logy 6 (2006) 1723–1728www.elsevier.com/locate/intimp

International Immunopharmaco

Inhibition of inducible nitric oxide synthase and cyclooxygenase-2expression by flavonoids isolated from Tanacetum microphyllum

José Antonio Guerra, María Francisca Molina, María José Abad ⁎,Angel María Villar, Paulina Bermejo

Department of Pharmacology, Faculty of Pharmacy, University Complutense, Ciudad Universitaria s/n, 28040, Madrid, Spain

Received 22 May 2006; received in revised form 31 July 2006; accepted 3 August 2006

Abstract

Plant flavonoids show anti-inflammatory activity both in vitro and in vivo. Some flavonoids have been reported previously toinhibit nitric oxide (NO) and prostaglandin E2 (PGE2) production by suppressing inducible nitric oxide synthase (iNOS) andcyclooxygenase-2 (COX-2) expression. The present study focuses on the effect of various naturally occurring flavonoids (santin,ermanin, centaureidin and 5,3′-dihydroxy-4′-methoxy-7-methoxycarbonylflavonol) on modulation of lipopolysaccharide (LPS)-induced iNOS and COX-2 expression in RAW 264.7 cells. Western blotting showed that all flavonoids suppressed the induction ofboth iNOS and COX-2. Ermanin and 5,3′-dihydroxy-4′-methoxy-7-methoxycarbonylflavonol were the most potent inhibitors. Thisstudy suggests that inhibition of iNOS and COX-2 expression by flavonoids may be one of the mechanisms responsible for theiranti-inflammatory effects, and that they may be potential agents for use in the treatment of inflammatory diseases.© 2006 Elsevier B.V. All rights reserved.

Keywords: Tanacetum microphyllum; Flavonoids; Nitric oxide; Inducible nitric oxide synthase; Prostaglandin E2; Cyclooxygenase-2

1. Introduction

Hundreds of herbal remedies have been used histori-cally in the maintenance of health and in the treatment ofdiseases. Today, such alternatives to over-the-countermedicines and prescriptions are still in demand. Onepotential use for herbals is the treatment of inflammatoryconditions [1–3]. A number of inflammatory mediatorsare released by cells in response to localized injury ortrauma. These mediators elicit or enhance particularfunctions of the inflammatory process and may bemonitored to assess an inflammatory response.

⁎ Corresponding author. Tel.: +34 91 3941871; fax: +34 913941726.

E-mail address: mjabad@farm.ucm.es (M.J. Abad).

1567-5769/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.intimp.2006.08.012

A body of evidence suggests that prostaglandins(PGs) and nitric oxide (NO) are involved in variouspathophysiological processes including inflammationand carcinogenesis, and that cyclooxygenase-2 (COX-2)and inducible nitric oxide synthase (iNOS) are mainlyresponsible for the production of large amounts of thesemediators [4,5]. iNOS is expressed in various cell types:in the macrophages it is induced by different inflamma-tory stimuli such as bacterial endotoxic lipopolysaccha-ride (LPS) and inflammatory cytokines [6]. PGs alsoplay a major role as mediators of the inflammatoryresponse. COX is the enzyme which converts arachi-donic acid (AA) to PGs. Like NOS, COX has been foundin more than two isoforms, of which inducible isoformCOX-2 is responsible for the production of largeamounts of pro-inflammatory PGs at the inflammatory

Fig. 1. Structure of flavonoids isolated from Tanacetum microphyllum.

1724 J.A. Guerra et al. / International Immunopharmacology 6 (2006) 1723–1728

site [7,8]. Thus, there are many efforts to developenzyme inhibitors or repressors of enzyme formationthat are selective for the inducible forms of theseenzymes, and do not affect the desirable activity oftheir respective constitutive isoforms.

Plants from the genus Tanacetum L. (Compositae)have been used in traditional medicine over a longperiod of time; of these Tanacetum parthenium (L.)Schultz Bip. is the most prominent species and a knownremedy for the treatment of various diseases, includingarthritis, fever, migraine, vertigo, menstrual disorders,stomach-ache, toothache, insect bites and psoriasis [9].Additionally, Tanacetum microphyllum DC., an endem-ic species of the Iberian Peninsula, is useful for treatingvarious inflammatory disorders [10]. The secondarymetabolites that mediate the pharmacological effect ofthis species are mainly flavonoids.

Flavonoids are natural compounds, chemically derivedfrom benzo-γ-pirone or flavone, widely distributed in theplant kingdom, which possess a number of biologicalactivities. These compounds are important not only for thenormal growth, development and defence of plants, butalso for animals, including humans. They are consideredimportant constituents of the human diet: in fact they arefound in fruits, vegetables and some beverages such as redwine, coffee and beer. Flavonoids may exert a wide rangeof therapeutical activities such as antihepatotoxic, anti-osteoporotic, vascular protective, antiulcer, antispasmod-ic, antisecretory and antidiarrhoeal properties [11,12].Plant flavonoids also affect inflammatory processes inmammals and possess both anti-inflammatory andimmunomodulatory activities in vitro and in vivo. Theseactions have been reported at the cellular level to involve,in part, inhibition of COX/lipoxygenase (LOX) andprotection against oxidation [13–15]. In addition to thedirect inhibition of AA-metabolizing enzymes, recentinvestigations also revealed that certain flavonoids inhibitthe induction of iNOS expression in macrophages, thusreducing NO production. Investigations have also shownrecently that several flavonoids inhibit the induction ofCOX-2, the inducible isoform of COX [16–19].

As part of our investigations of the biologicalproperties of flavonoids in the genus Tanacetum, wereported earlier the discovery of four flavonoids in theaerial parts of T. microphyllum: santin, ermanin, centaur-eidin and 5,3′-dihydroxy-4′-methoxy-7-methoxycarbo-nylflavonol [20,21]. In previous work, we demonstratedthe in vitro inhibition of AA metabolism by the com-pounds included in this study [22–24]; there are alsostudies on their anti-inflammatory effects in vivo [20,21].These results prompted us to start a more extensiveinvestigation of their possible inhibitory effects. To

develop new compounds that are inhibitors of eitheriNOS or COX-2 may require a better understanding ofinflammatory states. In the present study, we haveinvestigated the effects of the four flavonoids isolatedfrom T. microphyllum on the expression of iNOS andCOX-2 inRAW264.7macrophages stimulatedwith LPS.

2. Materials and methods

2.1. Materials

The flavonoids santin, ermanin, centaureidin and 5,3′-dihydroxy-4′-methoxy-7-methoxycarbonylflavonol (flavonol)were isolated from T. microphyllum as previously described

1725J.A. Guerra et al. / International Immunopharmacology 6 (2006) 1723–1728

[20,21] (Fig. 1). RPMI 1640medium, fetal bovine serum (FBS),penicillin and streptomycin were obtained from GIBCO/BRLLife Technologies (Grand Island, NY, USA). Polyclonalantibodies specific for iNOS and COX-2 were obtained fromSantaCruz Biotechnology (Santa Cruz, CA,USA). All materialsfor sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE) were obtained from Bio-Rad Laboratories (Her-cules, CA, USA). Other reagents were purchased from SigmaChemical Co. (St. Louis, MO, USA). LPS were from Escheri-chia coli 055:B5 (L2880, not less 500 EV per mg).

2.2. Cell culture

RAW 264.7 cells, a murine macrophage cell line, were ob-tained from theATCC (AmericanTypeCultureCollection). Cellswere maintained in RPMI 1640 medium containing 10% FBS,100 U/ml penicillin and 100 μg/ml streptomycin, and incubatedat 37 °C under 5% CO2. Flavonoids were dissolved in dimethyl-sulfoxide and further diluted in culture medium. Dimethylsulf-oxide was employed at a constant final concentration (less than0.1%) which did not affect the viability of macrophages.

2.3. Western blot analysis

Cells were plated at a density of 106/ml in 60 mm tissueculture dishes overnight to allow macrophage adherence.Confluent macrophages were incubated with or without E. coliLPS (0.5 μg/ml) in the absence or presence of differentflavonoids in various concentrations for 18 h. Cells werewashed twice with ice-cold phosphate-buffered saline (PBS)(1 ml/dish), and lysed in a buffer containing 10 mM Tris–HCl,pH 7.5, 150 mM NaCl, 5% NP40, 1 mM EGTA and 1 mM

Fig. 2. Effects of santin, ermanin, centaureidin and flavonol on LPS-induced iNflavonoids and LPS (0.5 μg/ml) for 18 h to induce protein expression. AfterMaterials and methods. Data of relative expression (arbitrary units) are exprein triplicate (one-way ANOVA test: ⁎pb0.05 compared with the control gro

EDTA, and disrupted with a Hielscher GMBH Disrupter. Thelysis buffer containing the disrupted cells was centrifuged at14,000 rpm and 4 °C for 10 min. Protein concentrations wereestimated by the BCA method using bovine serum albumin asstandard. Equal amounts of protein (50 μg) of the cell lysateswere dissolved in Laemmli's sample buffer [25], boiled for5 min, and subjected to SDS-PAGE (7.5% polyacrylamide).Western blotting was performed by transferring proteins from aslab gel to a sheet of polyvinylidene difluoride membrane at20 mA at room temperature. For iNOS and COX-2 detection,the filter was then blocked with PBS, 5% non-fat dried milk for1 h at room temperature and incubated with the first antibody(diluted 1:1000 in PBS for iNOS and 1:2000 for COX-2), 5%non-fat dried milk, 0.1% Tween 20 overnight at 4 °C. Bothfilters were then incubated with a secondary antibody (anti-rabbit IgG-horseradish peroxidase conjugate 1:2000 dilutionfor iNOS and anti-goat IgG-horseradish peroxidase conjugate1:5000 dilution for COX-2) for 1 h at room temperature.Subsequently, blots were extensively washed with PBS,developed using ECL-detection reagents and exposed tofilms. The protein bands of iNOS and COX-2 on films werescanned and densitometrically analysed by personal densitom-eter (Molecular Dynamics).

2.4. Data analysis

Results are expressed as means±SEM of at least threeindependent experiments assayed in triplicate. Inhibitoryconcentration 50% (IC50) values were calculated from at leastfour significant concentrations, using Pharm/PCS Pharmaco-logic Calculation System, Version 4 (Microcomputer SpecialistP.O.). Statistical analysis was performed using a Bonferonni

OS expression in RAW 264.7 macrophages. Cells were incubated withthe above treatment, cells were washed and harvested as described inssed as means±SEM of at least three independent experiments assayedup, and ⁎⁎pb0.01 compared with the control group).

Fig. 4. Effects of santin, ermanin, centaureidin and flavonol on LPS-induced iNOS and COX-2 expression in RAW 264.7 macrophages.iNOS and COX-2 expression was analyzed by Western blotting andthen quantitatively assessed using a densitometer: (A) santin andermanin on iNOS, (B) centaureidin and flavonol on iNOS, (C) santinand ermanin on COX-2, (D) centaureidin and flavonol on COX-2.

Fig. 3. Effects of santin, ermanin, centaureidin and flavonol on LPS-induced COX-2 expression in RAW 264.7 macrophages. Cells were incubatedwith flavonoids and LPS (0.5 μg/ml) for 18 h to induce protein expression. After the above treatment, cells were washed and harvested as described inMaterials and methods. Data of relative expression (arbitrary units) are expressed as means±SEM of at least three independent experiments assayedin triplicate (one-way ANOVA test: ⁎pb0.05 compared with the control group, and ⁎⁎pb0.01 compared with the control group).

1726 J.A. Guerra et al. / International Immunopharmacology 6 (2006) 1723–1728

post-test after one-way ANOVA test to determine significantdifferences in the study, and a p valueb0.05 was consideredsignificant.

3. Results

To determine if the inhibitory effects of flavonoids oninflammatory mediators such as NO and PGE2 were related toa modulation of iNOS and COX-2 induction, we evaluatedtheir expression by Western blot analysis.

3.1. Effect of flavonoids on LPS-induced iNOS expression

We determined whether these compounds affect theexpression of iNOS. Western blot analysis demonstrated thatunstimulated RAW 264.7 macrophages expressed only a smallamount of iNOS proteins. Flavonoid treatment alone did notaffect the basal expression of iNOS proteins (data not shown)and cell viability was N90% at the concentrations tested. UponLPS treatment (0.5 μg/ml) for 18 h, the expression of iNOSproteins increased in these cells. These experiments show aconcentration-dependent inhibitory activity by flavonoids onLPS-induced iNOS expression, as evident in Figs. 2 and 4. Theinhibitory profile of flavonoids examined here on iNOSinduction overlaps with their inhibitory activity on NOproduction previously reported [24]. From densitometricanalysis, ermanin and flavonol decreased LPS-induced iNOSexpression by 92±0.1% and 91±0.5%, respectively, whilesantin and centaureidin produced decreases of 71±0.4% and83±0.1% in iNOS expression. The IC50 values of ermanin,flavonol, santin and centaureidin at inhibiting this reactionwere 6.70 μM, 40.21 μM, 8.87 μMand 16.30 μM, respectively.

1727J.A. Guerra et al. / International Immunopharmacology 6 (2006) 1723–1728

3.2. Effect of flavonoids on LPS-induced COX-2 expression

We determined whether these compounds affect theexpression of COX-2. RAW 264.7 cells did not expressdetectable COX-2 proteins when incubated in the mediumwithout LPS for 18 h, and the basal level of COX-2 proteinswas not affected when incubated with flavonoids (data notshown). Cell viability was N90% at the concentrations tested.Upon LPS treatment (0.5 μg/ml) for 18 h, COX-2 proteinsdrastically increased in these cells. These experiments show aconcentration-dependent inhibitory activity by flavonoids onLPS-induced COX-2 expression, as evident in Figs. 3 and 4.The inhibitory profile of flavonoids examined here on COX-2induction overlaps with their inhibitory activity on PGE2

production previously reported [24]. From densitometricanalysis, ermanin and flavonol decreased LPS-induced COX-2 expression by 94±0.5% and 97±0.2%, respectively, whilesantin and centaureidin produced decreases of 74±0.3% and83±0.1% in COX-2 expression The IC50 values of ermanin,flavonol, santin and centaureidin at inhibiting this reactionwere 1.68 μM, 28.86 μM, 1.74 μMand 24.49 μM, respectively.

4. Discussion

As part of our investigations of the biologicalproperties of flavonoids in the genus Tanacetum, wereported earlier the discovery of four flavonoids in theaerial parts of T. microphyllum: santin, ermanin,centaureidin and flavonol [20,21]. The present investi-gation gives a scientific basis for using the Tanacetumspecies as a source for anti-inflammatory agents. Wehave previously studied these flavonoids as potentialanti-inflammatory agents and reported that some ofthem are able to control AA metabolite production invitro as well as in vivo, having a potential role inmodulating the inflammatory process [20–23]. Recent-ly, we have investigated the effects of these flavonoidson NO and PGE2 (COX-2) production in LPS-stimulated mouse peritoneal macrophages [24]. Resultsof this study showed that all the mentionated flavo-noids reduced iNOS and COX-2 expression in LPS-activated macrophages, and could be ranked accord-ing to their maximum value if inhibitory activity assantinbcentaureidinbermanin= flavonol for iNOS andsantinbcentaureidinbermaninb flavonol for COX-2.To obtain a clearer knowledge of the inhibitorymechanism of NO and PGE2 production, the effect ofthese compounds on iNOS and COX-2 expression wasexamined. Moreover, our results confirm or identifysome features concerning the inhibition of thesemediators by flavonoids.

The active flavonoids examined possess a C-2,3double bond and 5,7-dihydroxyl groups in the A ring

(santin, ermanin and centaureidin). However, the sub-stitution at position 7 by a carbomethoxy group maybe favourable for iNOS activity, as is apparent fromthe data of flavonol, which is, together with ermanin,the most effective flavonoid. From these results andthose previously reported [24], we postulate that thepresence of a methoxy group at the 6 position isdetrimental to iNOS and COX-2 activities. However,the activity is restored by 3′,4′-disubstitution in the Bring, as is apparent from the data of centaureidinversus santin.

Our study suggests that the modulation of COX-2and iNOS by these flavonoids may be important inthe prevention of inflammation, and may contributeto their anti-inflammatory effects. The profile andpotency of these compounds may have relevance forthe inhibition of the inflammatory response, repre-senting a new approach for the modulation ofdifferent inflammatory pathologies. Future perspec-tives should include understanding the molecularbasis for inhibitory effects of these specific herbalconstituents on pro-inflammatory cytokine geneexpression, and specifically the role of altered signaltransduction.

Acknowledgements

The technical assistance of Ms. Brooke-Turner isgratefully acknowledged.

References

[1] Bremner P, Heinrich M. Natural products as targeted modulatorsof the nuclear factor-κB pathway. J Pharm Pharmacol2002;54:453–72.

[2] Newman DJ, Cragg GM, Snader KM. Natural products assources of new drugs over the period 1981–2002. J Nat Prod2003;66:1022–37.

[3] Lee KH. Current developments in the discovery and design ofnew drug candidates from plant natural product leads. J Nat Prod2004;67:273–83.

[4] Hinz B, Brune K. Cyclooxygenase-2 10 years later. J PharmacolExp Ther 2002;300:367–75.

[5] Clark IA,AwburnMM,WhittenRO,Harper CG, LiombaNG, et al.Tissue distribution of migration inhibitory factor and induciblenitric oxide synthase in falciparum malaria and sepsis in Africanchildren. Malar J 2003;2:6–23.

[6] Aktan F. iNOS-mediated nitric oxide production and itsregulation. Life Sci 2004;75:639–53.

[7] Vane JR, Bakhle YS, Botting RM. Cyclooxygenase 1 and 2.Annu Rev Pharmacol Toxicol 1998;38:97–120.

[8] Simon LS. Role of regulation of cyclooxygenase-2 duringinflammation. Am J Med 1999;106:37S–42S.

[9] Ernst E, Pittler MH. The efficacy and safety of feverfew(Tanacetum parthenium L.): an update of a systematic review.Public Health Nutr 2000;3:509–14.

1728 J.A. Guerra et al. / International Immunopharmacology 6 (2006) 1723–1728

[10] Abad MJ, Bermejo P, Villar AM. The phytochemical andpharmacological investigations on Tanacetum microphyllum.Recent Res Dev Life Sci 2003;1:371–95.

[11] Di Carlo G, Mascolo N, Izzo AA, Capasso F. Flavonoids: old andnew aspects of a class of natural therapeutic drugs. Life Sci1999;65:337–53.

[12] Nijueldt RJ, Van Nood K, Van Hoorn DE, Boelens PG, VanNorren K, Van Leenwen PB. Flavonoids: a review of probablemechanisms of actions and potential application. Am J Clin Nutr2001;74:418–25.

[13] Van Hoorn DEC, Nijueldt RJ, Van Leenwen PAM, Hofman Z,M'Rabet L, et al. Accurate prediction of xanthine oxidaseinhibition based on the structure of flavonoids. Eur J Pharmacol2002;451:111–8.

[14] Sadik CD, Sies H, Schewe T. Inhibition of 15-lipoxygenase byflavonoids: structure-activity relations and mode of action.Biochem Pharmacol 2003;65:773–81.

[15] Smolinski AT, Pestka JJ. Modulation of lipopolysaccharide-induced proinflammatory cytokine production in vitro and invivo by the herbal constituents apigenin (chamomile), ginseno-side Rb1 (ginseng) and parthenolide (feverfew). Food ChemToxicol 2003;41:1381–90.

[16] Chi YS, Cheon BS, Kim HP. Effect of wogonin, a plant flavonefrom Scutellaria radix, on the supression of cyclooxygenase-2and the induction of inducible nitric oxide synthase inlipopolysaccharide-treated RAW 264.7 cells. Biochem Pharma-col 2001;61:1195–203.

[17] Raso GM, Meli R, Di Carlo G, Pacilio M, Di Carlo R. Inhibitionof inducible nitric oxide synthase and cyclooxygenase-2

expression by flavonoids in macrophage J774A.1. Life Sci2001;68:921–31.

[18] Shen SC, Lee WR, Lin HY, Huang HC, Ko CH, et al. In vitro andin vivo inhibitory activities of rutin, wogonin and quercetin onlipopolysaccharide-induced nitric oxide and prostaglandin E2

production. Eur J Pharmacol 2002;446:187–94.[19] Cho H, Yun CW, Park WK, Kong JY, Kim KS, et al. Modulation

of the activity of pro-inflammatory enzymes COX-2 and iNOSby chrysin derivatives. Pharmacol Res 2004;49:37–43.

[20] Abad MJ, Bermejo P, Villar A, Valverde S. Anti-inflammatoryactivity of two flavonoids from Tanacetum microphyllum. J NatProd 1993;56:1164–7.

[21] Martínez J, Silván AM, Abad MJ, Bermejo P, Villar A, SöllhuberM. Isolation of two flavonoids from Tanacetum microphyllum asPMA-induced ear edema inhibitors. J Nat Prod 1997;60:142–4.

[22] Abad MJ, Bermejo P, Villar A. The activity of flavonoidsextracted from Tanacetum microphyllum DC. (Compositae) onsoybean lipoxygenase and prostaglandin synthetase. Gen Phar-macol 1995;26:815–9.

[23] Silván AM, AbadMJ, Bermejo P, Villar A. Effects of compoundsextracted from Tanacetum microphyllum on arachidonic acidmetabolism in cellular systems. Planta Med 1998;64:200–3.

[24] Abad MJ, Bermejo P, Alvarez M, Guerra JA, Silván AM, VillarAM. Flavonoids and a sesquiterpene lactone from Tanacetummicrophyllum inhibit anti-inflammatory mediators in LPS-stimulated mouse peritoneal macrophages. Planta Med2004;70:34–8.

[25] Laemmli VK. Cleavage of structural proteins during the assemblyof the head of bacteriophage T4. Nature 1970;117:680–5.