The Effects of Natural Flavonoids on Lipoxygenase-Mediated Oxidation of Compounds with a Benzene Ring Structure—A New Possible Mechanism of Flavonoid Anti-Chemical Carcinogenesis and Other Toxicities

Abstract

Numerous studies have strongly suggested that flavonoids exhibit antimutagenic, anticarcinogenic, antiallergic, and anti-inflammatory properties, but the mechanism is still far from clear. In this study, the effect of natural flavonoid compounds, such as green tea polyphenol, epigallocatechin gallate, quercetin, and rutin on lipoxygenase-mediated co-oxidation of guaiacol, benzidine, paraphenylenediamine, and dimethoxybenzidine was investigated. Green tea polyphenol, epigallocatechin gallate, quercetin, and rutin can reduce the co-oxidation reaction speed of tested compounds mediated by soybean lipoxygenase and the production of oxidative products and free radical intermediates. Their median inhibition concentrations on guaiacol oxidation mediated by soybean lipoxygenase were 8.22 mg·L⁻¹, 17.8 μmol·L⁻¹, 41.5 μmol·L⁻¹, and 46.3 μmol·L⁻¹, respectively. These were all significantly lower than glutathione, dithiothreitol, butylated hydroxyanisole and gossypol. The data collected in this study suggest that flavonoids may have an anticarcinogenicity and antitoxicity effect through inhibition of oxidative activation.

Lipoxygenase (LOX), a nonheme iron-containing enzyme, universally exists in animals and plants. Studies have shown that, in addition to dioxidation of polyunsaturated fatty acid (PUFA), LOX demonstrates co-oxidation activity in exogenous compounds in the presence of (PUFA), fatty acid hydroperoxide, or hydrogen peroxide (Joseph, Srinivasan, and Kulkarni 1993; Roy and Kulkarni 1996; Perez-Gilabert, Sanchez-Ferrer, and Garcia-Carmona 1994). LOX generally shows relatively high activity in some extrahepatic tissues with little or no cytochrome P450 (e.g., lung, brain, heart, skin, uterus, placenta, fetal tissues, and blood cells) (Kulkarni 2001; Hover and Kulkarni 2000; Greenberg-Levy, Budowski, and Grossman 1993; Lomnitski, Sklan, and Grossman 1995; Straif et al. 2000; Tomimoto et al. 2002). Therefore, it is regarded as a likely substitute oxidative metabolic pathway for endogenous and exogenous compounds in extrahepatic tissues (Kulkarni 2001).

LOX-mediated oxidation of some endogenous and exogenous compounds can activate those compounds, which may be related to carcinogenicity and other toxicities. For example, LOX co-oxidation of benzo(a)pyrene-7,8-dihydrodiol (Joseph et al. 1994; Correa and Shiao 1994), aflatoxin B1 (Husgafvel-Pursiainen et al. 1995; Datta and Kulkarni 1994), and benzidine (Zenser et al. 2002; Yamazoe, Roth, and Kadlubar 1986; Roy and Kulkarni 1994) has been cited as a possible pathway for activation of these carcinogenic substances. The identification of compounds that inhibit the LOX-mediated oxidative process can be important in reducing toxicity and, specifically, carcinogenicity of certain substances. Recent studies have described the potential effects of modifier of the enzyme such as vitamin C, gossypol, nordihydroguaiaretic acid (NDGA), and butylated hydroxyanisole (BHA) (Joseph, Srinivasan, and Kulkarni 1993; Roy and Kulkarni 1996, 1999; Hu and Kulkarni 1998), on the LOX-mediated oxidative process of exogenous compounds.

Flavonoids are polyphenolic substances that are present in most plants—particularly seeds, fruit skin, bark, and flowers—and plant-based foods. Numerous studies have strongly suggested that flavonoids exhibit antimutagenic, anticarcinogenic, antiallergic, and anti-inflammatory properties (Ren et al. 2003; Maron 2004; Kris-Etherton et al. 2004; Kanazawa et al. 1998; Young, Hansen-Moller, and Oksbjerg 2004; Toker et al. 2004; Jankun, Selman, and Swiercz 1997). Because of their common occurrence, natural flavonoids such as tea polyphenols, EGCG, quercetin, and rutin have become the focus of a substantial amount of research (Hong et al. 2001; Hertog and Katan 1998; Bouriche et al. 2005). It has been reported that tea polyphenols (Hong et al. 2001), quercetin (Hertog and Katan 1998; Bouriche et al. 2005), and rutin (Bouriche et al. 2005) can inhibit the LOX-mediated dioxygenation of arachidonic acid, thus eliciting an anti-inflammatory effect and suppressing carcinogenesis in the colon. However, data on the effect of natural flavonoids on LOX-mediated co-oxidation are more limited. Only one study regarding lipoxygenase-catalyzed bioactivation of phenytoin was identified (Yu and Wells 1995).

In the present study, we used soybean lipoxygenase (SLO) as a model enzyme to investigate effects and mechanisms of tea polyphenols, quercetin, and rutin on lipoxygenase-mediated oxidation of various compounds having a benzene ring structure (e.g., benzidine). The hypothesis being tested was that the flavonoids tested would inhibit SLO-related oxidation of these potentially reactive substrates.

MATERIALS AND METHODS

Chemical

SLO (activity: 150,000 Sigma units per milligram of protein) was purchased from Sigma Aldrich Fluck Division (Seelze, Germany). Hydrogen peroxide (H₂O₂, 37%), benzidine, para-phenylenediamine (PPD), guaiacol, 3,3′-dimethoxybenzidine, Tetramethyl-p-phenylenediamine (TMPD), gossypol, reduced glutathione, butylated hydroxylanisole (BHA), dithiothreitol (DTT), EGCG, and rutin were also purchased from Sigma-Aldrich. Quercetin (>98%) was obtained from the National Institute for the Control of Pharmaceutical and Biological Products (NICPBP). Tea polyphenol (>95%) came from Hangzhou Zhongke natural plants company.

Basic Protocol

Tris HCL buffer (50 mmol·L⁻¹, pH 6.0) was added to, and mixed with a solution of 800 nmol·L⁻¹ of SLO and 20.00 mmol·L⁻¹ guaiacol. The mixture was maintained at 25°C for 2 min and then 1.00 mmol·L⁻¹H₂O₂ (final volume 1 ml) was added to begin the reaction. The reaction solution without enzymes served as the control. Temporal changes in absorbance by the oxidative product (tetramethyl benzcatechin) were observed at 470 nm, and the enzyme metabolism activity (nmol·min⁻¹·nmol⁻¹SLO) was calculated using molar absorptivity (26.6 × 10³ L·mol⁻¹·cm⁻¹) (Koduri and Tien 1995; Kulkarni and Cook 1988).

Inhibitory Test

Spectrophotometric Detection

Before adding substrate and H₂O₂, 800 nmol·L⁻¹ SLO in 50 mmol·L⁻¹ Tris buffer (pH 6.0) were mixed with the modifiers (gossypol, reduced glutathione, BHA, DTT, EGCG, or rutin) and incubated at 25°C for 3 min. Other procedures followed the above basic protocol. Temporal changes in absorbance of each substrate oxidative product were detected at the maximum absorption wavelength using a 2501PC ultraviolet spectrophotometer. Enzyme activity was calculated based on corresponding molar absorptivity, and the inhibition ratio was calculated using the enzyme activity of the reaction system with or without modifiers. The maximum absorption wavelength and molar absorptivity of each substrate oxidative product was 425 nm, 6.5 × 10⁴ L·mol⁻¹·cm⁻¹ of benzidine, 485 nm of PPD, 470 nm of 3,3′–dimethoxybenzidine, and 610 nm, 11.6 × 10³ L·mol⁻¹·cm⁻¹ of TMPD (Koduri and Tien 1995; Kulkarni and Cook 1988).

ESR Detection

Various modifiers were mixed with 800 nmol·L⁻¹ SLO in 50 mmol·L⁻¹ pH 6.0 Tris buffer. The solution was incubated at 25°C for 3 min, followed by the addition of 1.00 mmol·L⁻¹ PPD for an additional 2 min, and intensive mixing with 1.00 mmol·L⁻¹H₂O₂ (final volume = 1 ml). The representative spectrum was determined using a JES-FE1XG electron spin resonance (ESR) spectrometer. Analysis conditions were as follows: temperature: 20°C; field intensity: 3400 ± 250 T; magnification: 4 × 100; adjusting width: 2.5 G; response time: 0.3 s; microwave power: 40 mW; resonant frequency: 9.450 GHz; scan time: 16 min.

Statistical Analysis

Data are reported as the mean and standard deviation (X̄ ± SD). IC₅₀ valves were determined using the Probit method with SPSS10.0 software. Comparison of the means of two samples was performed with a t test and multisample mean comparison was performed with the Q test, Dunnett’s T3 single factor analysis of variance test. All statistical analyses were run at a 95% confidence level (α = .05).

RESULTS

Effects of Experimental Conditions on the SLO-Mediated Enzymatic Reaction

The optimal reaction condition was investigated using guaiacol as a typical substrate. The optimal SLO-mediated guaiacol, oxidative conditions were pH 6.0, 800.00 nmol·L⁻¹ SLO, 20.00 mmol·L⁻¹ guaiacol, and 3.00 mmol·L⁻¹ H₂O₂ (Figure 1). The V _max was 2.33 nmol·min⁻¹·nmol⁻¹ SLO (using the Lineweaver-Burke graphic method), and the K _M of guaiacol was 7.54 mmol·L⁻¹.

Effects of Natural Flavonoids on SLO-Mediated Guaiacol Oxidation

Natural flavonoids, including tea polyphenols, EGCG, quercetin, and rutin, reducers (GSH and DTT), free radical scavengers (BHA), and typical LOX inhibitors, can inhibit SLO-mediated oxidation of guaiacol. The inhibitory effect was manifested in a dose-dependent manner (Figure 2). Only the EGCG effect is illustrated in Figure 2; however, other modifiers demonstrated a similar effect. Based on the IC₅₀, the modifiers can be ranked as follows: EGCG < quercetin < rutin < GSH < DTT < BHA < gossypol. The IC₅₀ Values of natural flavonoids were significantly lower than those of frequently-used LOX activity modifiers, such as GSH, DTT, BHA, and gossypol. A similar result was also obtained by comparing with 95% confidence interval (CI) of the IC₅₀ Values of these modifiers (Table 1).

Results of spectrum scanning indicated that a time-dependent absorption peak emerged at 470 nm in SLO-mediated oxidation of guaiacol. That peak is similar to the absorption peak of 4-methylcatechol (an oxidative product of guaiacol) at 470 nm described previously (Kulkarni and Cook 1988) (Figure 3A ). The location of the reaction products absorption peak did not vary when different concentrations of modifiers were added to the reaction system. However, the height of the absorption peaks did decrease by different degrees (Figure 3B ). Only the EGCG scan is illustrated in Figure 3B ; however, scans from other modifiers were similar. This suggests that the decrease of time-dependent absorbance was induced by the decrease of SLO-mediated oxidative metabolism of guaiacol. The inhibition is likely caused by the modifiers, and the effect appears to be dose-dependent.

The Inhibitory Effect of Natural Flavonoids on SLO-Mediated Oxidation of Other Compounds with a Benzene Ring Structure

Natural flavonoids, such as tea polyphenols and its monomer EGCG, quercetin, and rutin also strongly inhibit the SLO-mediated oxidation of other compounds, including benzidine, a carcinogen. There was a dose-response relationship between the concentration of flavonoids and their oxidation inhibition. Analysis of variance indicated that when compared to quercetin, the IC₅₀ of EGCG (required for oxidation of four–benzene ring structure compounds) was significantly lower. The IC₅₀ of PPD was also lower than that of rutin. The EGCG IC₅₀ for oxidation of 3,3′-dimethoxybenzidine was lower than that of rutin, but not significantly so. We did not compare the benzidine oxidation IC₅₀ values of EGCG and rutin because the concentration (0.10 mmol·L⁻¹) of benzidine used in the inhibition experiment of rutin was lower than that used in other experiments. In addition, for the same flavonoid, there were significant differences among the oxidation IC₅₀ values of four different compounds (Table 2).

Results of ESR Detection

Results of ESR detection on SLO-mediated oxidation of PPD show that a free radical intermediate (radical cation of PPD) was generated during the oxidative process (Figure 4). Under the same experimental conditions, linewidth and line style with modifiers were similar to those without modifiers, indicating that the detected free radical was same or similar, and any decrease in spectrum peak height was reflective of the free radical intermediate. Given the same modifier concentration, the ESR peak height for EGCG and quercetin was relatively low, indicating that the inhibitory function was stronger than that of GSH, BHA, and gossypol.

DISCUSSION

LOX is a ubiquitous isozyme with iron ions as prosthetic group. Several studies indicate that the catalytic activity of SLO is similar to that of LOX in animal tissues (Kulkarni 2001), and therefore it has been widely used as a model enzyme in LOX research. It has been demonstrated that many kinds of LOX originating from various extrahepatic tissues of animals, including humans, can mediate co-oxidation of more than one hundred compounds (Kulkarni 2000, 2001). SLO can catalyze the oxidation of guaiacol and benzidine and displays characteristics typical of an enzymatic reaction in the presence of hydrogen peroxide. The results of this research are similar to those reported by Kulkarni and Cook (1988). Oxidative activation products (e.g., radicals of benzidine and benzidine diimine cation) generated in the LOX-mediated co-oxidation of guaiacol, benzidine, PPD, and dimethoxybenzidine are probably active forms of these xenobiotics and likely demonstrate substantial mutagenic, carcinogenic and other toxic characteristics (Hamaguchi and Tsutsui 2000; Hu and Kulkarni 2000; Vlckova et al. 1999; Wells et al. 1997; Savitsky et al. 1994).

This research suggests that tea polyphenols, EGCG, quercetin, and rutin could significantly decrease SLO-mediated co-oxidation of four compounds. Their effects were notably stronger than that of some commonly used LOX modifiers like GSH, DTT, gossypol, and BHA. The flavonoid-induced inhibition of oxidation was manifested not only in the decrease of oxidative end products and output speed, but also in a decrease in the generation of free radical intermediates. The decreases in the production rate and oxidative end products are probably due to the direct or indirect influence of flavonoids on certain metabolic links in the enzymatic oxidation of the tested compounds, thus affecting the generation of free radical intermediates. There may also be direct impacts on the intermediate products themselves. The actual mechanism that inhibits SLO-mediated oxidation of guaiacol and benzidine initiated by tea polyphenols, EGCG, quercetin, and rutin is not clear. Based on results of these studies and other research reported in the literature, we suspect that there may be two main mechanisms. First, the activity of SLO may be directly inhibited. Tea polyphenols, quercetin, and other flavonoids are reported to be direct inhibitors of LOX (Hong et al. 2001; Hu, Shen, and Pu 1999; Bouriche et al. 2005; Robak et al. 1988). They deactivate SLO by precipitating the apoenzyme, inhibiting enzymatic activity, and chelating Fe³⁺ on the enzyme active site (Morel, Lescoat, and Cillard 1994; Milic, Djilas, and Canadanovic-Brunet 1998). The sedimentation of the enzymatic agretope may be the main inhibitory mechanism (Kinjo et al. 2002). Second, flavonoids may remove free radicals generated as intermediate reaction products. Direct removal or neutralization may occur by the generation of a steady phenoxy free radical through dehydrogenation of the phenolic hydroxyl group. Results of ESR demonstrate that the free radical intermediate is generated during the oxidative process mediated by SLO. The decrease of free radical output through the addition of modifiers provides supportable evidence for the above inhibition mechanisms. In addition to inhibiting oxidative enzymes, flavonoids may also protect against the oxidative damage of chemical poisons in organisms by other mechanisms, for example, interacting with membrane (Oteiza et al. 2005).

This research suggests that the inhibition of SLO-mediated oxidation by tea polyphenols and EGCG with pyrogallol (especially EGCG with galloyl) on four compounds is stronger than that of quercetin and rutin with catechol structure. This result is coincident with other reported studies on the structure-activity relationship between the free radical capture activity of flavonoids and their structure (Kinjo et al. 2002). The specific effect trigger is the generation of two intramolecular hydrogen bonds after dehydrogenation of the pyrogallic acid-type phenolic hydroxyl group, so that the stabilization of the phenoxy-oxy radical is higher than that of the radical from the o-diphenolic hydroxyl group (Hongyu 1999).

Although all of the four kinds of flavonoids evaluated in the current investigation have strong inhibitory effects, the magnitude of the effect varies with the target compound. For example, rutin strongly inhibits (IC₅₀ < 50.00 μmol·L⁻¹) SLO-mediated oxidation of guaiacol, benzidine, and 3,3′-dimethoxybenzidine. But its inhibition of the oxidation of PPD is relatively weak, with an IC₅₀ of 461.9 μmol·L⁻¹. Tea polyphenols and EGCG also strongly inhibit SLO-mediated oxidation of PPD, but exhibit a very weak inhibition of the oxidation of TMPD (the IC₅₀ of tea polyphenols was as high as 0.90 g·L⁻¹, but no obvious inhibition was detected until the EGCG concentration increased to 500.00 μmol·L⁻¹). Differences in inhibition by flavonoids (like tea polyphenols) on LOX-mediated oxidation are significant between compounds. This may be due to the structural differences of these compounds or differences in the catalysis ability of SLO of the different compounds.

This research suggest that inhibition of the flavonoids on LOX-mediated co-oxidation of xenobiotics is a new possible mechanism of flavonoid anti–chemical cancinogenesis and other toxicities in the light of evidence that oxidative activation products generated in the LOX-mediated co-oxidation of guaiacol, benzidine, PPD, and dimethoxybenzidine are probably active forms of these xenobiotics and likely demonstrate substantial carcinogenic, teratogenic, mutagenic, and other toxic characteristics (Hamaguchi and Tsutsui 2000; Hu and Kulkarni 2000; Steele et al. 2000; Vlckova et al. 1999; Wells et al. 1997; Savitsky et al. 1994). Although this research shows that natural flavonoids significantly inhibit the oxidation of four compounds, additional in vivo studies are needed because laboratory conditions vary substantially from the internal environment.