Eur J Nutr DOI 10.1007/s00394-015-0862-9

ORIGINAL CONTRIBUTION

Comparing the metabolism of quercetin in rats, mice and gerbils Shu‑Lan Yeh · Yi‑Chin Lin · Yi‑Ling Lin · Chien‑Chun Li · Cheng‑Hung Chuang 

Received: 21 October 2014 / Accepted: 12 February 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  Purpose  Several species of rodents are used to investigate the metabolism of quercetin in vivo. However, it is unclear whether they are a proper animal model. Thus, we compared the metabolism of quercetin in Wistar rats (rats), Balb/c mice (mice) and Mongolian gerbils (gerbils). Methods  We determined the levels of quercetin metabolites, quercetin-3-glucuronide (Q3G), quercetin-3′-sulfate (Q3′S) and methyl-quercetin isorhamnetin (IH), in the plasma, lungs and livers of three species of animals by high-performance liquid chromatography after acute and/ or chronic quercetin administration. The metabolic enzyme activities in the intestinal mucosal membrane and liver were also investigated. Results  First, we found that after acute quercetin administration, the Q3′S level was the highest in gerbils. However, after long-term supplementation (20 weeks), Q3G was the dominant metabolite in the plasma, lungs and livers followed by IH and Q3′S in all animals, although the gerbils still had a higher Q3′S conversion ratio. The average concentrations of total quercetin concentration in the plasma of gerbils were the highest in both short- and long-term studies. The activities of uridine 5′-diphosphate-glucuronosyltransferase, phenolsulfotransferase and catechol-Omethyltransferase were induced by quercetin in a dose- and tissue-dependent manner in all animals. S.‑L. Yeh · Y.‑C. Lin · Y.‑L. Lin · C.‑C. Li  Institute of Nutritional Science, Chung Shan Medical University, No. 110 Sec. 1 Jianguo N. Rd, Taichung 402, Taiwan, ROC C.‑H. Chuang (*)  Department of Nutrition, Master Program of Biomedical Nutrition, Hungkuang University, No. 1018 Sec. 6 Taiwan Boulevard, Taichung 43302, Taiwan, ROC e-mail: [email protected]

Conclusions  Taken together, in general, after long-term supplementation the metabolism of quercetin is similar in all animals and is comparable to that of humans. However, the accumulation of quercetin and Q3′S conversion ratio in gerbils are higher than those in the other animals. Keywords  Quercetin-3-glucuronide · Quercetin-3′-sulfate · Uridine 5′-diphosphate- glucuronosyltransferase · Phenolsulfotransferase · Catechol-O-methyltransferase Abbreviations Q3G Quercetin-3-glucuronide Q3′S Quercetin-3′-sulfate IH Isorhamnetin QA Quercetin aglycone Rats Wistar rats Mice Balb/c mice Gerbils Mongolian gerbils UDPGT Uridine 5′-diphosphate-glucuronosyltransferase PST Phenolsulfotransferase COMT Catechol-O-methyltransferase UDPGA Uridine 5′-diphosphoglucuronic acid

Introduction Epidemiologic studies show that high quercetin intakes are associated with a lower risk of heart disease, stroke [1, 2], and several types of cancers including lung, breast and prostate [3, 4]. In vitro studies with quercetin aglycone (QA) show that quercetin possesses various bioactivities including antioxidative, anticancer and anti-inflammatory activities [5–7]. However, in vivo studies demonstrate that quercetin given orally is mostly metabolized to conjugated

13



metabolites, that is, sulfate-, glucuronide- and methyl-conjugated quercetin [8, 9], in the small intestine and liver. Little or no free QA is present in the human circulation [8, 10]. It has been reported that QA and quercetin metabolites may have different biological activities, such as antioxidant activity [11]. In addition, QA and quercetin-3′-sulfate (Q3′S), rather than quercetin-3-glucuronide (Q3G), inhibit receptor-mediated contractions of the isolated porcine artery [12]. Our previous study demonstrated that quercetin-metabolite-enriched plasma but not quercetin itself significantly increased peroxisome proliferator-activated receptor-γ expression in A549 lung cancer cells [13]. Thus, using an appropriate in vivo model to investigate the exact biological role of quercetin administrated orally is needed. Although several animal species have been used to investigate the bioactivities of quercetin in vivo, few studies compare the profiles of quercetin metabolites among animals. de Boer et al. [14] demonstrate that long-term exposure leads to a wide distribution of quercetin and quercetin metabolites in rat tissues with the highest total quercetin concentrations in the lungs; while there are high levels of quercetin and its metabolites in the kidneys and livers but not in the lungs in pigs exposed to quercetin [15]. However, quercetin metabolite profiles were not shown in these studies. Previously, we found that sulfate-, glucuronide- and methyl-conjugated quercetin analogues were present in the plasma of Mongolian gerbils (gerbils) after 2 h of quercetin administration [13], but the distribution in the organs and the quercetin metabolite profile after long-term supplementation was unclear. Thus, in this study, we compared the levels and the profiles of three major metabolites of quercetin, Q3G, Q3′S and methyl-quercetin (IH, determined as isorhamnetin equivalent), in Wistar rats (rats), Balb/c mice (mice) and gerbils. Quercetin (50 and 100 mg/kg body weight, once a week) was administrated to the animals by oral gavage for the first 5 weeks for time-course studies. Then, for the long-term feeding study, quercetin was administrated at the same dose (thrice/week) for an additional 20 weeks. The concentration of quercetin and its metabolites in the plasma, lungs and liver as well as the activities of phase II enzymes in the intestinal mucosal membrane and liver in the experimental animals were determined. We found that gerbils had the highest accumulation of total quercetin and Q3′S conversion ratio. However, similar to humans, the level of Q3G was higher than that of Q3′S in the plasma, lungs and livers of all the animals after long-term supplementation.

Materials and methods Chemicals and reagents All chemicals used were of reagent grade or higher. Quercetin (3,3′,4′,5,7-pentahydroxylflavone), adenosine-3′-

13

Eur J Nutr

phosphate-5′-phospho-sulfate, Helix pomatia enzyme mixture, s-adenosyl-l-methionine, uridine 5′-diphosphoglucuronic acid (UDPGA), sulfatase and β-glucuronidase were purchased from Sigma chemical (St. Louis, MO). Isorhamnetin was purchased from Extrasynthese (Cedex, France). Animals Adult male Wistar rats (220–250 g, n  = 21) and male BALB/c mice (20–25 g, n  = 21) were purchased from the animal Center of the National Science Council. Male Mongolian gerbils (50–55 g, n = 21) were obtained from the laboratory animal center at Taichung Veterans General Hospital (Taichung, Taiwan). All animals were 6–8 weeks old. The study protocol was approved by the Animal Research Committee at Hungkuang University (Approval No: 98012). Study design The animals were housed in hanging wire mesh cages in a temperature (25 ± 2 °C) and humidity (65 ± 5 %) controlled room with an alternating 12-h light/dark cycle. Upon arrival, animals were acclimated for 1 week, during which they were fed a standard rodent diet (Lab 5001, Purina Mills, St. Louis, MO) and water ad libitum. The diet contained 28.5 % protein, 13.5 % fat and 58.0 % carbohydrate as a percentage of the total energy. Based on the volume of diet consumed, the average calorie intake of mice, gerbils and rats was calculated to be about 60, 36 and 21 kcal/100 g body weight/day, respectively. The animals of each species were randomly assigned to the following three groups (n  = 7/group): (1) control group (lard only, C); (2) low-dose quercetin group (50 mg/kg body weight, LQ); (3) high-dose quercetin group (100 mg/kg body weight, HQ). Quercetin was dissolved in refined lard (Weilih Food, Taipei) before being fed to the animals by gavage each time. We chose the refined lard as a vehicle because the quercetin concentration in this oil was undetectable and markedly lower than that in soybean oil or in corn oil (approximately 0.075 μM and 0.009 μM, respectively). The animals in the control group received refined lard only. In the first 5 weeks, we performed a time-course study. Quercetin (50 or 100 mg/kg body weight for LQ and HQ group, respectively) was administrated to the animals by oral gavage once a week to determine the pharmacokinetic and the metabolites of quercetin in plasma. It took 5 weeks because we needed to collect enough blood to determine all parameters. During this period, blood samples were collected from the retro-orbital plexus after quercetin was given for 1–72 h. Then, based on the study of de Boer et al. [14], we performed a long-term feeding study. Quercetin was administrated at the same dose mentioned above three times/week for the additional 20 weeks to determine the concentrations of quercetin in the lungs and livers and the long-term toxicity

Eur J Nutr

of quercetin. During the 20-week experimental period, all of the animal body weights were recorded weekly. There were no significant differences in body weight among the groups in the same species (data not shown). At the end of the experiment, blood samples were collected from the retro-orbital plexus in a test tube containing heparin and were centrifuged (650×g, 15 min) to separate the plasma. After the blood was drawn, the animals were killed by CO2 asphyxiation. The livers, lungs and intestinal mucosa were collected. A portion of the lung or liver tissue from each animal was stored in 10 % formalin for histopathologic examination. The other tissues and plasma sample were stored at −80 °C until analysis. We did not find any toxic response by histopathologic examination in livers and lungs in all animals (data not shown). The plasma samples were analyzed within 1 week. Measurement of total quercetin and its metabolites We measured total or individual concentrations of quercetin and its metabolites, Q3G, Q3′S, and IH, in the plasma and tissues according to previous studies [14, 16]. Liver or lung tissues (0.3 g) were cut into small pieces and washed with PBS three times to avoid the systemic effect, and then, immediately homogenized in 300 μL potassium phosphate buffer (pH 7.0). The homogenized plasma or tissue samples (100 μL) were incubated with a 5 μL Helix pomatia enzyme mixture (~7500 U β-glucuronidase and ~750 U sulfatase in 0.5 M sodium acetate with 28 mM ascorbic acid, pH 5.0) at 37 °C for 2 h. Then, 200 μL of acetonitrile and 100 μL of 20 % H3PO4 were addedto the mixture to deproteinize. After centrifugation at 2300×g for 10 min at 10 °C, the supernatant was filtered and analyzed by HPLC with spectrophotometric detection at 370 nm [16]. The total quercetin concentration represents the sum of methyl- and non-methylquercetin concentrations. To determine the individual concentration of quercetin metabolites, samples were added to the same volume of sodium acetate buffer but without hydrolyzed enzyme. We prepared Q3G according the methods described by Chen et al. [17] as well Q3′S by the method described by Day et al. [8] as the standard compounds. The details have been described in our previous study [18].

For uridine 5′-diphosphate-glucuronosyltransferase (UDPGT), the tissue sample was mixed 1:1 (v/v) with 0.15 M Tris-HCl buffer (pH 7.6, containing 21 mM of d-saccharic acid, 1,4-lactone, 20 mM MgCl2 and 2 mM d, l-dithiothreitol, 2.5 mM UDPGA, 10 µM quercetin) and was incubated at 37 °C in a shaking water bath for 30 min. The amount conjugated with glucuronic acid was represented by the difference in quercetin levels between samples incubated with and without UDPGA. UDPGT activity was calculated using the formula: quercetin-conjugated/30× protein concentration (pmol/min mg protein). For phenolsulfotransferase (PST), the tissue sample was mixed 1:1 (v/v) with 0.1 M Tris-HCl buffer (pH 7.9, containing 2 mM MgCl2 and 20 mM of D, l-dithiothreitol, 1 mM adenosine-3′-phosphate-5′-phosphosulfate, 10 µM quercetin) and incubated at 37 °C in a shaking water bath for 30 min. The difference in quercetin content between the samples incubated with and without adenosine-3′-phosphate-5′phosphosulfate was assumed to be the amount conjugated with sulfate. PST activity was calculated using quercetin-conjugated/30× protein concentration (pmol/min mg protein). For Catechol-O-methyltransferase (COMT), the tissue sample was mixed 1:1 (v/v) with 0.1 M Tris-HCl buffer (pH 7.9, containing 20 mM MgCl2 and 2 mM D, l-dithiothreitol, 10 mM s-adenosyl-l-methionine, 10 µM quercetin) and was incubated at 37 °C in a shaking water bath for 30 min. The difference in quercetin content between the samples incubated with and without s-adenosyl-l-methionine was assumed to be the amount conjugated with the methyl group. COMT activity was calculated using methylquercetin concentration/30× protein concentration (pmol/ min mg protein). Quercetin content was determined with the HPLC method as described above. Statistical analysis Values are expressed as mean ± SD. Using one-way factorial analysis of variance (ANOVA) followed by Duncan’s multiple-range test, we compared group means. In addition, two-way ANOVA was performed to test the interaction of quercetin supplementation and animal species. P values <0.05 were considered statistically significant.

Determination of metabolic enzyme activities The phase II metabolic enzyme activities were determined according to the method modified from that of Piskula and Terao [19]. Briefly, liver and mucosa from the upper half of the small intestine were homogenized in 0.15 M Tris-HCl buffer, pH 7.6 (1:9, wt/v) with sonication and centrifuged at 1000×g for 15 min at 4 °C. The supernatants (tissue samples) were collected and stored at −80 °C until assayed. Protein concentrations of tissue samples were determined by the Lowry method [20].

Results Changes of plasma total quercetin and quercetin metabolites after a 5‑week supplementation Blood samples were collected at 1, 2, 3, 4, 6 and 72 h after oral administration of quercetin. As shown in Fig. 1, at about 2 h, the concentration of total quercetin in the plasma of all three species of animals significantly increased to a peak

13



To investigate the accumulation of quercetin in animals, the animals were supplemented with quercetin thrice/week for an additional 20 weeks. After an overnight fast, the animals were killed and the concentration of total quercetin was determined. As shown in Table 2, total quercetin concentrations in the plasma increased in a dose-dependent manner. The concentrations (about 4.5–6.5 μM) in the HQ treated group in all three species of animals were significantly higher than those in the control groups (P < 0.05). Similar to the finding in the time-course study, the total quercetin concentration in the gerbils and mice was higher than that of the rats. Regarding quercetin metabolite levels, the levels of Q3G was the highest in all three species of animals, followed by IH and Q3′S. The percentage of Q3G was more than 50 % in all animals. The gerbils still had a higher Q3′S concentration (1.8 μM) than the other two animals (0.6 and 0.7 μM for rats and mice, respectively). Quercetin metabolite levels in lungs and livers after additional 20 week supplementation As shown in Table 3, quercetin supplementation increased the levels of total quercetin, quercetin metabolites and QA

13

Total quercetin concentration (µ µM)

(A) C LQ (50 mg/kg) HQ (100 mg/kg)

8 c

6

b

b

4

b

b

a

a

b

b

2

a a

0

0

1

2

3

a a a

a a a

a

4

6

a a a

24

72

Time (h) 10

Total quercetin concentration (µ µM)

Quercetin metabolite levels in plasma after additional 20 weeks supplementation

10

(B)

C LQ (50 mg/kg) HQ (100 mg/kg)

c

8

6 b

4

b

b

2

b

ab

a a

a

a

a

a a a

a

a

0

18

Total quercetin concentration (µ µM)

(P < 0.05). The average concentration of total quercetin concentration at 2 h in the HQ groups of three species was in the following order: gerbils >mice, and rats (P < 0.05). In the HQ groups, the calculated maximum concentration (Cmax) of total quercetin for rats, mice and gerbils was 7.7, 8.5 and 14.2  μM, respectively, whereas the half-life was 1.6, 1.5 and 1.8 h, respectively. The levels of total quercetin almost returned to the baseline at 6 h. There was a dose effect of quercetin on the plasma concentration. Furthermore, we determined the concentration of each quercetin metabolite in the plasma in the HQ groups (Table 1). In the control group, after 1–4 h of vehicle administration, quercetin metabolite levels were not different (data not shown); thus, we presented the mean values of these metabolites in Table 1. Significant increases in Q3G and Q3′S levels were observed from 1–3 h after quercetin administration compared with the control groups of all animals. In the rats and mice, the increase in concentrations of Q3G tended to be earlier and higher than those of Q3′S. However, in the gerbils, Q3′S concentrations tended to be higher than those of Q3G at 1–4 h. In general, the concentrations of IH and QA at 1–4 h in the HQ treated group were not significantly different from the control group.

Eur J Nutr

16

0

1

2

3

4

6

Time (h)

a a a

a a a

24

(C)

72

C LQ (50 mg/kg) HQ (100 mg/kg)

14 c

12 b

10 8

b c

bc

bc

b bc

6 4

a a a

2

a

a

a

ab a a

ab a a

a

a a a

0 0

1

2

3

4

6

24

72

Time (h)

Fig. 1  Time-course study on concentration of total quercetin in plasma of Wistar rats (a), Balb/c mice (b) and Mongolian gerbils (c) after 1–72 h of oral administration of quercetin (LQ 50 mg/kg body weight or HQ 100 mg/kg body weight, once/week) in the first 5 weeks. The control group (c) was administered lard only. Values (mean ± SD, n = 6–7) not sharing a common letter are significantly different in the same group of each animal (P < 0.05)

Eur J Nutr Table 1  The concentration of quercetin -3-glucuronide (Q3G), quercetin-3′- sulfate (Q3′S), isorhamnetin (IH) and quercetin aglycone (QA) in the plasma of rats, mice and gerbils after 1–4 h of oral administration of high doses of quercetin (HQ, 100 mg/kg body weight) or the vehicle only (C) in the first 5 weeks

Group Rats  C  HQ

Mice  C  HQ

Values (mean ± SD, n = 6–7) in the same quercetin metabolite and the same animal not sharing a common letter are significantly different (P < 0.05). The values in the control group are mean value of 1–4 h

Time (h)

Q3G (μM) %1

1 2 3 4

0.44 ± 0.07a 1.75 ± 1.26ab 3.90 ± 1.39bc 4.71 ± 1.17c 4.33 ± 2.56bc

28 43 58 59 70

0.58 ± 0.20a 1.58 ± 0.82ab 2.11 ± 1.24ab 3.32 ± 1.32b 1.28 ± 0.93a

36 39 32 33 21

0.49 ± 0.02a 0.56 ± 0.12a 0.60 ± 0.12a 0.59 ± 0.09a 0.49 ± 0.10a

31 14 9 7 8

0.09 ± 0.01a 0.15 ± 0.14a 0.10 ± 0.04a 0.12 ± 0.07a 0.11 ± 0.01a

6 4 2 1 2

1 2 3 4

0.27 ± 0.02a 4.97 ± 0.11cd 5.62 ± 0.14d 3.66 ± 2.20bc 2.86 ± 0.65b

29 82 58 44 39

0.13 ± 0.12a 0.59 ± 0.18ab 3.45 ± 2.28bc 4.02 ± 2.17bc 3.73 ± 1.96c

14 10 36 49 50

0.38 ± 0.01bc 0.41 ± 0.06a 0.36 ± 0.01a 0.33 ± 0.02a 0.32 ± 0.02a

41 7 4 4 4

0.15 ± 0.07a 0.14 ± 0.04a 0.26 ± 0.12a 0.28 ± 0.09a 0.48 ± 0.02b

16 2 3 3 7

1 2 3

0.25 ± 0.04a 3.66 ± 0.76cd 4.38 ± 1.21d 3.06 ± 0.26bc

16 33 39 34

0.11 ± 0.09a 6.26 ± 1.66b 5.93 ± 2.55b 4.95 ± 2.55b

7 56 52 55

0.38 ± 0.01a 0.37 ± 0.06a 0.40 ± 0.07a 0.48 ± 0.16a

25 3 4 5

0.78 ± 0.11a 0.80 ± 0.23a 0.63 ± 0.27a 0.50 ± 0.19a

51 7 6 6

4

2.14 ± 0.39b

32

3.75 ± 2.34ab

55

0.37 ± 0.07a

6

0.51 ± 0.22a

8

Gerbils  C  HQ

1   % of sum (Q3G + Q3′S + IH + QA)

IH (μM) %

Q3′S (μM) %

QA (μM) %

Table 2  The concentration of total quercetin and quercetin -3-glucuronide (Q3G), quercetin-3′- sulfate (Q3′S) and isorhamnetin (IH) in the plasma of rats, mice and gerbils Group

Two-way ANOVA1

Plasm (μM) %2

Rats Total quercetin  C 1.75 ± 1.28aA  LQ 2.75 ± 1.01aAB  HQ 4.46 ± 1.04bA Q3G  C 0.98 ± 0.71aA  LQ 1.65 ± 0.68abA  HQ 2.22 ± 0.49bA Q3′S  C

aA

0.15 ± 0.08

Mice

%

Gerbils

%

– – –

1.70 ± 0.90aA 2.69 ± 1.13aA 5.80 ± 2.24bAB

– – –

2.17 ± 0.67aB 3.62 ± 0.89abB 6.59 ± 1.39bB

– – –

55 61 54

1.37 ± 0.42aA 2.11 ± 1.01abA 4.03 ± 0.61bB

56 64 66

1.03 ± 0.37aA 1.95 ± 0.58aA 3.80 ± 1.02aAB

52 55 55

aB

aC

 LQ  HQ IH  C  LQ

0.21 ± 0.12abA 0.56 ± 0.23bA

8 8 14

0.28 ± 0.01 0.32 ± 0.11abA 0.74 ± 0.09bB

11 10 12

0.42 ± 0.08 0.88 ± 0.07aB 1.81 ± 0.10bC

21 25 26

0.65 ± 0.31aA 0.84 ± 0.21aA

37 31

0.81 ± 0.22aA 0.88 ± 0.14aA

33 28

0.53 ± 0.12aA 0.72 ± 0.20aA

27 20

 HQ

1.32 ± 0.34aA

32

1.36 ± 0.21bA

22

1.27 ± 0.27aA

18

Quercetin

Species

Q × S

<0.001

<0.001

NS

0.003

<0.001

NS

0.03

NS

NS

0.04

NS

NS

The animals were administered quercetin (LQ: 50 mg/kg body weight or HQ: 100 mg/kg body weight, 3 times/week) or lard only (C) by gavage for 20 weeks Values (mean ± SD, n = 6–7) not sharing a common small letter are significantly different (P < 0.05) among groups in the same species; while values without a common capital letter are significantly different among the three kinds of animal fed with the same dose of quercetin 1

  P value for two-way ANOVA analysis. NS not significant (P > 0.05)

2

  % of sum (Q3G + Q3′S + IH)

13



Eur J Nutr

in the lungs of all animals in a dose-dependent manner. The total quercetin concentration in the mice and gerbils was significantly higher than in the rats (P < 0.05). The proportion of Q3G was the highest of the metabolites in all three species of animals. The levels of Q3′S were similar to or lower than those of IH in all animals. In the livers (Table 4), the concentrations of total quercetin and quercetin metabolites were increased by quercetin administration in a dose-dependent manner in all animals. The QA level increased in a quercetin dose-dependent manner only in the gerbils. The total quercetin level in the gerbils was still the highest. The proportion of Q3G was also the highest of all the metabolites in all animals. The IH levels seemed higher than those of Q3′S especially in the rats and mice.

level of quercetin and species in UDPGT activity was significant (two-way ANOVA, P = 0.014). Similar to the finding in the intestinal mucous membrane, quercetin supplementation increased all enzyme activities in a dose-dependent manner in livers (Table 5). Gerbils in the HQ group had the highest PST activity (311 pmol/min mg protein), followed by the mice (231 pmol/min mg protein) and rats (212 pmol/min mg protein) with HQ treatment (P < 0.05). The interaction between the level of quercetin and species in PST activity was significant (two-way ANOVA, P = 0.002). Rats tended to have the highest COMT activity.

Discussion UDPGT, PST and COMT activity in intestinal mucous membrane and liver

When using animals to investigate possible physiological roles of quercetin administrated orally in humans, it is essential to understand the metabolism of quercetin in animals. In the present study, we found that at about 2 h after quercetin was fed orally (50 and 100 mg/kg body weight) the total quercetin plasma concentrations in rats, mice and gerbils reached their peak (about 6–12 μM) in a dose-dependent manner and decreased to baseline at about 6 h (Fig. 1), despite situations being a little different with other quercetin metabolites (Table 1). These results were comparable to those found in humans. The levels of Cmax

To clarify the effect of quercetin supplementation on the phase II metabolic enzymes, the UDPGT, PST and COMT activity in the intestinal mucous and liver were analyzed. Quercetin supplementation increased UDPGT, PST and COMT activity in all three species of animals in a dosedependent manner (Table 5). The highest UDPGT activity was found in the mice (628 pmol/min mg protein), while the highest PST activity was observed in the gerbils (28 pmol/min mg protein). The interaction between the

Table 3  The concentration of total quercetin, quercetin -3-glucuronide (Q3G), quercetin-3′-sulfate (Q3′S), isorhamnetin and quercetin aglycone (QA) in lungs of Wistar rats, Balb/c mice and Mongolian gerbils The animals were administered quercetin (LQ: 50 mg/kg body weight or HQ: 100 mg/kg body weight, 3 times/week) or lard only (C, control) by gavage for 20 weeks Values (mean ± SD, n = 6–7) not sharing a common small letter are significantly different (P < 0.05) among groups of the same animals; while values without a common capital letter are significantly different among the three kinds of animals fed with the same dose of quercetin 1

  P value, two-way ANOVA. NS not significant (P > 0.05) 2   % of sum (Q3G + Q3′S + IH + QA)

13

Group

%2

Rats Total quercetin  C 10.3 ± 2.8aA  LQ 25.3 ± 10.4bA  HQ 26.9 ± 16.1bA QA  C 0.2 ± 0.1aA  LQ 0.6 ± 0.2abA  HQ 1.3 ± 0.6bA Q3G  C 6.6 ± 1.9aA  LQ 15.6 ± 2.7bA  HQ 17.9 ± 3.5bA Q3′S  C

Two-way ANOVA1

Lung (nmol/g tissue) Mice

%

Gerbils

%

– – –

15.0 ± 10.2aA – 30.2 ± 6.5bA – 39.2 ± 11.8bB –

2 3 5

0.6 ± 0.2aB 3 0.9 ± 0.4AB 3 1.2 ± 0.8bA 3

0.5 ± 0.2aB 1.1 ± 0.4abB 1.9 ± 0.7bB

4 4 5

64 66 63

10.5 ± 4.7aA 59 22.4 ± 3.8abA 70 30.2 ± 2.4bB 68

8.8 ± 3.2aA 19.8 ± 6.1abA 24.6 ± 4.7abB

62 63 58

aA

aB

Quercetin

Species

Q × S

0.001

NS

NS

0.04

NS

NS

0.001

0.04

NS

0.006

<0.001

NS

0.01

NS

NS

11.2 ± 8.7aA – 27.5 ± 19.9abA – 40.9 ± 28.9bB –

aB

 LQ  HQ IH  C  LQ

1.1 ± 0.1 11 2.5 ± 0.3bA 11 3.6 ± 1.0abA 13

2.7 ± 1.5 15 3.6 ± 1.1bAB 11 5.4 ± 2.7bB 12

2.1 ± 0.5 15 5.1 ± 1.7bAB 16 8.8 ± 0.9cC 21

2.5 ± 0.2aA 4.8 ± 1.4bA

24 20

4.1 ± 1.2aB 5.2 ± 1.3bA

23 16

2.7 ± 1.5aAB 5.6 ± 2.2abA

19 18

 HQ

5.8 ± 2.1bA

20

7.8 ± 2.5bA

18

6.9 ± 2.4bA

16

Eur J Nutr Table 4  The concentration of total quercetin, quercetin -3-glucuronide (Q3G), quercetin-3′-sulfate (Q3′S), isorhamnetin and quercetin aglycone (QA) in livers of Wistar rats, Balb/c mice and Mongolian gerbils The animals were administered quercetin (LQ: 50 mg/kg body weight or HQ: 100 mg/kg body weight, 3 times/week) or lard only (C, control) by gavage for 20 weeks Values (mean ± SD, n = 6–7) not sharing a common small letter are significantly different (P < 0.05) among groups of the same animals, while values without a common capital letter are significantly different among the three kinds of animals fed with the same dose of quercetin 1

  P value, two-way ANOVA. NS  not significant (P > 0.05) 2   % of sum (Q3G + Q3′S + IH + QA)

Group

%2

Rats Total quercetin  C 16.0 ± 4.8aB  LQ 27.5 ± 7.1abA  HQ 34.8 ± 8.7bB QA  C 1.1 ± 0.2aB  LQ 0.8 ± 0.5aA  HQ 0.9 ± 0.4aA Q3G  C 7.3 ± 3.2aA  LQ 12.9 ± 8.1aA  HQ 16.8 ± 9.4aA Q3′S  C  LQ  HQ IH  C  LQ  HQ

Two-way ANOVA1

Liver (nmol/g tissue) Mice

– – – 7 3 3

% aA

8.7 ± 5.8 – 24.3 ± 6.9bA – 27.4 ± 10.3bA – 0.3 ± 0.1aA 0.7 ± 0.3aA 0.7 ± 0.2aA

3 3 2

48 47 49

5.4 ± 4.1aA 56 15.4 ± 10.7bA 58 22.7 ± 6.3bAB 65

2.5 ± 0.6aAB 17 3.7 ± 0.8abA 14 5.3 ± 1.5bA 16

1.7 ± 0.3aA 18 3.8 ± 1.0abA 14 4.5 ± 0.4bA 13

aAB

aA

Gerbils

%

Quercetin

Species

Q × S

0.004

0.04

NS

NS

NS

NS

<0.001

0.01

NS

0.02

0.001

NS

<0.001

NS

NS

aB

22.3 ± 12.5 – 39.6 ± 10.0abB – 58.1 ± 24.3bC – 0.9 ± 0.4aB 1.7 ± 0.6aB 2.5 ± 1.1bB

4 4 4

13.4 ± 3.2aB 55 23.3 ± 7.8abB 55 35.2 ± 12.6bB 58 4.7 ± 1.1aB 8.8 ± 2.0abB 10.4 ± 2.2bB aB

19 21 17

4.2 ± 2.1 9.8 ± 2.1aB

28 36

2.3 ± 0.7 6.9 ± 3.1bA

24 26

5.2 ± 1.8 8.6 ± 3.1bB

22 20

11.3 ± 0.7aB

33

7.2 ± 2.2bA

21

12.6 ± 5.6bB

21

Table 5  The activity of uridine 5′-diphosphate-glucuronosyltransferase (UDPGT), phenolsulfotransferase (PST) and catechol-O-methyltransferase (COMT) in intestinal mucous membrane and liver of rats, mice and gerbils Group

UDPGT  C  LQ  HQ PST  C  LQ  HQ COMT  C  LQ  HQ

Intestine pmol/min mg protein

Liver pmol/min mg protein

Rats

Mice

Gerbils

Rats

257.1 ± 51.6ab 338.1 ± 42.6bcd 528.8 ± 74.9e

237.2 ± 52.5ab 373.8 ± 58.8 cd 627.8 ± 81.9e

199.2 ± 32.9a 285.8 ± 51.4abc 421.2 ± 61.2d

23.2 ± 8.9bc 36.0 ± 10.2abc 42.1 ± 10.9a

Mice

Gerbils

30.2 ± 7.6abc 44.7 ± 12.5a 48.3 ± 10.3a

19.2 ± 9.23c 33.75 ± 5.9abc 38.2 ± 11.2ab

14.2 ± 1.1ab 18.2 ± 0.5bc 20.9 ± 1.2c

11 ± 2.8a 13.6 ± 1.5a 19.7 ± 2.5c

12.3 ± 2.1a 18.6 ± 4.2c 28.3 ± 3.4d

134.2 ± 22.5a 179 ± 26.7ab 212 ± 13.5b

129.3 ± 34.9a 197.8 ± 19.9b 231.3 ± 30.1bc

141.9 ± 9.1a 272.9 ± 31.6cd 311.2 ± 50.4d

88.1 ± 7.1a 135.7 ± 11.8bc

112.6 ± 15.1ab 157.2 ± 26.8c

95.4 ± 1.9a 102.3 ± 23.6a

235.6 ± 84a 682.3 ± 83.2cd

224.9 ± 91.2a 499.5 ± 48.1b

324.1 ± 99.9a 538.8 ± 80.2bc

201.4 ± 14.7d

188.6 ± 21.7d

157.5 ± 17.8c

856.2 ± 113.7e

639.1 ± 51.4bcd

727.8 ± 104.8de

The animals were administered quercetin (LQ: 50 mg/kg body weight or HQ: 100 mg/kg body weight, 3 times/week) by gavage for 20 weeks. The control group (C) was administered lard only Values (mean ± SD, n = 6–7) in the same enzyme (in each tissue) not sharing a common small letter are significantly different (P < 0.05)

in the plasma and the half-life of quercetin metabolites are associated with quercetin doses, food patterns and subjects [10, 21–23]. In healthy subjects, a quercetin supplement of 500 mg three times daily for 7 days results in a terminal average half-life of 3.5 h; the Cmax of total quercetin

in the plasma ranges from 0.5–4 μM [24]. Graefe et al. [10] found that after intake of quercetin-4′-glucose (about 100 mg quercetin) the Cmax of total quercetin in the plasma of healthy volunteers is about 7 μM and is reached at 0.7 h. In addition, Mullen et al. [25] found that the time reaching

13



Cmax and half-life are different among metabolites ranging from 0.75–2.5 h and 1.7–5.5 h, respectively, in healthy humans. As we compared the metabolite profiles in the plasma and tissues of the three species of animals, we found that the gerbils converted quercetin to Q3′S more than the mice and rats. After 2 h of quercetin administration, the ratios of Q3′S/total quercetin in gerbils, mice and rats were about 52, 36 and 32%, respectively (Table 1). In addition, after long-term supplementation, the ratios of Q3′S in the gerbil tissues also tended to be higher than in the mice and rats, even though Q3G was the dominant metabolite in all animals after long-term supplementation of quercetin. A previous study performed in the UK showed that 1–3 h after the consumption of 270 g of fried onions, the Q3′S concentration is about twofold higher than the Q3G concentration in the plasma of healthy volunteers [25]. Recently, Nakamura et al. [23] showed that Q3G and methylated quercetin rather than Q3′S are the dominant metabolites in humans after consumption of onions only; however, the level of Q3′S increased after consumption of onions in combination with tofu. These authors pointed out the doses of quercetin and the other components in meals affect the metabolic conversion of quercetin. In addition, a study in which a quercetincontaining supplement was given for 3 months showed that after supplementation, the concentration of Q3G is about tenfold of the Q3′S concentration [26], whereas the level of isorhamnetin-3-glucuronide is the highest. The results of these studies imply that the profiles of quercetin metabolites in humans may vary, depending on many factors including the subjects. Thus, it may be difficult to conclude which species of animal is a better animal model for humans especially for acute effects. However, consistent with the finding of Cialdella-Kam et al. [26], we found that the ratio of methylated quercetin increased after long-term supplementation. Although we did not determine the level of isorhamnetin-3-glucuronide, it is possible that the metabolite is also present in the plasma and tissues of these animals. Herein, we unexpectedly found that after long-term supplementation with HQ, the levels of QA in the lungs of all the animals were significantly increased, despite the QA level in the plasma being similar to the control (Table 1). This finding suggests that quercetin-conjugated metabolites may be deconjugated and converted to quercetin in tissues. It has been suggested that quercetin metabolites may convert to quercetin in humans under some physiological conditions such as in activated macrophages [27, 28]. Our recent study also found that Q3G incubation with A549 cells, a human lung cell line, significantly increases the intracellular QA level and the activity of β-glucuronidase, which converts Q3G to QA [29]. The conversion of quercetin metabolites by lung cells may increase the physiological activity of oral quercetin in lungs. However, further studies

13

Eur J Nutr

are warranted to investigate the mechanism and the significance of the increase in QA in the lung tissues of these animals. In the present study, we found that all three animal species accumulated total quercetin in the lungs and livers at similar levels (25–40 nmol/g tissue and 24–58 nmol/g tissue, respectively). These levels are comparable to those reported by de Boer et al. [14] in Fisher 344 rats that after oral administration of quercetin (500 mg/kg body weight/ week) for 11 weeks, quercetin and its metabolites are widely distributed in tissues with the highest quercetin concentration in the lungs (15.3 nmol/g). In pigs, however, the liver, intestinal wall and kidneys rather than the lungs have the high levels of quercetin and its metabolites after long-term quercetin supplementation [15]. Little has been reported regarding the distribution of quercetin metabolites in the tissues of humans. Several epidemiological studies have shown that foods rich in quercetin may benefit the lungs, neurons and cardiovascular system, suggesting that quercetin or its metabolites are distributed to these tissues [11, 30, 31]. Our recent study also showed that quercetin and Q3G can enter and accumulate in human carcinoma cells [29]. UDPGT, PST and COMT, commonly present in the intestines and liver with several isoforms [32], are phase II enzymes which convert quercetin to glucuronated, sulfated and methylated metabolites, respectively [11]. The distribution and activity of these enzymes vary among subjects and are associated with tissues and dietary compounds [23, 33]. Among the animals we compared, gerbils tended to have a higher PST activity, mice had a higher UDPGT activity and rats had a higher COMT activity. The trend was associated with the converted ratio of quercetin metabolites in the plasma and tissues among these animals. It has been shown that quercetin increases UDPGT activity in the intestines and, to a lesser extent, in the liver of rats [33]. In the present study, we found that in the HQ group (but not LQ group) of each animal, especially in mice, the percentage of UDPGT activity increased by quercetin in the intestinal mucosal membrane tended to be higher than that in the liver (intestine vs. liver: 105 vs. 82 % for rats; 164 vs. 60 % for mice; 111 vs. 100 % for gerbil). It has been reported that UDPGT1A1 is expressed highly in the small intestine, with activities even greater than in the liver [34]. Although we did not determine which isoform responded to quercetin metabolism, it has been suggested that in rats, UDPGT1A1, 1A6 and 1A7 are highly expressed in the liver and intestines [35] and may mainly convert dietary compounds, including quercetin, to conjugated metabolites. Different from UDPGT, PST had more activity in the liver than in the intestinal mucosal membrane in all the animals after quercetin supplementation. A study of human liver Hep G2 cells supports our finding, which shows that the major biotransformation of quercetin is the conjugation

Eur J Nutr

to sulfate and, to a lesser extent, to glucuronic acid in Hep G2 cells [36]. In addition, the COMT activities in the liver were also higher than in the intestines, especially in rats. This finding is consistent with the findings of Lu et al. [37]. The study found that rat liver has higher COMT activity than in humans and mice. They also found that the small intestine has lower COMT activity than the liver induced by epigallocatechin gallate and epigallocatechin. In humans, the liver has the highest COMT activity, followed by the kidneys and intestines [38]. Thus, our findings suggest that the difference in activities of the three phase II enzymes between the intestines and liver in all three species of animals is comparable to that in humans. In summary, the present study demonstrates that after both short-term and long-term supplementation, gerbils have the highest accumulation of total quercetin and Q3′S conversion ratios. However, similar to humans, after longterm supplementation, the level of Q3G is higher than Q3′S in the plasma, lungs and livers of all the animals. In addition, the activities of the three phase II enzymes are induced by quercetin in a dose- and tissue-dependent manner in all three species of animals as in humans. These results provide more insight into the metabolism of quercetin in rats, mice and gerbils, especially in the activity of metabolic enzymes that is important in using animal models to study the physiological or pharmacological activity of quercetin. Acknowledgments  This study was supported by a Grant (NSC-982320-B- 241-001-MY3) from the National Science Council, Republic of China. Liquid chromatography-mass spectrometry was performed in the Instrument Center at Chung Shan Medical University, which is supported by the National Science Council, Ministry of Education and Chung Shan Medical University. Conflict of interest  The authors declare no financial conflict of interest.

References 1. Arts IC, Hollman PC (2005) Polyphenols and disease risk in epidemiologic studies. Am J Clin Nutr 81:317S–325S 2. Vauzour D, Rodriguez-Mateos A, Corona G et al (2010) Polyphenols and human health: prevention of disease and mechanisms of action. Nutrients 2:1106–1131 3. Knekt P, Järvinen R, Seppänen R et al (1997) Dietary flavonoids and the risk of lung cancer and other malignant neoplasms. Am J Epidemiol 146:223–230 4. Lakhanpal P, Rai DK (2007) Quercetin: a versatile flavonoid. Internet J Med Update 2:22–37 5. Ross JA, Kasum CM (2002) Dietary flavonoids: bioavailability, metabolic effects, and safety. Annu Rev Nutr 22:19–34 6. Xu R, Zhang Y, Ye X et al (2013) Inhibition effects and induction of apoptosis of flavonoids on the prostate cancer cell line PC-3 in vitro. Food Chem 138:48–53 7. Thilakarathna SH, Rupasinghe HP, Needs PW (2013) Apple peel bioactive rich extracts effectively inhibit in vitro human LDL cholesterol oxidation. Food Chem 138:463–470

8. Day AJ, Mellon F, Barron D et al (2001) Human metabolism of dietary flavonoids: identification of plasma metabolites of quercetin. Free Radic Res 35:941–952 9. Murota K, Terao J (2003) Antioxidative flavonoid quercetin: implication of its intestinal absorption and metabolism. Arch Biochem Biophys 417:12–17 10. Graefe EU, Wittig J, Mueller S et al (2001) Pharmacokinetics and bioavailability of quercetin glycosides in humans. J Clin Pharmacol 41:492–499 11. Terao J, Murota K, Kawai Y (2011) Conjugated quercetin glucuronides as bioactive metabolites and precursors of aglycone in vivo. Food Funct 2:11–17 12. Suri S, Liu XH, Rayment S et al (2010) Quercetin and its major metabolites selectively modulate cyclic GMP-dependent relaxations and associated tolerance in pig isolated coronary artery. Br J Pharmacol 159:566–575 13. Yeh SL, Yeh CL, Chan ST et al (2011) Plasma rich in quercetin metabolites induces G2/M arrest by up-regulating PPAR-γ expression in human A549 lung cancer cells. Planta Med 77:992–998 14. de Boer VC, Dihal AA, van der Woude H et al (2005) Tissue distribution of quercetin in rats and pigs. J Nutr 135:1718–1725 15. Bieger J, Cermak R, Blank R et al (2008) Tissue distribution of quercetin in pigs after long-term dietary supplementation. J Nutr 138:1417–1420 16. Day AJ, Bao Y, Morgan MR et al (2000) Conjugation position of quercetin glucuronides and effect on biological activity. Free Radic Biol Med 29:1234–1243 17. Chen YK, Chen SQ, Li X et al (2005) Quantitative regioselectivity of glucuronidation of quercetin by recombinant UDP-glucuronosyltransferases 1A9 and 1A3 using enzymatic kinetic parameters. Xenobiotica 35:943–954 18. Chang YZ, Lin HC, Chan ST et al (2012) Effects of quercetin metabolites on the enhancing effect of β-carotene on DNA damage and cytochrome P1A1/2 expression in benzo[a]pyreneexposed A549 cells. Food Chem 133:445–450 19. Piskula MK, Terao J (1998) Accumulation of (−)-epicatechin metabolites in rat plasma after oral administration and distribution of conjugation enzymes in rat tissues. J Nutr 128:1172–1178 20. Lowry OH, Rosebrough NJ, Farr AL et al (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275 21. Egert S, Wolffram S, Bosy-Westphal A et al (2008) Daily quercetin supplementation dose-dependently increases plasma quercetin concentrations in healthy humans. J Nutr 138:1615–1621 22. Lee J, Ebeler SE, Zweigenbaum JA et al (2012) UHPLC-(ESI) QTOF MS/MS profiling of quercetin metabolites in human plasma postconsumption of applesauce enriched with apple peel and onion. J Agric Food Chem 60:8510–8520 23. Nakamura T, Murota K, Kumamoto S et al (2014) Plasma metabolites of dietary flavonoids after combination meal consumption with onion and tofu in humans. Mol Nutr Food Res 58:310–317 24. Moon YJ, Wang L, DiCenzo R et al (2008) Quercetin pharmacokinetics in humans. Biopharm Drug Dispos 29:205–217 25. Mullen W, Edwards CA, Crozier A (2006) Absorption, excretion and metabolite profiling of methyl-, glucuronyl-, glucosyland sulpho-conjugates of quercetin in human plasma and urine after ingestion of onions. Br J Nutr 96:107–116 26. Cialdella-Kam L, Nieman DC, Sha W et al (2013) Dose response to 3 months of quercetin-containing supplements on metabolite and quercetin conjugate profile in adults. Br J Nutr 109:1923–1933 27. Williamson G, Manach C (2005) Bioavailability and bioefficacy of polyphenols in humans. II. Review of 93 intervention studies. Am J Clin Nutr 81:243S–255S 28. Kawai Y, Nishikawa T, Shiba Y et al (2008) Macrophage as a target of quercetin glucuronides in human atherosclerotic arteries:

13

implication in the anti-atherosclerotic mechanism of dietary flavonoids. J Biol Chem 283:9424–9434 29. Chan ST, Lin YC, Chuang CH et al (2014) Oral and intraperitoneal administration of quercetin decreased lymphocyte DNA damage and plasma lipid peroxidation induced by TSA in vivo. Biomed Res Int 2014:580626 30. Dajas F, Andrés AC, Florencia A (2013) Neuroprotective actions of flavones and flavonols: mechanisms and relationship to flavonoid structural features. Cent Nerv Syst Agents Med Chem 13:30–35 31. Sak K (2014) Site-specific anticancer effects of dietary flavonoid quercetin. Nutr Cancer 66:177–193 32. Kaminsky LS, Zhang QY (2003) The small intestine as a xenobiotic-metabolizing organ. Drug Metab Dispos 31:1520–1525 33. van der Logt EM, Roelofs HM, Nagengast FM et al (2003) Induction of rat hepatic and intestinal UDP-glucuronosyltransferases by naturally occurring dietary anticarcinogens. Carcinogenesis 24:1651–1656 34. Fisher MB, Vandenbranden M, Findlay K et al (2000) Tis sue distribution and interindividual variation in human

13

Eur J Nutr UDP- glucuronosyltransferase activity: relationship between UGT1A1 promoter genotype and variability in a liver bank. Pharmacogenetics 10:727–739 35. Grams B, Harms A, Braun S et al (2000) Distribution and inducibility by 3-methylcholanthrene of family 1 UDP-glucuronosyltransferases in the rat gastrointestinal tract. Arch Biochem Biophys 377:255–265 36. Bursztyka J, Perdu E, Pettersson K et al (2008) Biotransformation of genistein and bisphenol A in cell lines used for screening endocrine disruptors. Toxicol In Vitro 22:1595–1604 37. Lu H, Meng X, Yang CS (2003) Enzymology of methylation of tea catechins and inhibition of catechol-O-methyltransferase by (−)-epigallocatechin gallate. Drug Metab Dispos 31:572–579 38. Männistö PT, Kaakkola S (1999) Catechol-O-methyltransferase (COMT): biochemistry, molecular biology, pharmacology, and clinical efficacy of the new selective COMT inhibitors. Pharmacol Rev 51:593–628