QUE (5
160 mg/kg) clearly inactivated M. butyricum in incomplete Freund's adjuvant (Difco
decreased clinical signs of arthritis. Importantly, the dosage was Laboratories, Detroit, MI, USA). The injection was performed near
selected to be comparable to that administered to patients affected the tail base. The experiment included healthy animals (CO),
by prostatitis who received QUE as dietary supplement 1 g/day healthy animals treated with QUE (CO-Q) in an oral daily dose of
[22]. When the molecule was given by intracutaneous injection in 150 mg/kg b.w. (body weight) during 28 days, arthritic animals not
AA-induced rats at lower doses (5
60 mg/kg), the anti-arthritic treated (AA), arthritic animals treated with QUE (AA-Q) in an oral
effects were similar, while injection of relatively low doses daily dose of 150 mg/kg b.w. during 28 days. In each group 10 an-
(5
30 mg/kg) prior to AA induction significantly reduced arthritis imals were used. After the animals have been sacrificed under deep
signs, suggesting multiple approaches (different doses and modes ketamin/xylasine anesthesia, blood for plasma preparation and
of administrations) to exploit the clinical potentiality of QUE as an tissues for homogenate preparation (joint, lung and liver) were
anti-arthritic agent. Finally, analysis of cumulated arthritic scores taken at the end of the experiment (day 28). Tissue were imme-
clearly indicated that high oral doses were most efficient in diately frozen and stored at 80  C until analysis. Blood samples
reducing arthritic signs, followed by lower intracutaneous thera- were centrifuged at 2400
g for 15 min at 4  C and stored at 80  C
peutic or preventive QUE doses [22]. until analyses.
More recently, the study of the effects of QUE in RA was
extended to human subjects with contradictory results compared 2.2. Change of body weight
to pre-clinical studies. In a randomized controlled trial aimed to
investigate the efficacy of antioxidant supplementation in RA pa- Change of body weight (CBW; g) was measured on days 1, 14 and
tients, QUE was administered together with vitamin C 28. CBW was calculated as the difference of the body mass
(166 mg þ 133 mg/capsule, respectively) for 4 weeks in 26 subjects. measured on days 14 and 28 and the body weight measured at the
No changes in the levels of serum pro-inflammatory cytokines and beginning of the experiment (day 1).
C-reactive protein (CRP) in RA patients after supplementation were
observed [23]. In a more recent work, 51 women affected by RA 2.3. Arthritic score
were supplemented with 500 mg/day of QUE for 8 weeks. As in the
previous study, measurements of several markers of inflammation, The arthritic score was measured as the total score of hind paw
such as plasmatic total antioxidant capacity, malondialdehyde, volume (ml, max. points 8) plus paw diameter of forelimb (mm,
152 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157

max. points 5) plus diameter of scab in the site of M. butyricum liver tissues). Linoleic acid (99%, SigmaeAldrich, USA) was used as a
application, measured parallel to the spinal column (mm, max. substrate prepared in solubilized state in the concentration of
points 5) for each animal [28]. 0.2143
105e0.7143
105 M as previously described [4]. The
assay of LOX was monitored as an increase in the absorbance at
2.4. Measurement of C-reactive protein (CRP) in plasma 234 nm which reflects the formation of the hydroperoxylinoleic
acid. For the LOX activity assay, an UV/VIS Spectrometer Per-
For the determination of rat CRP concentration in plasma (mg/ kineElmer Lambda 35 (USA) was used. The reaction medium
ml), the ELISA kit from Immunology (Consultant Laboratories, Inc.) contained a 50 mM TriseHCl buffer (pH 7.0), 2.5 mL of the enzyme
was used. The reaction of secondary biotin-conjugated anti-rat CRP protein and solubilized linoleic acid.
antibody was evaluated by streptavidin-horseradish peroxidase
(HRP). The tetramethylebenzidine reaction with HRP bound to 2.9. Immunoblot analysis of heme oxygenase-1 (HO-1)
immune complex was measured at 450 nm (Microplate reader
Labsystems Multiskan RC). The results were calculated using the Protein samples from joint and lung homogenates were
standard calibration curve on internal standards. resolved on a 10% SDS-polyacrylamide gel and transferred onto
nitrocellulose membrane (Whatman GmbH, Dassel, Germany).
2.5. Measurement of monocyte chemotactic protein-1 (MCP-1) in Immunoblots were performed using primary antibodies for HO-1
plasma (1:1000; Abcam, Cambridge, UK) and b-actin was used as loading
control. Immunodetected proteins were visualized using ECL kit
For determination of MCP-1 concentration in plasma (pg/ml), (BioRad, Hercules, CA, USA).
Instant ELISA kit from eBioscience® was used. Assay procedures
were applied as described in the product manual. Rat chemokine
2.10. Preparation of whole cell extract for NF-kB determination
present in the samples binds to anti-rat chemokine antibodies
adsorbed to the microwells. The reaction of secondary biotin-
Whole cell extracts from joint, left lung and liver were prepared
conjugated anti-rat chemokine antibody was evaluated by
by using Active Motif nuclear extract kit (Carlsbad, CA, USA) as
streptavidin-HRP. Tetramethyl-benzidine reaction with HRP bound
described by the manufacturer. Protein concentration in whole cell
to immune complex was measured at 490 nm in comparison with
extracts was determined by the Bradford protein assay (BioRad)
reference wavelength 620 nm (microplate reader MRX II, Dynex,
using bovine serum albumin as standard. NF-kB activation was
USA). The results were calculated from standard calibration curve
monitored by TransAM NF-kB p65 Transcription Factor Assay Kit
on internal standards.
(Active Motif). Absorbance was measured at 450 nm using micro-
plate reader (Perkin Elmer Applied Biosystems). Results were
2.6. Measurement of interleukin 1b (IL-1b) in plasma
expressed as absorbance per milligram of total protein.
For the determination of IL-1b concentration in plasma, the
ELISA kit from R&D Systems Quantikine® was used. Assay pro- 2.11. Western blot analysis of extracellular signal-regulated protein
cedures were used as described in the product manual. Rat cytokine kinase (ERK)
present in the samples binds to anti-rat cytokine antibodies
absorbed in the microwells. The reaction of secondary biotin- Phosphorylation of ERK was analyzed by Western blotting.
conjugated anti-rat cytokine antibody is evaluated by HRP. The Samples were homogenized in lysis buffer (20 mM TriseHCl, pH 8,
tetramethylbenzidine reaction with HRP bound to immune com- 150 mM NaCl, 1% Triton X-100) containing protease inhibitor
plex was measured at 490 nm in comparison with the reference cocktail, 1 mM Na3VO4, 5 mM b-glycerophosphate, and protein
wavelength 620 nm (microplate reader MRX II). The results were concentration was determined by the method of Bradford [30].
calculated using the standard calibration curve on internal Sixty mg of total protein of each extract was separated by 12% SDS-
standards. polyacrylamide gel and transferred onto a nitrocellulose membrane
(Whatman). Membranes were incubated overnight at 4  C with a
2.7. Plasma antioxidant power primary antibody anti-phospho-44/42 ERK (1:1000, Cell Signaling,
Celbio, Milan, Italy) and then with the secondary antibody HRP-
The total antioxidant capacity was measured in the plasma of conjugated (BioRad) (1:5000). Immunodetected proteins were
rats treated as indicated above employing the ABTS (2,20 -azino- visualized using ECL assay kit (BioRad) following the manufac-
bis(3-ethylbenz-thiazoline-6-sulfonic acid)) radical cation decol- turer's recommended protocol. b-actin was used as the loading
orization. Briefly, a mixture containing ABTS solution (10 mM) and control.
hydrogen peroxide (2 mM) was stored in the dark at 4  C overnight
before use. The produced ABTSþ solution was diluted (1:10) to 2.12. Statistical analysis
obtain an absorbance of approximately 0.31 at 660 nm. In a cuvette
1000 mL of a buffer pH 3.6, 10 mL of plasma and finally 25 mL of ABTS Mean and S.E.M. values were calculated for each parameter in
were added before measurement of the absorbance (660 nm) after each group (8e10 animals in each experimental group). Statistically
5 min. A blank was run in each assay and determinations were significant differences among treated group, untreated group and
carried out in triplicate. Results were expressed in micromolar control groups were tested using parametric Analysis of Variance
equivalent of ascorbic acid, an antioxidant present in the plasma (ANOVA). Post hoc tests (TukeyeKramer (ANOVA)) were applied in
[29]. situation where differences among groups were significant at level
of significance a ¼ 0.05. After post hoc testing the following sig-
2.8. Tissue activity of 12/15-lipoxygenase (LOX) in liver and lung nificance designations were specified as follows: extremely sig-
nificant (P < 0.001), highly significant (P < 0.01), significant
Concentration of proteins in lung and liver homogenates was (P < 0.05), not significant (P > 0.05). Since, for measurement of
determined by using the Bradford method [30] and expressed in ABTS, HO-1 and ERK kinase samples from 4 to 5 animals were used,
mg/ml of enzyme preparation (cytosolic fraction from rat lung and unpaired Student's t-test was applied.
 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157 153

3. Results Table 2
Effect of quercetin on inflammatory markers assessed in plasma.

3.1. Effect of QUE on clinical parameters and on selected CO AA CO-Q AA-Q

inflammatory parameters in plasma IL-1b [pg/ml] 4.83 ± 2.22 44.72 ± 10.62*** 1.92 ± 0.92 15.64 ± 1.66þþ
CRP [mg/ml] 544 ± 30 727 ± 49** 421 ± 51 605 ± 23
The ability of QUE to ameliorate the clinical parameters MCP-1 [pg/ml] 3377 ± 387 5445 ± 468* 3602 ± 281 3732 ± 442þ
following induction of AA in rats was evaluated. As reported in Levels of IL-1b, interleukin-1b; CRP, C-reactive protein; and MCP-1, monocyte
Table 1, a significant decrease in body weight and an increase in chemotactic protein 1; were measured in plasma on the 28th day. Data are
arthritic score were observed at both days monitored in AA group. means ± SEM. *P < 0.01 vs. CO, **P < 0.01 vs. CO, ***P < 0.001 vs. CO, þP < 0.05 vs.
AA, þþP < 0.001 vs. AA. CO, control group; AA, adjuvant arthritis group; CO-Q,
The administration of QUE in healthy and in arthritis animals did
quercetin treated control group; and AA-Q, quercetin treated adjuvant arthritis
not change clinical parameters observed in this study. At day 28, the group.
arthritic score almost doubled in the AA group compared to control
and QUE treated groups and this increase persisted after adminis-
tration of QUE to AA animals. When selected markers of inflam-
mation (such as IL-1b, CRP and MCP-1) were measured, their level
was significantly increased as a result of arthritis induction, as ex-
pected (Table 2). QUE ameliorated all parameters with significant
difference achieved for IL-1b and MCP-1. The QUE treated group did
not show any significant change in inflammatory markers
compared to healthy control animals. A slight decrease in IL-1b and
CRP and a small increase in MCP-1 levels were assessed in plasma
following QUE mono-treatment (Table 2).

3.2. Quercetin increased plasma antioxidant power

We measured the antioxidant capacity in plasma of AA rats

following QUE treatment using ABTS assay. As reported in Fig. 1, the
total antioxidant capacity decreased of about 25e30% in AA group.
However, QUE treatment (AA-Q group) almost re-established the
basal levels, confirming the capacity of this molecule to counteract
the oxidative stress associated to AA induction. Moreover, healthy Fig. 1. Effect of quercetin on plasma antioxidant power. The total antioxidant capacity
animals treated with QUE (CO-Q group) showed a higher antioxi- was measured in the plasma of rats treated as described in Methods section employing
the ABTS assay. Data are means ± SEM. *P < 0.01 and **P < 0.001 vs CO; þP < 0.01 and
dant capacity compared to controls (CO) (Fig. 1). þþ
P < 0.001 vs AA. CO, control group; AA, adjuvant arthritis group; CO-Q, quercetin
treated control group; and AA-Q, quercetin treated adjuvant arthritis group.

3.3. Effect of QUE on activity of 12/15-LOX in liver and lung

homogenates administered with QUE (CO-QUE group) in comparison with the
control group (Fig. 2).
The enzyme 5-LOX catalyzes the conversion of arachidonic acid
into the leukotrienes, whose production has been associated to
inflammation in arthritis. Suppression of 5-LOX expression ame- 3.4. Effect of QUE on HO-1 expression
liorates clinical parameters in RA and AA [31,32]. A similar role can
be attributed to 15-LOX [33]. Therefore, the level of expression of HO-1 expression was evaluated as tissue marker of antioxidant
12/15-LOX were measured in lung and liver of rats treated with defense in joint and lung. The rationale for the selection of these
QUE. In our previous paper evaluating effect of pinosylvin in AA, we two tissues resides in the observation that this enzyme plays a
found that inflammatory pathological changes were present also in cytoprotective role not only in the lung [34], but also in the synovial
organs primarily not affected by AA as lung and liver [4]. For this tissue of RA patients, where it regulates joint inflammation [35]. AA
reason, also in this study these two organs were included in rats showed a different trend in HO-1 expression from the two
experimental evaluation. The effect of QUE on activity of 12/15-LOX organs analyzed. HO-1 protein level increased in joint (Fig. 3A), but
was comparable in both homogenates. Activity increased in decreased in the lung (Fig. 3B), whereas no significant changes
arthritic animals in comparison with healthy animals. After were detected in the liver (data not shown). In healthy animals,
administration of QUE a significant decrease to the control levels QUE did not affect HO-1. A protective effect was observed in both
was assessed in the AA group. A minor although significant joint and lung from AA animals receiving QUE. In fact, in lung from
decrease of 12/15-LOX activity was also observed in healthy rats AA-Q rats QUE restored HO-1 to control levels (Fig. 3B), while in

Table 1
Effect of quercetin on clinical markers in adjuvant arthritis.

CO AA CO-Q AA-Q

Body weight [g], (d 14) 218 ± 11 190 ± 7 214 ± 2 186 ± 7

Body weight [g], (d 28) 231 ± 8 187 ± 5*** 251 ± 2 184 ± 6
Arthritic score [points], (d 14) 10.0 ± 0.3 15.0 ± 0.4*** 10.0 ± 0.0 15.0 ± 0.6
Arthritic score [points], (d 28) 10.0 ± 0.1 20.0 ± 0.7*** 10.0 ± 0.0 21.0 ± 1.0

Clinical parameters, body weight and arthritic score were measured on the 14th and 28th day. Data are means ± SEM. ***P < 0.001 vs. CO. CO, control group; AA, adjuvant
arthritis group; CO-Q, quercetin treated control group; and AA-Q, quercetin treated adjuvant arthritis group.
154 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157

investigated [36,37], levels of p65 DNA binding were measured in

joint, lung and liver. A marked increase of NF-kB activation was
observed in tissues from AA animals (ranging from 1.87 to 2.21 fold
increase). In AA-Q rats QUE strongly inhibited NF-kB activation in
joint (55.2%), lung (57.8%) and liver (55.8%) while QUE did not
affect this parameter in healthy animals (CO-Q) (Fig. 4).

3.6. QUE reduced ERK activation in AA animals

To better investigate the protective effect of QUE in AA, we

measured the level of activation of ERK pathway. In fact, it has been
reported that ERK activation is enhanced in patients meeting RA
criteria compared to other diagnoses [38]. In addition, ERK in-
hibitors have been suggested as a treatment for RA patients,
including those who are nonresponsive to the anti-cytokine ther-
Fig. 2. Effect of quercetin on 12/15-LOX activity. Activity of LOX was assessed lung and apies [39]. As reported in Fig. 5, in AA rats a significant increase in
liver. Data are means ± SEM. ***P < 0.001vs CO, þþþP < 0.001 vs AA. CO, control group; p44/42 ERK was observed in joint and lung (panels A and B), while
AA, adjuvant arthritis group; CO-Q, quercetin treated control group; and AA-Q, quer-
not significant changes were detected in liver (data not shown).
cetin treated adjuvant arthritis group.
QUE treatment strongly decreased ERK phosphorylation in AA an-
imals, while in healthy animals QUE did not show any significant
joint further increased HO-1 expression (about 1.5 times) effect in both joint and lung. These results indicate that the acti-
compared to AA animals (Fig. 3A). vation of ERK may be involved, at least in part, in the QUE-mediated
protective effects observed in AA-animals.
3.5. Effect of QUE on the levels of NF-kB activation
4. Discussion
Since NF-kB activation plays a central role in inflammatory and
oxidative processes and its role in arthritis has been largely The present manuscript highlights the capacity of QUE to
improve the anti-inflammatory status and antioxidant defenses in
AA rats. At systemic level, QUE increases the plasma antioxidant
capacity (Fig. 1) at a dosage (150 mg/kg) corresponding to the
approximate daily intake of QUE glycosides (15e20 mg/die) within
a regular diet in the human population [15]. Considering the
complex series of biotransformation reactions which QUE un-
dergoes following its uptake and metabolism at intestinal and he-
patic levels [14,15], it is unlikely that the effects reported in Fig. 1
are attributable to QUE aglycone; probably, a mixture of QUE me-
tabolites differently conjugated (e.g., addition of methyl, sulfate
groups and glucuronic acid) are responsible for the results of ABTS
assay shown in Fig. 1. Although we are aware that the use of this
method is controversial, it has been previously applied to demon-
strate the plasma antioxidant capacity of QUE in rats [40]. In
addition, data exist in the literature on the possibility that QUE can
maintain its antioxidant properties in vivo, in different cellular
models, evidencing that the in vivo scavenging activity of QUE
generates different oxidation products [41e43]. Based on this
observation, we expected a rationale improvement of the

Fig. 3. Effect of quercetin on HO-1 expression in rat joint (A) and lung (B). Levels of Fig. 4. Effect of quercetin on NF-қB activation. Levels of NF-қB were assessed by
protein were measured by immunoblot and densities of the bands were quantified measuring p65 DNA binding in joint, lung and liver. Data are reported as percentage of
with an imaging densitometer. Protein quantification was expressed as a ratio to b- control values. CO joint 1.27 ± 0.02, CO lung 0.46 ± 0.04, CO liver 0.5 ± 0.06 Abs/mg of
actin. Data are means ± SEM. **P < 0.01 vs CO, ***P < 0.001 vs CO, þþP < 0.01 vs total protein. Data are means ± SEM. **P < 0.01 vs CO, ***P < 0.001 vs CO, þþP < 0.01 vs
AA, þþþP < 0.001 vs AA. CO, control group; AA, adjuvant arthritis group; CO-Q, AA, þþþP < 0.001 vs AA. CO, control group; AA, adjuvant arthritis group; CO-Q,
quercetin treated control group; and AA-Q, quercetin treated adjuvant arthritis group. quercetin treated control group; and AA-Q, quercetin treated adjuvant arthritis group.
 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157 155

More complex, but also fascinating is the potential mechanism

explaining the increased expression of HO-1 following QUE treat-
ment in AA-rats (Fig. 3). HO-1 is an oxidative stress-responsive
protein that shows anti-inflammatory and antioxidant activities
[47] not only in the lung [34], but also in the synovial tissue of RA
patients [35]. In our model, AA rats showed an increase in the
expression of HO-1 in the joint, which can be interpreted as a
stimulus of antioxidant defense. Kitamura [48] showed that the
levels of this protein are significantly elevated in human synovial
fluid and correlated with the plasma levels of CRP, a cytokine of the
acute phase of inflammation. This is in according with our data,
which show increased levels of CRP in AA rats. These studies could
support the idea that the local concentration of HO-1 in synovial
fluid may reflect the severity of inflammation in the joint [48]. In
addition, as suggested in the literature [49], QUE and other poly-
phenols can exert a protective effect against inflammation and
oxidative damage upregulating HO-1. Conversely, levels of HO-1
decreased in the lung of AA rats, suggesting that in this organ
oxidative stress overwhelmed the antioxidant defense. The reason
for the different response in joint and lung is not clear, but it may be
related to the cell type-specific effects of HO-1 [50]. Anyway, QUE is
able to restore the expression of lung HO-1 to control levels. A
similar protective effect of QUE on HO-1 was also observed in rats in
a model of lung injury by paraquat [51]. These results are of
particular importance since pharmacological up-regulation of HO-1
has been demonstrated to provide a protective response in in-
flammatory diseases [50], including arthritis [52].
Regarding the possible mechanism of action of QUE, various
hypotheses can be put forward. It is known that HO-1 is one of the
antioxidant enzymes whose expression is under the control of
Nrf2-Keap1 system [53,54]. Recent data suggest that one possible
mechanism explaining the antioxidant effects of phytochemicals
via Nrf2 activation may derive from their electrophilic nature [55];
in fact, dietary phytochemicals can be oxidized to electrophilic
Fig. 5. Effect of quercetin on ERK activation. Levels of p44/42 MAPK were measured by hydroquinones and quinones which, in turns, can react with and
immunoblot in rat joint (A) and lung (B). Densities of the bands corresponding to p44/ oxidize specific cysteine residues in Keap1, allowing the tran-
42 were quantified with an imaging densitometer. Quantification of proteins was scription factor Nrf2 to leave its docking and inactive position into
expressed as a ratio to b-actin. Data are means ± SEM. ***P < 0.001 vs
the cytoplasm and translocate to the nucleus. Examples of this
CO, þþþP < 0.001 vs AA. CO, control group; AA, adjuvant arthritis group; CO-Q,
quercetin treated control group; and AA-Q, quercetin treated adjuvant arthritis group. mechanism are already present in the literature [55,56] and can be
extended to QUE also [57,58]. In other words, we are hypothesizing
antioxidant defenses and anti-inflammatory status in those tissues that the increased oxidative stress in AA rats triggers a partial
more directly involved in RA pathogenesis, such as joint, lung and oxidation of QUE generating its electrophilic forms which are able
liver. Data reported in the present work go in this direction and the to interact and oxidize specific (and still unknown) cysteine resi-
beneficial effects of QUE resemble the pleiotropic and multifunc- dues among the 27 present on Keap1 causing its inactivation and,
tional nature of the molecules discussed elsewhere [14,15,44]. In consequentially, allowing Nrf2 activation. Studies are in progress to
fact, we hypothesized that the protection offered by QUE against AA verify experimentally this hypothesis.
in the different assays here reported (Figs. 2e5) can be explained The possibility that QUE can regulate NF-kB activity and inhibit
evoking the ability of the molecule to trigger different molecular inflammatory cytokine expression reducing inflammatory re-
targets. actions in different systems has been largely explored in the liter-
The ability of QUE to potentiate an anti-inflammatory response ature [59e61]. Involvement of NF-kB in RA has been demonstrated
in AA was supported by the ability of the molecule to inhibit 12/15- in several papers [4,62,63] and its inhibition by QUE has also been
LOX activity (Fig. 2). Our results are confirmed by recent observa- associated to protection against RA [10]. We confirmed this effect
tions demonstrating that several flavonoids may act as LOX in- measuring a reduction of NF-kB activation in all tissues examined in
hibitors by inhibiting the formation of LTB4. In the case of QUE, the AA rats following QUE treatment. Similar findings have been
molecule showed an IC50 of 4.0 ± 1.2 mM on the production of LTB4 observed in vivo in other animal models of inflammation [64,65]
by human neutrophils [45]. In addition, in vitro studies demon- and in agreement with previous data obtained in vitro [66,67].
strated the ability of QUE to directly bind and inhibit soybean LOX-1 Although the molecular mechanisms involved in the suppressive
with a LineweavereBurk double reciprocal plot attesting an un- effects of flavonoids on NF-kB are still not clear, several hypotheses
competitive inhibition type [45]. However, it must be mentioned can be suggested. One of the proposed mechanisms is the direct
that the positive role of 5-LOX in sustaining RA has been partially inhibition of the intracellular signaling pathways leading to the
questioned by a recent work where it was reported that products of activation of NF-kB. QUE can significantly reduce phosphorylation
the 5-LOX activity were not required for the development of disease and degradation of IkBa (inhibitor of kBa) and nuclear level of NF-
in Lyme model of arthritis [46]. For these reasons, the Authors kB [58,64]. Based on these observations, one might assume that
suggest caution when targeting 5-LOX as therapy for inflammatory QUE inhibits inflammatory responses, including the production of
diseases.
156 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157

inflammatory cytokines and activity of LOX, mainly through the separates pro-inflammatory processes induced by RA from the
suppression of NF-kB activation. anti-inflammatory cellular responses triggered by QUE. This may be
Together with NF-kB, also ERK are central regulators in inflam- a consequence of the relatively low dose employed in the present
matory processes, including the development of RA [68,69]. Acti- work designed to mimic dietary supplementation of QUE. However,
vation of NF-kB and MAPK induces the production of pro- the effect of QUE on healthy rats observed in the present study
inflammatory cytokines and MMPs in RA [18,62]. In a murine suggests a potential preventive use of the molecule. We hypothe-
model of collagen-induced RA, the selective MEK inhibitor, size that the chronic administration of QUE at a dose superim-
PD184352, inhibited paw edema and clinical arthritis scores in a posable to that applied in the present study “before” the induction
dose-dependent manner [70]. In agreement with this observation, a of AA may result in a better response to inflammation. In fact, target
different ERK inhibitor, FR180204, has been shown to be effective tissues may react more efficiently to the pro-inflammatory insult
against RA enforcing the search for potent ERK inhibitors in the since their anti-inflammatory and antioxidant defenses have been
therapy of RA [39]. In this view, we observed an activation of ERK 44/ already potentiated by QUE. In such scenario, the threshold can be
42 in AA rats which was reduced to basal level following QUE bypassed resulting in absence of disease or a mild form of arthritis.
treatment. The easiest explanation of this result is the well-known An alternative possibility is to associate QUE with methotrexate
capacity of QUE to act as a direct inhibitor of MEK, with a Ki of (MTX) in combination therapy to follow disease progression and
about 1e2 mM [44]. Structural studies indicate that QUE directly inflammation in arthritic rats. MTX in small doses became the most
binds with MEK-1 ex vivo and in vitro in a pocket separate from but frequently used disease-modifying anti-rheumatic drug in the
adjacent to the ATP-binding site of MEK-1. It is interesting that QUE therapy of RA. In AA rats, the combined administration of MTX with
exerted stronger inhibitory effects than PD098059, a well-known natural compounds (e.g., N-feruloylserotonin, pynosilvin, carno-
pharmacologic inhibitor of MEK-1 [44,71]. The capacity of QUE to sine) resulted in a potentiation of the therapeutic effect of MTX at
behave as a not-specific kinase inhibitor (reviewed in Refs. [15,44]) low dose with a significant improvement of all inflammatory
can be also evoked to explain the inactivation of NF-kB presented in markers measured [4,28,79]. Both possibilities will be investigated
Fig. 4. In fact, it has been shown that QUE inhibits both IKKa and in the near future.
IKKb with apparent IC50 values of 11 and 4 mM, respectively,
reducing the Vmax and increased the Km, indicating a mixed-type of 5. Conclusion
inhibitory mechanism [72]. On the other hand, it is also known that
QUE may attenuate in vitro TNF-a-stimulated inflammatory medi- This study demonstrated that QUE, orally administrated in a rat
ator production by suppressing the activation of the ERK-mediated model of AA, ameliorated all markers of inflammation and oxida-
NF-kB pathway that is mediated by cellular peroxides [73]. tive stress measured. The molecule achieved this goal modulating
Data discussed above suggest that the effect of QUE on the key processes involved in cellular antioxidant defenses, including
production of pro-inflammatory cytokines in AA rats could occur down-regulation of NF-kB pathway and inhibiting ERK phosphor-
through the blocking of the ERK signaling pathway and regulation ylation. The ability of QUE to trigger multiple cellular pathways is in
of NF-kB activation, indicating a protective role of QUE against agreement with its functional pleiotropy and results in an
inflammation generated by oxidative stress in AA. However, our improvement of inflammatory response and reduction of oxidative
results do not support an effect of this molecule in decreasing the stress in arthritis.
arthritic score becoming clinically relevant nor in preventing
weight loss in AA rats (Table 1). Conflict of interest
The pathophysiology of altered body weight in RA is complex
and probably multifactorial. It has been reported that high levels of The authors of this paper declare no conflict of interest.
proinflammatory cytokines, particularly TNF-a and IL-1b, lead to an
increased hepatic protein synthesis [74] and to an increase in Acknowledgments
muscle proteolysis via the ubiquitin-proteasome pathway, attrib-
uted in part to NF-kB activation [75,76]. Nevertheless, Roubenoff The study was supported by grants: VEGA 2/0045/11, VEGA 2/
et al. [77] correlated the weight loss with TNF-a production by 0044/15 and performed in the frame of two SAV-CNR bilateral
spleen mononuclear cells (r ¼ 0.68, P < 0.007), while a weaker projects coordinated by Dr. Katarina Bauerova (Slovakia) and Dr.
correlation was seen with IL-1 production (r ¼ 0.45, P < 0.04). In our Gian Luigi Russo (Italy) entitled: “In vitro and in vivo models of
experiment we have similarly shown that although the IL-1b arthritic processes to study the mechanisms of inflammation and
plasmatic levels were lowered in a significant way after 28-day of oxidative stress link-up: New perspectives for arthritis therapy”
QUE treatment, a correction of body weight loss was not observed. and “Phytochemicals in ameliorating rheumatoid arthritis therapy:
Apart from inflammation, observed weight loss is the result of a from preclinical studies to clinical applications.” This work was also
complex network of interconnected factors, like animal mobility partially supported by a grant from the Italian Ministry of Economy
and food intake and may participate in vicious circle that results in and Finance to the National Research Council for the project
poor disease outcomes. In adjuvant arthritis generated hind paw “Innovazione e Sviluppo del Mezzogiorno e Conoscenze Integrate
swelling caused pain and limited mobility of animals, which are  ed Innovazione del Made in Italy Agroalimentare
per Sostenibilita
unable to feed ad libitum. Moreover together with hind paw joints -Legge 191/2009”.
also the temporomandibular joints are swollen and damaged.
The lacking effect of QUE on clinical parameters raised conflic- References
tual data in the literature with one study suggesting that oral
administration of QUE (5
160 mg/kg) clearly decreases clinical [1] C.H. Kunsch, J.A. Sikorski, C.L. Sundell, Current medicinal chemistry e
immunology, Endocr. Metabol. Agents 5 (2005) 249e258.
signs of arthritis in AA Lewis rats [22], while a different report
[2] K. Bauerova, A. Bezek, Gen. Physiol. Biophy. 18 (1999) 15e20.
indicated that QUE (200 mg/kg per os) did not prevent arthritis [78]. [3] K. Bauerova, E. Paulovicova, D. Mihalova, F. Drafi, M. Strosova, C. Mascia,
Our results support the latter conclusion, with important and novel F. Biasi, J. Rovensky, J. Kucharska, A. Gvozdjakova, S. Ponist, Acta Biochim. Pol.
information. We hypothesize that the described protective effects 57 (2010) 347e354.
[4] K. Bauerova, A. Acquaviva, S. Ponist, C. Gardi, D. Vecchio, F. Drafi, B. Arezzini,
of QUE in ameliorating inflammatory and oxidative status of rat L. Bezakova, V. Kuncirova, D. Mihalova, R. Nosal, Autoimmunity 48 (2015)
tissues are not sufficient to overcome the imaginary threshold that 46e56.
 C. Gardi et al. / Archives of Biochemistry and Biophysics 583 (2015) 150e157 157

[5] S. Kundu, P. Ghosh, S. Datta, A. Ghosh, S. Chattopadhyay, M. Chatterjee, Free [41] U. Takahama, S. Hirota, T. Oniki, Arch. Oral Biol. 51 (2006) 629e639.
Radic. Res. 46 (2012) 1482e1489. [42] G. Lopez-Lopez, L. Moreno, A. Cogolludo, M. Galisteo, M. Ibarra, J. Duarte,
[6] L.K. Stamp, I. Khalilova, J.M. Tarr, R. Senthilmohan, R. Turner, R.C. Haigh, F. Lodi, J. Tamargo, F. Perez-Vizcaino, Mol. Pharmacol. 65 (2004) 851e859.
P.G. Winyard, A.J. Kettle, Rheumatology 51 (2012) 1796e1803. [43] A.W. Boots, H. Li, R.P. Schins, R. Duffin, J.W. Heemskerk, A. Bast, G.R. Haenen,
[7] M. Biniecka, A. Kennedy, C.T. Ng, T.C. Chang, E. Balogh, E. Fox, D.J. Veale, Toxicol. Appl. Pharmacol. 222 (2007) 89e96.
U. Fearon, J.N. O'Sullivan, Arthritis Res. Ther. 13 (2011) R121. [44] G.L. Russo, M. Russo, C. Spagnuolo, I. Tedesco, S. Bilotto, R. Iannitti, R. Palumbo,
[8] D. Khanna, G. Sethi, K.S. Ahn, M.K. Pandey, A.B. Kunnumakkara, B. Sung, Cancer Treat. Res. 159 (2014) 185e205.
A. Aggarwal, B.B. Aggarwal, Curr. Opin. Pharmacol. 7 (2007) 344e351. [45] D. Ribeiro, M. Freitas, S.M. Tome, A.M. Silva, G. Porto, E.J. Cabrita,
[9] J.H. Lee, G.H. Kim, J. Food Sci. 75 (2010) H212eH217. M.M. Marques, E. Fernandes, Eur. J. Med. Chem. 72 (2014) 137e145.
[10] J.J. Ji, Y. Lin, S.S. Huang, H.L. Zhang, Y.P. Diao, K. Li, Afr. J. Tradit. Complement. [46] V.A. Blaho, Y. Zhang, J.M. Hughes-Hanks, C.R. Brown, J. Immunol. 186 (2011)
Altern. Med. 10 (2013) 418e421. 3076e3084.
[11] J.A. Rothwell, M. Urpi-Sarda, M. Boto-Ordonez, C. Knox, R. Llorach, R. Eisner, [47] T. Nguyen, P. Nioi, C.B. Pickett, J. Biol. Chem. 284 (2009) 13291e13295.
J. Cruz, V. Neveu, D. Wishart, C. Manach, C. Andres-Lacueva, A. Scalbert, [48] A. Kitamura, K. Nishida, T. Komiyama, H. Doi, Y. Kadota, A. Yoshida, T. Ozaki,
Database 2012 (2012) bas031. Mod. Rheumatol. 21 (2011) 150e157.
[12] G.S. Kelly, Altern. Med. Rev. 16 (2011) 172e194. [49] I. Rahman, S.K. Biswas, P.A. Kirkham, Biochem. Pharmacol. 72 (2006)
[13] D.W. Lamson, M.S. Brignall, Altern. Med. Rev. 5 (2000) 196e208. 1439e1452.
[14] M. Russo, C. Spagnuolo, I. Tedesco, S. Bilotto, G.L. Russo, Biochem. Pharmacol. [50] A. Paine, B. Eiz-Vesper, R. Blasczyk, S. Immenschuh, Biochem. Pharmacol. 80
83 (2012) 6e15. (2010) 1895e1903.
[15] G.L. Russo, M. Russo, C. Spagnuolo, Food Funct. 5 (2014) 2393e2401. [51] H.K. Park, S.J. Kim, Y. Kwon do, J.H. Park, Y.C. Kim, Life Sci. 87 (2010) 181e186.
[16] M. Sato, T. Miyazaki, F. Kambe, K. Maeda, H. Seo, J. Rheumatol. 24 (1997) [52] M. Benallaoua, M. Francois, F. Batteux, N. Thelier, J.Y. Shyy, C. Fitting, L. Tsagris,
1680e1684. J. Boczkowski, J.F. Savouret, M.T. Corvol, S. Poiraudeau, F. Rannou, Arthritis
[17] J.K. Jackson, T. Higo, W.L. Hunter, H.M. Burt, Inflamm. Res. 55 (2006) 168e175. Rheumatol. 56 (2007) 2585e2594.
[18] M.S. Sung, E.G. Lee, H.S. Jeon, H.J. Chae, S.J. Park, Y.C. Lee, W.H. Yoo, Inflam- [53] K.T. Turpaev, Biochem. Biokhimiia 78 (2013) 111e126.
mation 35 (2012) 1585e1594. [54] H.K. Na, Y.J. Surh, Free Radic. Biol. Med. 67 (2014) 353e365.
[19] J. Huang, M. Zhu, Y. Tao, S. Wang, J. Chen, W. Sun, S. Li, J. Pharm. Pharmacol. 64 [55] H.J. Forman, K.J. Davies, F. Ursini, Free Radic. Biol. Med. 66 (2014) 24e35.
(2012) 1119e1127. [56] C. Gerhauser, Curr. Opin. Clin. Nutr. Metabol. Care 16 (2013) 405e410.
[20] T. Guardia, A.E. Rotelli, A.O. Juarez, L.E. Pelzer, Farmaco 56 (2001) 683e687. [57] S. Tanigawa, M. Fujii, D.X. Hou, Free Radic. Biol. Med. 42 (2007) 1690e1703.
[21] A.E. Rotelli, T. Guardia, A.O. Juarez, N.E. de la Rocha, L.E. Pelzer, Pharm. Res. 48 [58] C.H. Kang, Y.H. Choi, S.K. Moon, W.J. Kim, G.Y. Kim, Int. Immunopharmacol. 17
(2003) 601e606. (2013) 808e813.
[22] M. Mamani-Matsuda, T. Kauss, A. Al-Kharrat, J. Rambert, F. Fawaz, D. Thiolat, [59] S.Y. Cho, S.J. Park, M.J. Kwon, T.S. Jeong, S.H. Bok, W.Y. Choi, W.I. Jeong,
D. Moynet, S. Coves, D. Malvy, M.D. Mossalayi, Biochem. Pharmacol. 72 (2006) S.Y. Ryu, S.H. Do, C.S. Lee, J.C. Song, K.S. Jeong, Mol. Cell. Biochem. 243 (2003)
1304e1310. 153e160.
[23] S.C. Bae, W.J. Jung, E.J. Lee, R. Yu, M.K. Sung, J. Am. Coll. Nutr. 28 (2009) 56e62. [60] S. Martinez-Florez, B. Gutierrez-Fernandez, S. Sanchez-Campos, J. Gonzalez-
[24] F. Javadi, S. Eghtesadi, A. Ahmadzadeh, N. Aryaeian, M. Zabihiyeganeh, Gallego, M.J. Tunon, J. Nutr. 135 (2005) 1359e1365.
A.R. Foroushani, S. Jazayeri, Int. J. Prev. Med. 5 (2014) 293e301. [61] M.R. Indra, S. Karyono, R. Ratnawati, S.G. Malik, BMC Res. Notes 6 (2013) 275.
[25] E.O. Santos, L.M. Kabeya, A.S. Figueiredo-Rinhel, L.F. Marchi, M.F. Andrade, [62] A. Alghasham, Z. Rasheed, Autoimmunity 47 (2014) 77e94.
F. Piatesi, A.B. Paoliello-Paschoalato, A.E. Azzolini, Y.M. Lucisano-Valim, Int. [63] S.S. Makarov, Arthritis Res. 3 (2001) 200e206.
Immunopharmacol. 21 (2014) 102e111. [64] Y.C. Chang, M.H. Tsai, W.H. Sheu, S.C. Hsieh, A.N. Chiang, PLoS One 8 (2013)
[26] C. Manach, G. Williamson, C. Morand, A. Scalbert, C. Remesy, Am. J. Clin. Nutr. e80744.
81 (2005) 230Se242S. [65] M. Comalada, D. Camuesco, S. Sierra, I. Ballester, J. Xaus, J. Galvez, A. Zarzuelo,
[27] H. Matsuno, H. Nakamura, K. Katayama, S. Hayashi, S. Kano, K. Yudoh, Y. Kiso, Eur. J. Immunol. 35 (2005) 584e592.
Biosci. Biotechnol. Biochem. 73 (2009) 288e292. [66] Y.D. Min, C.H. Choi, H. Bark, H.Y. Son, H.H. Park, S. Lee, J.W. Park, E.K. Park,
[28] V. Kuncirova, S. Ponist, D. Mihalova, F. Drafi, R. Nosal, A. Acquaviva, C. Gardi, H.I. Shin, S.H. Kim, Inflamm. Res. 56 (2007) 210e215.
J. Harmatha, I. Hradkova, K. Bauerova, Fundam. Clin. Pharmacol. 28 (2014) [67] M.P. Nair, S. Mahajan, J.L. Reynolds, R. Aalinkeel, H. Nair, S.A. Schwartz,
616e626. C. Kandaswami, Clin. Vac. Immunol. 13 (2006) 319e328.
[29] O. Erel, Clin. Biochem. 37 (2004) 277e285. [68] Y. Okazaki, T. Sawada, K. Nagatani, Y. Komagata, T. Inoue, S. Muto, A. Itai,
[30] M.M. Bradford, Anal. Biochem. 72 (1976) 248e254. K. Yamamoto, J. Rheumatol. 32 (2005) 1440e1447.
[31] R.E. Gay, M. Neidhart, F. Pataky, S. Tries, S. Laufer, S. Gay, J. Rheumatol. 28 [69] G. Pearson, F. Robinson, T. Beers Gibson, B.E. Xu, M. Karandikar, K. Berman,
(2001) 2060e2065. M.H. Cobb, Endocr. Rev. 22 (2001) 153e183.
[32] K.R. Gheorghe, M. Korotkova, A.I. Catrina, L. Backman, E. af Klint, H.E. Claesson, [70] M.J. Thiel, C.J. Schaefer, M.E. Lesch, J.L. Mobley, D.T. Dudley, H. Tecle,
O. Radmark, P.J. Jakobsson, Arthritis Res. Ther. 11 (2009) R83. S.D. Barrett, D.J. Schrier, C.M. Flory, Arthritis Rheumatol. 56 (2007)
[33] M.Y. Wu, T.H. Lin, Y.C. Chiu, H.C. Liou, R.S. Yang, W.M. Fu, J. Cell. Biochem. 113 3347e3357.
(2012) 2279e2289. [71] D.X. Hou, T. Kumamoto, Antioxidants Redox Signal. 13 (2010) 691e719.
[34] S.W. Ryter, H.P. Kim, K. Nakahira, B.S. Zuckerbraun, D. Morse, A.M. Choi, An- [72] G.W. Peet, J. Li, J. Biol. Chem. 274 (1999) 32655e32661.
tioxidants Redox Signal. 9 (2007) 2157e2173. [73] C.S. Lee, E.B. Jeong, Y.J. Kim, M.S. Lee, S.J. Seo, K.H. Park, M.W. Lee, Int.
[35] H. Kobayashi, M. Takeno, T. Saito, Y. Takeda, Y. Kirino, K. Noyori, T. Hayashi, Immunopharmacol. 16 (2013) 481e487.
A. Ueda, Y. Ishigatsubo, Arthritis Rheum. 54 (2006) 1132e1142. [74] J. Walsmith, R. Roubenoff, Int. J. Cardiol. 85 (2002) 89e99.
[36] J. Campbell, C.J. Ciesielski, A.E. Hunt, N.J. Horwood, J.T. Beech, L.A. Hayes, [75] S. Acharyya, S.A. Villalta, N. Bakkar, T. Bupha-Intr, P.M. Janssen, M. Carathers,
A. Denys, M. Feldmann, F.M. Brennan, B.M. Foxwell, J. Immunol. 173 (2004) Z.W. Li, A.A. Beg, S. Ghosh, Z. Sahenk, M. Weinstein, K.L. Gardner, J.A. Rafael-
6928e6937. Fortney, M. Karin, J.G. Tidball, A.S. Baldwin, D.C. Guttridge, J. Clin. Invest. 117
[37] P.W. Tsao, T. Suzuki, R. Totsuka, T. Murata, T. Takagi, Y. Ohmachi, H. Fujimura, (2007) 889e901.
I. Takata, Clin. Immunol. Immunopathol. 83 (1997) 173e178. [76] J.E. Morley, D.R. Thomas, M.M. Wilson, Am. J. Clin. Nutr. 83 (2006) 735e743.
[38] D. de Launay, M.G. van de Sande, M.J. de Hair, A.M. Grabiec, G.P. van de Sande, [77] R. Roubenoff, L.M. Freeman, D.E. Smith, L.W. Abad, C.A. Dinarello, J.J. Kehayias,
K.A. Lehmann, C.A. Wijbrandts, L.G. van Baarsen, D.M. Gerlag, P.P. Tak, Arthritis Rheum. 40 (1997) 534e539.
K.A. Reedquist, Ann. Rheumatic Dis. 71 (2012) 415e423. [78] S. Ramos-Romero, F.J. Perez-Cano, E. Ramiro-Puig, A. Franch, M. Castell,
[39] M. Ohori, Drug News Perspect. 21 (2008) 245e250. Pharm. Res. 66 (2012) 207e212.
[40] G.C. Justino, M.R. Santos, S. Canario, C. Borges, M.H. Florencio, L. Mira, Arch. [79] F. Drafi, K. Bauerova, V. Kuncirova, S. Ponist, D. Mihalova, T. Fedorova,
Biochem. Biophys. 432 (2004) 109e121. J. Harmatha, R. Nosal, Interdiscip. Toxicol. 5 (2012) 84e91.