Pergamon 0306-3623(94)00242-8

Gen. Pharmac. Vol. 26, No. 4, pp. 815-819, 1995 Copyright © 1995 Elsevier Science Ltd

Printed in Great Britain. All rights reserved 0306-3623/95 $9.50 + 0.00

The Activity of Flavonoids Extracted from Tanacetum microphyllum DC. (Compositae)

on Soybean Lipoxygenase and Prostaglandin Synthetase

M A R I A JOSE A B A D , P A U L I N A B E R M E J O and A N G E L V I L L A R Department of Pharmacology, Faculty of Pharmacy, University Complutense, 28040 Madrid, Spain

(Received 29 July 1994)

A~traet--l . The in vitro effects of centaureidin and 5,Y-dihydroxy-4'-methoxy-7-carbomethoxyflavonol (Fig. 1), two anti-inflammatory flavonoids extracted from Tanacetum microphyllum DC., have been examined on both cyclooxygenase and lipoxygenase activity.

2. These flavonoids produced an inhibition of soybean lipoxygenase activity in a dose-dependent manner, with IC50 values (20 and 29 #M respectively) similar to the reference drug.

3. The IC50 values for the in vitro inhibition of cyclooxygenase activity by these flavonoids, were higher than those that produced lipoxygenase activity (318 and 60#M respectively).

4. These results suggest that the anti-inflammatory activity of our flavonoids may, at least in part, be due to the inhibition of leukotriene synthesis.

5. This is the first report of the biological activity in vitro of these compounds.

Key Words: Tanacetum microphyllum DC., flavonoids, anti-inflammatory, in vitro studies, soybean lipoxygenase, prostaglandin synthetase

INTRODUCTION

Plants belonging to the genus Tanacetum U (Com- positae) have been used for medicinal purposes for many centuries, showing several different biological activities. One of these species, Tanacetum parthenium

(L.) Schultz Bip., known by the common name "feverfew", has been used in folk medicine since the Middle Ages in the treatment of migraine, asthma, rheumatism and gynecological problems (Berry, 1984). Since the publication of two double-blind placebo controlled clinical trials in England, the interest in "feverfew" has spectacularly increased, and the drug has again become very popular in the treatment of migraine and arthritis (Johnson et al., 1985; Murphy et al., 1988). The compound regarded as mainly responsible for the activity is the ses- quiterpene lactone parthenolide (Groenewegen and Heptinstail, 1990).

In Spanish traditional medicine, Tanacetum micro-

phyllum DC., an endemic species of the Iberian Peninsula, is used for the treatment of inflammation and rheumatism, and as a cytoprotective agent. Pre-

viously it has been shown that extracts of this Spanish herb exert anti-inflammatory and gastroprotective actions in the rat (Abad et al., 1991). In search of the active principles contained in Tanacetum microphyl-

lure, we have investigated the dichloromethane ex- tract, the most active extract, and we have isolated two flavonoids, centaureidin and 5,Y-dihydroxy-4'- methoxy-7-carbomethoxyflavonol (Fig. 1).

The in vivo anti-inflammatory activity of these compounds was reported recently (Abad et al., 1993), and this may be relevant to the beneficial effects of the herb in inflammatory conditions. In our continuing study aimed at relating the traditional use of this plant to the active compounds present, we have now investigated the pharmacological activity of these flavonoids on two experimental models of inflam- mation in vitro.

MATERIALS AND METHODS

Soybean lipoxygenase assay

The experiments were performed according to the method of Sircar et al. (1983) as modified by Evans

815

816 Maria Jose Abad et al.

OH

cH3o.- -f -11- -o 3 OH O

OCH 3

OH

C H 3 O O C ~

OH O

OCH 3

2 Fig. I. Structure of centaureidin (I) and 5,3'-dihydroxy-4'-

methoxy-7-carbomethoxyflavonol (2).

(1987), using arachidonic acid (AA) as the substrate. The incubation mixture contained the following, AA 2mM, borax buffer 0.05 M pH 8.0 and enzyme lipoxygenase type I 1500 units. The enzymatic reac- tions were started with the addition of substrate (final volume 3 ml), and monitored spectrophotometrically by the appearance of a conjugated diene at 234 nm for 10min at 37°C. Absorbances were recorded on a VIS/UV spectrophotometer (Kontron, Model Uvikon-930).

Increasing amounts of test compounds were added to the standard assay system, and that concentration of drug required to reduce the rate of the enzyme reaction by 50% (IC50 value) was computed from the resulting dose-response curves. In every instance, flavonoids were dissolved in a final volume (10/d) of ethanol. Nordihydroguaiaretic acid (NDGA), a known inhibitor of soybean lipoxygenase, dissolved in a 1% N,N-dimethylformamide--borax buffer 0.05 M pH 8.0 solution, was used as a reference drug. Percent inactivation of enzyme activity by test drugs was determined by comparison with a vehicle control rate of enzyme activity (in the absence of inhibitor). Each test drug and control was run in quintuplicate at each concentration and the results averaged.

Prostaglandin synthetase assay

The method used was essentially the one described by Takeguchi and Sih (1972) and Fort et al. (1985), using AA as substrate. Portions of frozen bovine seminal vesicles (BSV) (80g), obtained from the Municipal Slaughterhouse (Madrid, Spain), were allowed to thaw and homogenized in 150mi of

phosphate buffer 0.1 M pH 8.0 on a Ultra Turrax homogenizer. The 100,000g pellet were prepared by the method of Takeguchi et aL (1971), using a refrigerated centrifuge (Sorwall, Model RC-5B), and an ultracentrifuge (Kontron, Model Centrikon T2080). On the day of the experiment, bovine seminal vesicular mitochondrial fraction (BSVM), containing prostaglandin synthetase with cyciooxygenase ac- tivity, were resuspended in 2ml final volume (20 mg/ml of microsomal protein) of Tris-HC 1 buffer 0.05 M pH 8.4, containing 0.01% Trit6n X-100

BSVM (6mg protein/0.3 ml) were pre-incubated with cofactor solution 5.8 mM L-adrenaline (0.8 ml) in Tris-HC1 buffer 0.05 M pH 8.4 and test com- pounds (in 10/zl of ethanol) at 37°C for 4 min. After incubation, the prostaglandin synthetase activity was started by the addition of substrate AA 2 mM and followed by measuring the change in the absorbance at 480 nm between 2-5 min.

Indomethacin, suspended in the same buffer, was used as a reference drug. Results are expressed as percentages of the increase in absorbance noted for the control under identical conditions (in the absence of inhibitor). Each drug was tested in quintuplicate at each concentration and on several different micro- somal preparations.

Drugs

Centaureidin and 5,3'-dihydroxy-4'-methoxy-7- carbomethoxyflavonol (Fig. 1) were isolated and purified from T. microphyllum as previously described (Abad et al., 1993). All the chemicals used were obtained from Sigma Chemical Co., St Louis, MO, U.S.A.

Statistical analysis

The parameters in all the samples were tested for statistical analysis with the aid of Mann-Whitney U-test. Probability values (p) of less than 0.05 were taken to indicate statistical significance. Regression analysis was used to calculate IC50, defined as the concentration of inhibitor necessary for 50% inhi- bition of the enzyme reaction.

RESULTS

Effect o f soybean lipoxygenase

The lipoxygenase inhibitor NDGA, which puta- tively blocks the formation of 5-1ipoxygenase prod- ucts, exerted a significant effect in this test system, with an ICs0 value of 44/~M (Fig. 2). Figure 2 also shows that centaureidin and 5,3'-dihydroxy-4'- methoxy-7-carbomethoxyflavonol produced concen- tration-dependent inhibition of soybean lipoxy- genase, with IC50 values (20 and 29/~M,

120 --

100 --

80 -- ~ ~ ~ ~

40 .~ , J , + Centaureidin

20 , -~-x- • Flavonol

o Z I I Z I Z Z Z I I 0 10 20 30 40 50 60 70 80 90 100

Concent ra t ion (10-6M)

Fig. 2. Inhibition of soybean lipoxygenase by nordihydro- guaiaretic acid (NDGA) and flavonoids extracted from T. microphyllum DC. Each point represents the

mean + SEM of five determinations.

Flavonoids on lipoxygenase and prostaglandin synthetase 817

*P ~< 0 .05;**P ~ 0.01 Table 1. IC50 values of flavonoids extracted from T. microphyllum DC

Soybean lipoxygenase Prostaglandin synthetase Flavonoid 1C5o (#M) IC~0 (#M)

Centaureidin 20 318 Flavonol 29 60

respectively) slightly smaller than that of the refer- ence drug.

Effect o f prostaglandin synthetase

In this test system, indomethacin, a classical non- steroidal anti-inflammatory drug (NSAID), potently inhibited release of the cyclooxygenase metabolites, as shown in Fig. 3. The flavonoid 5,3'-dihydroxy- 4'-methoxy-7-carbomethoxyflavonol, also produced concentration-dependent inhibition of prostaglandin synthetase activity, but only exerted significant effect (p <0.01) at higher concentrations (Fig. 3). How- ever, there was no evidence for significant inhibition of prostaglandin synthetase by centaureidin, at con- centrations up to 100 ~tM (data not shown). These flavonoids required a higher IC50 than for lipoxy- genase (IC50 value for prostaglandin release of 318 and 60 pM, respectively) (Table 1).

DISCUSSION

Flavonoids isolated from T. microphyllum inhib- ited significantly the carrageenin-induced paw edema in mice (Abad et aL, 1993). The pharmacological

120 F ~'P ~< 0'05;*~'P ~< 0.01

100 [-- ~'~m

60 - ~ . / J • ,ndomethacin

40 }--£+ m ' ~ + Centaureidin l / / / • Flavonol

20 ~ - / . f

o l 0 ' l I I I I L I I I b 0 1 2 3 4 5 6 7 8 9 10

Concentration (10-4M) Fig. 3. Inhibition of prostaglandin synthetase by indo- methacin and flavonoids extracted from T. microphyllum DC. Each point represents the mean __+ SEM of five determi-

nations.

testing of these compounds on in vitro models, was undertaken in order to check if they were the com- pounds responsible for the anti-inflammatory non- ulcerogenic activity exhibited by the crude extract. This is the first report of the biological action in vitro of centaureidin and 5,3"-dihydroxy-4'-methoxy-7- carbomethoxyflavonol and of their possible mech- anism of action.

We demonstrate in the present paper that these flavonoids significantly blocked the synthesis of soy- bean iipoxygenase products in vitro. These results also indicate that centaureidin and 5,3'-dihydroxy-4'- methoxy-7-carbomethoxyflavonol, exerted concen- tration-dependent cyclooxygenase inhibitory effects in microsomal systems in vitro, but with a high IC50 value. In view of the results reached in this pre- liminary investigation, we can suggest that these flavonoids may owe their anti-inflammatory effect to a different mechanism of interactions with prosta- glandins (PGs).

In the last few years, there has been considerable interest in the role of the cyclooxygenase and lipoxy- genase products of AA metabolism in both acute and chronic inflammatory diseases and as cytoprotective and antisecretory agents in the stomach. The lipoxy- genase products hydroxy fatty acids (HETEs) and ieukotrienes can be chemotactic for leukocytes during inflammation, and are considered to be involved in the initiation and maintenance of these inflammatory diseases (Piper, 1983). PGs, for which cyclooxygenase is the key enzyme, potentiate the effects of other mediators (histamine, serotonin, etc.) in inducing vasodilatation, pain and increasing capillary per- meability. Furthermore, PGs have been shown to prevent the formation of ulcers by a mechanism independent of their antisecretory properties and to protect the gastric mucosa (Williams, 1983). The combination of anti-inflammatory and anti- ulcerogenic actions is unusual but favourable, never- theless some natural products exhibit both properties, such as flavonoids (Alcaraz and Hoult, 1985a,b) and sesquiterpene lactones (Hall et al., 1980; Giordano et al., 1990).

Many reports provide data on the interference of flavonoids with mechanisms involved in the inflam- matory process, thereby raising a different hypothesis on how these compounds act. According to Vane's hypothesis (1971), inhibition of cyclooxygenase

818 Maria Jose Abad et al.

activity is the major mechanism to explain the anti- inflammatory actions of classical NSAIDs.

Flavonoids can have influences on cyclooxygenase activity. The minimum structural requirement for activation is a phenol possessing a polar side chain in either meta- or para-position (Baumann et al., 1979), and it has been postulated that flavonoids possessing the o-dihydroxyl group (pyrocatechin) are able to inhibit cyclooxygenase (Wurm et al., 1982). Corre- sponding with these hypotheses, centaureidin and 5, 3'-dihydroxy-4'- methoxy-7-carbomethoxy- flavonol did not inhibit cyclooxygenase, but the results obtained in this study showed that these flavonoids inhibit prostaglandin synthethase prod- ucts, although the concentrations necessary were rather high. In spite of that, 5,Y-dihydroxy- 4'methoxy-7-carbomethoxyflavonol is more active on prostaglandin synthetase activity than centaureidin, and these data suggest that the carbomethoxy group at 7 position, whose presence has been found only in some models of isoflavones (Mabry et al., 1970), might be responsible for this action.

Classical NSAIDs, such as aspirin and indo- methacin, cannot influence lipoxygenase activity at doses capable of inhibiting cyclooxygenase and inflammatory reaction. Several flavonoids, such as quercetin and baicalein, inhibit lipoxygenase from different sources (Yamamoto et al., 1984; Masters et al., 1985).

The minimum structural requirement for inhibition of soybean lipoxygenase is the presence of the keto group at 4 position, with the absence of substitution at 2' position (Wurm et al., 1982). Corresponding to this hypothesis, centaureidin and 5,Y-dihydroxy-4'- methoxy-7-carbomethoxyflavonol are enzyme inhibi- tors, as we have demonstrated in the present paper.

These data suggest that the inhibition of leuko- triene synthesis may, at least in part, be responsible for the anti-inflammatory action of our flavonoids. These compounds, mainly 5,3'-dihydroxy-4'- methoxy-7-carbomethoxyflavonol, at high concen- trations, have a slight inhibitory effect on prostaglandin synthetase activity. Drugs that cause selective inhibition of lipoxygenase (with therapeutic value in anaphylaxis) as well as agents that would be dual inhibitor of both cyclooxygenase and lipoxy- genase, are capable of controlling the inflammatory conditions with similar properties to the anti- inflammatory corticosteroids, but are devoid of the steroid-related toxicity (Higgs and Vane, 1983).

These results strongly suggest that centaureidin and 5,3'dihydroxy-4-methoxy-7-carbomethoxyflavo- nol, two active components of T. microphyllum, may be useful in the treatment of inflammatory diseases. However, the exact mechanism of action of these

flavonoids on the inflammatory process was not determined, and further work is obviously required to assess the effectiveness of interaction with other pro- inflammatory biochemical pathways and their poss- ible structure-activity relationships, so that the mode of action of these compounds can be clarified.

REFERENCES

Abad M. J., Bermejo P. and Villar A. (1991) Anti- inflammatory and anti-ulcerogenic activities of the organic extracts of Tanacetum microphyllum DC. in rats. Phytother. Res. 5, 179-181.

Abad M. J., Bermejo P., Villar A. and Valverde S. (1993) Anti-inflammatory activity of two flavonoids from Tanacetum microphyllum. J. Nat. Prod. 56, 1164-I 167.

Alcaraz M. J. and Hour J. R. S. (1985a) Actions of flavonoids and the novel anti-inflammatory flavone, hypolaetin-8-glucoside, on prostaglandin biosynthesis and inactivation. Biochem. Pharmac. 34, 2477-2482.

Alcaraz M. J. and Hour J. R. S. (1985b) Effects of hypolaetin-8-glucoside and related flavonoids on soybean lipoxygenase and snake venom phospholipase A 2. Arch. Int. Pharmacodyn. 278, 4-12.

Baumann J., Von Bruchhausen F. and Wurm G. (1979) A structure-activity study on the influence of phenolic com- pounds and bioflavonoids on rats renal prostaglandin synthetase. Naunyn Schmiedeberg's Arch. Pharmac. 307, 73-78.

Berry M. I. (1984) Feverfew faces the future. Pharm. J. 232, 611-614.

Evans A. T. (1987) Actions of cannabis constituents on enzymes of arachidonate metabolism: anti-inflammatory potential. Biochem. Pharmac. 36, 2035-2037.

Fort M., Farr~ A. J., Colombo M. and Guti6rrez B. (1985) Correlaci6n entre actividad antidiarreica y actividad antiinflamatoria (edema de carragenina) con 18 anti- inflamatorios. Rev. Farm. Clin. Exp.. 2, 272.

Giordano O. S., Guerreiro E., Pestchanker M. J., Guzmfin J., Pastor D. and Guardia T. (1990) The gastric cytopro- tective effect of several sesquiterpene lactones. J. Nat. Prod. 53, 803-809.

Groenewegen W. A. and Heptinstall S. (1990) A comparison of the effects of an extract of feverfew and parthenolide, a component of feverfew, on human platelet activity in vitro. J. Pharm. Pharmac. 42, 553-557.

Hall I. H., Starnes J. and Lee K. H. (1980) Mode of action of sesquiterpene lactones as anti-inflammatory agents. J. Pharm. Sci. 69, 537-543.

Higgs G. A. and Vane J. R. (1983) Inhibition of cyclo- oxygenase and lipoxygenase. Br. Med. Bull. 39, 265-272.

Johnson E. S., Kadam N. P., Hylands D. M. and Hylands P. J. (1985) Efficacy of feverfew (Tanacetum parthenium) as prophylactic treatment of migraine. Br. Med. J. 291, 569-573.

Mabry T. J., Markham K. R. and Thomas M. B. (1970) The NMR spectra of flavonoids. In The Systematic Identifi- cation of Flavonoids, pp. 275-343. Springer-Verlag, Berlin.

Masters D. J., Spruce K. E., Vickers V. C. and McMillan R. M. (1985) Measurement of arachidonate metabolism using cyanopropyl minocolumns: effects of cyclo- oxygenase and lipoxygenase inhibitors. Br. J. Pharmac. 84, 45.

Murphy J. J., Heptinstall S. and Mitchell J. R. A. (1988) Randomised double-blind placebo-controlled trial of feverfew in migraine prevention. Lancet 2, 189-192.

Piper P. J. (1983) Pharmacology of leukotrienes. Br. Med. Bull. 39, 255-261.

Sircar J. C., Schwender C. F. and Johnson E. A. (1983) Soybean lipoxygenase inhibition by nonsteroidal anti- inflammatory drugs. Prostaglandins 25, 393-396.

Flavonoids on lipoxygenase and prostaglandin synthetase 819

Takeguchi C. and Sih C. J. (1972) A rapid spectrophoto- metric assay for prostaglandin synthetase: application to the study of non-steroidal anti-inflammatory agents. Prostaglandins 2, 169-184.

Takeguchi C., Kohno E. and Sih C. J. (1971) Mechanism of prostaglandin biosynthesis.l. Characterization and assay of bovine prostaglandin synthetase. Biochemistry 111, 2372-2376.

Vane J. R. (1971) Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature 231, 232-235.

Williams T. J. (1983) Interactions between prostaglandins, leukotrienes and others mediators of inflammation. Br. Med. Bull. 39, 239-242.

Wurm G., Baumann J. and Geres U. (1982) Beeinflussung des arachidossS.urestoffwechsels durch flavonoide. Deutsche Apotheker Zeitung 122, 2062-2068.

Yamamoto S., Yoshimoto T., Furukawa M., Horie T. and Watanabe-Kohno S. J. (1984) Arachidonate 5-1ipoxy- genase and its new inhibitors~ Allergy clin. Immun. 74, 349-352.