- Email: [email protected]
Oxidation of the flavonol fisetin by polyphenol oxidase
Biochimica et Biophysica Acta 1425 (1998) 534^542
Oxidation of the £avonol ¢setin by polyphenol oxidase Mercedes Jime¨nez *, Josefa Escribano-Cebria¨n, Francisco Garc|¨a-Carmona Departamento de Bioqu|¨mica y Biolog|¨a Molecular A, Unidad Docente de Biolog|¨a, Facultad de Veterinaria, Universidad de Murcia, E-30080 Murcia, Spain Received 9 July 1998; received in revised form 18 September 1998; accepted 23 September 1998
Abstract The present study demonstrates the antiradical efficiency of fisetin, a flavonol widely distributed in fruits and vegetables, by its ability to react with two different free radicals, ABTSc and DPPHc . The polyphenolic nature of fisetin led us to consider whether it might be oxidised by polyphenol oxidase (PPO), and the results reported show that it can be oxidised by PPO extracted and partially purified from broad bean seeds. The reaction was followed by recording spectral changes with time, with maximal spectral changes being observed at 282 nm (increase in absorbance) and at 362 nm (decrease). The presence of two isosbectic points (at 265 and 304 nm) suggested that only one absorbent product was formed. These spectral changes were not observed in the absence of PPO. The oxidation rate varied with the pH, reaching its highest value at pH 5.5. The fisetin oxidation rate increased in the presence of sodium dodecyl sulfate, an activator of polyphenol oxidase. Maximal activity was obtained at 0.87 mM sodium dodecyl sulfate. The following kinetic parameters were determined: Vmax = 49 WM/ min, Km = 0.6 mM, Vmax /Km = 8.2U1032 min31 . Flavonol oxidation was inhibited by selective PPO inhibitors such as cinnamic acid (a classical competitive inhibitor, Ki = 1.4 mM) and 4-hexylresorcinol, which behaved as a slow-binding inhibitor. The results reported show that fisetin oxidation was strictly dependent on the presence of polyphenol oxidase. ß 1998 Elsevier Science B.V. All rights reserved. Keywords: Polyphenol oxidase; Flavonol; Fisetin ; Antioxidant
1. Introduction There is a considerable amount of evidence that points to an association between diets rich in fresh fruits and vegetables and a decreased risk of cardiovascular disease and certain forms of cancer. It is generally assumed that the active dietary constituents Abbreviations: ABTS , 2,2P-azino-bis(3-ethylbenzthiazoline-6sulfonic acid); ABTSc , (2,2P-azino-bis(3-ethylbenzthiazoline-6sulfonic acid) radical; DPPHc , 2,2P-diphenyl-1-picrylhydrazyl radical * Corresponding author. Fax: +34-968-364147; E-mail: [email protected]
contributing to these protective e¡ects are the antioxidants (vitamins, carotenoids, sterols, polyphenols). Natural polyphenolic antioxidants are increasingly attracting the attention of food scientists because of their bene¢cial e¡ects on human health and food preservation. Oxidation is one of the most important processes of food deterioration because it may a¡ect food safety, colour, £avour and texture. Antioxidants may protect food quality by preventing the oxidative deterioration of lipids [1]. Recent works highlight the role of one class of natural phenols, known as £avonoids, and their bene¢cial e¡ects on human health due to their wide dis-
0304-4165 / 98 / $ ^ see front matter ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 8 ) 0 0 1 0 8 - 1
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
tribution in edible plants and in plant-derived foodstu¡s, which make a substantial contribution to the human diet. Flavonoids have multiple chemical and biological actions, including antioxidant activity. Their antioxidant properties reside in their radical scavenging activity. The position and the degree of hydroxylation are of primary importance in determining the strength of their antioxidant activity. It is known that £avonoids with a 3P,4P dihydroxy con¢guration possess antioxidant activity [2^4]. Fisetin is a particularly interesting £avonoid because of its presence in human foods and its biological activity. That it has antioxidant activity is shown by the fact that it inhibited human low-density lipoprotein (LDL) oxidation in vitro [5]. Fisetin shows a dose- and time-dependent increase in lipolysis, which is synergistic with epinephrine on L-adrenergic receptors [6]. Fisetin has also been reported as antiin£ammatory. It is a potent inhibitor of lysosomal enzyme secretion and arachidonic acid release in rat neutrophils [7]. It also inhibits mammalian 5-lipoxygenase and cyclooxygenase enzymes [8]. The in vivo e¡ect of ¢setin on the inhibition of a£atoxin B1 carcinogenicity and, hence, in the detoxi¢cation processes, has been described [9]. Fisetin completely inhibited thrombin- and cathepsin G-induced platelet aggregation [10]. Its antimutagenicity [11] and its antiproliferative e¡ects on certain malignant cells are thought to be due to the inhibition of phosphatidylinositol 3-kinase and protein kinase C [12]. Fisetin is a potent inhibitor of the human P-form phenolsulfonyltransferase, suggesting it may play a role as chemopreventive agent in sulfation-induced carcinogenesis [13]. Moreover, the protection of plasmid pBR322 DNA by ¢setin against single-stranded break induced by single molecular oxygen has also been shown [14]. The polyphenolic nature of ¢setin and, in particular, the fact that it is an o-dihydroxylated compound led us to suggest its potential oxidation by polyphenol oxidase (PPO) (EC 1.14.18.1), which is widely distributed in nature and responsible for melanisation in animals [15,16] and browning in plants [17]. The enzyme catalyses two distinct reactions involving molecular oxygen: (a) the o-hydroxylation of monophenols and (b) the oxidation of odiphenols to o-quinones. The PPO oxidation of monophenols and o-diphenols with simple chemical
535
structures has been extensively studied, although some of these substrates are not even natural compounds. Investigations carried out in vitro suggest that PPO may participate in the degradation of natural phenols with complex structures, such as anthocyanins in strawberries [18] and grapes [19], and the £avonols present in tea leaves [20]. These compounds are not directly oxidised by the enzyme but by the quinones formed by PPO from catechol, chlorogenic acid or catechin [18]. In spite of the increasing interest shown in the antioxidant properties of ¢setin, there have been no studies analysing its radical scavenging activity or its potential oxidation by polyphenol oxidase. Therefore, the aim of this work was to study the PPOcatalysed oxidation of ¢setin, which is shown, by the two di¡erent methods used in this paper to evaluate its antiradical capacity, to be an e¡ective radical scavenger. The results obtained con¢rm that PPO is capable of oxidising ¢setin directly. The spectrophotometric and kinetic characteristics of this oxidation were investigated, and the speci¢city of the reaction was demonstrated by the use of speci¢c PPO inhibitors. The oxidation of ¢setin by PPO in vitro poses questions about the possible involvement of the enzyme in the oxidation of natural polyphenols in vivo which would cause the loss of antioxidant activity of fruits and vegetables. 2. Materials and methods 2.1. Reagents Broad bean seeds were purchased from a local market in the city of Murcia. Fisetin hydrate (3,3P,4P,7-tetrahydroxy£avone), DPPHc (2,2P-diphenyl-1-picrylhydrazyl radical) and 4-hexylresorcinol were purchased from Aldrich (Madrid, Spain). ABTS (2,2P-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid), cinnamic acid and horseradish peroxidase (HRP, type VI; Rz = 3.2) were obtained from Sigma Quimica (Madrid, Spain). TX-114 was obtained from Fluka Quimica (Madrid, Spain) and condensed three times as described [21] but using 100 mM sodium phosphate bu¡er (pH 7.3). The detergent phase of the third condensation had a TX-
BBAGEN 24701 19-11-98
536
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
114 concentration of 25% (w/v) and was used as the stock detergent solution for all of the experiments. 2.2. Measurement of the antiradical activity of ¢setin The antiradical activity of ¢setin was measured by the disappearance of 2,2P-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) radical (ABTSc ) [22] ABTSc was generated in a reaction medium (1 ml) containing 2 mM ABTS , 35 WM H2 O2 and 31 nM horseradish peroxidase (EC 1.11.1.7) in 50 mM sodium acetate bu¡er, pH 5.0. ABTSc accumulation was followed by measuring the increase in absorbance at 420 nm, a bandwidth at which ¢setin does not show any appreciable absorption. ABTSc concentration was calculated by using an O 420 = 31 100 M31 cm31 . Once ABTS oxidation had ceased because of the complete depletion of H2 O2 (end point of the reaction), ¢setin was added at di¡erent concentrations. After addition, the absorbance at 420 nm decreased until constant absorbance was asymptotically reached. For each ¢setin concentration tested, the percentage of ABTSc remaining was determined and the values transferred to another graph showing the percentage of residual ABTSc as a function of the molar ratio of ¢setin to ABTSc . Another method used to determine the free radical scavenging activity of antioxidants was their reaction with the free radical DPPHc , according to the procedure described by Brand-Williams et al. [23]. The assay medium contained 60 WM DPPHc in methanol and di¡erent concentrations of a methanol ¢setin solution. The reaction was monitored by following the decrease in absorbance at 515 nm until the reaction reached a plateau. The percentage of DPPHc remaining as a function of the molar ratio of ¢setin to DPPHc was calculated similarly to the way used for the ABTSc method. 2.3. Enzyme extraction Polyphenol oxidase was extracted in its latent state from the chloroplast membranes of broad bean (Vicia faba) seeds. All procedures were carried out at 4³C. Chloroplasts were prepared by homogenising 50 g of broad bean seeds in 150 ml of sodium phos-
phate bu¡er containing 0.33 M sorbitol, 2 mM EDTA, 1 mM MgCl2 , and serine protease inhibitors, which were added immediately before use (1 mM phenylmethanesulfonyl £uoride (PMSF) and 1 mM benzamidine hydrochloride). The slurry was ¢ltered through eight layers of gauze and centrifuged at 600Ug for 2 min. The pellet was discarded, and the supernatant was centrifuged at 20 000Ug for 30 min. The resultant pellet was resuspended in a solution containing 50 ml of 10 mM sodium phosphate bu¡er (pH 7.3) and kept at 4³C for 20 min. Then, the solution was centrifuged at 20 000Ug for 20 min, thus pelleting the chloroplast membranes. These membranes were resuspended with 20 ml of 1.5% (w/v) TX-114 in 100 mM phosphate bu¡er (pH 7.3) for 30 min at 4³C. After high-speed centrifugation (60 000Ug for 20 min), this light-green extract yielded a clear supernatant with PPO activity. This was subjected to temperature phase partitioning by adding TX-114 at 4³C to give a ¢nal concentration of 8% (w/v). The mixture was kept at 4³C for 15 min in a Tectron thermostatic bath and then warmed to 37³C. After 15 min, the solution became spontaneously turbid due to the formation, aggregation and precipitation of large micelles composed of detergent, hydrophobic proteins and the remaining chlorophylls. This solution was centrifuged at 5000Ug for 10 min at room temperature. The detergent-rich phase was discarded and the clear supernatant containing the soluble PPO brought to 45% saturation with saturated (NH4 )2 SO4 (pH 7.0), and kept overnight, under continuous stirring, at 4³C. Then, the solution was centrifuged at 120 000Ug for 30 min at 4³C and the pellet was discarded. Saturated (NH4 )2 SO4 (pH 7.0) was added to the clean supernatant to give 85% saturation and stirred for 1 h at 4³C. The precipitate obtained between 45% and 85% was collected by centrifugation at the same rotor speed and dissolved in a minimal volume of 10 mM sodium phosphate bu¡er, pH 7.3. The salt content was removed by dialysis against 10 mM sodium phosphate bu¡er, pH 7.3. The solution thus obtained was used as enzyme source. To avoid any possible activation of the enzyme by endogenous proteases, PMSF and benzamidine hydrochloride were added before and after dialysis to give a ¢nal concentration of 1 mM.
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
2.4. Enzyme assays Unless otherwise noted, the spectrophotometric assays were performed at 25³C in a reaction medium (1 ml) containing 75 WM ¢setin and 0.87 mM sodium dodecyl sulfate (SDS) in 0.1 M sodium phosphate bu¡er, pH 5.5. The reaction was initiated by the addition of polyphenol oxidase. Fisetin oxidation was monitored by measuring the increase in absorbance at 282 nm or the decrease at 362 nm. Di¡erences between the extinction coe¤cients (vO), corresponding to the di¡erence between absorption due to substrate and that due to reaction products, were determined for the oxidation process. The vO values were calculated from a calibration curve of ¢setin (0^12 WM) quantitatively oxidised in the presence of an excess of sodium periodate at di¡erent pH values. The increases in absorbance at 282 nm or the decreases at 362 from t = 0 to constant absorbance values were plotted against ¢setin concentration. The values found, at pH 5.5, were vO282 = 5345 M31 cm31 (r = 0.997) and vO362 = 18 261 M31 cm31 (r = 0.999). pH studies were carried out using 0.1 M sodium acetate and sodium phosphate bu¡ers from pH 3.5 to pH 7, in the presence of 0.87 mM SDS. The pH of the assay solution was determined at room temperature, using a Crison micro pH 2002 meter. After catalysis, the pH of the assay solution was again measured.
537
bu¡er, pH 5.5. The reaction was started by adding the enzyme. 2.7. Other methods The protein content was determined according to the Bradford Bio-Rad protein assay using bovine serum albumin as a standard [24]. 3. Results and discussion Flavonoids are, in general, strong antioxidants, their antioxidant potential residing in their radical scavenging capacity. Several methods to evaluate antiradical activity have been reported [25,26] with those involving the reaction with the ABTSc [22] or DPPHc [23] radicals receiving most attention. We used both methods to demonstrate that ¢setin is an e¡ective radical scavenger. The so-called ABTSc method is based on measuring the destruction of the ABTSc radical due to its reaction with an antioxidant compound [22]. The ABTSc was ¢rst generated in the assay medium by
2.5. Optimal activation by SDS Activation studies were carried out in a reaction medium (1 ml) containing 75 WM ¢setin and di¡erent detergent concentrations in 0.1 M sodium phosphate bu¡er, pH 5.5. The concentration of SDS varied from 0 to 5.2 mM. 2.6. Inhibition studies Inhibition of ¢setin oxidation by polyphenol oxidase was followed at 372 nm and 25³C by using 4hexylresorcinol and cinnamic acid as speci¢c inhibitors of the enzyme. Unless otherwise stated, the reaction medium (1.0 ml) contained ¢setin (18^90 WM), 4-hexylresorcinol (0^1.5 mM) or cinnamic acid (0^5 mM) and 0.87 mM SDS in 0.1 M sodium phosphate
Fig. 1. Disappearance of ABTSc as a function of the number of moles of ¢setin/mol ABTSc . The percentage was calculated by the formula ((vAmax 3vA)/vAmax )U100, where vAmax is the maximum decrease in absorbance obtained at a high molar ratio and vA the maximum decrease in absorbance at each molar ratio. Inset: structure of ¢setin.
BBAGEN 24701 19-11-98
538
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
the horseradish peroxidase (HRP)/H2 O2 -mediated oxidation of ABTS . After ABTSc formation had ceased (due to the complete depletion of H2 O2 ), different concentrations of ¢setin were added to the reaction medium. Following the addition of ¢setin, the disappearance of ABTSc was followed by measuring the decrease in absorbance at 420 nm until a plateau was reached (data not shown). For each ¢setin concentration used, the percentage of ABTSc remaining was determined, and the values plotted against the molar ratio of ¢setin to ABTSc (Fig. 1). The concentration of antioxidant needed to decrease the initial radical concentration by 50% (e¤cient concentration = EC50 (mol/l antioxidant6mol/l ABTSc ) is a widely used parameter to measure antioxidant power [23]. The lower the EC50 , the higher the antioxidant power. This ratio, calculated from the results obtained in Fig. 1, was 0.04. The antiradical activity of ¢setin was also tested by observing its reaction with the stable free radical DPPHc . A similar procedure to that previously explained for the ABTSc was used (data not shown). From the results obtained (data not shown) an EC50 value of 0.05 was determined. Thus, the EC50 values obtained by both methods were very similar. Moreover, this EC50 value calculated for ¢setin was comparable to that determined for other polyphenols with a very high antiradical activity, such as gentisic and gallic acids [23]. The antioxidant activity of ¢setin is related to its structure (shown in Fig. 1). The o-dihydroxylation of the B ring contributes to the antioxidant activity [2^ 4]. This ¢setin hydroxylation suggests its potential oxidation by polyphenol oxidase. Fisetin oxidation by PPO extracted from broad bean seeds was followed by observing changes in the UV-visible spectrum with time (Fig. 2). Maximal spectral changes were observed at 362 nm (decreases in absorbance) and at 282 nm (increases in absorbance). The presence of two isosbectic points at 265 and 304 nm (arrowheads) suggests that only one absorbent product is formed during the course of ¢setin oxidation by PPO. These spectral changes were not observed in the absence of the enzyme and, therefore, they were considered to be the result of enzyme activity. The nature of the spectral changes during ¢setin oxidation indicated that the formation of reaction
Fig. 2. Consecutive spectra, at 25³C, obtained in the oxidation of ¢setin by broad bean seed polyphenol oxidase. The assay medium (1.0 ml) contained 75 WM quercetin and 0.87 mM SDS in 0.1 M sodium phosphate bu¡er, pH 5.5. The reaction was started by addition of the enzyme (33 Wg/ml). Scan speed was at 1-min intervals for 10 min.
product was proportional to time during the ¢rst 10 min of the reaction. This observation and the dependence of these spectral changes on the presence of the enzyme suggest that such changes are a reliable measure of ¢setin oxidation. Therefore, £avonol oxidation by PPO was routinely assayed by measuring either the decrease in absorbance at 362 nm or the increase at 282 nm vs. reaction time. In order to characterise ¢setin oxidation by PPO, the e¡ect of pH was analysed (not shown). First, the vO were determined, as described in Section 2, for all the di¡erent pH values assayed. These vO values were seen to be pH-dependent, and this variation was taken into account when calculating the ¢setin oxidation rate at the di¡erent pH values. The oxidation rate varied with pH, the optimal value being 5.5. Fisetin oxidation was monitored by measuring the decrease in absorbance at 362 nm. Similar results were obtained when the activity was monitored by following the increase in absorbance at 282 nm. Polyphenol oxidase can be found in an inactive or latent state. To obtain the enzyme in its latent form, a very mild extraction method must be used to avoid the modi¢cation or oxidation of PPO produced by
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
Fig. 3. E¡ect of SDS on the rate of ¢setin oxidation catalysed by polyphenol oxidase. The reaction medium, at 25³C, contained 75 WM ¢setin and 33 Wg/ml enzyme in 0.1 M sodium phosphate bu¡er, pH 5.5. The ¢setin oxidation rate was monitored by measuring the decrease in absorbance at 362 nm.
acetone powers [27]. The method used to extract and partially purify polyphenol oxidase from broad bean seeds was used to purify PPO in latent state from di¡erent sources [28]. Latent PPO can be released from latency or activated by a variety of treatments or agents including acid or basic shock, proteolytic attack, phospholipid release [17], polyamines [29], polyglucans [30], divalent cations [31] and anionic detergents [17,32,33]. SDS activation is particularly interesting because few enzymes are known to be activated by this anionic detergent. Thus, PPO is activated at high SDS concentrations, which would denature many other enzymes [32]. The activation brought about by several seemingly unrelated substances is a phenomenon often attributed to conformational changes of the enzyme molecule, solubilisation of the enzyme or the removal of an inhibitor. In the case of SDS, it was shown that the active site is blocked in absence of the detergent but open in its presence [17]. Fisetin oxidation by polyphenol oxidase extracted from broad bean seeds also increased in the presence of SDS (Fig. 3). Scans of the SDS concentration carried out at pH 5.5 showed the optimum SDS concentration to be 0.87 mM, since higher detergent
539
concentrations led to a decrease in enzymatic activity. To further study ¢setin oxidation by PPO, the dependence of the oxidation rate on £avonol concentration was examined. The enzyme showed Michaelis^Menten type kinetics (data not shown), although enzyme saturation by the substrate was not reached in the concentration range assayed. It was not possible to further increase ¢setin concentration because of its high absorbance at the measurement wavelength, despite the fact that spectrophotometric cells with a path length of 5 mm (half of the standard length) were used to decrease the initial absorbance values. Hanes^Woolf plots for the kinetic data of ¢setin oxidation resulted in linear relationships (not shown), from which the kinetic parameters were evaluated (Vmax = 49 WM/min, Km = 0.6 mM, Vmax / Km = 8.2U1032 min31 ). The spectral changes for ¢setin oxidation shown in Fig. 2 were completely dependent on the presence of the active enzyme. To further prove the speci¢city of the £avonol oxidation by PPO, the e¡ect of selective inhibitors was examined. A considerable number of polyphenol oxidase inhibitors is known, one large group being constituted by compounds, such as benzoic acid and their derivatives, which are structurally analogous to phenolic substrates and which generally show competitive inhibition with respect to these substrates. Other recently recognised PPO inhibitors include 4-substituted resorcinols, which are also structurally related to phenolic substrates. 4-Hexylresorcinol has been shown to be e¡ective in preventing black spot formation in shrimps [34] and for browning control in di¡erent fruits [35]. 4-Hexylresorcinol and other 4-substituted resorcinols are also very effective inhibitors of mushroom tyrosinase [36,37], inhibiting the enzyme in a non-classical manner and classi¢ed as slow competitive inhibitors [37]. An inhibition study of ¢setin oxidation was, therefore, carried out with 4-hexylresorcinol. A prolonged transient phase was observed for the inhibition of PPO (data not shown), and thus, 4-hexylresorcinol presents the characteristics of a competitive slowbinding inhibitor of polyphenol oxidase according to Williams and Morrison classi¢cation [38]. Fig. 4 shows the e¡ect of the inhibitor on the initial (A) and steady-state (B) rates, respectively, at di¡erent concentrations of substrate. Initial velocities (Vo ) de-
BBAGEN 24701 19-11-98
540
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
(Fig. 4B), a value of 0.05 mM was determined for 0 the overall dissociation constant (K i ). All these e¡ects of 4-hexylresorcinol on PPO were similar to those obtained for the slow inhibition of tyrosinase by L-mimosine [39], tropolone [40] and kojic acid [41]. A mechanism to explain the behaviour of these time-dependent inhibitors has been previously postulated [39]. This involves the rapid formation of a complex between the inhibitor and the oxy form of the enzyme, a complex that subsequently undergoes a relatively slow reversible reaction. According to the kinetic analysis of this mechanism the following equation was obtained: kapp k36
k6 I K i 1 S=K m I
Fig. 5 shows 1= kapp 3k36 vs. 1/[I], for three different ¢setin concentrations. From the intercept with the ordinate axis (1/k6 ), a value of 0.024 s31 was obtained for k6 ,the constant for the slow transition process of the inhibitor^enzyme complex. An inhibition study of ¢setin oxidation was also carried out with cinnamic acid, a commonly used tyrosinase inhibitor that has been identi¢ed as a competitive inhibitor for PPO from di¡erent sources [42]. Flavonol oxidation was also inhibited by the
Fig. 4. Dixon plots for the inhibitory e¡ect of 4-hexylresorcinol on initial (A) and steady-state (B) rates of ¢setin oxidation catalysed by PPO. The reaction medium contained £avonol and inhibitor at the indicated concentrations, 0.87 mM SDS and 94 Wg/ml enzyme. The ¢setin concentrations used were: (b) 90, (a) 75, (F) 60, (E) 45, (R) 30 and (O) 18 WM.
creased with inhibitor concentration, indicating that an enzyme^inhibitor complex is rapidly formed with an apparent dissociation constant (Ki ) of 0.3 mM. The steady-state rates (Vs ) also decreased as 4-hexylresorcinol concentration rose, and when Dixon plots were made at di¡erent ¢setin concentrations
Fig. 5. Graphical calculation of k6 for the inhibition of ¢setin oxidation by 4-hexylresorcinol. Experimental conditions were as in Fig. 4. Fisetin concentrations were: (b) 90, (E) 45 and (R) 30 WM.
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
presence of cinnamic acid in the reaction medium, although this inhibitor behaved as a classical competitive inhibitor. From the Dixon plots (data not shown), a Ki value of 1.4 mM was determined. This paper therefore shows the antioxidative potential of ¢setin since the £avonol acted as an e¡ective radical scavenger, as proved by its reaction with the free radicals ABTSc and DPPHc . These results agree with previous studies reporting the e¤ciency of ¢setin in inhibiting human LDL oxidation in vitro [5]. We can also conclude that ¢setin is oxidised by polyphenol oxidase. This oxidation is completely dependent on the presence of active enzyme because the reaction was inhibited by speci¢c PPO inhibitors such as 4-hexylresorcinol and cinnamic acid. There have been reports suggesting that PPO could play a role in the degradation of polyphenols with complex structures, such as anthocyanin pigments from grapes and strawberries [18,19]. However, direct oxidation by PPO did not appear to be the main decolourisation route. Quinones and the intermediary compounds formed during the oxidation by PPO of catechol, catechin or chlorogenic acid might be responsible for the destruction of anthocyanins either through oxidation or co-polymerisation. The fact that PPO oxidises ¢setin in vitro poses questions about the enzyme's potential involvement in the oxidation of natural polyphenols in vivo. Future studies should be aimed at ascertaining whether or not PPO is involved in the loss of antioxidant activity due to the oxidation of phenolic compounds, particularly during the improper handling or storage of fruits and vegetables. Acknowledgements This work was supported by a grant from the Comisio¨n Interministerial de Ciencia y Tecnolog|¨a (Spain), DGES project PB95-1024. References [1] R.A. Jacob, Nutr. Res. 15 (1995) 755^766. [2] C.A. Rice-Evans, N.J. Miller, G. Paganga, Free Radic. Biol. Med. 20 (1996) 933^956.
541
[3] S.A. van Acker, D.J. van der Berg, M.N. Tromp, D.H. Gri¤oen, W.P. van Bennekom, W.J. van der Vijgh, A. Bast, Free Radic. Biol. Med. 20 (1996) 331^342. [4] S.A. van Acker, M.J. Groot, D.J. van der Berg, M.N. Tromp, G. Donne-Op den Kelder, W.J. van der Vijgh, A. Bast, Chem. Res. Toxicol. 9 (1996) 1305^1312. [5] I. Myara, I. Pico, B. Vedie, N. Moath, J. Pharmacol. Toxicol. Methods 30 (1993) 69^73. [6] U.R. Kappusamy, N.P. Das, Biochem. Pharmacol. 47 (1994) 521^529. [7] M. Tordera, M.L. Ferrandiz, M.J. Alcaraz, Z. Naturforsch. C 49 (1994) 235^240. [8] M.J. Laughton, P.J. Evans, M.A. Moroney, J.R. Hault, B. Halliwell, Biochem. Pharmacol. 42 (1991) 1673^1681. [9] V.S. Aboobaker, A.D. Balgi, R.K. Bhatacharya, In vivo 8 (1994) 1095^1098. [10] R.N. Puri, R.W. Colman, Anal. Biochem. 210 (1993) 50^57. [11] R. Edenhareder, I. Petersdor¡, R. Rauscher, Mutat. Res. 287 (1993) 261^274. [12] G. Agullo, L. Gamet-Payrastre, S. Manenti, C. Viala, C. Remesy, H. Chap, B. Payrastre, Biochem. Pharmacol. 53 (1997) 1649^1657. [13] E.A. Eaton, U.K. Walle, A.J. Lewis, T. Hudson, A.A. Wilson, T. Walle, Drug. Metab. Dispos. 24 (1996) 232^237. [14] T.P. Devasagayam, M. Subramanian, B.B. Singh, R. Ramanathan, N.P. Das, J. Photochem. Photobiol. B 30 (1995) 97^103. [15] G. Prota, J. Invest. Dermatol. 100 (1993) 156S^161S. [16] V. del Marmol, F. Beermaann, FEBS Lett. 381 (1996) 165^168. [17] C.W.G. van Gelder, W.H. Flurkey, H.J. Wichers, Phytochemistry 45 (1997) 1309^1323. [18] P. Wesche-Ebeling, M.W. Montgomery, J. Food Sci. 55 (1990) 731^734. [19] P. Sarni-Manchado, V. Cheynier, M. Moutounet, Phytochemistry 45 (1997) 1365^1369. [20] A. Finger, J. Sci. Food Agric. 66 (1994) 293^305. [21] E. Nun¬ez-Delicado, R. Bru, A. Sa¨nchez-Ferrer, F. Garc|¨aCarmona, J. Chromatogr. B. Biomed. Appl. 680 (1996) 105^ 112. [22] J. Escribano, M.A. Pedren¬o, F. Garc|¨a-Carmona, R. Mun¬oz, Phytochem. Anal. 9 (1998) 124^127. [23] W. Brand-Williams, M.E. Cuvelier, C. Berset, Lebensm. Wiss u. Technol. 28 (1995) 25^30. [24] M.M. Bradford, Anal. Biochem. 72 (1976) 248^254. [25] T. Hayashi, K. Sawa, M. Kawasaki, M. Arisawa, M. Shimizu, N. Morita, J. Nat. Prod. 51 (1998) 345^348. [26] J. Robak, R.J. Gryglewski, Biochem. Pharmacol. 37 (1998) 837^841. [27] S. Chazarra, J. Cabanes, J. Escribano, F. Garc|¨a-Carmona, J. Agric. Food Chem. 44 (1996) 984^988. [28] A. Sa¨nchez-Ferrer, R. Bru, F. Garc|¨a-Carmona, Anal. Biochem. 184 (1990) 279^282. [29] M. Jime¨nez, F. Garc|¨a-Carmona, Biochem. Int. 25 (1991) 861^868. [30] M. Jime¨nez, F. Garc|¨a-Carmona, Plant. Physiol. Biochem. 31 (1993) 541^546.
BBAGEN 24701 19-11-98
542
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
[31] M. Jime¨nez, F. Garc|¨a-Carmona, Phytochem. Anal. 4 (1993) 149^151. [32] A.M. Moore, W.H. Flurkey, J. Biol. Chem. 265 (1990) 4982^4990. [33] M. Jime¨nez, F. Garc|¨a-Carmona, Arch. Biochem. Biophys. 331 (1996) 15^22. [34] R. Iyengar, C.W. Bohmont, A. McEvily, J. Food Compos. Anal. 4 (1991) 148^157. [35] V.H. Frankos, D.F. Schmitt, L.C. Haws, A.J. McEvily, R. Iyengar, S.A. Miller, Regul. Toxicol. Pharmacol. 14 (1991) 202^212. [36] J.A. McEvily, R. Iyengar, W.S. Otwell, Crit. Rev. Food Sci. Nutr. 32 (1992) 253^273.
[37] M. Jime¨nez, F. Garc|¨a-Carmona, J. Agric. Food Chem. 45 (1997) 2061^2065. [38] J.W. Williams, J.E. Morrison, Methods Enzymol. 63 (1979) 437^467. [39] J. Cabanes, F. Garc|¨a-Ca¨novas, J. Tudela, J.A. Lozano, F. Garc|¨a-Carmona, Phytochemistry 26 (1987) 917^919. [40] E. Valero, M. Garc|¨a-Moreno, R. Varo¨n, F. Garc|¨a-Carmona, J. Agric. Food Chem. 39 (1991) 1043^1046. [41] J. Cabanes, S. Chazarra, F. Garc|¨a-Carmona, J. Pharm. Pharmacol. 46 (1994) 982^985. [42] A. Sa¨nchez-Ferrer, F. Laveda, F. Garc|¨a-Carmona, J. Agric. Food Chem. 41 (1993) 1219^1224.
BBAGEN 24701 19-11-98
Oxidation of the £avonol ¢setin by polyphenol oxidase Mercedes Jime¨nez *, Josefa Escribano-Cebria¨n, Francisco Garc|¨a-Carmona Departamento de Bioqu|¨mica y Biolog|¨a Molecular A, Unidad Docente de Biolog|¨a, Facultad de Veterinaria, Universidad de Murcia, E-30080 Murcia, Spain Received 9 July 1998; received in revised form 18 September 1998; accepted 23 September 1998
Abstract The present study demonstrates the antiradical efficiency of fisetin, a flavonol widely distributed in fruits and vegetables, by its ability to react with two different free radicals, ABTSc and DPPHc . The polyphenolic nature of fisetin led us to consider whether it might be oxidised by polyphenol oxidase (PPO), and the results reported show that it can be oxidised by PPO extracted and partially purified from broad bean seeds. The reaction was followed by recording spectral changes with time, with maximal spectral changes being observed at 282 nm (increase in absorbance) and at 362 nm (decrease). The presence of two isosbectic points (at 265 and 304 nm) suggested that only one absorbent product was formed. These spectral changes were not observed in the absence of PPO. The oxidation rate varied with the pH, reaching its highest value at pH 5.5. The fisetin oxidation rate increased in the presence of sodium dodecyl sulfate, an activator of polyphenol oxidase. Maximal activity was obtained at 0.87 mM sodium dodecyl sulfate. The following kinetic parameters were determined: Vmax = 49 WM/ min, Km = 0.6 mM, Vmax /Km = 8.2U1032 min31 . Flavonol oxidation was inhibited by selective PPO inhibitors such as cinnamic acid (a classical competitive inhibitor, Ki = 1.4 mM) and 4-hexylresorcinol, which behaved as a slow-binding inhibitor. The results reported show that fisetin oxidation was strictly dependent on the presence of polyphenol oxidase. ß 1998 Elsevier Science B.V. All rights reserved. Keywords: Polyphenol oxidase; Flavonol; Fisetin ; Antioxidant
1. Introduction There is a considerable amount of evidence that points to an association between diets rich in fresh fruits and vegetables and a decreased risk of cardiovascular disease and certain forms of cancer. It is generally assumed that the active dietary constituents Abbreviations: ABTS , 2,2P-azino-bis(3-ethylbenzthiazoline-6sulfonic acid); ABTSc , (2,2P-azino-bis(3-ethylbenzthiazoline-6sulfonic acid) radical; DPPHc , 2,2P-diphenyl-1-picrylhydrazyl radical * Corresponding author. Fax: +34-968-364147; E-mail: [email protected]
contributing to these protective e¡ects are the antioxidants (vitamins, carotenoids, sterols, polyphenols). Natural polyphenolic antioxidants are increasingly attracting the attention of food scientists because of their bene¢cial e¡ects on human health and food preservation. Oxidation is one of the most important processes of food deterioration because it may a¡ect food safety, colour, £avour and texture. Antioxidants may protect food quality by preventing the oxidative deterioration of lipids [1]. Recent works highlight the role of one class of natural phenols, known as £avonoids, and their bene¢cial e¡ects on human health due to their wide dis-
0304-4165 / 98 / $ ^ see front matter ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 8 ) 0 0 1 0 8 - 1
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
tribution in edible plants and in plant-derived foodstu¡s, which make a substantial contribution to the human diet. Flavonoids have multiple chemical and biological actions, including antioxidant activity. Their antioxidant properties reside in their radical scavenging activity. The position and the degree of hydroxylation are of primary importance in determining the strength of their antioxidant activity. It is known that £avonoids with a 3P,4P dihydroxy con¢guration possess antioxidant activity [2^4]. Fisetin is a particularly interesting £avonoid because of its presence in human foods and its biological activity. That it has antioxidant activity is shown by the fact that it inhibited human low-density lipoprotein (LDL) oxidation in vitro [5]. Fisetin shows a dose- and time-dependent increase in lipolysis, which is synergistic with epinephrine on L-adrenergic receptors [6]. Fisetin has also been reported as antiin£ammatory. It is a potent inhibitor of lysosomal enzyme secretion and arachidonic acid release in rat neutrophils [7]. It also inhibits mammalian 5-lipoxygenase and cyclooxygenase enzymes [8]. The in vivo e¡ect of ¢setin on the inhibition of a£atoxin B1 carcinogenicity and, hence, in the detoxi¢cation processes, has been described [9]. Fisetin completely inhibited thrombin- and cathepsin G-induced platelet aggregation [10]. Its antimutagenicity [11] and its antiproliferative e¡ects on certain malignant cells are thought to be due to the inhibition of phosphatidylinositol 3-kinase and protein kinase C [12]. Fisetin is a potent inhibitor of the human P-form phenolsulfonyltransferase, suggesting it may play a role as chemopreventive agent in sulfation-induced carcinogenesis [13]. Moreover, the protection of plasmid pBR322 DNA by ¢setin against single-stranded break induced by single molecular oxygen has also been shown [14]. The polyphenolic nature of ¢setin and, in particular, the fact that it is an o-dihydroxylated compound led us to suggest its potential oxidation by polyphenol oxidase (PPO) (EC 1.14.18.1), which is widely distributed in nature and responsible for melanisation in animals [15,16] and browning in plants [17]. The enzyme catalyses two distinct reactions involving molecular oxygen: (a) the o-hydroxylation of monophenols and (b) the oxidation of odiphenols to o-quinones. The PPO oxidation of monophenols and o-diphenols with simple chemical
535
structures has been extensively studied, although some of these substrates are not even natural compounds. Investigations carried out in vitro suggest that PPO may participate in the degradation of natural phenols with complex structures, such as anthocyanins in strawberries [18] and grapes [19], and the £avonols present in tea leaves [20]. These compounds are not directly oxidised by the enzyme but by the quinones formed by PPO from catechol, chlorogenic acid or catechin [18]. In spite of the increasing interest shown in the antioxidant properties of ¢setin, there have been no studies analysing its radical scavenging activity or its potential oxidation by polyphenol oxidase. Therefore, the aim of this work was to study the PPOcatalysed oxidation of ¢setin, which is shown, by the two di¡erent methods used in this paper to evaluate its antiradical capacity, to be an e¡ective radical scavenger. The results obtained con¢rm that PPO is capable of oxidising ¢setin directly. The spectrophotometric and kinetic characteristics of this oxidation were investigated, and the speci¢city of the reaction was demonstrated by the use of speci¢c PPO inhibitors. The oxidation of ¢setin by PPO in vitro poses questions about the possible involvement of the enzyme in the oxidation of natural polyphenols in vivo which would cause the loss of antioxidant activity of fruits and vegetables. 2. Materials and methods 2.1. Reagents Broad bean seeds were purchased from a local market in the city of Murcia. Fisetin hydrate (3,3P,4P,7-tetrahydroxy£avone), DPPHc (2,2P-diphenyl-1-picrylhydrazyl radical) and 4-hexylresorcinol were purchased from Aldrich (Madrid, Spain). ABTS (2,2P-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid), cinnamic acid and horseradish peroxidase (HRP, type VI; Rz = 3.2) were obtained from Sigma Quimica (Madrid, Spain). TX-114 was obtained from Fluka Quimica (Madrid, Spain) and condensed three times as described [21] but using 100 mM sodium phosphate bu¡er (pH 7.3). The detergent phase of the third condensation had a TX-
BBAGEN 24701 19-11-98
536
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
114 concentration of 25% (w/v) and was used as the stock detergent solution for all of the experiments. 2.2. Measurement of the antiradical activity of ¢setin The antiradical activity of ¢setin was measured by the disappearance of 2,2P-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) radical (ABTSc ) [22] ABTSc was generated in a reaction medium (1 ml) containing 2 mM ABTS , 35 WM H2 O2 and 31 nM horseradish peroxidase (EC 1.11.1.7) in 50 mM sodium acetate bu¡er, pH 5.0. ABTSc accumulation was followed by measuring the increase in absorbance at 420 nm, a bandwidth at which ¢setin does not show any appreciable absorption. ABTSc concentration was calculated by using an O 420 = 31 100 M31 cm31 . Once ABTS oxidation had ceased because of the complete depletion of H2 O2 (end point of the reaction), ¢setin was added at di¡erent concentrations. After addition, the absorbance at 420 nm decreased until constant absorbance was asymptotically reached. For each ¢setin concentration tested, the percentage of ABTSc remaining was determined and the values transferred to another graph showing the percentage of residual ABTSc as a function of the molar ratio of ¢setin to ABTSc . Another method used to determine the free radical scavenging activity of antioxidants was their reaction with the free radical DPPHc , according to the procedure described by Brand-Williams et al. [23]. The assay medium contained 60 WM DPPHc in methanol and di¡erent concentrations of a methanol ¢setin solution. The reaction was monitored by following the decrease in absorbance at 515 nm until the reaction reached a plateau. The percentage of DPPHc remaining as a function of the molar ratio of ¢setin to DPPHc was calculated similarly to the way used for the ABTSc method. 2.3. Enzyme extraction Polyphenol oxidase was extracted in its latent state from the chloroplast membranes of broad bean (Vicia faba) seeds. All procedures were carried out at 4³C. Chloroplasts were prepared by homogenising 50 g of broad bean seeds in 150 ml of sodium phos-
phate bu¡er containing 0.33 M sorbitol, 2 mM EDTA, 1 mM MgCl2 , and serine protease inhibitors, which were added immediately before use (1 mM phenylmethanesulfonyl £uoride (PMSF) and 1 mM benzamidine hydrochloride). The slurry was ¢ltered through eight layers of gauze and centrifuged at 600Ug for 2 min. The pellet was discarded, and the supernatant was centrifuged at 20 000Ug for 30 min. The resultant pellet was resuspended in a solution containing 50 ml of 10 mM sodium phosphate bu¡er (pH 7.3) and kept at 4³C for 20 min. Then, the solution was centrifuged at 20 000Ug for 20 min, thus pelleting the chloroplast membranes. These membranes were resuspended with 20 ml of 1.5% (w/v) TX-114 in 100 mM phosphate bu¡er (pH 7.3) for 30 min at 4³C. After high-speed centrifugation (60 000Ug for 20 min), this light-green extract yielded a clear supernatant with PPO activity. This was subjected to temperature phase partitioning by adding TX-114 at 4³C to give a ¢nal concentration of 8% (w/v). The mixture was kept at 4³C for 15 min in a Tectron thermostatic bath and then warmed to 37³C. After 15 min, the solution became spontaneously turbid due to the formation, aggregation and precipitation of large micelles composed of detergent, hydrophobic proteins and the remaining chlorophylls. This solution was centrifuged at 5000Ug for 10 min at room temperature. The detergent-rich phase was discarded and the clear supernatant containing the soluble PPO brought to 45% saturation with saturated (NH4 )2 SO4 (pH 7.0), and kept overnight, under continuous stirring, at 4³C. Then, the solution was centrifuged at 120 000Ug for 30 min at 4³C and the pellet was discarded. Saturated (NH4 )2 SO4 (pH 7.0) was added to the clean supernatant to give 85% saturation and stirred for 1 h at 4³C. The precipitate obtained between 45% and 85% was collected by centrifugation at the same rotor speed and dissolved in a minimal volume of 10 mM sodium phosphate bu¡er, pH 7.3. The salt content was removed by dialysis against 10 mM sodium phosphate bu¡er, pH 7.3. The solution thus obtained was used as enzyme source. To avoid any possible activation of the enzyme by endogenous proteases, PMSF and benzamidine hydrochloride were added before and after dialysis to give a ¢nal concentration of 1 mM.
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
2.4. Enzyme assays Unless otherwise noted, the spectrophotometric assays were performed at 25³C in a reaction medium (1 ml) containing 75 WM ¢setin and 0.87 mM sodium dodecyl sulfate (SDS) in 0.1 M sodium phosphate bu¡er, pH 5.5. The reaction was initiated by the addition of polyphenol oxidase. Fisetin oxidation was monitored by measuring the increase in absorbance at 282 nm or the decrease at 362 nm. Di¡erences between the extinction coe¤cients (vO), corresponding to the di¡erence between absorption due to substrate and that due to reaction products, were determined for the oxidation process. The vO values were calculated from a calibration curve of ¢setin (0^12 WM) quantitatively oxidised in the presence of an excess of sodium periodate at di¡erent pH values. The increases in absorbance at 282 nm or the decreases at 362 from t = 0 to constant absorbance values were plotted against ¢setin concentration. The values found, at pH 5.5, were vO282 = 5345 M31 cm31 (r = 0.997) and vO362 = 18 261 M31 cm31 (r = 0.999). pH studies were carried out using 0.1 M sodium acetate and sodium phosphate bu¡ers from pH 3.5 to pH 7, in the presence of 0.87 mM SDS. The pH of the assay solution was determined at room temperature, using a Crison micro pH 2002 meter. After catalysis, the pH of the assay solution was again measured.
537
bu¡er, pH 5.5. The reaction was started by adding the enzyme. 2.7. Other methods The protein content was determined according to the Bradford Bio-Rad protein assay using bovine serum albumin as a standard [24]. 3. Results and discussion Flavonoids are, in general, strong antioxidants, their antioxidant potential residing in their radical scavenging capacity. Several methods to evaluate antiradical activity have been reported [25,26] with those involving the reaction with the ABTSc [22] or DPPHc [23] radicals receiving most attention. We used both methods to demonstrate that ¢setin is an e¡ective radical scavenger. The so-called ABTSc method is based on measuring the destruction of the ABTSc radical due to its reaction with an antioxidant compound [22]. The ABTSc was ¢rst generated in the assay medium by
2.5. Optimal activation by SDS Activation studies were carried out in a reaction medium (1 ml) containing 75 WM ¢setin and di¡erent detergent concentrations in 0.1 M sodium phosphate bu¡er, pH 5.5. The concentration of SDS varied from 0 to 5.2 mM. 2.6. Inhibition studies Inhibition of ¢setin oxidation by polyphenol oxidase was followed at 372 nm and 25³C by using 4hexylresorcinol and cinnamic acid as speci¢c inhibitors of the enzyme. Unless otherwise stated, the reaction medium (1.0 ml) contained ¢setin (18^90 WM), 4-hexylresorcinol (0^1.5 mM) or cinnamic acid (0^5 mM) and 0.87 mM SDS in 0.1 M sodium phosphate
Fig. 1. Disappearance of ABTSc as a function of the number of moles of ¢setin/mol ABTSc . The percentage was calculated by the formula ((vAmax 3vA)/vAmax )U100, where vAmax is the maximum decrease in absorbance obtained at a high molar ratio and vA the maximum decrease in absorbance at each molar ratio. Inset: structure of ¢setin.
BBAGEN 24701 19-11-98
538
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
the horseradish peroxidase (HRP)/H2 O2 -mediated oxidation of ABTS . After ABTSc formation had ceased (due to the complete depletion of H2 O2 ), different concentrations of ¢setin were added to the reaction medium. Following the addition of ¢setin, the disappearance of ABTSc was followed by measuring the decrease in absorbance at 420 nm until a plateau was reached (data not shown). For each ¢setin concentration used, the percentage of ABTSc remaining was determined, and the values plotted against the molar ratio of ¢setin to ABTSc (Fig. 1). The concentration of antioxidant needed to decrease the initial radical concentration by 50% (e¤cient concentration = EC50 (mol/l antioxidant6mol/l ABTSc ) is a widely used parameter to measure antioxidant power [23]. The lower the EC50 , the higher the antioxidant power. This ratio, calculated from the results obtained in Fig. 1, was 0.04. The antiradical activity of ¢setin was also tested by observing its reaction with the stable free radical DPPHc . A similar procedure to that previously explained for the ABTSc was used (data not shown). From the results obtained (data not shown) an EC50 value of 0.05 was determined. Thus, the EC50 values obtained by both methods were very similar. Moreover, this EC50 value calculated for ¢setin was comparable to that determined for other polyphenols with a very high antiradical activity, such as gentisic and gallic acids [23]. The antioxidant activity of ¢setin is related to its structure (shown in Fig. 1). The o-dihydroxylation of the B ring contributes to the antioxidant activity [2^ 4]. This ¢setin hydroxylation suggests its potential oxidation by polyphenol oxidase. Fisetin oxidation by PPO extracted from broad bean seeds was followed by observing changes in the UV-visible spectrum with time (Fig. 2). Maximal spectral changes were observed at 362 nm (decreases in absorbance) and at 282 nm (increases in absorbance). The presence of two isosbectic points at 265 and 304 nm (arrowheads) suggests that only one absorbent product is formed during the course of ¢setin oxidation by PPO. These spectral changes were not observed in the absence of the enzyme and, therefore, they were considered to be the result of enzyme activity. The nature of the spectral changes during ¢setin oxidation indicated that the formation of reaction
Fig. 2. Consecutive spectra, at 25³C, obtained in the oxidation of ¢setin by broad bean seed polyphenol oxidase. The assay medium (1.0 ml) contained 75 WM quercetin and 0.87 mM SDS in 0.1 M sodium phosphate bu¡er, pH 5.5. The reaction was started by addition of the enzyme (33 Wg/ml). Scan speed was at 1-min intervals for 10 min.
product was proportional to time during the ¢rst 10 min of the reaction. This observation and the dependence of these spectral changes on the presence of the enzyme suggest that such changes are a reliable measure of ¢setin oxidation. Therefore, £avonol oxidation by PPO was routinely assayed by measuring either the decrease in absorbance at 362 nm or the increase at 282 nm vs. reaction time. In order to characterise ¢setin oxidation by PPO, the e¡ect of pH was analysed (not shown). First, the vO were determined, as described in Section 2, for all the di¡erent pH values assayed. These vO values were seen to be pH-dependent, and this variation was taken into account when calculating the ¢setin oxidation rate at the di¡erent pH values. The oxidation rate varied with pH, the optimal value being 5.5. Fisetin oxidation was monitored by measuring the decrease in absorbance at 362 nm. Similar results were obtained when the activity was monitored by following the increase in absorbance at 282 nm. Polyphenol oxidase can be found in an inactive or latent state. To obtain the enzyme in its latent form, a very mild extraction method must be used to avoid the modi¢cation or oxidation of PPO produced by
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
Fig. 3. E¡ect of SDS on the rate of ¢setin oxidation catalysed by polyphenol oxidase. The reaction medium, at 25³C, contained 75 WM ¢setin and 33 Wg/ml enzyme in 0.1 M sodium phosphate bu¡er, pH 5.5. The ¢setin oxidation rate was monitored by measuring the decrease in absorbance at 362 nm.
acetone powers [27]. The method used to extract and partially purify polyphenol oxidase from broad bean seeds was used to purify PPO in latent state from di¡erent sources [28]. Latent PPO can be released from latency or activated by a variety of treatments or agents including acid or basic shock, proteolytic attack, phospholipid release [17], polyamines [29], polyglucans [30], divalent cations [31] and anionic detergents [17,32,33]. SDS activation is particularly interesting because few enzymes are known to be activated by this anionic detergent. Thus, PPO is activated at high SDS concentrations, which would denature many other enzymes [32]. The activation brought about by several seemingly unrelated substances is a phenomenon often attributed to conformational changes of the enzyme molecule, solubilisation of the enzyme or the removal of an inhibitor. In the case of SDS, it was shown that the active site is blocked in absence of the detergent but open in its presence [17]. Fisetin oxidation by polyphenol oxidase extracted from broad bean seeds also increased in the presence of SDS (Fig. 3). Scans of the SDS concentration carried out at pH 5.5 showed the optimum SDS concentration to be 0.87 mM, since higher detergent
539
concentrations led to a decrease in enzymatic activity. To further study ¢setin oxidation by PPO, the dependence of the oxidation rate on £avonol concentration was examined. The enzyme showed Michaelis^Menten type kinetics (data not shown), although enzyme saturation by the substrate was not reached in the concentration range assayed. It was not possible to further increase ¢setin concentration because of its high absorbance at the measurement wavelength, despite the fact that spectrophotometric cells with a path length of 5 mm (half of the standard length) were used to decrease the initial absorbance values. Hanes^Woolf plots for the kinetic data of ¢setin oxidation resulted in linear relationships (not shown), from which the kinetic parameters were evaluated (Vmax = 49 WM/min, Km = 0.6 mM, Vmax / Km = 8.2U1032 min31 ). The spectral changes for ¢setin oxidation shown in Fig. 2 were completely dependent on the presence of the active enzyme. To further prove the speci¢city of the £avonol oxidation by PPO, the e¡ect of selective inhibitors was examined. A considerable number of polyphenol oxidase inhibitors is known, one large group being constituted by compounds, such as benzoic acid and their derivatives, which are structurally analogous to phenolic substrates and which generally show competitive inhibition with respect to these substrates. Other recently recognised PPO inhibitors include 4-substituted resorcinols, which are also structurally related to phenolic substrates. 4-Hexylresorcinol has been shown to be e¡ective in preventing black spot formation in shrimps [34] and for browning control in di¡erent fruits [35]. 4-Hexylresorcinol and other 4-substituted resorcinols are also very effective inhibitors of mushroom tyrosinase [36,37], inhibiting the enzyme in a non-classical manner and classi¢ed as slow competitive inhibitors [37]. An inhibition study of ¢setin oxidation was, therefore, carried out with 4-hexylresorcinol. A prolonged transient phase was observed for the inhibition of PPO (data not shown), and thus, 4-hexylresorcinol presents the characteristics of a competitive slowbinding inhibitor of polyphenol oxidase according to Williams and Morrison classi¢cation [38]. Fig. 4 shows the e¡ect of the inhibitor on the initial (A) and steady-state (B) rates, respectively, at di¡erent concentrations of substrate. Initial velocities (Vo ) de-
BBAGEN 24701 19-11-98
540
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
(Fig. 4B), a value of 0.05 mM was determined for 0 the overall dissociation constant (K i ). All these e¡ects of 4-hexylresorcinol on PPO were similar to those obtained for the slow inhibition of tyrosinase by L-mimosine [39], tropolone [40] and kojic acid [41]. A mechanism to explain the behaviour of these time-dependent inhibitors has been previously postulated [39]. This involves the rapid formation of a complex between the inhibitor and the oxy form of the enzyme, a complex that subsequently undergoes a relatively slow reversible reaction. According to the kinetic analysis of this mechanism the following equation was obtained: kapp k36
k6 I K i 1 S=K m I
Fig. 5 shows 1= kapp 3k36 vs. 1/[I], for three different ¢setin concentrations. From the intercept with the ordinate axis (1/k6 ), a value of 0.024 s31 was obtained for k6 ,the constant for the slow transition process of the inhibitor^enzyme complex. An inhibition study of ¢setin oxidation was also carried out with cinnamic acid, a commonly used tyrosinase inhibitor that has been identi¢ed as a competitive inhibitor for PPO from di¡erent sources [42]. Flavonol oxidation was also inhibited by the
Fig. 4. Dixon plots for the inhibitory e¡ect of 4-hexylresorcinol on initial (A) and steady-state (B) rates of ¢setin oxidation catalysed by PPO. The reaction medium contained £avonol and inhibitor at the indicated concentrations, 0.87 mM SDS and 94 Wg/ml enzyme. The ¢setin concentrations used were: (b) 90, (a) 75, (F) 60, (E) 45, (R) 30 and (O) 18 WM.
creased with inhibitor concentration, indicating that an enzyme^inhibitor complex is rapidly formed with an apparent dissociation constant (Ki ) of 0.3 mM. The steady-state rates (Vs ) also decreased as 4-hexylresorcinol concentration rose, and when Dixon plots were made at di¡erent ¢setin concentrations
Fig. 5. Graphical calculation of k6 for the inhibition of ¢setin oxidation by 4-hexylresorcinol. Experimental conditions were as in Fig. 4. Fisetin concentrations were: (b) 90, (E) 45 and (R) 30 WM.
BBAGEN 24701 19-11-98
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
presence of cinnamic acid in the reaction medium, although this inhibitor behaved as a classical competitive inhibitor. From the Dixon plots (data not shown), a Ki value of 1.4 mM was determined. This paper therefore shows the antioxidative potential of ¢setin since the £avonol acted as an e¡ective radical scavenger, as proved by its reaction with the free radicals ABTSc and DPPHc . These results agree with previous studies reporting the e¤ciency of ¢setin in inhibiting human LDL oxidation in vitro [5]. We can also conclude that ¢setin is oxidised by polyphenol oxidase. This oxidation is completely dependent on the presence of active enzyme because the reaction was inhibited by speci¢c PPO inhibitors such as 4-hexylresorcinol and cinnamic acid. There have been reports suggesting that PPO could play a role in the degradation of polyphenols with complex structures, such as anthocyanin pigments from grapes and strawberries [18,19]. However, direct oxidation by PPO did not appear to be the main decolourisation route. Quinones and the intermediary compounds formed during the oxidation by PPO of catechol, catechin or chlorogenic acid might be responsible for the destruction of anthocyanins either through oxidation or co-polymerisation. The fact that PPO oxidises ¢setin in vitro poses questions about the enzyme's potential involvement in the oxidation of natural polyphenols in vivo. Future studies should be aimed at ascertaining whether or not PPO is involved in the loss of antioxidant activity due to the oxidation of phenolic compounds, particularly during the improper handling or storage of fruits and vegetables. Acknowledgements This work was supported by a grant from the Comisio¨n Interministerial de Ciencia y Tecnolog|¨a (Spain), DGES project PB95-1024. References [1] R.A. Jacob, Nutr. Res. 15 (1995) 755^766. [2] C.A. Rice-Evans, N.J. Miller, G. Paganga, Free Radic. Biol. Med. 20 (1996) 933^956.
541
[3] S.A. van Acker, D.J. van der Berg, M.N. Tromp, D.H. Gri¤oen, W.P. van Bennekom, W.J. van der Vijgh, A. Bast, Free Radic. Biol. Med. 20 (1996) 331^342. [4] S.A. van Acker, M.J. Groot, D.J. van der Berg, M.N. Tromp, G. Donne-Op den Kelder, W.J. van der Vijgh, A. Bast, Chem. Res. Toxicol. 9 (1996) 1305^1312. [5] I. Myara, I. Pico, B. Vedie, N. Moath, J. Pharmacol. Toxicol. Methods 30 (1993) 69^73. [6] U.R. Kappusamy, N.P. Das, Biochem. Pharmacol. 47 (1994) 521^529. [7] M. Tordera, M.L. Ferrandiz, M.J. Alcaraz, Z. Naturforsch. C 49 (1994) 235^240. [8] M.J. Laughton, P.J. Evans, M.A. Moroney, J.R. Hault, B. Halliwell, Biochem. Pharmacol. 42 (1991) 1673^1681. [9] V.S. Aboobaker, A.D. Balgi, R.K. Bhatacharya, In vivo 8 (1994) 1095^1098. [10] R.N. Puri, R.W. Colman, Anal. Biochem. 210 (1993) 50^57. [11] R. Edenhareder, I. Petersdor¡, R. Rauscher, Mutat. Res. 287 (1993) 261^274. [12] G. Agullo, L. Gamet-Payrastre, S. Manenti, C. Viala, C. Remesy, H. Chap, B. Payrastre, Biochem. Pharmacol. 53 (1997) 1649^1657. [13] E.A. Eaton, U.K. Walle, A.J. Lewis, T. Hudson, A.A. Wilson, T. Walle, Drug. Metab. Dispos. 24 (1996) 232^237. [14] T.P. Devasagayam, M. Subramanian, B.B. Singh, R. Ramanathan, N.P. Das, J. Photochem. Photobiol. B 30 (1995) 97^103. [15] G. Prota, J. Invest. Dermatol. 100 (1993) 156S^161S. [16] V. del Marmol, F. Beermaann, FEBS Lett. 381 (1996) 165^168. [17] C.W.G. van Gelder, W.H. Flurkey, H.J. Wichers, Phytochemistry 45 (1997) 1309^1323. [18] P. Wesche-Ebeling, M.W. Montgomery, J. Food Sci. 55 (1990) 731^734. [19] P. Sarni-Manchado, V. Cheynier, M. Moutounet, Phytochemistry 45 (1997) 1365^1369. [20] A. Finger, J. Sci. Food Agric. 66 (1994) 293^305. [21] E. Nun¬ez-Delicado, R. Bru, A. Sa¨nchez-Ferrer, F. Garc|¨aCarmona, J. Chromatogr. B. Biomed. Appl. 680 (1996) 105^ 112. [22] J. Escribano, M.A. Pedren¬o, F. Garc|¨a-Carmona, R. Mun¬oz, Phytochem. Anal. 9 (1998) 124^127. [23] W. Brand-Williams, M.E. Cuvelier, C. Berset, Lebensm. Wiss u. Technol. 28 (1995) 25^30. [24] M.M. Bradford, Anal. Biochem. 72 (1976) 248^254. [25] T. Hayashi, K. Sawa, M. Kawasaki, M. Arisawa, M. Shimizu, N. Morita, J. Nat. Prod. 51 (1998) 345^348. [26] J. Robak, R.J. Gryglewski, Biochem. Pharmacol. 37 (1998) 837^841. [27] S. Chazarra, J. Cabanes, J. Escribano, F. Garc|¨a-Carmona, J. Agric. Food Chem. 44 (1996) 984^988. [28] A. Sa¨nchez-Ferrer, R. Bru, F. Garc|¨a-Carmona, Anal. Biochem. 184 (1990) 279^282. [29] M. Jime¨nez, F. Garc|¨a-Carmona, Biochem. Int. 25 (1991) 861^868. [30] M. Jime¨nez, F. Garc|¨a-Carmona, Plant. Physiol. Biochem. 31 (1993) 541^546.
BBAGEN 24701 19-11-98
542
M. Jime¨nez et al. / Biochimica et Biophysica Acta 1425 (1998) 534^542
[31] M. Jime¨nez, F. Garc|¨a-Carmona, Phytochem. Anal. 4 (1993) 149^151. [32] A.M. Moore, W.H. Flurkey, J. Biol. Chem. 265 (1990) 4982^4990. [33] M. Jime¨nez, F. Garc|¨a-Carmona, Arch. Biochem. Biophys. 331 (1996) 15^22. [34] R. Iyengar, C.W. Bohmont, A. McEvily, J. Food Compos. Anal. 4 (1991) 148^157. [35] V.H. Frankos, D.F. Schmitt, L.C. Haws, A.J. McEvily, R. Iyengar, S.A. Miller, Regul. Toxicol. Pharmacol. 14 (1991) 202^212. [36] J.A. McEvily, R. Iyengar, W.S. Otwell, Crit. Rev. Food Sci. Nutr. 32 (1992) 253^273.
[37] M. Jime¨nez, F. Garc|¨a-Carmona, J. Agric. Food Chem. 45 (1997) 2061^2065. [38] J.W. Williams, J.E. Morrison, Methods Enzymol. 63 (1979) 437^467. [39] J. Cabanes, F. Garc|¨a-Ca¨novas, J. Tudela, J.A. Lozano, F. Garc|¨a-Carmona, Phytochemistry 26 (1987) 917^919. [40] E. Valero, M. Garc|¨a-Moreno, R. Varo¨n, F. Garc|¨a-Carmona, J. Agric. Food Chem. 39 (1991) 1043^1046. [41] J. Cabanes, S. Chazarra, F. Garc|¨a-Carmona, J. Pharm. Pharmacol. 46 (1994) 982^985. [42] A. Sa¨nchez-Ferrer, F. Laveda, F. Garc|¨a-Carmona, J. Agric. Food Chem. 41 (1993) 1219^1224.
BBAGEN 24701 19-11-98