Effects of naturally occurring dihydroflavonols from Inula viscosa on inflammation and enzymes involved in the arachidonic acid metabolism

Life Sciences 81 (2007) 480 – 488 www.elsevier.com/locate/lifescie

Effects of naturally occurring dihydroflavonols from Inula viscosa on inflammation and enzymes involved in the arachidonic acid metabolism Victoriano Hernández, M.Carmen Recio ⁎, Salvador Máñez, Rosa M. Giner, José-Luis Ríos Departament de Farmacologia, Facultat de Farmacia, Universitat de València, Av. Vicent Andrés Estellés s/n. 46100 Burjassot, Spain Received 16 January 2007; accepted 13 June 2007

Abstract The anti-inflammatory properties of three flavanones isolated from Inula viscosa, sakuranetin, 7-O-methylaromadendrin, and 3-acetyl-7-Omethylaromadendrin, have been tested both in vitro and in vivo. Acute inflammation in vivo was induced by means of topical application of 12-Otetradecanoylphorbol 13-acetate (TPA) to mouse ears or by subcutaneous injection of phospholipase A2 (PLA2) into mouse paws. The test compounds were evaluated in vitro for their effect on both the metabolism of arachidonic acid and on the release and/or activity of enzymes involved in the inflammatory response such as elastase, myeloperoxidase (MPO), and protein kinase C (PKC). The most active compounds in vivo against PLA2-induced paw oedema were 7-O-methylaromadendrin (ED50 = 8 mg/kg) and sakuranetin (ED50 = 18 mg/kg). In contrast, the most potent compound against TPA-induced ear oedema was 3-acetyl-7-O-methylaromadendrin (ED50 = 185 μg/ear), followed by sakuranetin (ED50 = 205 μg/ear). In vitro, the latter compound was the most potent inhibitor of leukotriene (LT) B4 production by peritoneal rat neutrophils (IC50 = 9 μM) and it was also the only compound that directly inhibited the activity of 5-lipoxygenase (5-LOX). 3-Acetyl-7-Omethylaromadendrin also inhibited LTB4 production (IC50 = 15 μM), but had no effect on 5-LOX activity. The only flavanone that inhibited the secretory PLA2 activity in vitro was 7-O-methylaromadendrin. This finding may partly explain the anti-inflammatory effect observed in vivo, although other mechanisms such as the inhibition of histamine release by mast cells may also be implicated. Sakuranetin at 100 μM was found to inhibit elastase release, although this result is partly due to direct inhibition of the enzyme itself. At the same concentration, 7-Omethylaromadendrin only affected the enzyme release. Finally, none of the flavanones exhibited any effect on MPO or PKC activities. Taken together, these findings indicate that sakuranetin may be a selective inhibitor of 5-LOX. © 2007 Elsevier Inc. All rights reserved. Keywords: Inula viscose; Dihydroflavonols; Anti-inflammatory activity; LTB4

Introduction Inula viscosa (L.) Aiton (Dittrichia viscosa (L.) Greuter), Asteraceae, is an herbaceous plant known for its antiseptic properties and its effectiveness against skin inflammations. We have previously reported the isolation and spectroscopic identification of different sesquiterpenoids and flavonoids from the dichloromethane extract taken from the aerial parts of this plant and we have also published our findings on the pharmacological activity of these compounds against both acute and chronic models of inflammation, along with data concerning the interaction of the compounds with certain mediators (Máñez et al., 1999; Hernández et al., 2001, 2005). However, while there ⁎ Corresponding author. Tel.: +34 963 543283; fax: +34 963544973. E-mail address: [email protected] (M.C. Recio). 0024-3205/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2007.06.006

is a great deal of interest in the pharmacological activities of flavonoids as anti-inflammatory agents, very few studies have examined the analogous effects of flavanones (see reviews by Middleton et al., 2000; Pietta, 2000; Heijnen et al., 2002; Kim et al., 2004). In fact, Máñez et al. (1999) were the first to describe the anti-inflammatory effect in vivo of two dihydroflavonols, namely 7-O-methylaromadendrin and 3-O-acetylpadmatin. Now, as part of our ongoing study of the flavonoids of I. viscosa, we describe the underlying mechanisms of the antiinflammatory effects of 3,5,4′-trihydroxy-7-methoxy flavanone (7-O-methylaromadendrin, Fig. 1, 1 R = OH) and two other related substances identified as 3-acetoxy-5,4′-dihydroxy-7methoxyflavanone (3-acetyl-7-O-methylaromadendrin, 2 R = OCOCH 3) and 5,4′-dihydroxy-7-O-methoxyflavanone (sakuranetin 3 R = H). In order to delve deeper into their pharmacological activity, we have carried out several assays on

V. Hernández et al. / Life Sciences 81 (2007) 480–488

481

Germany); ethanol 96°, disodium monophosphate, potasium biphosphate, potassium chloride, calcium chloride, magnesium chloride and sodium acetate from Panreac (Barcelona, Spain); [γ-32P] ATP from New England Nuclear (Boston, MA, USA); Nonidet P 40 from USB (Cleveland, OH, USA); kit for enzyme immunoassay of prostaglandin (PG) E2 and secretory PLA2 activity determination by colorimetric assay were purchased from Cayman Chemical Company (Ann Arbor, MI, USA). The rest of the reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Fig. 1. Chemical structures of flavanones isolated from Inula viscosa.

the effects of the compounds on arachidonate metabolism and leukocyte mediator release, as well as on models of acute inflammation in mice. Material and methods Isolation and characterization of the anti-inflammatory flavanones Inula viscosa (L.) Aiton was collected in the La Albufera Natural Park, Sueca (Valencia, Spain) in November 1999, and a voucher specimen deposited at the herbarium of the Department of Pharmacology (University of Valencia, Spain). Dried flowering aerial parts (880 g) were macerated with acetone (3 × 15 L) at room temperature. The combined concentrated extract (94.7 g) was suspended in water and extracted successively with n-hexane, CH2Cl2 and EtOAc. The CH2Cl2-soluble fraction (40 g) was fractionated with MeOH, on a Sephadex LH-20 column (100 × 5 cm) collecting 200 mL portions, which were analysed by silica gel thin-layer chromatography and reunified to give fractions I–VI. Fraction V (7400 mg) was submitted to column chromatography on silica gel (50 × 3 cm) eluting with CH2Cl2–EtOAc mixtures (99:1 to 9:1, in portions of 100 mL) to yield the reunified fractions V1–V10. Fraction V2 eluted with CH2Cl2–EtOAc (99:1) was constituted by a single pure compound (1, 155 mg). Fraction V4 eluted CH2Cl2– EtOAc (98:2), after precipitation with CH2Cl2 gave compound 2 (92 mg). Compound 3 was obtained from fraction V6, eluted with CH2Cl2–EtOAc (96:4). Purity of compounds was assessed by thin layer chromatography and nuclear magnetic resonance (1H and 13C NMR), and their structure was spectroscopically confirmed in accordance with bibliographic data (Grande et al., 1985; Máñez et al., 1999). Compounds assayed were identified as 5,4′-dihydroxy-7-O-methoxyflavanone or sakuranetin (1), 3acetoxy-5,4′-dihydroxy-7-methoxyflavanone or 3-acethyl-7-Omethylaromadendrin (2) and 3,5,4′-trihydroxy-7-methoxyflavanone or 7-O-methylaromadendrin (3). Chemicals Tween 80 was purchased from Fluka Chemika-Biochemika (Buchs, Switzerland); acetone, dimethyl sulphoxide, dithiothreitol, ethyleneglycol tetraacetic acid (EGTA), HEPEs, leupeptin, methanol, phenylmethanesulphonyl fluoride and trifluoroacetic acid (HPLC grade) from Merck (Darmstadt,

Animals Groups of six Swiss female mice weighing 25–30 g were used. For in vitro experiments, polymorphonuclear leukocytes were obtained from Wistar rats weighing 250–300 g. All animals were fed on a standard diet ad libitum. Housing conditions and all in vivo experiments were approved by the institutional Ethics Committee of the Faculty of Pharmacy, University of Valencia (Spain), according to the guidelines established by the European Union on Animal Care (CEE Council 86/609). Cytotoxicity assay Cytotoxicity of the flavonoids was measured with a modified 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide (MTT) colorimetric assay (Mosmann, 1983). Rat peritoneal neutrophils or mouse macrophages were exposed to 5 μl of the products (100 μM, final concentration) for 30 min at 37 °C, and then incubated with MTT. After 3–4 h at 37 °C, the formazan product was solubilised in dimethylsulfoxide. Absorbance was measured at 490 nm using a Labsystems Multiskan MCC/340 plate reader. A decrease in absorbance indicated a decrease in cell viability. Inhibition of leukotriene B4 production from rat polymorphonuclear leukocytes Rat peritoneal leukocytes (95% viability, Trypan blue exclusion test) were collected 4 h after the i.p. injection of 5% glycogen in Dulbecco's phosphate-buffered saline (PBS) according to Safayhi et al. (1995). For 5-lipoxygenase (5LOX) product formation from endogenous arachidonic acid, leukocytes were stimulated at 37 °C for 5 min with the calcium ionophore A23187 (1.9 μM). The cells were incubated in the presence of the test compounds at a final concentration of 100 μM. The reaction was stopped with MeOH/1 N HCl (97:3). Compounds or the reference drug zileuton were added 5 min prior to initiation. Experiments were performed in triplicate. The 5-LOX products were quantified after the addition of prostaglandin B2 as an internal standard. Separation of products was performed by isocratic elution from LiChrospher C18column (250 × 4 mm, 5 μM) (Merck) with MeOH/water/ trifluoracetic acid (74:26:0.007). The detection wavelength was set at 280 nm in order to detect prostaglandin B2 and leukotriene. Percentage inhibition was computed by

482

V. Hernández et al. / Life Sciences 81 (2007) 480–488

Fig. 2. Effect of flavanones on leukotriene B4 production from endogenous arachidonic acid in calcium ionophore-stimulated rat peritoneal polymorphonuclear leukocytes. Concentrations assayed ranged from 5 to 100 μM. Data are expressed as means ± S.E.M. percentages of metabolite formation with respect to controls: n = 3–6.

comparing values in drug groups with the value of dimethyl sulfoxide controls. 5-Lipoxygenase assay on homogenized rat peritoneal polymorphonuclear leukocytes According to the method of Sailer et al. (1996), slightly modified, the cells were washed with phosphate buffered saline, polymorphonuclear leukocytes pellet was suspended in phosphate buffered 50 mM with ethylenediamaminetetraacetic acid (EDTA) 1 mM. Then the cells were lysed at 2 × 107 cells/ml by sonication (3 × 5 s, Branson Sonifier), operating on ice, and were subjected to sequential centrifugation at 10,000 g and 20,000 g for 10 and 30 min (4 °C). Aliquots from the supernatant corresponding to 1 × 107 cells were taken and suspended in phosphate buffered 50 mM with EDTA 1 mM. Then 10 μl of ATP 100 mM (final concentration 1 mM) and 10 μl of calcium 180 mM (final concentration 1.8 mM) were added and incubated in a bath at 37 °C. The compounds dissolved in dimethyl sulfoxide at final concentration ranging from 20 to 60 μM were incubated with the supernatants for 5 min. Finally, arachidonic acid (final concentration 20 μM) was added and the enzyme was allowed to act on the substrate for 5 min. Zileuton was used as the reference drug, at concentrations between 0.5 and 5 μM. All incubations including controls were carried out in the presence of 0.5% dimethyl sulfoxide. Experiments were performed in triplicate. Elastase release from human neutrophils/elastase activity assay Following Barret's protocol (1981), 1.25 × 106 human leukocytes were suspended in buffered Hanks' balanced solution salt containing calcium and magnesium, and were incubated with the test compounds at a final concentration of 100 μM for 5 min at 37 °C. Human neutrophils were stimulated by the addition of 12-O-tetradecanoylphorbol-13-acetate (TPA) (65 μg/ml). Fifteen minutes later the samples were taken out of the bath and placed on ice. After centrifugation, 200 μl aliquots

of supernatant were introduced in a 96-well microtiter plate. Five microliters of N-tertbutoxycarbonyl-L-alanin-p-nitrophenyl ester (Boc–Ala–OPhNO2) were added to each well and then incubated at 37 °C for 30 min. Finally the absorbance was measured at 414 nm. α1-Antitrypsin (90 μg/ml) was used as a positive control (Hernández et al., 2005). In order to evaluate the effect of the flavonoids on the activity of the enzyme released, we employed the method described above, but in this case, human polymorphonuclear cells were first stimulated with TPA (65 μg/ml) for 30 min and then the supernatant aliquots were incubated with test compounds at a final concentration of 100 μM for 15 min. Assay of myeloperoxidase from human neutrophils Cells (4 × 107) were suspended in Hanks' balanced solution salt and stimulated with TPA (65 μg/ml). After incubation at 37 °C for 30 min, the sample was centrifuged and enzyme activity in the supernatant was determined using a colorimetric method and a Labsystems Multiskan MCC/340 plate reader set to measure absorbance at 620 nm. Details of the method have been described earlier (Góngora et al., 2003). Prostaglandin E2 production in RAW 264.7 macrophages Macrophages RAW 264.7 were cultured as previously described (De León et al., 2003) in DMEM medium containing 2 mM L-glutamine (Gibco), 100 U/ml penicillin (Gibco), 100 μg/ Table 1 Effect of flavanones on 5-lipoxygenase total activity Compound

%LTB4 production a

Control Sakuranetin (20 μM) 7-O-Methylaromadendrin 3-Acetyl-7-O-methylaromadendrin

100 ± 3 42 ± 6 b b100 c N100 c

a Data are expressed as means ± S.E.M. n = 3 independent experiments, and every experiment was performed in duplicate. b P b 0.01 with respect to the control group (Dunnett's t-test). c Not significant.

V. Hernández et al. / Life Sciences 81 (2007) 480–488 Table 2 Effect of flavanones on elastase release and elastase activity Compound

Control Sakuranetin 7-O-Methylaromadendrin 3-Acetyl-7-Omethylaromadendrin a b c d

Release assay a

Activity assay b

Absorbance at 414 nm

Absorbance at 414 nm

0.096 ± 0.008 0.038 ± 0.004 c 0.048 ± 0.004 c 0.074 ± 0.006 d

0.095 ± 0.007 0.066 ± 0.002 c 0.108 ± 0.005 d –

Assay of protease release from neutrophils n = 4. Assay of protease activity of the released protease from neutrophils n = 4. P b 0.01 with respect to the control group (Dunnett's t-test). Not significant.

ml streptomycin and 10% fetal bovine serum (Gibco). Cells were removed from the tissue culture flask using a cell scraper and resuspended at a final concentration of 1 × 106 cells/ml. Macrophages at 1 × 106 cells/ml were coincubated in 96-well culture plate (200 μl) with 1 μg/ml of lipopolysaccharide (LPS) at 37 °C for 24 h in the presence of test compounds (final concentration: 25, 10 and 1 μM) or vehicle. The prostaglandin E2 production was determined in culture supernatant by a specific enzyme immunoassay kit (Prostaglandin E2 Express EIA kit) from Cayman Chemical, employed according to the manufacturer's instructions. Secretory phospholipase A2 assay Non-enzymatic controls, 100% enzymatic activity controls and the test compounds were assayed by triplicate in a 96-well microtiter plate, following the manufacturer's instructions. Test compounds and the reference drug 4-(4-octadecylphenyl)-4oxobutenoic acid were tested at a final concentration of 100 μM. Phospholipase A2 was tested at the final concentration of 4.4 ng/ ml, after an initial incubation with the inhibitors. The incubation was for 30 min in the case of bee venom phospholipase A2 and for 2-3 h in the case of phospholipase A2 from Naja mossambica venom. Purification of mouse brain protein kinase C Protein kinase C was partially purified from mouse brain by DEAE-cellulose (DE52) column chromatography (DíazGuerra and Boscá, 1990). In brief, mouse brain was placed

483

in a homogenization buffer at 0–4°C and homogenized with a Polytron, and the medium kept on ice, adding 0.1% Nonidet P 40. The homogenized brain was centrifuged and the supernatant fraction was applied to a DE52 column and eluted with a 0–300 mM linear salt gradient in the equilibrated buffer. Protein concentration was determined by the method of Bradford. The enzyme was eluted with 3 ml of 150 mM sodium chloride. Protein kinase C assay Protein kinase C activity was determined at 37 °C in a 180 μl volume reaction mixture containing: 80 μl of the protein kinase C solution, 80 μl of the reaction mixture containing ATP 50 μM (0.1–0.05 μCi), magnesium chloride 10 μM, dithiothreitol, myelin basic protein 50 μg/ml, HEPES 20 mM, phosphatidylserine 10 μg/ml and TPA 20 nM. The reaction began with the addition of 20 μl of the mixture (γ-32P)-ATP (3000 Ci/mol)/ magnesium chloride) (Díaz-Guerra and Boscá, 1990). For basal phosphorylations the composition of the medium was similar, except that it did not contain phosphatidylserine or TPA, but it had ethyleneglycoltetraacetic acid. Flavonoids (final concentration 1 μM) were dissolved in dimethyl sulfoxide and added immediately prior to ATP addition. Dimethyl sulfoxide concentration in the assay mixture did not exceed 0.25%. After 5 min the reaction was stopped by the addition of trichloroacetic acid and the phosphorylated protein was collected on nitrocellulose filters. Radioactivity was determined by liquid scintillation counting. Phospholipase A2-induced paw oedema in mouse Details of the method (Neves et al., 1993) were described previously (Giner-Larza et al., 2001). Phospholipase A2 from N. mossambica (1.4 U in 25 μl sterile saline) was injected s.c. into the right hind paw. The test compounds (50 mg/kg) and the reference drug cyproheptadine (5 mg/kg) were injected i.p. 30 min before phospholipase A2. Products and reference drug were dissolved in tween 80-ethanol-saline (1:1:10). The right and left paw volumes were measured on a plethysmometer (Ugo Basile) 30 and 60 min after inflammation induction. The 50% inhibitory dose against phospholipase A2induced paw oedema was calculated by administering the compounds at different doses ranging from 80 to 5 mg/kg.

Fig. 3. Effect of flavanones (final concentration 25 μM) on PGE2 production in RAW 264.7 macrophages stimulated by LPS. Data are expressed as means ± S.E.M. n = 3.

484

V. Hernández et al. / Life Sciences 81 (2007) 480–488

Fig. 4. Effect of flavanones and 4-(4-octadecylphenyl)-4-oxobutenoic acid (OBBA) at final concentration of 100 μM on phospholipase A2 activity. Results of diheptanoyl-1,2-dithio-phosphatidylcholine cleavage by phospholipase A2 are expressed in absorbance of coloured compound 5,5′dithio-bis-(2-nitrobenzoic acid) at 414 nm every 2 min. Data are expressed as means ± S.E.M. n = 3. ⁎P b 0.05; ⁎⁎P b 0.01 (Dunnett's t-test).

the concentration or dose/effect regression lines obtained from at least four values.

TPA-induced mouse ear oedema Oedema was induced by topical application of 2.5 μg of TPA per ear. Test compounds and the standard drug indomethacin were applied topically (0.5 mg/ear), simultaneously with TPA (Giner et al., 2000). Flavonoids and indomethacin were dissolved in acetone. The 50% inhibitory dose was determined by applying the flavonoids at four different doses, ranging from 600 to 75 μg/ear. Ear thickness was measured before TPA application and 4 h after, and oedema is expressed as the increase in thickness due to inflammation. Statistical analysis Data are expressed as means ± S.E.M. Statistical evaluation was carried out by one-way analysis of variance followed by Dunnett's t-test for multiple comparisons. Statistical significance is expressed as ⁎P b 0.05; ⁎⁎P b 0.01. The inhibitory concentrations or inhibitory doses 50% were calculated from Table 3 Anti-inflammatory effect of the flavanones on acute TPA-induced ear oedema Product

▵Ear thickness a (μm ± S.E.M.)

I.R. b

ID50 (mg/ear) c

Control Sakuranetin 7-O-Methylaromadendrin 3-Acetyl-7-Omethylaromadendrin Indomethacin

264 ± 12 96 ± 8 d 95 ± 12 d 78 ± 17 d

– 64 64 70

– 0.205 e 0.373 f 0.185 g

76 ± 12 d

71

0.125 h

a Ear thickness expressed as the mean difference between thickness before and after challenge ± S.E.M. n = 6. b Inhibition ratio percentage at 0.5 mg/ear with respect to the control treated only with TPA. c 50% inhibitory dose. d P b 0.01 with respect to the control group (Dunnett's t-test). e Coefficient of determination (r2) for the linear regression = 0.9868, P = 0.0066 (ANOVA test, significant). f 2 r = 0.9522, P = 0.0242. g 2 r = 0.9655, P = 0.0174. h 2 r = 0.9995, P = 0.0231.

Results Effect on LTB4 generation by stimulated rat peritoneal leukocytes None of the tested compounds had cytotoxicity at 100 μM in the MTT test (viability N 95%). The production of leukotriene B4 in vitro by rat peritoneal leukocytes was estimated using a high pressure liquid chromatography method. In these conditions, zileuton at 1 μM inhibited the LTB4 production by 50%. Fig. 2 shows the concentration-dependent inhibitory effect of the flavanones on LTB4 production. The effect of apigenin, the flavone used as a reference, is also shown. Sakuranetin turned out to be the most potent compound, with an IC50 of 9 μM (coefficient of determination (r2) = 0.9968, P = 0.0351, considered significant). 3-Acetyl-7-O-methylaromadendrin gave an IC50 = 15 μM (r2 = 0.9485, P = 0.0050, considered extremely significant). 7-O-methylaromadendrin was less active, with an IC50 of 62 μM (r2 = 0.9984, P = 0.0007, considered extremely significant). The IC50 for apigenin was 14 μM (r2 = 0.9755, P = 0.053, considered significant). Effect on 5-lipoxygenase activity in homogenized rat peritoneal polymorphonuclear leukocytes Percentages of inhibition are shown in Table 1. The inhibition of 5-LOX total activity is expressed as percentages with respect to the control, which includes the 6-all-trans isomers of LTB4, LTB4 and 5(S)-hydroxyeicosatetraenoic acid. According to the results of this assay, sakuranetin directly inhibited the activity of 5-LOX. At 20 μM the flavanone decreased the production of LTB4 isomers (58% inhibition). In the same experimental conditions, the reference drug zileuton, a specific inhibitor of 5-LOX, at 1 μM reduced the enzyme activity by 84%. The other two flavanones did not affect the 5-LOX activity (Table 1).

V. Hernández et al. / Life Sciences 81 (2007) 480–488

485

Fig. 5. Effect of flavanones and cyproheptadine (50 and 5 mg/kg, i.p., 30 min before) on hind paw oedema caused in the mouse by phospholipase A2. Data are expressed as means ± S.E.M. n = 5–6 animals. ⁎⁎P b 0.01 (Dunnett's t-test).

Effect on elastase activity and elastase release from human neutrophils Results are expressed in absorbance of coloured compound pnitrophenol at 414 nm formed by elastase from the cleavage of Boc-Ala-OPhNO2. Sakuranetin and 7-O-methylaromadendrin at 100 μM, reduced the elastase release 60% and 50%, respectively. This effect seemed to be purely on the release in the case of 7-Omethylaromadendrin whereas sakuranetin, besides having reduced the release, exerted a moderate direct inhibition on the enzyme activity (30%) (Table 2). Effect on myeloperoxidase activity from human neutrophils None of the flavanones tested at 100 μM inhibited the MPO activity. Effect on prostaglandin E2 production in RAW 264.7 murine macrophages The tested compounds showed no cytotoxicity at 25 μM in the MTT test (viability N 95%). None of the flavanones tested inhibited the PGE2 production in RAW 264.7 mouse macrophages (Fig. 3). Effect of flavonoids on secretory phospholipase A2 activity Only 7-O-methylaromadendrin significantly inhibited the enzyme activity 2 min after the addition of substrate (31% inhibition). However, this inhibitory effect decreased later (Fig. 4). Effect on protein kinase C activity None of the flavanones tested at 1 μM inhibited the PKC activity. Effect on mouse ear oedema induced by TPA All the flavanones produced a significant dose-dependent decrease in TPA-induced oedema, in the same range as that of topically-administered indomethacin (Table 3).

Effect on phospholipase A2-induced mouse paw oedema All the flavonoids (50 mg/kg, i.p.), reduced paw oedema 60 min after enzyme injection (Fig. 5). Sakuranetin and 7-Omethylaromadendrin showed the strongest effect (76% and 69% inhibition, respectively), while 3-acethyl-7-O-methylaromadendrin reduced the inflammatory lesion by about 50% at 60 min. The dose-response relationship study of sakuranetin and 7-O-methylaromadendrin yielded for sakuranetin an ID50 of 18 mg/kg (r2 = 0.9996, P = 0.0128, considered significant) and for 7-O-methylaromadendrin an ID50 of 8 mg /kg (r2 = 0.9279, P = 0.0367, considered significant) 1 h after PLA2 injection. Discussion We present here the pharmacological activity of three flavanoids isolated from the dichloromethanic extract of I. viscosa, which has already been demonstrated to contain the main compounds that are implicated in the anti-inflammatory activity of this medicinal plant (Máñez et al., 1999; Hernández et al., 2001, 2005). The three flavanones are characterized not only by the absence of a double bond between C2-C3 in the C ring, but also by the constant presence of a 5-hydroxy-7-methoxy substitution pattern at the A ring and a single 4'-hydroxyl group at ring B. Inflammation is the immunological mechanism by which the body fights infection or injury from bacteria, viruses, and other pathogens. Excessive or inadequate activation of the system can have serious effects. Since aspirin and other non-steroidal antiinflammatory drugs were found to inhibit COX, which produces prostaglandins, the biosynthetic cascade of arachidonic acid (AA) has been the subject of intense research. This is due to the fact that the AA previously liberated from membrane phospholipids can be metabolized via the COX pathway into PGs and thromboxane A2, or via the LOX pathway to hydroperoxyeicosatetranoic acids (HpETEs), HETEs, and LTs. These molecules are intimately involved in inflammation, asthma, and allergies, as well as in many other physiological and pathological processes. When we examined the inhibitory effects of sakuranetin, 7O-methylaromadendrin, and 3-acetyl-7-O-methylaromadendrin on arachidonate metabolism in rat peritoneal leukocytes

486

V. Hernández et al. / Life Sciences 81 (2007) 480–488

stimulated with the cation ionophore A23187, we observed that all three compounds greatly reduced the production of LTB4. Similar behaviour has been reported for methylated flavonoid aglycones such as cirsiliol and sideritoflavone (Alcaraz and Ferrándiz, 1987; Laughton et al., 1991) and for flavanones such as pinocembrin (Sala et al., 2003). These results are somewhat of a surprise because the chemical structure of these compounds is quite different from that of the typical 5-LOX inhibitors, mainly flavonols and prenylated flavones (Chi et al., 2001). Classically, the anti-inflammatory activity of flavonoids has been attributed to their antioxidant activity (Middleton et al., 2000). However, recent studies have speculated that the hydrogen-donating antioxidant activity of these inhibitors is unlikely to be the sole explanation for their effects at the cellular level (Rice-Evans, 2001). Sakuranetin, 7O-methylaromadendrin, and 3-acetyl-7-O-methylaromadendrin were found to be effective inhibitors of LTB4 synthesis in intact cells but not so in studies with leukocyte homogenates, in which it is supposed that 5-LOX acts without interference of other cellular factors. In fact, sakuranetin was found to be the only compound that inhibited the LTB4 synthesis in leukocyte homogenates at 20 μM, a higher concentration than that observed in intact cells. The result indicated that this compound may inhibit the production of LTB4 indeed by directly inhibiting 5-LOX, but also by interacting with other proteins, such perhaps the cofactor 5-lipoxygenase activating protein (FLAP), or different protein kinases (such as protein kinase A, protein kinase C, protein tyrosine kinases and by mitogen-activated protein kinase (MAPK) kinase) that can modulate the 5-LOX activity (Peters-Golden and Brock, 2003). We must note that other flavonoids have been reported as inhibitors of protein kinase activity (Ferriola et al., 1989; Manthey, 2000; Máñez and Recio, 2002). Nevertheless, we have observed that our flavanones are inactive against the PKC isolated from mouse brain at 1 μM. This result is in accordance with that obtained by Ferriola et al. (1989), who observed that PKC activity was inhibited by different flavones and flavonols, while the flavanone hesperetin was inactive. More studies are thus necessary in order to examine the possible influence of protein phosphorylation in the reduction of LTB4 synthesis by sakuranetin in intact cells and homogenates. The key regulatory enzyme in the production of PGs, a group of powerful proinflammatory signalling molecules, is COX, which catalyzes the conversion of AA to PGG2 and PGH2 (Dubois et al., 1998). COX occurs in two isoforms: one is the constitutive or COX-1 and the other is the inducible or COX-2 (Patrignani et al., 2005; Grosser, 2006). Several studies have shown that the anti-inflammatory effect of flavonoids is due to the inhibition of COX, mostly COX-1 (Kim et al., 2004). The active flavonoids include flavones, flavonols, C-8 prenylated flavonoids, and biflavones. However, in one study that used a subchronic skin inflammation model, the compound wogonin was found to lower the mRNA levels of COX-2 (Chi et al., 2003). Takano-Ishikawa et al. (2006) recently reported the structureactivity relationship of the inhibitory effect of flavonoids on lipopolysaccharide (LPS)-induced prostaglandin production in macrophages. These results suggest that flavones and flavonols

are clearly more potent than flavanones. The flavanone eriodictylol at 40 μM inhibited PGE2 production with weak inhibition of COX-2 expression. It seems that this inhibitory effect is derived from another mechanism. Recently, Zhang et al. (2006), in an in vitro bioassay-guided analysis of the MeOH extract of Populus davidiana (Salicaceae), have shown inhibitory activity against COX-1 and COX-2. Continuous phytochemical study of this extract led the isolation of sakuranetin, among other flavonoids. This substance showed only moderate inhibition against COX-1 without effect on COX-2 activity. In our research, this result has been observed again: the flavanone and the two dihydroflavonols assayed at 25 μM had no inhibitory effect on LPS-induced PGE2 production in mouse macrophages RAW 246.7. Neutrophils are the most abundant bodyguard cells. In the tissue, they act as phagocytic cells and also release a variety of reactive oxygen species and proteases. Neutrophil elastase, for example, is one of the proteolytic enzymes capable of degrading fibrous elastin, an important extracellular matrix protein with a mechanical function in the lungs, arteries, skin, and ligaments. In addition, elastase cleaves other matrix proteins including fibronectin, laminin, and cartilage proteolgycans, along with other proteins with important biological functions, including collagen type I. These defense mechanisms paradoxically lead to inflammation and may be central to the pathogenesis of a range of chronic diseases such as asthma, emphysema, fibrosis, and rheumatoid arthritis (Barret et al., 1998). A better understanding of the influences on human neutrophil elastase activity is thus essential. In this context, several reports have demonstrated the effect of a variety of naturally occurring phenolic compounds on this activity (Melzig et al., 2001; Sartor, 2002). Meloni et al. (1995) observed the weak inhibition of human neutrophil elastase by 3′-hydroxyfarrerol, which acts through a reversible, noncompetitive inhibition mode (Middleton et al. 2000). Sartor et al. (2002) suggested that either the galloyl moiety or a hydroxyl group at C3 and three hydroxyl groups on the B ring – one at C4′ and a C2-C3 double bond – play a crucial role in the inhibition of pure elastase from human leukocytes. Among the compounds that have been studied, taxifolin, a flavanone with an O-catechol group in the B ring and three hydroxyl groups at C3, C5, and C7, showed no activity at 100 μM in the assay used. In contrast, naringenin, a flavanone with hydroxyl groups at C4′, C5, and C7, presented a very low inhibitory potency in the same assay. This finding indicates that the double bond between C2 and C3 in the chromane system enhances the inhibitory activity against neutrophil elastase (Melzig et al., 2001). We have found no reports on the ability of flavanones to inhibit enzyme release by TPA from human neutrophils; this is thus the first time that the effect of this type of flavonoid on this particular phenomenon has been reported. From our results, it seems that the chemical features previously highlighted by various authors have no influence on the process of intracellular exocytosis, at least not in this case. The most active flavanone was sakuranetin, which has neither a C2-C3 double bond nor a hydroxyl group at C3. Moreover, the cellular exocytosis is enhanced by sakuranetin's moderate anti-elastase activity. In contrast, the dihydroflavonol

V. Hernández et al. / Life Sciences 81 (2007) 480–488

7-O-methylaromadendrin inhibited only elastase release, albeit in a range similar to that of sakuranetin. More studies on the effects of flavanones on elastase release under different conditions are needed to understand this special behaviour more fully. Whatever the mechanisms involved, however, it is likely that this effect contributes to the anti-inflammatory activity of these compounds. While the effect of the dihydroflavonol 7-O-methylaromadendrin on the oedema induced by a single application of TPA has previously been published (Máñez et al., 1999), this is the first report on the in vivo anti-inflammatory activities of sakuranetin and 3-acetyl-7-O-methylaromadendrin. All three test compounds were assayed to determine whether they exhibited any dosedependent anti-inflammatory effects on either the ear oedema induced by TPA or the paw oedema induced by PLA2. The potency of all three compounds in the TPA test was lower than that of other previously studied flavanones such as pinocembrin (Sala et al., 2003). The most potent compound of the three was the acetylated dihydroflavanol, a fact that can be explained in pharmacokinetic terms since in this model of inflammation the site of action (ear) coincides with the site of TPA application. Many groups of potential anti-inflammatory agents, including H1histamine antagonists, corticoids, and inhibitors of PLA2, COX, or 5-LOX, are effective against TPA-induced ear oedema (De Young et al., 1989). There are thus several different options for determining the mechanism of action of a given active compound. One of these mechanisms involves examining the activity of protein kinase C, a TPA-activated enzyme implicated in the regulation of a great number of inflammatory processes (Nishizuka, 1995). In the assay we carried out, the possibility that the test compounds could interfere with PKC activation was ruled out because when they were incubated with the isolated enzyme in vitro (final concentration 1 μM), none of the compounds inhibited PKC activity. The oedema induced by subcutaneous injection of PLA2 is produced by several types of inflammatory mediators, with mast cell degranulation apparently playing a major role (Cirino et al., 1989; Hartman et al., 1991). The three flavonoids were found to diminish the oedema induced by this enzyme, probably because they prevent the release of histamine or serotonin. This hypothesis is supported by the fact that most flavonoids inhibit histamine release from human mast cells (Middleton et al., 2000). In addition, for 7-O-methylaromadendrin, the most potent substance in this test, the inhibition of PLA2 observed in vitro may play a role in this effect as it was the only compound to show activity against this enzyme. Furthermore, both 7-Omethylaromadendrin and sakuranetin at 100 μM act through exocytosis to inhibit elastase release. This finding may explain why these two compounds showed the highest degree of activity in the in vivo model of paw oedema induced by PLA2. In addition, several studies have demonstrated that flavonoids can reduce the levels of intracellular calcium and induce the phosphorylation of some proteins through the activity of an isoenzyme of PKC independent of calcium or phorbol ester (Wang et al., 1999). This effect leads to secretion regulation. Various studies on the structure-pharmacological activity relationships have established that the most important features

487

of active flavonoids are the presence of a O-3',4'dihydroxy structure in the B ring and a free 3-hydroxy group in the C-ring (Pietta, 2000; Heijnen et al., 2002). In addition, the spatial conformation and the lipophilicity of the molecule must be taken into account (Heijnen et al., 2002). In the case of flavanones, Pouget et al. (2001) reported that the substitution by a methoxy group at position 7 and/or 5 increases the antiproliferative effect on MCF-7 cells. The 4-keto functionality appears to be essential for this activity. In contrast, flavanone, 2′-hydroxyflavanone, 4′-hydroxyflavanone, and 6′-hydroxyflavanone showed the most significant dose-dependent inhibition of TPA-induced proliferative effects on NIH3T3 cells (Ko et al., 2002). The authors proposed that the blocking of TPAinduced intracellular signalling responses might be involved in the antipromotive mechanisms of flavanones. Chen et al. (2004) have reported that the 4′-hydroxy structure in the B ring and the 5,7-metahydroxy arrangement in the A ring both contribute to the anti-inflammatory activity through the regulation of cell adhesion processes such as the inhibition of ICAM-1 expression. These features are all present in the flavanoids isolated from I. viscosa and may justify their anti-inflammatory activity, although more studies are necessary, especially on the effect of these compounds on proinflammatory mediator expression. In conclusion, of the three compounds studied, sakuranetin was the most active in the in vitro models, inhibiting the production of LTB4, acting directly on the 5-LOX enzyme, and regulating secretory processes such as elastase release. A great many activities of the flavonoids have traditionally been associated with their antioxidative properties. Nevertheless, they can also act through different mechanisms, as seen from our results with sakuranetin, which lacks the structural features necessary for antioxidative activity. One possible mechanism that has been previously reported is that of kinase inhibition, which, while not demonstrated in the case of PKC, may be possible for other enzymes with analogous functions. In addition, our results indicate a possible non-redox inhibition of lipoxygenases, as well as a blockage of some proteins implicated in exocytotic mechanisms. Acknowledgements The authors wish to thank the Spanish government for its financial support (DGESIC, PM98-0206). We are indebted to the Centre de Transfusions de la Comunitat Valenciana (València, Spain) for its generous supply of human blood. The authors thank Dr. L. Boscá for his expert technical assistance in the PKC assay. References Alcaraz, M.J., Ferrándiz, M.L., 1987. Modification of the arachidonic metabolism by flavonoids. Journal of Ethnopharmacology 21, 209–229. Barret, A.J., 1981. Leukocyte elastase. Methods in Enzymology 80, 581–588. Barret, A.J., Rawlings, N.D., Woessner, F.F., 1998. Handbook of Proteolytic Enzymes. Academic Press, London, p. 54. Cirino, G., Peers, S.H., Wallace, J.L., Flower, R.J., 1989. A study of phospholipase A2-induced oedema in rat paw. European Journal of Pharmacology 166, 505–510.

488

V. Hernández et al. / Life Sciences 81 (2007) 480–488

Chen, C.-C., Chow, M.-P., Huang, W.-C., Lin, Y.-C., Chang, Y.-Y., 2004. Flavonoids inhibit tumor necrosis factor -α-induced up-regulation of intracellular adhesion molecule-1 (ICAM-1) in respiratory epithelial cells through activator protein-1 and nuclear factor-κB: Structure-activity relationships. Molecular Pharmacology 66, 683–693. Chi, Y.S., Jong, H., Son, H.K., Chang, H.W., Kang, S.S., Kim, H.P., 2001. Effects of naturally occurring prenylated flavonoids on arachidonic acid metabolizing enzymes: cyclooxygenases and lipoxygenases. Biochemical Pharmacology 62, 1185–1191. Chi, Y.S., Lim, H., Park, H., Kim, H.P., 2003. Effect of wogonin, a plant flavone from Scutellaria radix, on skin inflammation: in vivo regulation of inflammation associated gene-expression. Biochemical Pharmacology, 66, 1271–1278. De León, E.J., Alcaraz, M.J., Domínguez, J.N., Charris, J., Terencio, M.C., 2003. 1-(2,3,4-trimethoxyphenyl)-3-(3(2-chloroquinolinyl))-2-propen-1one, a chalcone derivative with analgesic, anti-inflammatory and immunomodulatory properties. Inflammation Research 52, 246–257. De Young, L.M., Kheifets, J.B., Ballaron, S.J., Young, J.M., 1989. Edema and cell infiltration in the phorbol ester-treated mouse ear are temporally separate and can be differentially modulated by pharmacologic agents. Agents and Actions 26, 335–341. Díaz-Guerra, M.J., Boscá, L., 1990. Lack of translocation of protein kinase C from the cytosol to the membranes in vasopressin-stimulated hepatocytes. Biochemical Journal 269, 163–168. Dubois, R.N., Abramson, S.B., Crofford, L., Gupta, R.A., Simon, L.S., Van De Putte, L.B., Lipsky, P.E., 1998. Cyclooxygenase in biology and disease. FASEB Journal 12, 1063–1073. Ferriola, P.C., Cody, V., Middleton, E., 1989. Protein kinase C inhibition by plant flavonoids: kinetic mechanism and structure-activity relationships. Biochemical Pharmacology 38, 1617–1624. Giner, R.M., Villalba, M.L., Recio, M.C., Máñez, S., Cerdá-Nicolás, M., Ríos, J.L., 2000. Anti-inflammatory glycoterpenoids from Scrophularia auriculata. European Journal of Pharmacology 389, 243–252. Giner-Larza, E.V., Máñez, S., Recio, M.C., Giner, R.M., Prieto, J.M., CerdáNicolás, M., Ríos, J.L., 2001. Oleanonic acid, a 3-oxotriterpene from Pistacia, inhibits leukotriene synthesis and has anti-inflammatory activity. European Journal of Pharmacology 428, 137–143. Góngora, L., Máñez, S., Giner, R.M., Recio, M.C., Schinella, G., Ríos, J.L., 2003. Inhibition of xanthine oxidase by phenolic conjugates of methylated quinic acid. Planta Medica 69, 396–401. Grande, M., Piera, F., Cuenca, A., Torrres, P., Bellido, I.S., 1985. Flavonoids from Inula viscosa. Planta Medica 51, 414–419. Grosser, T., 2006. Pharmacology of selective COX-2 inhibition. Journal of Thrombosis and Haemostasis 96, 393–400. Hartman, D.A., Tomchek, L.A., Lugay, J.R., Lewin, A.C., Chau, T.T., Carlson, R.P., 1991. Comparison of antiinflammatory and antiallergic drugs in the melittin-and D49 PLA2-induced mouse paw edema models. Agents and Actions 34, 84–88. Heijnen, C.G.M., Haenen, G.R.M.M., Oostveen, R.M., Stalpers, E.M., Bast, A., 2002. Protection of flavonoids against lipid peroxidation: the structure activity relationship revisited. Free Radical Research 36, 575–581. Hernández, V., Recio, M.C., Máñez, S., Prieto, J.M., Giner, R.M., Ríos, J.L., 2001. A mechanistic approach to the in vivo anti-inflammatory activity of sesquiterpenoid compounds isolated from Inula viscosa. Planta Medica 67, 726–731. Hernández, V., Recio, M.C., Máñez, S., Giner, R.M., Ríos, J.L., 2005. Antiinflammatory profile of dehydrocostic acid, a novel sesquiterpene acid with a pharmacophoric conjugated diene. European Journal of Pharmaceutical Sciences 26, 162–169. Kim, H.P., Son, K.H., Hyeun, W.C., Kang, S.S., 2004. Anti-inflammatory plant flavonoids and cellular action mechanisms. Journal of Pharmacological Sciences 96, 229–245. Ko, C.H., Shen, S.C., Lin, H.Y., Hou, W.C., Lee, W.R., Yang, L.L., Chen, Y.C., 2002. Flavanones structure-related inhibition on TPA-induced tumor promotion through suppression of extracellular signal-regulated protein kinases: involvement of prostaglandin E2 in anti-promotive process. Journal of Cellular Physiology 193, 93–102.

Laughton, M.J., Evans, P.J., Moroney, M.A., Hoult, J.R., Halliwell, B., 1991. Inhibition of mammalian 5-lipoxygenase and cyclooxygenase by flavonoids and phenolic dietary additives. Relationship to antioxidant activity and to iron ion-reducing ability. Biochemical Pharmacology 42, 1673–1681. Manthey, J.A., 2000. Biological properties of flavonoids pertaining to inflammation. Microcirculation 7, S29–S34. Máñez, S., Recio, M.C., 2002. Modulation of protein phosphorylation by natural products. In: Atta-Ur-Rahman (Ed.), Studies in Natural Products Chemistry: Bioactive Natural Products27 (part H). Elsevier Science Publishers, Amsterdam, pp. 819–890. Máñez, S., Recio, M.C., Gil, I., Gómez, C., Giner, R.M., Waterman, P.G., Ríos, J.L., 1999. A glycosyl analogue of diacylglycerol and other antiinflammatory constituents from Inula viscosa. Journal of Natural Products 62, 601–604. Meloni, F., Ballabio, P., Gorrini, M., De Amici, M., Marena, C., Malandrino, S., Luisetti, M., 1995. Effects of 3′-hydroxyfarrerol (IdB 1031), a novel flavonoid agent, on phagocyte products. Inflammation 19, 689–699. Melzig, M.F., Löser, B., Ciesielski, S., 2001. Inhibition of neutrophil elastase activity by phenolic compounds from plants. Pharmazie 56, 967–969. Middleton Jr., E., Kandaswami, C., Theoharides, T.C., 2000. The effects of plant flavonoids on mammalian cells: implications for inflammation, heart disease, and cancer. Pharmacological Reviews 52, 673–751. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: applications to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55–63. Neves, P.C.A., Neves, M.C.A., Bella Cruz, A., Sant' Ana, A.E.G., Yunes, R.A., Calixto, J.B., 1993. Differential effects of Mandevilla velunita compounds on paw oedema induced by phospholipase A2 and phospholipase C. European Journal of Pharmacology 243, 213–219. Nishizuka, Y., 1995. Protein kinase and lipid signalling for sustained cellular responses. FASEB Journal 9, 489–496. Pietta, P.G., 2000. Flavonoids as antioxidants. Journal of Natural Products 63, 1035–1042. Patrignani, P., Tacconelli, S., Sciulli, M.C., Capone, M.L., 2005. New insights into COX-2 biology and inhibition. Brain Research Reviews 48, 352–359. Pouget, C., Lauthier, F., Simon, A., Fagnere, C., Basly, J.-P., Delaye, C., Chulia, A.-J., 2001. Flavonoids: structural requirements for antiproliferative activity on breast cancer cells. Bioorganic and Medicinal Chemistry Letters 11, 3095–3097. Peters-Golden, M., Brock, T.G., 2003. 5-Lipoxygenase and FLAP. Prostaglandins, Leukotrienes and Essential Fatty Acids 69, 99–109. Rice-Evans, C., 2001. Flavonoid antioxidants. Current Medicinal Chemistry 8, 797–807. Safayhi, H., Sailer, S., Havens, M., 1995. Mechanism of 5-lipoxygenase inhibition by acetil-11-keto-β-boswellic acid. Molecular Pharmacology 47, 1212–1216. Sailer, E.R., Subramanian, L.R., Rall, B., Hoernlein, R.F., Ammon, H.P., Safayhi, H., 1996. Acetyl-11-keto-beta-boswellic acid (AKBA): structure requirements for binding and 5-lipoxygenase inhibitory activity. British Journal of Pharmacology 117, 615–618. Sala, A., Recio, M.C., Schinella, G., Máñez, S., Giner, R.M., Cerdá-Nicolás, M., Ríos, J.L., 2003. Assessment of the anti-inflammatory activity and scavenger activity of tiliroside. European Journal of Pharmacology 461, 53–61. Sartor, L., Pezzato, E., Dell'Aica, I., Caniato, R., Biggin, S., Garbisa, S., 2002. Inhibition of matrix-proteases by polyphenols: chemical insights for antiinflammatory and anti-invasion drug design. Biochemical Pharmacology 64, 229–237. Takano-Ishikawa, Y., Goto, M., Yamaki, K., 2006. Structure-activity relations of inhibitory effects of various flavonoids on lipopolysaccharide-induced prostaglandin E2 production in rat peritoneal macrophages: comparison between subclasses of flavonoids. Phytomedicine 3, 310–317. Wang, L., Correia, I., Basu, S., Theoharides, T.C., 1999. Calcium and phorbol ester effect on the mast cell phosphoprotein induced by cromolyn. European Journal of Pharmacology 371, 241–249. Zhang, X.-F., Hung, T.Y., Phuong, T.P., Ngoa, T.M., Min, B.-S., Song, K.-S., Seong, Y.H., Bae, K., 2006. Anti-inflmamatory activity of flavonoids from Populus davidiana. Archives of Pharmacal Research 29, 1102–1108.