Metabolites of dietary quercetin: Profile, isolation, identification, and antioxidant capacity

journal of functional foods 11 (2014) 121–129

Available at www.sciencedirect.com

ScienceDirect j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j ff

Metabolites of dietary quercetin: Profile, isolation, identification, and antioxidant capacity Wieslaw Wiczkowski *, Dorota Szawara-Nowak, Joanna Topolska, Katarzyna Olejarz, Henryk Zielin´ski, Mariusz K. Piskuła Institute of Animal Reproduction and Food Research of the Polish Academy of Sciences in Olsztyn, Tuwima 10, 10-748 Olsztyn, Poland

A R T I C L E

I N F O

A B S T R A C T

Article history:

The study of the profile, isolation and antioxidant capacity of pig quercetin metabolites after

Received 6 June 2014

onion dry skin intake was performed. Onion dry skin contain very high content of querce-

Received in revised form 18

tin, which occurred mainly in aglycone form (87%). Urine was collected from seventeen pigs

September 2014

before and within 2-6 hours after intake of onion dry skin providing quercetin. Quercetin

Accepted 24 September 2014

metabolites profile was analyzed by HPLC-MS/MS. Upon quercetin intake, quercetin was ab-

Available online

sorbed and occurred in pig urine in methylated, glucuronided, sulfated and combined derivatives. Among 12 quercetin metabolites identified, quercetin monoglucuronides were

Keywords:

predominant. Quercetin metabolites were isolated and purified by Amberlite XAD-16 column

Onion

chromatography and on HPLC C18 semi-preparative column. Quercetin metabolites mixture

Quercetin

showed higher radical-scavenging activities compared to native quercetin derivatives of onion

Pig quercetin metabolites

skin. Results indicated that the profile of quercetin metabolites in pig reflects that in a humans.

Isolation

Moreover, quercetin metabolites present in systemic circulation after quercetin intake still

HPLC-MS/MS

acted as antioxidants.

Antioxidant activity

1.

Introduction

In order to understand the biological activity of dietary quercetin, it is important to explore its fate in the organism. Since the bioavailability studies allow the determination of real exposure of the organism to quercetin, their results are an essential topic for other experiments aiming at the explanation of quercetin physiological functions (Ting, Jiang, Ho, & Huang, 2014). There have been a number of publications on the bioavailability of quercetin (Manach, Williamson, Morand,

© 2014 Elsevier Ltd. All rights reserved.

Scalbert, & Remesy, 2005). So far, it has been shown that quercetin is absorbed and occurs in animal and human blood plasma and urine in the form of conjugated derivatives (Terao, Murota, & Kawai, 2011). The extent of absorption, bioavailability and metabolism is highly dependent on the nature of quercetin derivatives consumed as well as on the food matrix used (Wiczkowski et al., 2008). In this respect, the pathways of quercetin fate after intake are not fully characterized and understood yet; however, a number of phenomena related to quercetin absorption, bioavailability and metabolism have already been described (Manach et al., 2005). Methylated, glucuronidated, sulfated, and combined conjugates of quercetin after the intake

* Corresponding author. Institute of Animal Reproduction and Food Research of the Polish Academy of Sciences in Olsztyn, Tuwima 10, 10-748 Olsztyn, Poland. Tel.: +48 895234606; fax: +48 895240124. E-mail address: [email protected] (W. Wiczkowski). http://dx.doi.org/10.1016/j.jff.2014.09.013 1756-4646/© 2014 Elsevier Ltd. All rights reserved.

122

journal of functional foods 11 (2014) 121–129

of food rich in quercetin glucosides and aglycone have been found (Terao et al., 2011). There have also been numerous studies on the biological activity of quercetin, where it is implied that quercetin present in food products has a protective potential against chronic degenerative diseases. Nonetheless, the mechanisms of this action are not completely explained. It is noteworthy that in most of the studies only quercetin 3-glucuronide was used (Bansal et al., 2012; Terao et al., 2011; Williamson, Barron, Shimoi, & Terao, 2005). To the best of our knowledge, in the previous studies related to the biological properties of quercetin metabolites, the strategy with dietary-relevant profile of quercetin metabolites has never been used. Although, as mentioned above, earlier studies indicated that after the intake of quercetin, the profile of quercetin metabolites in biological fluids was noted to consist of several compounds with different concentration (Graf et al., 2005; Mullen, Boitier, Stewart, & Crozier, 2004). Therefore, the main problem identified is connected with obtaining the mixture of quercetin metabolites, reflecting the profile of quercetin compounds in humans after quercetin consumption, as well as measuring its biological activities. Taking the above into account, determination of quercetin metabolites with their profile being close to that in human tissues, and subsequent isolation of these compounds are essential for further investigations explaining the physiological function of quercetin metabolites. Since previous studies indicated that onion is one of the best sources of bioavailable quercetin (de Vries, Hollman, Amersfoort, Olthof, & Katan, 2001; Hollman et al., 1997) and that aglycone of quercetin is better bioavailable than quercetin glucosides when consumed in the form of food matrix (Wiczkowski et al., 2008), in this study, dry skin of onion was used as a source of dietary quercetin. Because the digestive tract of a pig is similar to that of a human, this animal was chosen as the investigation model for production of quercetin metabolites. Four different assays were used for determination of antioxidant capacity of quercetin metabolites mixture in comparison to native quercetin derivatives found in onion skin.

2.

Material and methods

2.1.

Reagents

Tsushida Tsukuba (Japan) were used for identification and calculation.

2.2.

Dry skin was obtained from yellow common onion (Allium cepa L., Hercules variety) kindly provided by Yara Poland (Szczecin, Poland) and was used as quercetin source. The bulbs of onion were peeled off from the dried outer leaves. Next, dry skin (approximately 5 kg) was dried in the laboratory oven at a temperature of 40 °C. Upon drying, the samples obtained were pulverized and stored at −20 °C until analysis.

2.3.

Determination of quercetin in onion dry skin

Extraction and analysis of Q derivatives in onion were carried out as described previously by Wiczkowski et al. (2008). About 0.05 g of freeze-dried and pulverized onion dry skin was extracted by using 30 s sonication (VC 750, Sonics & Materials, Newtown, CT, USA) with 1 mL of 80% methanol. Subsequently, the mixture was vortexed for 30 s, again sonicated and vortexed, and centrifuged (Centrifuge 5415R, Eppendorf, Germany) for 10 min (13,200 × g at 4 °C). Supernatant was collected in a 5 mL flask. This step was repeated five times. Finally, before the analysis, the extract was centrifuged (20 min, 13,000 × g, 4 °C) and 5 µL of the extract was submitted to HPLCDAD system (Shimadzu, Kyoto, Japan) equipped with a 150 × 2.1 mm i.d. XBridge C18 3.5 µm column (Waters, Milford, MA, USA). The HPLC system consisted of two pumps (LC-10 ADVP), DAD detector (SPD-M10 AVP) set at 360 nm, autosampler (SIL-10 ADVP), column oven (CTO-10 ASVP) and system controller (SCL-10 AVP). All chromatographic determinations were performed at 45 °C with the flow rate of 0.23 mL/min. Q derivatives were eluted in gradient system composed of water/ formic acid (99:1, v/v, phase A) and acetonitrile/formic acid (99:1, v/v, phase B). Gradient was composed of: 10–20% B (0– 14 min), 20–80% B (14–25 min), 80–10% B (25–26 min), and 10% B (26–45 min). Compounds were identified by comparison of their retention times and UV-visible spectrum with standards and their quantity was calculated from HPLC-DAD peak area at 360 nm against appropriate external standard. The calibration curve (the range of 0.3–45 µM) was linear with a correlation coefficient of 0.98.

2.4. Reagents including acetonitrile, methanol, formic acid, ethyl acetate, sodium carbonate were purchased from Merck KGaA (Darmstadt, Germany). 2,2′-Azinobis(3-ethylbenzothiazoline6-sulfonic acid) diammonium salt (ABTS), 2,2-diphenyl-1picrylhydrazyl (DPPH), 6-hydroxy-2,5,7,8-tetramethylchroman2-carboxylic acid (Trolox) and Sulfatase type H-5 were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Hydrophilic condition (ACW) and lipophilic condition (ACL) kits (model no. 400.801) for the photochemiluminescence (PCL) assay were received from Analytik Jena AG (Jena, Germany). Water was purified with a Mili-Q system (Millipore, Bedford, MA, USA). Quercetin (Q), quercetin 3-glucoside (Q3G), isorhamnetin (iR), isorhamnetin 3-glucoside (iR3G) (Extrasynthese, Genay, France) and quercetin 4′,3-diglucoside (Q3,4′G) and quercetin 4′-glucoside kindly provided (Q4′G) by Dr. T.

Plant material

Animals, diets and study design

The study was conducted in accordance with the study protocol approved by the Ethical Committee of the University of Warmia and Mazury in Olsztyn (no 73/2010/N). Seventeen 6-month-old female pigs of Large White Landrace, each weighing approximately 100 kg, were used. Animals were kept in a light and temperature-controlled animal room of the Pig Breeding Farm in Dobrzejowice (Poland) in separate cages with a free access to tap water. Pigs followed a quercetin-free diet for 1 day prior to the experiment. After overnight fast, onion dry skin preparations providing 15 mg of Q (calculated as Q aglycone) per kg of body weight were mixed with pigs’ fodder and fed to each animal. Consumption of that mix was approximately 95%. Collection of urine was carried out before and within the period of 2–6 h after the Q intake. Finally, 7 L and 12 L of urine

journal of functional foods 11 (2014) 121–129

before and after the intake was collected, and 120 mL and 200 mL of 0.5% ascorbic acid were added as the preserving antioxidant agent, respectively. After being divided into 50 µL, 5 mL and 1 L portions, urine was frozen and stored at −70 °C until analyses.

2.5. Determination of quercetin and isorhamnetin aglycone in pig urine Analysis of Q and iR was determined according to Wiczkowski et al. (2008) on HPLC after enzymatic hydrolysis of their conjugates and extraction from urine. A 50 µL urine was mixed and incubated at 37 °C for 75 min with 50 µL of 0.2 mmol/L acetate buffer pH 5.5 containing 750 units of β-glucuronidase and 50 units of sulfatase from Sulfatase H-5. Released aglycones of Q and iR were extracted with 0.9 mL of ethyl acetate by triplicate 30 s sonication and 60 s vortexing. After centrifugation, supernatant was evaporated to dryness with a stream of nitrogen at 37 °C and dissolved in 100 µL of 60% methanol containing 0.1% formic acid. Next, 5 µL sample was subjected to HPLC-MS/MS (LC-20 Prominence, Shimadzu – QTRAP 5500, AB SCIEX, Vaughan, ON, Canada) analysis on 75 × 3.0 mm i.d. Kinetex C18 2.6 µm column (Phenomenex, Torrance, CA, USA) kept at 45 °C. Mobile phase flow rate was set to 0.3 mL/min and worked in gradient system composed of water/formic acid (99.9:0.1, v/v, phase A) and acetonitrile/formic acid (99.9:0.1, v/v, phase B). Gradient was as follows: 30–80% B (0–10 min), 80– 30% B (10–11 min) and 30% B (11–30 min). Qualitative and quantitative analyses were performed using MRM (Multiple Reaction Monitoring) method with the presence of the respective parent and daughter ion pairs, 301.1–151.1 m/z for Q and 315.1– 300.1 m/z for iR. Optimal MS/MS analyses were achieved under the following conditions: curtain gas: 20 L/min, collision gas: 9 L/min, ionspray voltage: −4500 V, temperature: 550 °C, 1 ion source gas: 55 L/min, 2 ion source gas: 70 L/min, declustering potential: −115 and −125 V, entrance potential: −11 V, collision energy: −30 and −40 eV, collision cell exit potential: −12 and −15 V, respectively.

2.6. Determination of quercetin metabolites profile in pig urine after onion dry skin intake Urine metabolites of Q were determined according to the modified method of Wiczkowski, Romaszko, and Piskuła (2010). Extraction of Q metabolites was carried out using Sep-Pak C18 cartridges (J.T. Baker, Phillipsburg, NJ, USA). After thawing, urine samples were diluted twice in 1% formic acid aqueous solution, centrifuged (20 min, 4 °C, 13,200 × g) and applied to SepPak C18 cartridges conditioned with methanol and 1% formic acid aqueous solution. After loading samples, cartridges were washed with 1% formic acid aqueous solution, followed by the elution of quercetin derivatives with 5% formic acid in methanol. The eluents were evaporated to dryness with a stream of nitrogen at 37 °C and dissolved in 60% methanol containing 0.1% formic acid. After centrifugation (20 min, 13,200 × g, 4 °C) 5 µL of the samples were submitted to HPLCMS/MS system equipped with a 150 × 2.1 mm i.d. XBridge C18 3.5 µm column (Waters). The HPLC system (LC-20 Prominence, Shimadzu) consisted of two pumps (LC-20AD), autosampler (SIL-20ACHT), column oven (CTO-10ASVP), degasser

123

(DGU-20A3) and system controller (CBM-20A) coupled with mass spectrometer (QTRAP 5500, AB SCIEX) consisting of a triple quadrupole, ion trap and ion source of electro-spray ionization (ESI) and controlled by the Analyst 1.5.1 software. All chromatographic determinations were performed at 45 °C with the flow rate of 0.23 mL/min. Quercetin metabolites were eluted in gradient system composed of water/formic acid (99.9:0.1, v/v, phase A) and acetonitrile/formic acid (99.9:0.1, v/v, phase A). Gradient was as follows: 30–80% B (0–10 min), 80–30% B (10–11 min) and 30% B (11–30 min). Qualitative analysis was performed based on the scanning of the biological matrix in negative mode. Also, scanning of fragmentation ions derived from the selected parent ion was conducted for the observation of all the ions formed by the disintegration of the parent ion. In addition, scanning of precursor ion and neutral particles was performed. Quercetin metabolites were identified based on the comparison of their retention times and MS/MS fragmentation spectrum (m/z values) with the previously published data (Graf et al., 2005; Mullen et al., 2004) and available standards. Quantitative analysis was made using Multiple Reaction Monitoring (MRM) method for appropriate external standards, Q4′,3G for Q and iR diglucuronide, Q3G for Q glucuronide and Q sulfate, iR3G for Q glucuronide and iR sulfate. The calibration curve (the range of 1–100 nM) was linear with a correlation coefficient of 0.97. Optimal identification of quercetin metabolites was achieved under the following conditions: curtain gas: 20 L/min, collision gas: 9 L/min, ionspray voltage: −4500 V, temperature: 550 °C, 1 ion source gas: 55 L/min, 2 ion source gas: 70 L/min, declustering potential: −50–120 V, entrance potential: −10 V, collision energy: −30–70 eV, collision cell exit potential: −10–45 V.

2.7.

Isolation

Q metabolites isolation from pig urine was carried out according to the modified method of Wiczkowski, Szawara-Nowak, and Topolska (2013) with the following procedure: thawing the urine, concentration of urine, purification of Q metabolites on ion-exchange gel, and isolation of Q metabolites on C18 semipreparative column. After thawing in a room temperature without exposure to light, the urine was concentrated with a rotary evaporator (Rotavapor R-200, Büchi, Postfach, Switzerland) at 35 °C from 12 L to around 100 mL, and lyophilized (Labconco Lyph-Lock 6, Kansas City, MO, USA). Then the sample was dissolved in 40 mL of 1% formic acid. Isolation of Q metabolites was done with column chromatography method on Amberlite XAD-16 ion exchange gel (Rohm & Hass, Philadelphia, PA, USA). The extract (about 10 mL) was loaded on a column (diameter: 6.5 cm, length: 43 cm) stabilized with 3000 mL of 1% formic acid in methanol and 3000 mL of 1% formic acid aqueous solution. After loading the sample, the column was washed with 2000 mL of 1% formic acid aqueous solution, and then fractions were eluted with 1350 mL of 0.1% formic acid in methanol. Based on the intensity of color reaction with the Folin–Ciocalteu reagent (125 µL of eluent, 125 µL of Folin– Ciocalteu reagent and deionization water (1/1), 250 µL of saturated sodium carbonate solution (Na2CO3), 2 mL of deionization water) the eluent collected was divided into six fractions (I–VI). The fractions were analyzed by HPLC-MS/MS method presented above in Section 2.6. Chromatography column was

124

journal of functional foods 11 (2014) 121–129

regenerated by washing with 3000 mL of 1% formic acid in methanol and 3000 mL of 1% formic acid aqueous solution. Finally, the fraction V, which contained Q metabolites was used for further separation.

2.8. Purification of isolated quercetin metabolites on semipreparative column Fraction V was concentrated on a rotary evaporator at 35 °C and lyophilized (Labconco Lyph-Lock 6), dissolved in 12 mL of water containing 0.1% formic acid, divided into 1 mL volume aliquots, and centrifuged (20 min, 13,000 × g, 4 °C). Purification of Q metabolites was done with HPLC-UV/Vis on semipreparative column C18 250 × 10 mm, 5 µm (XBridge Prep, Waters). The HPLC-UV/Vis system consisted of two pumps (LC10AD), column oven (CTO-10ASVP), degasser (DG-4400), UV/ Vis detector (SPD-10AV) set at 360 nm, and system software (Chroma, Warsaw, Poland). All chromatographic determinations were made at 45 °C with the flow rate of 1.5 mL/min and injection volume of 900 µL, in gradient system composed of water/formic acid (99:1, v/v) and acetonitrile/formic acid (99:1, v/v). Gradient was as follows: 1–1% B (0–40 min), 1–40% B (40– 45 min), 40–80% (45–70 min), 80–1% (70–75 min) and 1% B (75– 120 min). As a result of the analysis, three fractions (A-V, B-V, C-V) were separated and analyzed by HPLC-MS/MS method presented in Section 2.6. Fraction C-V, which contained Q metabolites was collected, evaporated to dryness and used for the antioxidative capacity analysis.

2.9. Preparation of isolated quercetin metabolites and standards of quercetin derivative solutions Appropriate amounts of purified mixture of Q metabolites, standards of Q derivatives were dissolved in 80% methanol. The preparation of Q metabolites is a mixture of different Q conjugates and assuming that their molar extinction coefficients differ from that of Q less up 10%, the concentration of the mixture of Q metabolites and Q derivative standards were set up with a spectroscopic method using molar extinction coefficient for Q (λ = 370 nm, ε = 20892.96) (Wiczkowski, Nemeth, Bucin´ski, & Piskuła, 2003) to exactly 1 mM.

2.10.

Antioxidant capacity

2.10.1. Trolox Equivalent Antioxidant Capacity (TEAC) assay The method presented by Re et al. (1999), with a minor modification, was used. The analysis was conducted using a spectrophotometer UV-160 1PC (Shimadzu). The ABTS•+ solution was diluted with 80% methanol to an absorbance of 0.70 ± 0.02 at 734 nm. Next, 1.48 mL of the ABTS•+ solution and 0.02 mL of the isolated metabolites of Q or solutions of Q derivatives standards or Trolox solution were mixed, and the absorbance was measured at 734 nm immediately after 6 min at 30 °C with appropriate solvent blanks. The antioxidant assay was carried out in triplicate for each sample. The TEAC of 80% methanol solutions of Q metabolites mixture and solution of Q derivatives standards was calculated using Trolox standard curve on the basis of the percent inhibition of the absorbance of the ABTS•+ solution at 734 nm. The antioxidant

capacity was expressed as Trolox equivalents based on the calibration curve with the range of 0.05–2.5 mM.

2.10.2. Determination of antioxidative capacity in lipophilic (ACL) and hydrophilic (ACW) conditions by the photochemiluminescence (PCL) method The PCL method was used to measure the antioxidant activity of isolated and purified metabolites of Q and standards of Q derivatives with a Photochem apparatus (Analytik Jena, Leipzig, Germany) against superoxide anion radicals generated from luminol (a photosensitizer) when exposed to UV light. The antioxidant activity of samples was analyzed using ACW and ACL kits, and the protocol of measurement was provided by the manufacturer. The assay was carried out as previously described by Zielinska, Wiczkowski, and Piskula (2008). The 80% methanol solution of Q metabolites preparation and standards of Q derivatives prior to the analysis were centrifuged by 10 min at 16,000 × g, and at 4 °C. The antioxidant assay was carried out in triplicate for each sample. The antioxidant capacity was expressed as Trolox equivalents based on the calibration curve with the range of 0.1–3.00 nM.

2.10.3. DPPH assay The DPPH• scavenging activity was determined using 80% methanol solution of Q metabolites preparation and standards of Q derivatives as described previously by Zielinska et al. (2008). The Trolox standard solutions (concentration 0.1– 2.0 mM) in 80% methanol were assayed under the same conditions. Subsequently, the DPPH• scavenging activity of the samples was expressed as Trolox equivalents on the basis of the reduction in the absorbance of the DPPH• solution at 515 nm. The measurements were carried out using a spectrophotometer UV-160 (Shimadzu).

2.11.

Statistical analysis

The data are presented as mean ± SD for triplicate analysis. The results were subjected to one-way analysis of variation (ANOVA) with Fisher’s Least Significant Difference test. P < 0.05 was considered significant. The Pearson Correlation test was used. The statistical analysis was performed using Statistica 6 (Stat Soft, Tulsa, OK, USA).

3.

Results and discussion

Derivatives of Q in onion dry skin were analyzed using the HPLC-DAD method. The compounds were identified by the means of a comparison of their retention times, UV-Vis spectra, the previous data (Wiczkowski et al., 2008) and presented in Table 1. Q and its four derivatives: Q3,4′G, Q3G, Q4′G, and iR4′G were found in onion dry skin as reported previously (Albishi, John, Al-Khalifa, & Shahidi, 2013; Wiczkowski et al., 2003, 2008). The total content of Q in onion dry skin determined by HPLCDAD was very high (22.18 ± 0.43 mg of Q/g of dry matter). One of the first reports indicating that onion dry skin is a very rich source of Q was published almost 40 years ago (Herrmann, 1976). In the study cited, the concentration of Q in onion dry

journal of functional foods 11 (2014) 121–129

Table 1 – Content and composition of quercetin derivatives in onion dry skin. Compound

mg/g dma

% of contribution in total quercetin content

Quercetin Isorhamnetin 4′-glucoside Quercetin 4′-glucoside Quercetin 3-glucoside Quercetin 3,4′-diglucoside Total

19.43 ± 0.31 0.09 ± 0.02 2.52 ± 0.11 0.09 ± 0.02 0.05 ± 0.01 22.18 ± 0.43

87.60 0.41 11.36 0.41 0.23

a

All values were expressed in milligrams of each quercetin derivatives per gram of dry matter of onion dry skin. Data are expressed as means ± SD (n = 3).

skin reached even the level of 6.5% dry matter. Similarly, Horbowicz and Bakowski (2000) as well as Wiczkowski et al. (2003) found that dry skin of onion is by far the richest in Q, where this compound may reach even the level of 4.5% of their weight. Other study (Patil & Pike, 1995) showed that dry skin of red onion (2.9% of Q) is richer in Q than dry skin of white onion (0.01% of Q). Shallot, another vegetable from the Allium genus, is also characterized by high levels of Q in dry skin (Wiczkowski et al., 2008). The differences in the contents of Q in the studies cited may stem from the fact that the content of this flavonoid in raw material is affected by varietal differences, cultivation agro-technology, growing season and postharvest treatment (Wiczkowski & Piskula, 2004). For example, the report of Horbowicz and Bakowski (2000) shows that the average level of Q in onions from the vegetation season characterized by high insolation, high temperatures and low precipitation, was higher than from the year with high precipitation and low temperature. As presented in Table 1, Q aglycone comprised 87.6% of total onion dry skin Q and glucoside derivatives of Q represent 12.4% of the total content of Q. Among them, Q4′G showed the highest content followed by Q3G, iR4′G, and Q3,4′G. Similar to our study, other reports (Lanzotti, 2006; Slimestad, Fossen, & Vagen, 2007; Wiczkowski et al., 2008) demonstrated high participation of Q aglycone in the total content of Q in onion dry skin, as well as the domination of Q4′G among the Q glucosides. Taking into consideration (i) a considerable number of studies that have shown onion as an excellent source of Q (de Vries et al., 2001; Hollman et al., 1997), (ii) indication that when Q and its derivatives are provided for intake in a form of natural sources Q aglycone is more bioavailable than its glucosides (Wiczkowski et al., 2008), as well as (iii) that onion dry skin is very rich in Q aglycone (Wiczkowski et al., 2003; Horbowicz & Bakowski, 2000), we conducted, for the first time, a study on the isolation of Q metabolites from pig urine after the intake of onion dry skin. The investigation of Lesser, Cermak, and Wolffram (2004) and Cermak, Landgraf, and Wolffram (2003) showed that depending on the diet used and the form of Q, the maximum concentration of Q in pig blood plasma after Q administration was in the interval between 45 and 210 min. Moreover, the maximum con-

125

centration of other metabolites of flavonoids in urine after oral intake was in the interval between 2 and 6 h (Manach et al., 2005; Wiczkowski et al., 2010). Based on these results, the urine was collected before (0 h) and between 2 and 6 h after onion dry skin consumption. In the first step, the determination of free Q and iR after enzymatic hydrolysis of urine before and after exposure to Q was performed. Since aglycone of Q and iR was not found in the urine after overnight fast, it indicates that the washout phase with Q-free diet preceding the study was sufficient. After the hydrolysis of urine collected after onion dry skin intake, aglycones of Q and iR were found, with the content of Q being more than that of iR (Fig. 1). This indicated that after consumption of the experimental fodder, at least part of Q from onion dry skin was absorbed and metabolized to conjugated compounds. In the urine, another methylated derivative of Q, namely 4′-O-methylated quercetin (tamarixetin), was not found. Similarly, in biological fluids of rats, Graf et al. (2005) observed only iR. On the other hand, previous studies of Lesser et al. (2004) and Cermak et al. (2003) pointed to the presence of both methylated forms of Q in blood plasma of pig after the intake of Q. However, there is no direct answer on which hydroxyl group of C ring catechol moiety is preferred during the process of methylation by catechol-O-methyltransferase (COMT, EC 2.1.1.6). It is suggested that it depends on animal species and/or on the dose of Q used, yet it may also occur randomly (Wiczkowski & Piskula, 2004). Moreover, the investigations of Morand, Manach, Crespy, and Remesy (2000) and Manach et al. (1999) showed that in blood plasma of rats both methylated derivatives of Q were presented as well. Contrary, in humans, the process of O-methylation plays a much smaller role in the metabolism of Q. There are reports demonstrating that after the intake of Q, besides Q aglycone, only iR is present in human blood plasma (Day et al., 2001; Graefe et al., 2001; Manach et al., 1998). Next, analysis of Q metabolites in urine was carried out by means of HPLC-MS/MS with mode of scanning precursor ions (301 m/z for Q and 315 for iR) and 12 metabolites were found. In the next step, scanning of fragmentation ions derived from the selected parent ion for observation of all the ions formed by the disintegration of the parent ion as well as neutral particles was performed. Finally, also 12 metabolites of Q were identified (Table 2), based on the comparison of their retention time and MS/MS fragmentation spectrum with the data published previously (Graf et al., 2005; Mullen et al., 2004). Our study is the first that presents the full profile/composition of quercetin metabolites in pig urine after intake of rich in quercetin onion dry skin. The HPLC-MS/MS TIC chromatograms of Q metabolites were characterized by five major peaks and five minor peaks (Fig. 2). Peak 6 was predominant in the chromatographic profile and its MS/MS analysis revealed a pseudomolecular ion at m/z 477 [M]− and fragment ions at m/z 301 [M-176]−. The m/z 301 [M]− corresponded to Q aglycone and was produced after a loss of 176 mass unite (mu) from the pseudomolecular ion. The neutral loss of 176 mu corresponded to one molecule of dehydro-glucuronic acid. Finally, taking into account these results and the previous data (Graf et al., 2005; Mullen et al., 2004), peak 6 was identified as Q monoglucuronide. What is more, another peak (4), which gave early retention, had a very similar MS/MS spectrum.

126

journal of functional foods 11 (2014) 121–129

Q

7.0e6

7.28

6.5e6 6.0e6 5.5e6

Intensity, cps

5.0e6 4.5e6 4.0e6 3.5e6

8.34

3.0e6

iR

2.5e6 2.0e6 1.5e6 1.0e6 5.0e5 0.0

0

2

4

6

8

10

12

14 Time, min

16

18

20

22

24

26

28

Fig. 1 – Chromatogram of HPLC-MS/MS with MRM mode of quercetin (Q) and isorhamnetin (iR) of hydrolyzed pig urine after onion dry skin intake.

Therefore, peaks 4 and 6 may be isomers with a different position of the glucuronyl group on the aglycone moiety. Identical phenomenon associated with the presence of peaks with a similar pattern of MS/MS fragmentation was found for two pairs of peaks: 1, 2 and 5, 7. In case of peaks 1 and 2, they were characterized by the pseudomolecular ion at 653 ([M]− m/z), and two fragment ions at m/z 477 [M-176]− and m/z 301 [M-2x176]−. Two neutral losses of 176 mu were found in this structure, therefore these peaks were tentatively identified as Q diglucuronides, but with a different position of glucuronidation. In case of peaks 5 and 7, they were characterized by the pseudomolecular ion of 491 ([M]− m/z) and one fragment ion m/z 315 [M-176]−. The m/z 315 [M]− corresponded to iR aglycone and was produced after a loss of 176 mu from the pseudomolecular ion. On the basis of these data and previous information (Graf et al., 2005; Mullen et al., 2004), peaks 5 and 7 were identified as iR monoglucuronides but again with a different position of

glucuronidation. Peak 3 was identified with the pseudomolecular ion at m/z 667 [M]− with fragment ions at m/z 491.0 [M-176]− and m/z 315 [M-2x176]−. This indicates that it can be assigned to iR diglucuronide. Another peak was formed by two compounds, 8 and 9. Compound 8 yielded pseudomolecular ion at m/z 557 [M]− and fragment ions at m/z 477 [M-80]− and m/z 301 [M-80–176]−. The m/z 477 [M]− corresponded to Q monoglucuronide and was produced after a loss of 80 mu from the pseudomolecular ion. The neutral loss of 80 mu corresponded to one molecule of sulfate. These data are characteristic of Q monoglucuronide sulfate (Graf et al., 2005; Mullen et al., 2004). When it comes to compound 9, it was identified with pseudomolecular ion at m/z 571 [M]− and fragment ions at m/z 491 [M-80]− and m/z 315 [M-80–176]−. The m/z 491 [M]− corresponded to iR monoglucuronide and was produced after a loss of 80 mu from the pseudomolecular ion and was identified as iR monoglucuronide sulfate. Also another peak was formed by

Table 2 – The MS data and composition of quercetin metabolites in pig urine after intake of onion dry skin. Compounds

[M]− (m/z)

MS/MS (m/z)

% of contribution in total content of quercetin metabolites

MRM for analysis (m/z)

Quercetin diglucuronide Quercetin diglucuronide Methylated quercetin diglucuronide Quercetin monoglucuronide Methylated quercetin monoglucuronide Quercetin monoglucuronide Methylated quercetin monoglucuronide Quercetin monoglucuronide sulfate Methylated quercetin monoglucuronide sulfate Quercetin sulfate Methylated quercetin sulfate Quercetin Isorhamnetin

653 653 667 477 491 477 491 557 571 381 395 301 315

477/301 477/301 491/315 301 315 301 315 477/301 491/315 301 315 179/151 300

4.7 8.8 5.7 12.7 11.7 31.7 6.9 1.6 0.7 9.8 4.0 1.7

653–301 653–301 667–315 477–301 491–315 477–301 491–315 557–301 571–315 381–301 395–315 301–151 315–300

127

Intensity, cps

journal of functional foods 11 (2014) 121–129

1.8e9 1.7e9 1.6e9 1.5e9 1.4e9 1.3e9 1.2e9 1.1e9 1.0e9 9.0e8 8.0e8 7.0e8 6.0e8 5.0e8 4.0e8 3.0e8 2.0e8 1.0e8 0.0

6

10, 11 4

5

2 7 3 1

12

8, 9 2

4

6

8

10

12

14 Time, min

16

18

20

22

24

26

28

Fig. 2 – Chromatogram of HPLC-MS/MS with scanning precursor ions (301 m/z for quercetin and 315 for isorhamnetin) of hydrolyzed pig urine after onion dry skin intake (1 and 2 – quercetin diglucuronide; 3 – isorhamnetin diglucuronide; 4 and 6 – quercetin monoglucuronide; 5 and 7 – isorhamnetin glucuronide; 8 – quercetin monoglucuronide sulfate; 9 – isorhamnetin monoglucuronide sulfate; 10 – quercetin sulfate; 11 – isorhamnetin sulfate; 12 – isorhamnetin).

two compounds, 10 and 11. Peak 10 revealed the pseudomolecular ion at m/z 381 [M]− and fragment ions at m/z 301 [M-80]−, while peak 11 had the MS/MS spectra of m/z 395 [M]− and m/z 315 [M-80]−. These peaks were putatively described as Q sulfate and iR sulfate, respectively. The last peak (12) gave MS/MS spectra of m/z 315 [M]− and m/z 300 [M-15]− which corresponded to iR aglycone. Peaks 1 and 2; 4 and 6; 5 and 7 had a very similar MS/MS spectrum but different retention time. As indicated in the studies of Graf et al. (2005) and Mullen et al. (2004), different position of glucuronidation in both Q and iR molecule is possible, resulting in different retention times. Therefore, several mono- and di-glucuronated structures with different positions of conjugation are possible (Barrington et al., 2009). Therefore compounds from each pair were designated as Q diglucuronide, Q monoglucuronide and iR monoglucuronide, respectively. Among the 12 Q metabolites identified in pig urine, the predominant were two Q monoglucuronides (peaks 6 and 4), followed by two iR monoglucuronides (peaks 5 and 7) (Fig. 2, Table 2). The Q metabolites excreted with urine were mainly represented by the glucuronated form (57.9%). Summing up, the glucuronidation was the major metabolic route of Q, while methylation and sulfation played a minor role in the metabolism of this compound in pigs (Table 2). In the report on the profile of Q metabolites after the intake of Q-containing food (Day et al., 2001), three Q metabolites, Q 3-glucuronide, iR 3-glucuronide and Q 3′-sulfate were found as major compounds in human blood plasma after onion consumption. In other study, Graf et al. (2005) identified six metabolites of Q (four different Q diglucuronides, Q monoglucuronide sulfate, methylated Q monoglucuronide sulfate) in blood plasma of rats which ingested radiolabeled Q 4′-glucoside. On the other hand, Mullen et al. (2004) detected 18 derivatives of Q in human urine after consumption of red onion. In their report, three conjugated forms of Q were also identified as main metabolites of Q. The profile of metabolites obtained in our study was

similar to that, except for five metabolites present at a very low level and containing glucose in their structure (which was not confirmed by other studies till now). This indicates that the profiles of Q metabolites in humans and pigs are similar. The results of several experiments on the antioxidant activity of Q metabolites have been published (Dueñas, Surco-Laos, Gonzalez-Manzano, González-Paramás, & Santos-Buelga, 2011; Pashikanti, de Alba, Boissonneault, & Cervantes-Laurean, 2010), yet there are no studies determining the antioxidant capacity of the pool of Q metabolites isolated from the biological system. Therefore, our study is the first to provide data of the antioxidant capacity of the mixture of Q metabolites reflecting the profile of Q metabolites in vivo after onion consumption. Four analytical strategies were applied: assay of TEAC, DPPH, PCL ACL and PCL ACW. Among them, two in vitro scavenging capacity assays (TEAC, DPPH) against stable, non-biological radicals were used. Moreover, two systems (ACL, ACW) based on biological oxidant as superoxide radical anion were applied. The selected methods are commonly accepted, validated and standardized (Magalhaes, Segundo, Reis, & Lima, 2008). In addition, the PCL methods use superoxide radical anion which is generated in cells under physiological conditions and may play harmful role as precursor of more reactive oxygen species, contributing to chronic degenerative diseases (Zielin´ska & Zielin´ski, 2011). The data on antioxidant capacity varied depending on the method used (Table 3). The antioxidant capacity of Q metabolites preparation derived from TEAC, ACL, ACW and DPPH assays were 3.54 ± 0.09, 3.45 ± 0.04, 1.98 ± 0.02 and 1.85 ± 0.02 mM Trolox, respectively. Moreover, the antioxidant capacity of the mixture of isolated and purified Q metabolites (1 mM) was compared to native Q compounds (each at 1 mM) occurring in onion. In all systems with standardized Q concentration, mixture of Q metabolites showed the highest antioxidant activity followed by Q aglycone and Q3G, while Q3,4′G showed the lowest antioxidant activity. The order of scavenging activity for TEAC,

128

journal of functional foods 11 (2014) 121–129

Table 3 – The antioxidant capacity of standardized for 1 mM concentration of quercetin metabolites mixture and standards of quercetin derivatives determined by four different assays. Compounds

mM Troloxa Antioxidant activity assays

Quercetin metabolites mixture Quercetin aglycone Isorhamnetin Quercetin 3,4′-diglucoside Quercetin 4′-glucoside Isorhamnetin 3-glucoside Quercetin 3-glucoside a

TEAC

DPPH

ACW PCL

ACL PCL

3.54 ± 0.09a 2.43 ± 0.02b 1.13 ± 0.01d 0.13 ± 0.01g 0.65 ± 0.02e 0.52 ± 0.04f 1.53 ± 0.02c

1.85 ± 0.05a 1.68 ± 0.01b 0.75 ± 0.06c 0.22 ± 0.01e 0.41 ± 0.03d 0.23 ± 0.01e 1.60 ± 0.05b

3.45 ± 0.04a 2.34 ± 0.02b 1.66 ± 0.02c 0.26 ± 0.02g 0.31 ± 0.01f 0.42 ± 0.01e 1.34 ± 0.02d

1.98 ± 0.02e 0.93 ± 0.02b 0.35 ± 0.01d 0.02 ± 0.00e 0.05 ± 0.00d 0.03 ± 0.00de 0.50 ± 0.01c

Data are expressed as means ± SD (n = 3). All values were expressed as mM of Trolox. Means in columns followed by the different letters are significantly different (P < 0.05).

DPPH and PLC ACW was as follows: Q metabolites mixture > Q > Q3G > iR > Q4′G > iR3G > Q3,4′G. Only when comprising data from the PCL ACL assay, the order of analyzed compounds was changed – iR3G showed higher ability to scavenge the superoxide anion radical than Q4′G. The antioxidant activity of Q metabolites preparation analyzed by TEAC was highly correlated with that provided by ACW (r = 0.975) and ACL (r = 0.973). Also ACW was highly correlated with ACW (r = 0.959), while DPPH correlated with TEAC (r = 0.914), ACL (r = 0.877) and ACW (r = 0.840). When comparing the antioxidant activity of Q conjugates, the presence or lack of the conjugation moiety in the molecular structure of Q is crucial. In the study (Dueñas et al., 2011), where the antioxidant activity of Q and its metabolites was compared, Q aglycone was characterized by higher antioxidant activity than that of each individual metabolite of Q, e.g. Q 3′-sulfate, Q 4′-sulfate, iR 3-sulfate, iR and Q 3-glucuronide. The authors concluded that substitution of the hydroxyl groups of Q by conjugate residues had resulted in a decrease in the antioxidant capacity when compared to the aglycone form. Similarly, results of the study by Ioku, Tsushida, Takei, Nakatani, and Terao (1995) have shown that Q4′G and Q3G reveal weaker antioxidant properties than the free form of Q. In our report, it was shown, for the first time, that the mixture of Q metabolites showed stronger antioxidant activity against applied radicals than Q, iR and their glucosides occurring in onion. In the experiment, Q derivatives (Q3G, Q4′G, Q3,4′G, iR3G, iR) were used as an equivalent of chemical structure of molecules of Q metabolites. The presence of Q metabolites with strong antioxidative capacity in the mixture of Q metabolites may be a suitable explanation of this phenomenon. Firstly, unlike in the study of Dueñas et al. (2011), antioxidant capacity measurements presented here are concerned with the mixture of Q metabolites reflecting the profile of Q metabolites in vivo after onion consumption, including 12 different metabolites of Q. Secondly, previous studies of the antioxidant properties of main metabolites of Q have shown that these properties depend on the position of the conjugation. The report of Janisch, Williamson, Needs, and Plumb (2004) demonstrated that antioxidant activity of selected Q metabolites was in the order of Q 7-glucuronide > Q aglycone > Q 3-glucuronide = Q3G > Q 4′-glucuronide > iR 3-glucuronide > Q 3-sulfate. Furthermore, the investigation of Nadsume et al. (2004) showed that also

7-glucuronide of epicatechin demonstrate the highest antioxidant properties among catechin metabolites. Summing up, the above studies indicated that both forms, 7-glucuronide and 3-glucuronide, exhibited high efficiency as antioxidants. Finally, two monoglucuronated forms of Q, which may include Q 7-glucuronide and/or Q 3-glucuronide, predominated in the mixture obtained (more than 44% of total Q metabolites). Therefore, taking into account these elucidations, higher antioxidant capacity of Q metabolites pool compared to Q aglycone is possible. What is more, other explanation of this phenomenon, without excluding above arguments, may be the effect of synergy when Q metabolites act in the pool. It is also plausible that the interactions of Q metabolites in vivo are as important as their total content. In conclusion, in order to understand the functions of quercetin metabolites in the humans, it is important to know their chemical structures which determine their physiological activity. The results of this study indicate that the quercetin metabolites present in systemic circulation formed in detoxification processes after quercetin consumption may still act as antioxidants. It is important to note that dry skin of onion is a rich source of bioavailable quercetin and can be utilized as natural food ingredient and/or dietary antioxidant, especially when the prophylaxis of diseases associated with oxidative damage is considered.

Acknowledgment The research was supported by the National Science Centre, Poland (project 798/N-COST/2010/0).

REFERENCES

Albishi, T., John, J. A., Al-Khalifa, A. S., & Shahidi, F. (2013). Antioxidative phenolic constituents of skins of onion varieties and their activities. Journal of Functional Foods, 2013(5), 1191–1203. Bansal, P., Paula, P., Kunwarc, A., Jayakumard, S., Nayaka, P. G., Priyadarsinic, K. I., & Unnikrishnan, M. K. (2012). Radioprotection by quercetin-3-O-rutinoside, a flavonoid

journal of functional foods 11 (2014) 121–129

glycoside – A cellular and mechanistic approach. Journal of Functional Foods, 4, 924–932. Barrington, R., Williamson, G., Bennett, R. N., Davis, B. D., Brodbelt, J. S., & Kroon, P. A. (2009). Absorption, conjugation and efflux of the flavonoids, kaempferol and galangin, using the intestinal CaCo-2/TC7 cell model. Journal of Functional Foods, 1, 74–87. Cermak, R., Landgraf, S., & Wolffram, S. (2003). The bioavailability of quercetin in pigs depends on the glycoside moiety and on dietary factors. Journal of Nutrition, 133, 2802–2807. de Vries, J. H. M., Hollman, P. C. H., Amersfoort, I., Olthof, M. R., & Katan, M. B. (2001). Red wine is a poor source of bioavailable flavonols in men. Journal of Nutrition, 131, 745–748. Day, A. J., Mellon, F., Barron, D., Sarrazin, G., Morgan, M. R. A., & Williamson, G. (2001). Human metabolism of dietary flavonoids: Identification of plasma metabolites of quercetin. Free Radical Research, 35, 941–952. Dueñas, M., Surco-Laos, F., Gonzalez-Manzano, S., GonzálezParamás, A. M., & Santos-Buelga, C. (2011). Antioxidant properties of major quercetin metabolites. European Food Research and Technology, 232, 103–111. Graefe, E. U., Witting, J., Mueller, S., Rierhling, A. K., Uehleke, B., Drewelow, B., Pforte, H., Jacobasch, G., Derendorf, H., & Veit, M. (2001). Pharmacokinetics and bioavailability of quercetin glycosides in humans. Journal of Clinical Pharmacology, 41, 492– 499. Graf, B. A., Mullen, W., Caldwell, S. T., Hartley, R. C., Duthie, G. G., Lean, M. E. J., Crozier, A., & Edwards, C. A. (2005). Disposition and metabolism of [2-14C]quercetin-4’-glucoside in rats. Drug Metabolism and Disposition, 33, 1036–1043. Herrmann, K. (1976). Flavonols and flavones in food plants; a review. Journal of Food Technology, 11, 433–448. Hollman, P. C. H., van Trijp, J. M. P., Buysman, M. N. C. P., van de Gaag, M. S., Mengelers, M. J. B., de Vries, J. H. M., & Katan, M. B. (1997). Relative bioavailability of the antioxidant flavonoid quercetin from various foods in man. FEBS Letters, 418, 152– 156. Horbowicz, M., & Bakowski, J. (2000). Wpływ niektórych czynników na zawartos´c´ kwercetyny w cebuli. Zeszyty Naukowe Akademii Rolniczej im. H. Kołła˛taja w Krakowie, 364, 97– 99, in Polish. Ioku, K., Tsushida, T., Takei, Y., Nakatani, N., & Terao, J. (1995). Antioxidative activity of quercetin and quercetin monoglucosides in solution and phospholipid bilayers. Biochimica et Biophysica Acta, 1234, 99–104. Janisch, K. M., Williamson, G., Needs, P., & Plumb, G. W. (2004). Properties of quercetin conjugates: Modulation of LDL oxidation and binding to human serum albumin. Free Radical Research, 38(8), 877–884. Lanzotti, V. (2006). The analysis of onion and garlic. Journal of Chromatography. A, 1112, 3–22. Lesser, S., Cermak, R., & Wolffram, S. (2004). Bioavailability of quercetin pigs is influenced by the dietary fat content. Journal of Nutrition, 134, 1508–1511. Magalhaes, L. M., Segundo, M. A., Reis, S., & Lima, J. L. F. C. (2008). Methodological aspects about in vitro evaluation of antioxidant properties. Analytica Chimica Acta, 613, 1–19. Manach, C., Morand, C., Crespy, V., Demingné, C., Texier, O., Régérat, F., & Rémésy, C. (1998). Quercetin is recovered in human plasma as conjugated derivatives which retain antioxidant properties. FEBS Letters, 426, 331–336. Manach, C., Texier, O., Morand, C., Crespy, V., Régérat, F., Demingné, C., & Rémésy, C. (1999). Comparison of the bioavailability of quercetin and catechin in rats. Free Radical Biology and Medicine, 11(12), 1259–1266. Manach, C., Williamson, G., Morand, C., Scalbert, A., & Remesy, C. (2005). Bioavailability and bioefficacy of polyphenols in

129

humans. I. Review of 97 bioavailability studies. American Journal of Clinical Nutrition, 81, 243S–255S. Morand, C., Manach, C., Crespy, V., & Remesy, C. (2000). Quercetin 3-O-β-glucoside is better absorbed than other quercetin forms and is not present in rat plasma. Free Radical Research, 33, 667– 676. Mullen, W., Boitier, A., Stewart, A. J., & Crozier, A. (2004). Flavonoid metabolites in human plasma and urine after the consumption of red onions: Analysis by liquid chromatography with photodiode array and full scan tandem mass spectrometric detection. Journal of Chromatography. A, 1058, 163–168. Nadsume, M., Osakabe, N., Yasuda, A., Baba, S., Takunaga, T., Kondo, K., Osawa, T., & Terao, J. (2004). In vitro antioxidative activity of (-)-epicatechin glucuronide metabolites present in human and rat plasma. Free Radical Research, 38(12), 1341– 1348. Pashikanti, S., de Alba, D. R., Boissonneault, G. A., & CervantesLaurean, D. (2010). Rutin metabolites: Novel inhibitors of nonoxidative advanced glycation end products. Free Radical Biology and Medicine, 48, 656–663. Patil, B. S., & Pike, M. L. (1995). Distribution of quercetin content in different rings of various coloured onion (Allium cepa L.) cultivars. Journal of Horticultural Sciences, 70, 643–650. Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & RiceEvans, C. A. (1999). Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biology and Medicine, 26, 1231–1237. Slimestad, R., Fossen, T., & Vagen, I. M. (2007). Onions: A source of unique dietary flavonoids. Journal of Agricultural and Food Chemistry, 55, 10067–10080. Terao, J., Murota, K., & Kawai, Y. (2011). Conjugated quercetin glucuronides as bioactive metabolites and precursors of aglycone in vivo. Food & Function, 2, 11–17. Ting, Y., Jiang, Y., Ho, C.-T., & Huang, Q. (2014). Common delivery systems for enhancing in vivo bioavailability and biological efficacy of nutraceuticals. Journal of Functional Foods, 7, 112– 128. Wiczkowski, W., Nemeth, K., Bucin´ski, A., & Piskuła, M. K. (2003). Bioavailability quercetin from fresh scales and dry skin of onion in rats. Polish Journal of Food and Nutrition Sciences, 12/53, 95–99. Wiczkowski, W., & Piskula, M. K. (2004). Food flavonoids. Polish Journal of Food and Nutrition Sciences, 13(54), 101–114. Wiczkowski, W., Romaszko, E., & Piskuła, M. K. (2010). Bioavailability of cyanidin glucosides from natural chokeberry (Aronia melanocarpa) juice with dietary-relevant dose of anthocyanins in humans. Journal of Agricultural and Food Chemistry, 58, 12130–12136. Wiczkowski, W., Romaszko, J., Bucin´ski, A., Szawara-Nowak, D., Honke, J., Zielin´ski, H., & Piskuła, M. K. (2008). Quercetin from shallot (Allium cepa var. Aggregatum) is more bioavailable than its glucosides. Journal of Nutrition, 138, 885–889. Wiczkowski, W., Szawara-Nowak, D., & Topolska, J. (2013). Red cabbage anthocyanins: Profile, isolation, identification, and antioxidant activity. Food Research International, 51, 303–309. Williamson, G., Barron, D., Shimoi, K., & Terao, J. (2005). In vitro biological properties of flavonoid conjugates found in vivo. Free Radical Research, 39, 457–469. Zielin´ska, D., & Zielin´ski, H. (2011). Antioxidant activity of flavone C-glucosides determined by updated analytical strategies. Food Chemistry, 124, 672–678. Zielinska, D., Wiczkowski, W., & Piskula, M. K. (2008). Determination of the relative contribution of quercetin and its glucosides to the antioxidant capacity of onion by cyclic voltammetry and spectrophotometric methods. Journal of Agricultural and Food Chemistry, 56, 3524–3531.