Antioxidant activity and inhibition of human neutrophil oxidative burst mediated by arylpropionic acid non-steroidal anti-inflammatory drugs

Article abstract of DOI:10.1248/bpb.29.1659

It has been suggested that the anti-inflammatory activity of some non-steroidal anti-inflammatory drugs (NSAIDs) may be partly due to their ability to scavenge reactive oxygen species (ROS) and reactive nitrogen species (RNS), as well as to inhibit the respiratory burst of neutrophils triggered by various activating agents. Therefore, the aim of the present work was to evaluate and compare the potential scavenging activity for an array of ROS (O₂^.-, H₂O₂, HO^., ROO^. and HOCl) and RNS (^.NO and ONOO^-) using in vitro non-cellular screening systems as well as a cellular screening system (human neutrophil oxidative burst), mediated by the arylpropionic acid derivatives (APAs) NSAIDs ibuprofen, flurbiprofen, fenoprofen, fenbufen, ketoprofen, naproxen and indoprofen. The results obtained in the present work demonstrate that under the present experimental conditions, many of the studied APA NSAIDs showed O₂^.- scavenging activity (fenbufen≈ flurbiprofen≈indoprofen≈ketoprofen), H₂O₂ (ketoprofen≈indoprofen≈fenbufen>flurbiprofen>naproxen), HO ^. (fenoprofen≈ibuprofen>fenbufen≈flurbiprofen> ketoprofen>indoprofen≈naproxen), ^.NO (indoprofen> naproxen), ONOO^- (indoprofen>naproxen>fenoprofen≈flurbiprofen≈ ibuprofen), as well as inhibit myeloperoxidase (MPO) activity (indoprofen) and scavenge human neutrophil derived ROS (ketoprofen>indoprofen >fenbufen>flurbiprofen). The observed effects, if confirmed in vivo, may strongly contribute to the antiinflammatory therapeutical activity observed with these NSAIDs.

Full text of DOI:10.1248/bpb.29.1659

August 2006
Biol. Pharm. Bull. 29(8) 1659—1670 (2006)
1659
Antioxidant Activity and Inhibition of Human Neutrophil Oxidative Burst
Mediated by Arylpropionic Acid Non-steroidal Anti-inflammatory Drugs
David COSTA, Luís MOUTINHO, José Luís Fontes Costa LIMA, and Eduarda FERNANDES*
REQUIMTE, Departamento de Química-Física, Faculdade de Farmácia, Universidade do Porto; Rua Aníbal Cunha, 164,
4099–030 Porto, Portugal. Received January 15, 2006; accepted April 5, 2006
It has been suggested that the anti-inflammatory activity of some non-steroidal anti-inflammatory drugs
(NSAIDs) may be partly due to their ability to scavenge reactive oxygen species (ROS) and reactive nitrogen
species (RNS), as well as to inhibit the respiratory burst of neutrophils triggered by various activating agents.
Therefore, the aim of the present work was to evaluate and compare the potential scavenging activity for an
ꢀ
ꢀ
array of ROS (O˙ , H O , HO˙, ROO˙ and HOCl) and RNS (˙NO and ONOO ) using in vitro non-cellular
2
2
2
screening systems as well as a cellular screening system (human neutrophil oxidative burst), mediated by the
arylpropionic acid derivatives (APAs) NSAIDs ibuprofen, flurbiprofen, fenoprofen, fenbufen, ketoprofen,
naproxen and indoprofen. The results obtained in the present work demonstrate that under the present ex-
ꢀ
perimental conditions, many of the studied APA NSAIDs showed O2˙ scavenging activity (fenbufenꢀ
flurbiprofenꢀindoprofenꢀketoprofen), H O2 (ketoprofenꢀindoprofenꢀfenbufenꢁflurbiprofenꢁnaproxen),
2
HO˙ (fenoprofenꢀibuprofenꢁfenbufenꢀflurbiprofenꢁketoprofenꢁindoprofenꢀnaproxen), ˙NO (indoprofenꢁ
ꢀ
naproxen), ONOO (indoprofenꢁnaproxenꢁfenoprofenꢀflurbiprofenꢀibuprofen), as well as inhibit myeloper-
oxidase (MPO) activity (indoprofen) and scavenge human neutrophil derived ROS (ketoprofenꢁindo-
profenꢁfenbufenꢁflurbiprofen). The observed effects, if confirmed in vivo, may strongly contribute to the anti-
inflammatory therapeutical activity observed with these NSAIDs.
Key words non-steroidal antiinflammatory drug; reactive species; human neutrophils; oxidative burst; myeloperoxidase
Non-steroidal anti-inflammatory drugs (NSAIDs) are various activating agents. In fact, it has been demonstrated
amongst the most widely prescribed drugs for the treatment that at high (antirheumatic) concentrations, several NSAIDs
1
0,17—21)
of pain, fever, and inflammation. The arylpropionic acid de- are potent scavengers of ROS and RNS,
inhibit the
rivatives (APAs) (e.g. fenbufen, fenoprofen, flurbiprofen, oxidative burst induced by a range of stimuli in human neu-
5
,21,22)
23)
ibuprofen, indoprofen, ketoprofen and naproxen, Fig. 1) con- trophils
stitute an important group of NSAIDs widely used for the idation.
treatment of pain and inflammation and represent the drugs
and bovine neutrophils, and prevent LDL ox-
24)
The scavenging activities of NSAIDs from the APA group
of choice commonly used in the management of muscu- against ROS and RNS, as well as their putative inhibition of
loskeletal traumatisms, rheumatoid arthritis, osteoarthritis, neutrophil burst have only been sparsely and incompletely
ankylosing spondylitis, acute gouty arthritis and dysmenor- evaluated, which makes it very difficult to compare the rela-
1—3)
rhoea.
Chemically, these derivatives are weak acids and tive antioxidant potency among them. Although the effective-
by virtue of a chiral carbon atom on the propionic acid side ness has already been reported for some NSAIDs (as previ-
25)
chain, exist as enantiomer pairs with (S)-isomers being pri- ously reported ), it is possible to present original and rele-
marily responsible for inhibition of prostaglandin production
and inflammatory events, although the majority of these
4)
acids continue to be employed as racemate.
Whereas low doses of NSAIDs markedly inhibit the in
vitro and in vivo synthesis of prostaglandin, higher doses are
required for an anti-inflammatory effect in vivo, suggesting
that at higher concentrations other biological activities con-
tribute to the final anti-inflammatory effects mediated by
5)
these compounds. Accordingly, it is well known that other
factors beyond prostaglandin synthesis, contribute strongly to
the pathophysiology of inflammatory processes, as is the case
of overproduction of reactive oxygen species (ROS), namely
ꢀ
superoxide radical (O˙ ), hydrogen peroxide (H O ), hy-
2
2
2
droxyl radical (HO˙), peroxyl radical (ROO˙) and hypochlor-
6—9)
ous acid (HOCl)
as well as nitrogen reactive species
(
(
RNS), namely nitric oxide (˙NO) and peroxynitrite anion
ꢀ
10—12)
ONOO ).
ROS and RNS are generated by different
cell types, with especial emphasis to neutrophils during the
1
3—16)
respiratory burst.
Thus, it has been suggested that the
anti-inflammatory activity of some NSAIDs may also be
partly due to their ability to scavenge ROS and RNS, as well
Fig. 1. Chemical Structures of Naproxen, Ketoprofen, Flurbiprofen,
as inhibit the respiratory burst of neutrophils triggered by Ibuprofen, Fenoprofen, Indoprofen and Fenbufen
∗
To whom correspondence should be addressed. e-mail: egracas@ff.up.pt
© 2006 Pharmaceutical Society of Japan

1
660
Vol. 29, No. 8
vant data by comparing a large panel of drugs using opti- hibited this effect. Tiron was used as a positive control. Ef-
ꢀ
mised systems and establishing complete dose-effect curves, fects are expressed as the percentage inhibition of the O˙ -
2
with determination of median effective concentrations induced NBT reduction to diformazan. Each study corre-
(
IC s). Therefore, the aim of the present work was focused sponds to four experiments, performed in triplicate.
50
on evaluating and comparing the potential scavenging activ-
H O Scavenging Assay H O scavenging activity was
2 2 2 2
ꢀ
ity for an array of ROS (O˙ , H O , HO˙, ROO˙ and HOCl) measured by chemiluminescence in a microplate reader
2
2
2
ꢀ
and RNS (˙NO and ONOO ) using in vitro non-cellular (Synergy HT, BIO-TEK) by monitoring the oxidation of luci-
20)
screening systems as well as a cellular screening system genin by H O , as previously reported. The reaction mix-
2
2
(
human neutrophil oxidative burst), mediated by the APAs tures in the sample wells contained the following reagents at
ibuprofen, flurbiprofen, fenoprofen, fenbufen, ketoprofen, the indicated final concentrations (in a final volume of
naproxen and indoprofen.
250 ml): 50 mM Tris buffer, pH 7.4, lucigenin (3 mM), tested
compound at various concentrations, dissolved in buffer and
H O (2%). The resulting chemiluminescence was followed
MATERIALS AND METHODS
2
2
for 5 min. Trolox was used as a positive control. Effects are
Materials The following reagents were obtained from expressed as the percentage inhibition of the H O -elicited
2
2
Sigma Chemical Co. (St. Louis, U.S.A.): Flurbiprofen, feno- lucigenin oxidation. Each study corresponds to four experi-
profen, fenbufen, ketoprofen, naproxen, indoprofen, ibupro- ments, performed in triplicate.
fen, nitroblue tetrazolium chloride (NBT), b-nicotinamide
HO˙ Scavenging Assay HO˙ scavenging activity was
adenine dinucleotide (NADH), phenazine methosulfate measured spectrophotometrically using a spectrophotometer
19)
(PMS), 30% hydrogen peroxide, ferric chloride anhydrous, (Lambda 45, Perkin-Elmer) as previously reported. HO˙
ethylenediaminetetraacetic acid (EDTA) disodium salt, radicals were generated by a Fenton system (ascorbic acid–
ascorbic acid, trichloroacetic acid, thiobarbituric acid (TBA), FeCl –EDTA). When exposed to HO˙, the sugar deoxyribose
3
deoxyribose, sodium hypochlorite solution with 4% availa- is degraded to malonaldehyde, which generates a pink chro-
ble chlorine (NaOCl), 5,5ꢂ-dithiobis(2-nitrobenzoic acid) mogen on heating with thiobarbituric acid (TBA), at low pH.
(DTNB), mannitol, cysteine, lipoic acid, carboxy-PTIO, Reaction mixtures contained the following reagents at the in-
sodium borohydride, sodium bicarbonate, fluorescein sodium dicated final concentrations (in a final volume of 1 ml):
salt, dihydrorhodamine 123 (DHR), 4,5-diaminofluorescein 10 mM phosphate buffer, pH 7.4, deoxyribose (2.8 mM), H O
2
2
(DAF-2), sodium nitroprusside dihydrate (NPS), Histopaque (1.42 mM), tested compounds in various concentrations,
1
077, Histopaque 1119, Dulbecco’s Phosphate Buffer saline, FeCl –EDTA (20 mM, 100 mM) and ascorbic acid (50 mM).
3
without calcium chloride and magnesium chloride (PBS), The iron salt was premixed with the chelator dissolved in
Trypan blue solution 0.4%, luminol, phorbol myristate water before addition to the reaction mixture. All other com-
acetate (PMA), tiron, diphenyleneiodonium chloride (DPI) ponents were dissolved in 10 mM phosphate buffer, pH 7.4.
and myeloperoxidase (MPO) from human leukocytes (EC After incubation at 37 °C for 1 h, 1 ml of trichloroacetic acid
1
.11.1.7). 4-Aminobenzoic acid hydrazide (ABAH) was ob- (2.8%, w/v) and 1 ml of TBA (1%, w/v) were added and the
®
tained from Calbiochem (Postfach, Germany). Sodium mixture heated in a water bath at 100 °C for 15 min. Ab-
chloride, sodium acetate, ethanol, sulphuric acid, hydrochlo- sorbance of the resulting solution was measured at 532 nm.
ric acid, sodium hydroxide, sodium carbonate and sodium ni- This assay was also performed without ascorbic acid or
trite were obtained from Merck (Darmstadt, Germany). Luci- EDTA to evaluate a possible pro-oxidant and/or iron chela-
genin, a,aꢂ-azodiisobutyramidine dihydrochloride (AAPH), tion activity. Mannitol was used as a positive control. Effects
dimethylsulfoxide (DMSO), TRIS-hydroxymethylamino- are expressed as the percentage inhibition of the deoxyribose
methane (TRIS), potassium chloride and Trolox were sup- degradation to malonaldehyde with the second-order rate
6)
plied by Fluka Chemie GmbH (Steinheim, Germany).
constants (Ks) being calculated according to Halliwell et al.
Measurement of ROS and RNS Scavenging Activity Each study corresponds to four experiments, performed in
ꢀ
Using in Vitro Non-cellular Screening Systems. O˙
duplicate.
ROO˙ Scavenging Assay ROO˙ scavenging activity was
2
ꢀ
Scavenging Assay O˙ scavenging activity was measured
2
spectrophotometrically in a microplate reader (Synergy HT, measured by monitoring fluorescence decay in a microplate
ꢀ
BIO-TEK) by monitoring the O˙ -induced reduction of ni- reader (Synergy HT, BIO-TEK), due to the oxidation of fluo-
2
troblue tetrazolium chloride (NBT) at 560 nm for 2 min, as rescein by ROO˙ generated by a,aꢂ-azodiisobutyramidine
19)
ꢀ
previously reported. O˙ radicals were generated by a b- dihydrochloride (AAPH) thermo-decomposition, according
2
nicotinamide adenine dinucleotide (NADH)/phenazine to a procedure termed the Oxygen Radical Absorbance Ca-
19)
methosulfate (PMS) system. The reaction mixture in the pacity (ORAC) assay, as previously reported. The reaction
sample wells contained the following reagents at the indi- mixtures in the sample wells contained the following
cated final concentrations (in a 300 ml final volume): NADH reagents dissolved in 75 mM phosphate buffer, pH 7.4, at the
(
166 mM), NBT (43 mM), tested compound, dissolved in indicated final concentrations (in a final volume of 200 ml):
DMSO, at various concentrations and PMS (2.7 mM). No di- fluorescein (61 nM), tested compound at various concentra-
rect effect was observed between DMSO and O˙ in the tions and AAPH (19 mM). Fluorescence measurements were
ꢀ
2
present assay conditions. All other components were dis- carried out at a constant temperature of 37 °C with excitation
solved in 19 mM phosphate buffer, pH 7.4. The assay was and emission wavelengths of 485 nm and 528 nm respec-
performed at room temperature. The selective effect of su- tively. Effects are expressed as the relative Trolox equivalent
peroxide on the reduction of NBT to diformazan was con- ORAC value which is calculated by the following equation,
19)
firmed by using superoxide dismutase, which completely in- where AUC represents the area under the curve :

August 2006
relative ORAC value
1661
3
7 °C. Effects are expressed as the percentage inhibition of
DAF-2 oxidation. Carboxy-PTIO was used as a positive con-
trol. Each study corresponds to four experiments, performed
ꢃ[(AUCsampleꢀAUCblank)/(AUCTroloxꢀAUCblank)]
ꢄ(molarity of Trolox/molarity of compound)
Each study corresponds to four experiments, performed in in triplicate.
ꢀ
ꢀ
triplicate.
ONOO Scavenging Assay. Synthesis of ONOO
ꢀ
HOCl Scavenging Assay. Production of HOCl Solu- Synthesis of ONOO was essentially performed as described
29)
tion HOCl was prepared immediately before use by adjust- before by Beckman et al. Briefly, an acidic solution (HCl
ing the pH of a 1% (v/v) solution of NaOCl to 6.2 with di- 0.7 M) of 0.6 M H O was mixed with NaNO (0.66 M) on ice
2
2
2
luted sulphuric acid. The concentration of HOCl was further for one second and the reaction quenched with ice-cold 3 M
determined spectrophotometrically using a spectrophotome- NaOH. Residual H O was removed by mixing with granular
2
2
ꢀ
ter (Lambda 45, Perkin-Elmer) at 235 nm and the molar ab- MnO pre-washed with NaOH 3 M. The ONOO stock solu-
2
ꢀ1
ꢀ1 26)
sorption coefficient of 100 M cm
was appropriately diluted with 50 mM phosphate buffer, pH of the solution collected for the experiment. ONOO concen-
.4. tration was determined by measuring the absorbance at 302
For each assay, HOCl tion was filtered and then frozen (ꢀ20 °C) with the top layer
ꢀ
7
ꢀ1
ꢀ1
Synthesis of 5-Thio-2-nitrobenzoic Acid (TNB) TNB nm (eꢃ1670 M ·cm ). The typical yield of freshly pre-
was prepared according to a described procedure with mod- pared ONOO ranged from 60—80 mM. Higher concentra-
ifications. Briefly, 50 mM sodium borohydride was added to a tions (ꢁ200 mM) of ONOO can be obtained by freeze frac-
7)
ꢀ
ꢀ
1
mM solution of 5,5ꢂ-dithiobis(2-nitrobenzoic acid) (DTNB) tionation. However, in the present study only freshly pre-
ꢀ
in 50 mM phosphate buffer, pH 6.6, containing 5 mM EDTA. pared ONOO solution was used to minimise nitrite ion con-
The solution was incubated at 37 °C for 30 min. TNB con- tamination. Prior to each study, the concentration of the
centration was determined by measuring the absorbance at ONOO stock was determined spectrophotometrically in
ꢀ
4
12 nm in a spectrophotometer (Lambda 45, Perkin-Elmer) 0.1 M NaOH.
ꢀ1
ꢀ1
ꢀ
using the molar absorption coefficient of 13600 M ·cm .
For each assay, TNB was appropriately diluted with 50 mM ONOO scavenging activity was measured by monitoring
phosphate buffer, pH 7.4.
Measurement of ONOO
Scavenging Activity
ꢀ
ꢀ
the ONOO -induced oxidation of dihydrorhodamine 123
Measurement of HOCl Scavenging Activity HOCl (DHR) to rhodamine 123 by fluorimetry, as previously re-
19)
scavenging activity was measured using a spectrophotometer ported. A stock solution of 2.89 mM DHR in dimethyl-
Lambda 45, Perkin-Elmer) monitoring the oxidation of TNB formamide was purged with nitrogen and stored at ꢀ20 °C.
(
19)
to DTNB, as previously reported. Reaction mixtures con- Working solutions of DHR diluted from the stock solution
tained the following reagents at the indicated final concentra- were placed on ice in the dark immediately before the deter-
tions (in a final volume of 1 ml): TNB (70 mM), tested com- minations. Buffer (90 mM sodium chloride, 50 mM sodium
pound at various concentrations, dissolved in ethanol or phosphate and 5 mM potassium chloride, pH 7.4) was purged
buffer, and HOCl (25 mM). The assay was performed at room with nitrogen and placed on ice before use. At the outset of
temperature. No direct effect was observed between ethanol the experiments, 100 mM diethylenetriaminepentaacetic acid
and HOCl in the present assay conditions. Absorbance was (DTPA) was added to the buffer. The assay was performed
measured at 412 nm, 5 min after the addition of HOCl. at 37 °C. Reaction mixtures contained the following reagents
Lipoic acid was used as a positive control. Effects are ex- at the indicated final concentrations (in a final volume
pressed as the percentage inhibition of TNB oxidation to of 1.5 ml): DHR (50 mM), tested compound, dissolved in
ꢀ
DTNB. Each study corresponds to four experiments, per- DMSO, at various concentrations and ONOO (600 nM) in
formed in duplicate.
0.1 M NaOH. In the present assay conditions, less that 10% of
˙NO Scavenging Assay ˙NO scavenging activity was the effect was found for DMSO itself, which was subtracted
measured by monitoring the ˙NO-induced oxidation of 4,5- in the results obtained for the compounds. Cysteine was used
diaminofluorescein (DAF-2) to triazolofluorescein by fluo- as a positive control. The fluorescence signal induced by
1
9)
ꢀ
rimetry as previously reported. In this assay, DAF-2 does DHR reacting with ONOO was measured using a spectro-
not react directly with ˙NO but rather with the oxidized form fluorimeter (LS-50B, Perkin-Elmer) with excitation and
27)
of ˙NO, involving N O . One milligram of DAF-2 in 0.55 emission wavelengths of 500 nm and 536 nm respectively
2
3
ml DMSO was diluted with 50 mM phosphate buffer (pH 7.4) and excitation and emission slit widths of 2.5 nm and 3.0 nm
to 1/400-fold. Reaction mixtures contained the following respectively. Background and final fluorescent intensities
reagents at the indicated final concentrations (in a final vol- were measured 5 min after treatment. The effects are ex-
ume of 1.5 ml): DAF-2 (3.14 mM), sodium nitroprusside pressed as the percentage inhibition of DHR oxidation. In a
(
NPS) (20 mM) and tested compounds in various concentra- parallel set of experiments the assays were performed in the
tions, dissolved in DMSO. No direct effect was observed be- presence of 25 mM NaHCO to simulate physiological condi-
3
tween DMSO and ˙NO in the present assay conditions. All tions with high CO concentrations in vivo. This evaluation is
2
other components were dissolved in 50 mM phosphate buffer, important because under physiological conditions, the reac-
ꢀ
pH 7.4. The reaction mixture was incubated for 10 min at tion of ONOO with bicarbonate is predominant, with the re-
ꢀ
3
7 °C, under tungsten light. Light is of prime importance for action rate constant of CO with ONOO being very rapid
2
28)
4
ꢀ1 ꢀ1 30)
the sustained release of ˙NO by NPS. The fluorescence sig- (k ꢃ3–5.8ꢄ10 M ·s ). Thus, the reactivity of the puta-
nal caused by DAF-2 reacting with ˙NO was measured using tive scavengers for ONOO should be able to match or ex-
2
ꢀ
a spectrofluorimeter (LS-50B, Perkin-Elmer) with excitation ceed that of bicarbonate. Each study corresponds to four ex-
and emission wavelengths of 495 nm and 521 nm respec- periments, performed in triplicate.
tively and excitation and emission slit widths of 8 nm, at
Measurement of PMA-Induced Oxidative Burst in

1
662
Vol. 29, No. 8
21)
Human Neutrophils This study was performed using neu- procedure.
Reaction mixtures contained the following
trophils isolated from healthy human volunteers. Essentially, reagents at the indicated final concentrations (in a final vol-
6
the study was based on the stimulation of human neutrophils ume of 200 ml): neutrophils (final suspensionꢃ5ꢄ10
with phorbol 12-myristate 13-acetate (PMA) with subsequent cells/ml), tested compounds at various concentrations (0—
ꢀ7
measurement of the neutrophil-ROS-generating-capacity. For 2 mM), lucigenin (125 mM) and PMA (1.6ꢄ10 M). Kinetic
this purpose, chemiluminescence methods, using luminol readings were initiated immediately after cell stimulation. All
and lucigenin as probes, were adapted to a microplate reader measurements were carried out at 37 °C and taken at the peak
(Synergy HT, BIO-TEK).
of the curve. In all experiments, this peak was observed at
Isolation of Neutrophils Venous blood was collected around 5 min. Effects are expressed as the percentage inhibi-
ꢀ
from healthy human volunteers by antecubital venipuncture, tion of lucigenin oxidation. Tiron (scavenger of O˙ ) and
2
into EDTA vacuum tubes. Whole venous blood was sepa- diphenyleneiodonium chloride (DPI) (inhibitor of NADPH
rated into cell types using Histopaque solutions 1077 and oxidase) were used as positive controls. Each study corre-
1
119 in 12 ml polypropylene centrifuge tubes, as previously sponds to four experiments, performed in triplicate.
21)
reported. Briefly, 3 ml of Histopaque 1077 was carefully
Luminol Oxidation by Purified-Myeloperoxidase—
layered on top of 3 ml Histopaque 1119 in a 12 ml Generated HOCl The oxidation of luminol by purified
polypropylene tube. 6 ml of the collected blood was decanted MPO-generated HOCl was performed according to a previ-
21)
on top of this discontinuous density gradient. The tube was ously described procedure. Reaction mixtures contained
centrifuged at 800 g for 30 min at 20 °C. On completion of the following reagents at the indicated final concentrations
centrifugation, the neutrophils were carefully removed using (in a final volume of 250 ml): NaCl (100 mM), luminol (250
a Pasteur pipette. The neutrophil pellet was removed and mM), tested compounds at various concentrations (0—2 mM),
doubled in volume using PBS (this reduces the viscosity of MPO (0.5 U/ml) and H O (100 mM). Kinetic readings were
2
2
the Histopaque-neutrophil suspension so that the cells can initiated immediately after addition of H O . All measure-
2
2
be centrifuged without the need for high g forces) before ments were carried out at a constant temperature of 37°C and
the neutrophils were centrifuged at 700 g for 5 min at 4 °C. were taken at the peak of the curve. In all experiments, this
The supernatant was decanted and a mixture of 1.25 ml peak was observed at around 7 to 9 min. The inhibitory ef-
PBSꢅ5.25 ml sterile distilled water added to the neutrophil fects of the tested compounds are expressed as the percent-
pellet to lyse any remaining red blood cells. The tube was age inhibition of luminol oxidation. ABAH was used as a
gently inverted for 1 min after which isotonicity was re-es- positive control. Each study corresponds to four experiments,
tablished by adding 2.2 ml of NaCl 3%. This suspension was performed in triplicate.
then subjected to a further centrifugation at 700 g for 5 min
at 4 °C, after which the supernatant was decanted and the RESULTS
neutrophil pellet re-suspended in PBS. The obtained cell sus-
pensions contained more than 99% neutrophils and the con-
trol of their viability showed more than 95% of the cells ex- cellular Systems. O˙ Scavenging Activity Figure 2
cluding Trypan blue (0.4%). Isolated neutrophils were main- shows the results obtained in the O2˙ scavenging assay. It
ROS and RNS Scavenging Activity Using in Vitro Non-
ꢀ
2
ꢀ
tained in ice until use. Neutrophils were used from one vol- was observed that flurbiprofen, fenbufen, ketoprofen and in-
ꢀ
unteer per experiment.
doprofen prevented the O˙ -induced reduction of NBT in a
2
Oxidation of Luminol by Neutrophils—Generated ROS concentration dependent manner, indoprofen being the most
The measurement of neutrophil burst was undertaken by potent, with an IC50 of 3387ꢆ191 mM, followed by fenbufen,
chemiluminescence, by monitoring ROS-induced oxidation ketoprofen and flurbiprofen with equivalent potencies (IC s
5
0
21)
of luminol, according to a previously described procedure.
of 3481ꢆ268, 3631ꢆ633 and 3753ꢆ282 mM, respectively)
Reaction mixtures contained the following reagents at the in- (meanꢆS.E.M.) (Table 1). No activity was observed for
dicated final concentrations (in a final volume of 200 ml): fenoprofen up to 1250 mM and for naproxen and ibuprofen up
6
neutrophils (final suspensionꢃ1ꢄ10 cells/ml), tested to 5000 mM. The positive control (Tiron) provided an IC of
5
0
compounds at various concentrations (0—2 mM), luminol 167ꢆ7 mM (meanꢆS.E.M.) (Table 1).
ꢀ7
(
500 mM) and PMA (1.6ꢄ10 M). Cells and tested com-
H O Scavenging Activity Figure 3 shows the results
2 2
pounds were pre-incubated for 5 min at 37 °C before the ad- obtained in the H O scavenging assay. It was observed that
2
2
dition of luminol and PMA and the measurements were car- fenbufen, flurbiprofen, ketoprofen, naproxen and indoprofen
ried out at 37 °C. Kinetic readings were initiated immediately prevented the H O -induced oxidation of lucigenin in a con-
2
2
after cell stimulation. Measurements were taken at the peak centration dependent manner, ketoprofen being the most
of the curve. In all experiments, this peak was observed at potent with an IC50 of 2847ꢆ142 mM, followed by indo-
around 5 min. Effects are expressed as the percentage inhibi- profen, fenbufen, flurbiprofen and naproxen (IC s: 3050ꢆ
5
0
tion of luminol oxidation. Lipoic acid (a HOCl scavenger) 87ꢀ3220ꢆ175ꢀ3698ꢆ355ꢁ5343ꢆ199 mM, respectively)
and 4-aminobenzoic acid hydrazide (ABAH) (a myeloperoxi- (meanꢆS.E.M.) (Table 1). No activity was observed for
dase (MPO) inhibitor) were used as positive controls. Each fenoprofen and ibuprofen up to 5 mM. The positive control
study corresponds to four individual experiments, performed (Trolox) provided an IC50 of 2535ꢆ94 mM (meanꢆS.E.M.)
in triplicate in each experiment.
(Table 1).
HO˙ Scavenging Activity Figure 4 shows the results ob-
Oxidation of Lucigenin by Neutrophils—Generated
ꢀ
O˙
The measurement of neutrophil burst was measured tained in the HO˙ scavenging assay. All assayed compounds
2
ꢀ
by chemiluminescence through monitoring the O˙ -induced prevented the HO˙-induced degradation of deoxyribose into
2
oxidation of lucigenin, according to a previously described malonaldehyde in a concentration dependent manner. The

August 2006
1663
ꢀ
Fig. 2. O˙ Scavenging Activity of Flurbiprofen, Fenbufen, Ketoprofen and Indoprofen
2
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
Fig. 3. H O Scavenging Activity of Fenbufen, Flurbiprofen, Ketoprofen, Naproxen and Indoprofen
2
2
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
Table 1. ROS Scavenging Effects (IC , ks and ORAC Activity; MeanꢆS.E.M.) of Fenbufen, Fenoprofen, Flurbiprofen, Ibuprofen, Indoprofen, Ketopro-
5
0
fen, Naproxen, Mannitol, Trolox, Tiron GSH and Ascorbic Acid
ꢀ
O˙
H O
2
HO˙
ROO˙
2
2
Tested
compound
ꢀ
1
ꢀ1
a)
IC50 (mM)
IC50 (mM)
IC50 (mM)
ks (M ·s
)
ORAC activity
1
1
1
1
1
1
1
0
0
0
0
0
0
0
Fenbufen
Fenoprofen
Flurbiprofen
Ibuprofen
Indoprofen
Ketoprofen
Naproxen
Mannitol
Trolox
3481ꢆ268
NB
3753ꢆ282
NA
3387ꢆ191
3631ꢆ633
NA
—
—
167ꢆ7
—
—
3220ꢆ175
NA
3698ꢆ355
NA
3050ꢆ87
2847ꢆ142
5343ꢆ199
—
2535ꢆ94
—
93ꢆ16
52ꢆ2
100ꢆ1
61ꢆ11
320ꢆ27
258ꢆ12
309ꢆ64
3272ꢆ215
—
(4.9ꢆ0.5)ꢄ10
(2.8ꢆ0.2)ꢄ10
(2.7ꢆ0.3)ꢄ10
(8.3ꢆ1.6)ꢄ10
(2.6ꢆ0.6)ꢄ10
(4.4ꢆ0.6)ꢄ10
(1.2ꢆ0.2)ꢄ10
0.16ꢆ0.10
0.09ꢆ0.06
0.14ꢆ0.05
0.02ꢆ0.01
0.04ꢆ0.02
0.02ꢆ0.01
0.36ꢆ0.03
—
1
0
(0.2ꢆ0.03)ꢄ10
—
—
—
—
1
—
Tiron
GSH
Ascorbic acid
—
—
—
—
—
0.34ꢆ0.10
0.25ꢆ0.07
NA, no activity was found up to 5000 mM. NB, no activity was found up to 1250 mM. a) mM Trolox equivalent/mM compound.

1
664
Vol. 29, No. 8
Fig. 4. HO˙ Scavenging Activity of Flurbiprofen, Fenoprofen, Fenbufen, Ketoprofen, Ibuprofen, Indoprofen and Naproxen
Each point represents the values obtained from four experiments, performed in duplicate (meanꢆS.E.M.).
order of potencies observed was fenoprofenꢀibuprofenꢁ Table 2. RNS Scavenging Effects (IC , MeanꢆS.E.M.) of Fenbufen,
5
0
Fenoprofen, Flurbiprofen, Ibuprofen, Indoprofen, Ketoprofen, Naproxen,
Carboxy-PTIO and Cysteine
fenbufenꢀflurbiprofenꢁketoprofenꢁnaproxenꢀindoprofen
(
IC s of 52ꢆ2ꢀ61ꢆ11ꢁ93ꢆ16ꢀ100ꢆ1ꢁ258ꢆ12ꢁ309ꢆ
5
0
6
4ꢀ320ꢆ27 mM, respectively) (meanꢆS.E.M.) (Table 1).
IC50 (mM)
The second-order rate constants (ks) calculated were
1
0
10
10
ꢀ
(4.9ꢆ0.5)ꢄ10 , (2.8ꢆ0.2)ꢄ10 , (2.0ꢆ0.2)ꢄ10 , (8.3ꢆ
Tested
ONOO
1
0
10
10
compound
1.6)ꢄ10 , (2.6ꢆ0.6)ꢄ10 , (3.3ꢆ0.3)ꢄ10 and (1.2ꢆ0.2)
˙NO
1
0
ꢀ1 ꢀ1
Absence
NaHCO3
Presence of
25 mM NaHCO3
ꢄ10
M
·s , for fenbufen, fenoprofen, flurbiprofen,
ibuprofen, indoprofen, ketoprofen and naproxen respectively
(
Table 1). The positive control (mannitol) provided an IC
Fenbufen
ꢇ5000
NA
NB
ꢇ5000
ꢇ5000
ꢇ5000
1497ꢆ130
NB
3944ꢆ367
—
5.2ꢆ0.1
NB
NA
ꢇ5000
ꢇ5000
1174ꢆ214
NB
1997ꢆ365
—
4.1ꢆ0.3
5
0
1
0
ꢀ1 ꢀ1
of 3272ꢆ215 mM and a ks of (0.20ꢆ0.03)ꢄ10
M
·s
Fenoprofen
Flurbiprofen
Ibuprofen
Indoprofen
Ketoprofen
Naproxen
ꢇ5000
ꢇ5000
968ꢆ104
ꢇ5000
3192ꢆ352
1.8ꢆ0.4
—
(meanꢆS.E.M.) (Table 1).
The assay performed without ascorbic acid or EDTA did
not indicate any pro-oxidant and/or iron chelation activity, re-
spectively, for the NSAIDs studied, since no increased mal-
onaldehyde formation was observed in their presence.
ROO˙ Scavenging Activity The results obtained from
the ROO˙ scavenging assay are shown in Table 1. It can be
observed that ROO˙ is effectively scavenged by naproxen,
while the other compounds were much less active against this
Carboxy-PTIO
Cysteine
NA, no activity was found up to 2500 mM. NB, no activity was found up to 5000 mM.
ROS. Nevertheless, the activity observed for all assayed did not attain 50% inhibition up to 5 mM (Table 1). The posi-
compounds was concentration-dependent. The order of po- tive control (carboxy-PTIO) yielded an IC50 of 1.8ꢆ0.4 mM
tencies was found to be naproxenꢁfenbufenꢁflurbiprofen (meanꢆS.E.M.) (Table 2).
ꢀ
ꢁ
fenoprofenꢁindoprofenꢁketoprofenꢃibuprofen (ORAC
ONOO Scavenging Activity The results obtained in
ꢀ
values of 0.36ꢆ0.03, 0.16ꢆ0.10, 0.14ꢆ0.05, 0.09ꢆ0.06, the ONOO scavenging assay in both the absence and pres-
.04ꢆ0.02, 0.02ꢆ0.01 and 0.02ꢆ0.01, respectively) ence of 25 mM NaHCO are presented in Figs. 6, 7 and Table
0
3
(
meanꢆS.E.M.) (Table 1). The positive controls GSH and 2. In the absence of NaHCO , it was observed that the oxida-
3
ꢀ
ascorbic acid yielded ORAC values of 0.34ꢆ0.10 and tion of DHR by ONOO was prevented by indoprofen,
.25ꢆ0.07 (meanꢆS.E.M.), respectively (Table 1). naproxen, flurbiprofen, ibuprofen and fenoprofen. On the
HOCl Scavenging Activity No scavenging activity other hand, fenbufen and ketoprofen exerted no effect up to
0
against HOCl was observed for any of the NSAIDs studied. 5 mM. The most effective compounds were shown to be: in-
The positive control (lipoic acid) provided an IC50 of doprofenꢁnaproxen (IC s: 1497ꢆ130 and 3944ꢆ367 mM,
50
5
9.5ꢆ4.7 mM (meanꢆS.E.M.).
respectively) (meanꢆS.E.M.)ꢁꢁflurbiprofenꢀibuprofenꢀ
˙
NO Scavenging Activity The results obtained in the fenoprofen (37, 25 and 23% of inhibitory effect at 5000 mM,
˙NO scavenging assay are presented in Fig. 5 and Table 2. It respectively). In the presence of NaHCO , the inhibitor activ-
3
was observed that indoprofen and naproxen are efficient ˙NO ities of indoprofen and naproxen were significantly increased
scavengers (Fig. 5), while the other NSAIDs have lower but (the resulting IC s were 1174ꢆ214 and 1997ꢆ365 mM, re-
5
0
effective activities, except fenoprofen for which no activity spectively) (meanꢆS.E.M.). The percentage inhibition for
was found up to 2.5 mM. The order of potencies found was the other effective compounds at the maximum concentration
indoprofenꢁnaproxen (IC s: 968ꢆ104 and 3192ꢆ852 mM of 5 mM was 40, 29 and 27% inhibition for flurbiprofen,
5
0
respectively) (meanꢆS.E.M.) (Table 2). The other NSAIDs ibuprofen and fenoprofen, respectively). The positive control

August 2006
1665
Fig. 5. ˙NO Scavenging Activity of Indoprofen and Naproxen
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
ꢀ
Fig. 6. ONOO Scavenging Activity of Fenoprofen, Flurbiprofen, Ibuprofen, Indoprofen and Naproxen
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
ꢀ
Fig. 7. ONOO Scavenging Activity of Flurbiprofen, Fenoprofen, Indoprofen, Ibuprofen and Naproxen in the Presence of 25 mM of Bicarbonate
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).

1
666
Vol. 29, No. 8
cysteine, provided an IC of 5.2ꢆ0.1 mM and 4.1ꢆ0.3 mM, in Table 3. Inhibition of PMA-Induced Neutrophil Burst by Fenbufen, Feno-
5
0
profen, Flurbiprofen, Ibuprofen, Indoprofen, Ketoprofen, Naproxen,
Diphenyleneiodonium Chloride (DPI), Tiron, 4-Aminobenzoic Acid
the absence and presence of NaHCO , respectively (Table 2).
3
Inhibition of Neutrophil Burst The measurement of
neutrophil burst was performed in two complementary as-
says. In the first assay the neutrophil burst was measured by
chemiluminescence, through monitoring the ROS-induced
oxidation of luminol (Fig. 8, Table 3). It can be observed that
from all the compounds tested, only fenbufen was shown to
be effective in preventing the neutrophil burst, but even here
only a weak inhibitory activity was observed (46% inhibitory
effect at the maximum concentration of 500 mM).
(
ABAH) and Lipoic Acid (IC ; MeanꢆS.E.M.)
50
IC50 (mM)
Tested
compound
Neutrophils
luminol
Neutrophils
lucigenin
MPO
luminol
Fenbufen
46% inhib.
at 500 mM
NB
36% inhib.
at 500 mM
NB
NA
Fenoprofen
NB
NB
Flurbiprofen
NB
30% inhib.
at 500 mM
NB
539ꢆ32
392ꢆ66
NB
—
1.1ꢆ0.1
—
The positive controls ABAH and lipoic acid provided
IC s of 18ꢆ4 and 943ꢆ271 mM (meanꢆS.E.M.), respec-
5
0
Ibuprofen
Indoprofen
Ketoprofen
Naproxen
ABAH
DPI
Lipoic acid
Tiron
NB
NB
NB
NB
18ꢆ4
—
NB
4070ꢆ505
NB
NB
0.34ꢆ0.01
—
tively.
In a second assay, the neutrophil burst was also measured
ꢀ
by chemiluminescence, through monitoring the O˙ -induced
2
oxidation of lucigenin (Fig. 9, Table 3). In this assay, indo-
profen, ketoprofen, fenbufen and flurbiprofen displayed in-
hibitory activity in a concentration dependent manner, while
fenoprofen, ibuprofen and naproxen showed no effect. Keto-
profen and indoprofen were the most potent (IC s of
943ꢆ271
—
119ꢆ10
—
88ꢆ10
NA, no activity was found up to 500 mM. NB, no activity was found up to 1000 mM.
5
0
Fig. 8. Inhibitory Effect (%) of Fenbufen on Luminol Oxidation Mediated by Neutrophil-Generated ROS
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
Fig. 9. Inhibitory Effect (%) of Fenbufen, Indoprofen, Ketoprofen and Flurbiprofen on Lucigenin Oxidation Mediated by Neutrophil-Generated O2
Each point represents the values obtained from four experiments, performed in duplicate (meanꢆS.E.M.).

August 2006
1667
Fig. 10. Inhibitory Effect (%) of Indoprofen on Luminol Oxidation Mediated by Myeloperoxidase (from Human Leukocytes)-Generated HOCl
Each point represents the values obtained from four experiments, performed in triplicate (meanꢆS.E.M.).
3
92ꢆ66 and 539ꢆ32 mM, respectively) (meanꢆS.E.M.). radicals formed at the inflammation sites. Hiller and Wil-
32)
Fenbufen and flurbiprofen only displayed weak effects (36% son showed that various NSAIDs tested could scavenge
and 30% inhibition, respectively, at the maximum concentra- HO˙ generated in a free solution by pulse radiolysis at an al-
tion of 500 mM). The IC values found for the positive con- most diffusion-controlled rate (k values between 5ꢄ10 and
trols, DPI (inhibitor of NADPH oxidase), tiron (scavenger of 10
O˙ ) were 1.1ꢆ0.1 and 88ꢆ10 mM (meanꢆS.E.M.), respec- ibuprofen, flurbiprofen and naproxen were effective HO˙
tively (Table 3).
9
5
0
2
1
0
ꢀ1 ꢀ1
M
·s ). Subsequently, it was shown that ketoprofen,
ꢀ
2
scavengers, using a methodology followed in the present
6
,33,34)
Luminol Oxidation by MPO-Generated HOCl Figure study.
Our results agree with those found elsewhere and
10 and Table 3 show the results obtained in the oxidation of further indicate that the other APA NSAIDs fenbufen, feno-
luminol by MPO-generated HOCl. Is this assay an effective profen and indoprofen are also good HO˙ scavengers. Con-
activity was only observed for indoprofen, with an IC value sidering that the direct reactivity of the aryl ring with HO˙ is
5
0
35)
of 4070ꢆ505 mM (meanꢆS.E.M.). The IC values found for very important for this type of activity and that the tested
the positive control, ABAH and lipoic acid were 0.34ꢆ0.01 compounds contain two phenyl groups in their chemical
5
0
and 119ꢆ10 mM (meanꢆS.E.M.), respectively (Table 3).
structure (expect ibuprofen, which only contains one phenyl
group), this may explain the potent scavenging activity
found. Notwithstanding, some compounds are capable of
redox cycling the metal ion required for hydroxyl generation,
DISCUSSION
Although the inhibition of prostaglandin synthesis consti- thus increasing radical production, exhibiting a pro-oxidant
36)
tutes the primary mechanism behind the anti-inflammatory activity. With a view to evaluating the pro-oxidant potential
action of NSAIDs, it has been suggested that the activity of of the APA NSAIDs, we modified the method by omitting
these compounds may be also partly due to their ability to ascorbic acid. Additionally, the assay performed in the ab-
scavenge ROS and RNS and to inhibit the respiratory burst sence of EDTA is also useful because it indicates the poten-
1
0,18—21)
of neutrophils triggered by various activator agents.
tial of the tested compounds to chelate iron ions. Fe(III)
As mentioned in the introduction, these activities have only chelators will decrease the amount of thiobarbituric-reactive
been sparsely and inadequately evaluated for NSAIDs from substances formed from deoxyribose as a result of Fe(III) re-
37)
the APA group, which prompted us to undertake a more sys- moval from the sugar. However, the assay performed with-
tematic evaluation of their relative potencies. In considering out ascorbic acid or EDTA did not indicate any pro-oxidant
the results obtained from the present assays using in vitro and/or iron chelation activity respectively, for the studied
screening systems, the studied APA NSAIDs were shown to NSAIDs, since no increased malonaldehyde formation was
be scavengers of ROS and/or RNS and in some cases, in- observed in their presence.
hibitors of the PMA-induced burst of human neutrophils, but
only within the mM to mM range. At first sight, it could be in- macrophages, eosinophils, and neutrophils generate O˙ ,
ferred that these effects, which were only obtained at high with NADPH oxidase playing an important role.
concentrations, would not be important for the therapeutic portance of O˙ scavenging activity in the therapy of inflam-
effect of these APA NSAIDs. However, by looking at the lit- mation has been well established. In the present study, it
During inflammation, the activation of mast cells,
ꢀ
2
38,39)
The im-
ꢀ
2
10)
erature, it can be observed that the effective concentrations in was observed that flurbiprofen, fenbufen, ketoprofen and in-
ꢀ
in vitro studies do vary, depending on the methodology used. doprofen prevented the O˙ -induced reduction of NBT in a
2
Furthermore, the effective concentrations are within those concentration dependent manner, indoprofen being the most
expected in vivo during anti-inflammatory therapy with APA potent, while no activity was observed for fenoprofen and
2,31)
NSAIDs,
at least for HO˙. HO˙ is one of the most reactive naproxen.

1
668
Vol. 29, No. 8
Much of the damage caused by increased generation of non-cellular assays. Indeed, a higher concentration of ABAH
ꢀ
O˙ and H O in the inflammatory sites arises from their (specific MPO inhibitor) is required for an effective in-
2
2
2
transition metal-dependent conversion into the highly reac- hibitory effect using neutrophils (IC50 of 18ꢆ4 mM) com-
4
0)
41)
tive HO˙ and into HOCl by MPO catalysis. In the pres- pared to that for purified MPO (IC of 0.34ꢆ0.01 mM).
5
0
ent study, the scavenging effects observed for H O were
rather similar to those obtained for O˙ with the exception of functions has previously been tested in vitro and in vivo. The
The interaction of APA NSAIDs with human neutrophil
2
2
ꢀ
2
naproxen, which exerted no effect against the later. On the aggregation and degranulation of human neutrophils in re-
other hand, no scavenging activity against HOCl was ob- sponse to the chemo attractant N-formyl-methionyl-leucyl-
served for any of the studied NSAIDs. The lack of HOCl phenylalanine (fMLP) was shown to be inhibited at therapeu-
4
5,46)
scavenging activity agrees with a previous reported study, in tic concentrations by ibuprofen,
flurbiprofen and keto-
45)
which ibuprofen, naproxen, flurbiprofen and ketoprofen did profen. On the other hand, ibuprofen had no effect on
not display scavenging activity for HOCl, using taurine chlo- fMLP-induced O generation at therapeutic concentrations
ꢀ
46)
2
42)
rination as the detecting system.
Alternatively, NSAIDs may also prevent the HOCl damag- these cells.
and failed to show any effect against PMA stimulation of
4
6,47)
Notably, neutrophils from patients taking
ing effects by inhibiting its production by MPO. To verify therapeutic doses of ibuprofen showed profiles of inhibited
46)
this possibility, the putative inhibitory effect of the tested responses to fMLP similar to those observed in vitro.
APA NSAIDs on MPO activity was tested using both the pu- Therefore, even if APA NSAIDs do not prevent the genera-
rified enzyme and its activity in human neutrophils. Using tion of ROS, their inhibition of aggregation and degranula-
the purified enzyme, an effective inhibitory activity was only tion of human neutrophils may also contribute to their anti-
observed for indoprofen. Notably, this is the only APA inflammatory effects.
NSAID bearing an indolic structure. Other indolic NSAIDs
The ROO˙ scavenging activity may be of extreme value
like indomethacin and acemetacin have been shown to be since, during the inflammatory processes, the walls of af-
42)
strong MPO inhibitors. The absence of an inhibitory effect fected tissue cells involved become amenable to lipid peroxi-
against purified MPO had already been observed for ibupro- dation, explaining the increase of lipid peroxidation marker
42)
19)
fen, ketoprofen and flurbiprofen. However, these authors compounds in inflammatory diseases. It was observed that
observed an inhibitory effect for naproxen, which was not ROO˙ is effectively scavenged by naproxen, while the other
confirmed in the present study. The reason for this discrep- compounds were much less active against this ROS. Of all
ancy is not clear as yet, though different methodologies were compounds tested, naproxen is the only one bearing a naph-
used to evaluate MPO activity.
thalene moiety. Curiously, the b-blocker propranolol, which
The majority of the neutrophil burst and the purified is another compound with a non-hydroxylated naphthalene
human MPO assays as well as those performed in the present moiety, has been shown to be a potent scavenger of ROO˙,
27)
study use the oxidation of luminol or lucigenin as the meas- with an ORAC value similar to that of naproxen. Although
ured endpoints. While luminol can be oxidized by several the chemical reactivity between ROO˙ and the naphthalene
1
6)
ꢀ 16,43,44)
ROS,
lucigenin is oxidized mainly by O2˙
or moiety is not clear as yet, it may be of great interest to fur-
2
0)
H O . When the neutrophil burst was measured by moni- ther study it in the future.
2
2
ꢀ
ꢀ
toring the O˙ -induced oxidation of lucigenin, the inhibitory
RNS, namely ˙NO and ONOO are also produced in the
2
activity of flurbiprofen, fenbufen, ketoprofen, and indopro- inflammatory sites and may contribute to the deleterious ef-
fen, was displayed in a concentration dependent manner with fects of inflammation. In fact, ˙NO is produced by nitric
ketoprofen and indoprofen being the most potent, while oxide synthases (NOS), which are typically induced during
48)
fenoprofen, ibuprofen and naproxen exerted no effect. Thus, inflammation. Possible pro-inflammatory effects of ˙NO
these results may be explained, not as an inhibitory effect include augmentation of vascular permeability of inflamed
concerning neutrophil activation but rather as a reflection of tissues, the generation of other destructive free radicals
ꢀ
ꢀ
ꢀ
a scavenging activity against neutrophil derived O˙ and/or (namely ONOO by reaction with O˙ ), the induction of cy-
2
2
H O since the cellular and non-cellular assays are almost su- clooxygenase as well as angiogenic and inflammatory cy-
2
2
perimposed. Noteworthy, the observed IC s are much lower tokines, activation of matrix metalloprotease and induction
5
0
49)
ꢀ
in the neutrophil assays (down to 10 times lower), which is a of chondrocyte apoptosis. ONOO itself is a relatively
strong reminder that the concentrations tested in vitro may long-lived cytotoxicant with strong oxidizing properties to-
not be as important as the identification of the potential to wards various cellular constituents, including sulphydryls,
50)
scavenge reactive species. When the oxidation of luminol lipids, amino acids, and nucleotides. In the present study,
was used as the measured endpoint, it was observed that indoprofen was shown to be a potent scavenger of ˙NO and
ꢀ
from all the tested compounds, only fenbufen was effective ONOO as was naproxen, although to a lesser extent, while
in preventing the neutrophil burst, but even here only a weak fenbufen, fenoprofen, flurbiprofen, ibuprofen and ketoprofen
inhibitory activity was observed. These results agree with only exerted residual effects. Again, it appears that the in-
those previously reported for ibuprofen, flurbiprofen, keto- dolic structure of indoprofen is a determining factor for its
25)
profen and naproxen. It is generally agreed that luminol RNS scavenging activity, in line with that observed for other
19)
51)
chemiluminescence in neutrophils results from intra and indolic NSAIDs. In a previous study, the scavenging ef-
extra cellular events and depends mainly on the reactions of fect of various NSAIDs for ˙NO was directly evaluated in
1
6,25)
the MPO–H O –Cl system (namely HOCl).
The absence vitro, by ESR spectrometry, using carboxy-PTIO as the ˙NO
2
2
of effect observed for indoprofen (shown to be inhibitor of detecting reagent and NOC-7 as the ˙NO donor. In this study,
MPO in the non-cellular assay) using neutrophils could be both ibuprofen and ketoprofen displayed potent scavenging
explained by a different sensitivity between the cellular and effects, well within therapeutic concentrations (IC s of 48
5
0

August 2006
1669
and 67 mM respectively), while naproxen only had low activ-
R. I., Herman R., Rider L., Kimmel S., Weissmann G., Proc. Natl.
Acad. Sci. U.S.A., 82, 7227—7231 (1985).
Halliwell B., Gutteridge J. M. C., Aruoma O. I., Anal. Biochem., 165,
51)
ity within the tested concentrations. The differences in the
effective concentrations observed, compared to the present
study, strongly suggests that in vitro screening systems are
much more important for determining relative potencies
among putative scavengers than for extrapolating effective
concentrations. In accordance with this, ketoprofen and
ibuprofen were demonstrated to protect B65 neuronal cells (a
rat neuroblastoma cell line) against ˙NO induced cell death,
6
)
2
15—219 (1987).
7
8
)
)
Ching T. L., Jong J., Bast A., Anal. Biochem., 218, 377—381 (1994).
Ohyashiki T., Nunomara M., Katoh T., Biochim. Biophys. Acta, 1421,
131—139 (1999).
9
1
1
)
Barja G., J. Bioenerg. Biomemb., 34, 227—233 (2002).
0) Halliwell B., Hoult J. R., Blake D. R., FASEB, 2, 2867—2873 (1988).
1) Salvemini D., Wang Z. Q., Bourdon D. M., Stern M. K., Currie M. G.,
Manning P. T., Eur. J. Pharmacol., 303, 217—220 (1996).
51)
again within therapeutic concentrations (333 nM—333 mM).
12) Cuzzocrea S., Zingarelli B., Costantino G., Szabó A., Salzman A. L.,
It is worth mentioning that the interference with ˙NO by
NSAIDs may also be due to the inhibition of inducible NO
synthase (iNOS), which was already demonstrated for APA
NSAIDs. Indeed, in an in vitro model involving RAW 264.7
macrophages, it was demonstrated that iNOS mRNA expres-
Caputi A. P., Szabó C., Br. J. Pharmacol., 121, 1065—1074 (1997).
1
1
1
1
3) Halliwell B., Cell Biol. Int. Rep., 6, 529—542 (1982).
4) Weiss J. R., New Engl. J. Med., 320, 365—376 (1989).
5) Yang K. D., Hill H. R., J. Pediatrics, 119, 343—354 (1991).
6) Hasegawa H., Suzuki K., Nakaji S., Sugawara K., J. Immunol. Meth-
ods, 210, 1—10 (1997).
5
2,53)
sion is suppressed by flurbiprofen and ibuprofen,
an ef- 17) Dallegri F., Patrone F., Ballestrero A., Ottonello L., Ferrando F., Sac-
chetti C., Int. J. Tissue React., 12, 107—111 (1990).
8) Fernandes E., Toste S., Lima J. L. F. C., Reis S., Free Radic. Biol.
Med., 35, 1008—1117 (2003).
9) Fernandes E., Costa D., Toste S. A., Lima J. L. S. C., Reis S., Free
Radic. Biol. Med., 37, 1895—1905 (2004).
fect related to the inhibition of transcription factor activation
1
1
2
2
2
like NF-kB and AP-1, resulting in diminished formation of
5
3,54)
pro-inflammatory factors like iNOS and TNF-a.
How-
ever, the putative beneficial effects resulting from lower lev-
els of ˙NO should be interpreted cautiously since the possible
depletion of physiological ˙NO concentrations may also be
harmful for the patient. ˙NO, at physiological levels is mainly
involved in homeostatic biochemical and physiological
0) Costa D., Gomes A., Reis S., Lima J. L. F. C., Fernandes E., Life Sci.,
7
6, 2841—2848 (2005).
1) Costa D., Marques A. P., Reis R. L., Lima J. L. F. C., Fernandes E.,
Free Radic. Biol. Med., 40, 632—640 (2006).
2) Twomey B., Dale M., Biochem. Pharmacol., 43, 413—418 (1992).
processes such as signal transduction, neurotransmission, 23) Landoni M. F., Cunningham F. M., Lees P., Vet. Rec., 137, 428—431
(
1995).
smooth muscle relaxation, peristalsis, gastroprotective ef-
fects, inhibition of platelet aggregation, blood pressure mod-
ulation, immune system control as well as learning and mem-
2
4) Zapolska-Downar D., Zapolski-Downar A., Bukowska H., Galka H.,
Naruszewicz M., Life Sci., 65, 2289—2303 (1999).
5) Parij N., Nagy A. M., Fondu P., Nève J., Eur. J. Pharmacol., 352,
299—303 (1998).
2
55,56)
ory.
Therefore, a possible variation of ˙NO physiological
levels by APA NSAIDs could potentially affect these physio- 26) Aruoma O. I., Gen. Pharmac., 28, 269—272 (1997).
2
7) Gomes A., Fernandes E., Lima J. L. F. C., J. Fluorescenc., 16, 119—
39 (2006).
8) Ullrich T., Oberle S., Abate A., Schroder H., FEBS Lett., 406, 66—68
1997).
logical processes.
In conclusion, the results obtained in the present work
demonstrate that under the present experimental conditions,
1
2
(
many of the studied APA NSAIDs showed to be scavengers 29) Beckman J. S., Chen J., Ischiropoulos H., Crow J. P., Methods Enzy-
ꢀ
of O2˙ (fenbufenꢀflurbiprofenꢀindoprofenꢀketoprofen),
mol., 223, 229—240 (1994).
0) Whiteman M., Ketsawatsakul U., Halliwell B., Ann. N.Y. Acad. Sci.,
3
H O
(ketoprofenꢀindoprofenꢀfenbufenꢁflurbiprofenꢁ
2
2
9
62, 242—259 (2002).
naproxen), HO˙ (fenoprofenꢀibuprofenꢁfenbufenꢀflurbi-
profenꢁketoprofenꢁindoprofenꢀnaproxen), ˙NO (indopro-
fenꢁnaproxen), ONOO (indoprofenꢁnaproxenꢁfenopro-
3
3
1) Schulz M., Schmoldt A., Pharmazie, 58, 447—474 (2003).
2) Hiller K. O., Wilson R. L., Biochem. Pharmacol., 32, 2109—2111
(1983).
ꢀ
3
3
3
3
3
3) Aruoma O. I., Halliwell B., Xenobiotica, 18, 459—470 (1988).
4) Parij N., Nagy A. M., Neve J., Free Radic. Res., 23, 571—579 (1995).
5) Ng T. B., Liu F., Zhao L., J. Neural. Transm., 107, 1243—1251 (2000).
6) Li C., Xie B., J. Agric. Food Chem., 48, 6362—6366 (2000).
7) Moran J. F., Klucas R. V., Grayer R. J., Abian J., Becana M., Free
Radic. Biol. Med., 22, 861—870 (1997).
fenꢀflurbiprofenꢀibuprofen), to inhibit MPO activity (indo-
profen) and to scavenge human neutrophil derived ROS
(ketoprofenꢁindoprofenꢁfenbufenꢁflurbiprofen). The ob-
served effects, if confirmed in vivo, may strongly contribute
to the anti-inflammatory therapeutical activity of these
NSAIDs.
38) Barnes P. J., Free Radic. Biol. Med., 9, 235—243 (1990).
9) Hampton M. B., Kettle A. J., Winterbourn C. C., Blood, 92, 3007—
3
3
017 (1998).
Acknowledgements The authors acknowledge the finan-
4
0) Halliwell B., Gutteridge J. M., Mol. Aspects Med., 8, 89—193 (1985).
cial support given by Reitoria da Universidade do Porto and 41) Winterbourn C. C., Biochim. Biophys. Acta, 840, 204—210 (1985).
Caixa Geral de Depósitos (REIT-EF 2005/2006). David 42) Neve J., Parij N., Moguilevsky N., Eur. J. Pharmacol., 417, 37—43
(
2001).
Costa acknowledges FCT and FSE his PhD grant
SFRH/BD/10483/2002).
4
4
3) Gyllenhammar H., J. Immunol. Methods, 97, 209—213 (1987).
4) Li Y., Zhu H., Kuppusamy P., Roubaud V., Zweier J. L., Trush M. A., J.
Biol. Chem., 273, 2015—2023 (1998).
(
REFERENCES
45) Smith R. J., Iden S. S., Biochem. Pharmacol., 29, 2389—2395 (1980).
46) Kaplan H. B., Edelson H. S., Korchak H. M., Given W. P., Abramson
1
2
3
)
)
)
Bird H. A., Drugs Aging, 12, 87—95 (1998).
Davies N. M., Clin. Pharmacokinet., 34, 101—154 (1998).
Patel B. K., Hanna-Brown M., Hadley M. R., Hutt A. J., Electrophore- 48) Cuzzocrea S., Riley D. P., Caputi A. P., Salvemini D., Pharmacology
sis, 25, 2625—2656 (2004).
Reviews, 53, 135—159 (2001).
Panico A. M., Cardil V., Vittorio F., Ronsisvalle G., Scoto G. M., Par- 49) Miyasaka N., Hirata Y., Life Sci., 61, 2073—2081 (1997).
S., Weissmann G., Biochem. Pharmacol., 33, 371—378 (1984).
47) Nielsen V. G., Webster R. O., Immunopharmacol., 13, 61—71 (1987).
4
)
)
enti C., Gentile B., Morrone R., Nicolosi G., Il Farmaco, 58, 1339—
344 (2003).
Abramson S., Korchak H., Ludewig R., Edelson H., Haines K., Levin
50) Briviba K., Klotz L. O., Sies H., Methods Enzymol., 301, 391—411
(1999).
51) Asanuma M., Nishibayashi-Asanuma S., Miyazaki I., Kohno M.,
1
5

1
5
670
Vol. 29, No. 8
Ogawa N., J. Neurochem., 76, 1895—1904 (2001).
2) Aeberhard E. E., Henderson S. A., Arabolos N. S., Griscavage J. M.,
54) Tegeder I., Niederberger E., Israr E., Gühring H., Brune K., Euchen-
hofer C., Grösch S., Geisslinger G., FASEB J., 15, 595—597 (2001).
Castro F. E., Barrett C. T., Ignarro L. J., Biochem. Biophys. Res. Com- 55) Moncada S., Palmer R. M. J., Higgs E. A., Pharmacology Reviews, 43,
mun., 208, 1053—1059 (1995).
109—142 (1991).
53) Hinz B., Brune K., Rau T., Pahl A., Pharm. Res., 18, 151—156
56) Martin M. J., Jimenez M. D., Motilva V., Curr. Pharm. Design, 7,
(2001).
881—908 (2001).