Abstract
Quercetin-4′-glucoside is a major flavonol in onions, and this study investigated the absorption and fate of radiolabeled quercetin-4′-glucoside in rats. Rats ingested [2-14C]quercetin-4′-glucoside and the distribution of radioactivity throughout the body was determined after 0.5, 1, 2, and 5 h. The gastrointestinal tract, liver, kidney, and plasma were extracted, and radiolabeled components were identified and quantified using high-performance liquid chromatography with on-line radioactivity detection and tandem mass spectrometry. Two hours after dosing, all the [2-14C]quercetin-4′-glucoside had been metabolized. More than 85% of the ingested radioactivity was present in the gastrointestinal tract at all time points with ∼6% being absorbed and present in blood and internal organs, primarily the liver and kidneys. More than 95% of the absorbed radioactivity was in the form of >20 different methylated glucuronated and/or sulfated quercetin conjugates. Five hours after ingestion, the main radiolabeled metabolites were quercetin diglucuronides in the gut, liver, and kidneys and glucuronyl sulfates of methylated quercetin in plasma. The main site of quercetin metabolism seemed to be the gastrointestinal tract. Quercetin metabolites may have a major influence on the gut mucosal epithelium and on colonic disease.
There has been much interest in the role of nonvitamin antioxidants in plant foods and the prevention of heart disease and cancer. Molecules such as quercetin have much greater antioxidant potential in vitro than vitamins A, C, and E (Rice-Evans et al., 1996; Shi et al., 2001). Epidemiological and laboratory studies suggest that intake of dietary quercetin, which in plants exists almost exclusively as glycoside conjugates, may reduce the incidence of coronary heart disease and certain cancers, including colon cancer (Geleijnse et al., 2002; Knekt et al., 2002). Numerous in vitro and ex vivo experiments suggest the mechanisms may include delayed oxidation of low-density lipoprotein, inhibition of platelet aggregation, anti-inflammatory properties, modulation of enzyme activities associated with carcinogen activation and detoxification, prevention of oxidative DNA damage and modulation of gene expression, and apoptosis and malignant transformation (Duthie et al., 2000; Middleton et al., 2000). However, it is still not established whether quercetin glycosides are absorbed from the gastrointestinal (GI) tract intact or whether they are metabolized before, during, or after absorption. Studies with ileostomy patients indicated the disappearance of more than 50% of quercetin glucosides from onion before the terminal ileum (Hollman et al., 1995; Walle et al., 2000). However, this may have been the result of metabolism in the gut rather than absorption. Less than 2% of the ingested quercetin was found in plasma (Manach et al., 1998; Erlund et al., 2000; Moon et al., 2000) and less than 7% in urine (Graefe et al., 2001). If quercetin glycosides are metabolized in the GI tract, the resultant metabolites may then be absorbed, but equally they may remain in the gut where they could play a role in the maintenance of GI health due to their proximity to epithelial cells in the small and large intestine. The activity of potential metabolites may not be the same as the parent compounds and thus it is essential that the exact fate of ingested flavonols be determined.
Quercetin-4′-glucoside is a major flavonol conjugate in onions (Tsushida and Suzuki, 1995), and in the present study, [2-14C]quercetin-4′-glucoside was fed to rats to investigate the dynamics of quercetin absorption, metabolite formation, and accumulation in body tissues and plasma. We have reported previously that 1 h after ingestion of 14C-labeled quercetin-4′-glucoside, only 6.4% of the dose was absorbed from the GI tract of rats. All absorbed radiolabeled quercetin had undergone metabolic modification (Mullen et al., 2002). However, the fate of the metabolites during the normal passage of food through the small intestine and into the colon was unknown. Consequently, this article reports on the dynamics of quercetin metabolism, absorption, and biotransformation in rat plasma and tissues over a 5-h period after ingestion of [2-14C]quercetin-4′-glucoside.
Materials and Methods
Chemicals. [2-14C]Quercetin-4′-O-β-d-glucoside (Fig. 1), specific activity 138.8 MBq/mmol, was synthesized using procedures based on the previously reported method for the synthesis of [2-13C]quercetin-4′-O-β-d-glucoside (Caldwell et al., 2000), except that the intermediate ester was not purified by filtration through alumina. Standards of quercetin and kaempferol were purchased from Sigma Chemical (Poole, Dorset, UK). Isorhamnetin was obtained from Apin Chemicals Ltd. (Abingdon, Oxon, UK). A sample of quercetin-3-glucuronide was kindly supplied by Prof. Gary Williamson (Food Research Institute, Norwich, UK). Quercetin-4′-glucoside was generously provided by Dr. T. Tsushida (National Food Research Institute, Ibaraki, Japan). All other chemicals were of analytical grade and solvents were of HPLC grade purchased from Rathburn Chemicals Ltd. (Walkerburn, Peebleshire, UK).
Animals and Oral Administration of [2-14C]Quercetin-4′-Glucoside. After an overnight fast, 12 male rats (Rowett Hooded Lister strain; mean weight 430 ± 4 g) were each offered 1 g of stock rat feed (CRM; Special Diet Services, Witham, Essex, UK) containing 3.26 mg (7.6 mg/kg) of [2-14C]quercetin-4′-glucoside to give a radioactivity dose of 58.5 × 106 dpm. The rats consumed all of the ration within 2 min.
Sample Collection and Preparation. After 0.5, 1, 2, and 5 h, three rats were terminally anesthetized with isofluorane, and blood removed by cardiac puncture was collected in heparinized evacuated tubes (BD Biosciences, Oxford, UK). Plasma was obtained by centrifugation at 1500g for 10 min at 4°C. The pelleted red blood cells were resuspended in 9 ml of phosphate-buffered saline, pH 7.4. Livers were perfused in situ with chilled 0.15 M KCl and then removed along with brain, heart, kidney, lung, muscle, spleen, and testes. The GI tract, including its contents, was removed intact. All samples were frozen in liquid nitrogen and stored at –80°C before lyophilization after which they were ground to a powder using a mortar and pestle and again stored at –80°C.
Measurement of Radioactivity. Aliquots of the lyophilized tissue, plasma, and red blood cells were treated with tissue solubilizer (National Diagnostics, Hull, UK) for 3 h at 50°C in a shaking water bath. With the exception of red blood cells, which were bleached using 1.75 ml of a 25% solution of sodium hypochlorite, the solubilization treatment produced clear solutions. Aliquots of the solubilized tissue were added to scintillation fluid (Optiflow Safe One, Fisons, Longhborough, UK), and the radioactivity was determined using a Wallac 1409 liquid scintillation counter (Pharmacia Diagnostics AB, Uppsala, Sweden).
Extraction of Samples. Lyophilized tissues from rats sacrificed at the same time point were pooled and extracted as described previously (Mullen et al., 2002). Briefly, aliquots of the pooled, lyophilized tissues were extracted three times with 50% methanol in 0.1 M phosphate buffer, pH 7.0, containing 20 mM sodium diethydithiocarbamate as an antioxidant. After centrifugation, the three methanolic supernatants were combined, and the methanol was removed in vacuo. The remaining aqueous phase was adjusted to pH 3 and partitioned three times with an equal volume of ethyl acetate. The ethyl acetate extracts were combined and reduced to dryness in vacuo, and then resuspended and stored in methanol at –80°C. Metabolites present in the aqueous phase were further purified using a C18 Sep Pak cartridge (Waters, Milford, MA), which was eluted with methanol. The eluent was reduced to dryness in vacuo, resuspended in methanol, and stored at –80°C. To precipitate proteins, plasma was treated twice for 10 min with 2.5 volumes of acetone. After centrifugation the two acetone extracts were combined and reduced to the aqueous phase in vacuo, and ethyl acetate and aqueous extracts were obtained as described above. More than 85% of the radioactivity, which was originally present in the tissues/plasma, went into solution during the extraction process. Before analysis by HPLC-radiocounting (RC)-tandem mass spectrometry (MS-MS), the ethyl acetate extracts and the aqueous extracts were combined, and aliquots were dissolved in HPLC mobile phase.
Analysis by HPLC-RC-MS-MS. Aliquots of plasma and tissue extracts containing ≈30,000 dpm of radioactivity were analyzed on a P4000 liquid chromatograph fitted with an AS 3000 autosampler and with the initial detection by a UV6000 diode array absorbance monitor scanning from 250 to 700 nm (Thermo Electron Corporation, Waltham, MA). Separation of the radiolabeled compounds was carried out using a 250 × 4.6 mm i.d. 4 μM Synergy RP-Max column (Phenomenex, Macclesfield, UK), at 40°C, eluted with a 60-min gradient of 5 to 40% acetonitrile in 1% aqueous formic acid at a flow rate of 1 ml/min. After passing through the flow cell of the absorbance monitor, the column eluate was split by 50% and simultaneously directed to a radioactivity monitor (Reeve Analytical model 9701; Lab Logic, Sheffield, UK), and a Finnigan LCQ Duo tandem mass spectrometer with an electrospray interface operating in negative ion mode and scanning from 150 to 2000 atomic mass units.
Results
Distribution of Radioactivity in Body Tissues and Plasma. Distribution of radioactivity in rats was determined at 0.5, 1, 2, and 5 h after the ingestion of [2-14C]quercetin-4′-glucoside. At all time points, most (86–93%) of the ingested radioactivity was still present in the GI tract, which included the stomach, small and large intestine, and their contents (Table 1). At 30 min, 7.2% of the ingested radioactivity was present in internal organs and blood, indicating early absorption from the upper GI tract. The absorbed radioactivity was distributed throughout all analyzed organs, but the amounts present in the spleen, red blood cells, and the brain were minute. The maximum amount detected in the brain, for instance, was only 1000 dpm (8.2 nmol), which represents 0.0017% of the ingested radioactivity. Highest levels were detected in plasma with a peak concentration of 2.9% of the ingested radioactivity (14 μM) at 0.5 h. The total radioactivity present in internal organs and plasma was highest at 0.5 h and decreased thereafter during the time course of the experiment.
Separation and Identification of Radiolabeled Metabolites. HPLC-RC analysis of methanolic extracts of the rat feed confirmed that the only radiolabeled compound present was [2-14C]quercetin-4′-glucoside, and no breakdown products had formed before dosing. The HPLC-RC profiles obtained from tissue and plasma extracts revealed a complex pattern of radiolabeled metabolites (Fig. 2). Using negative ion, electrospray, tandem mass spectrometry, 22 of 26 detected metabolites were identified as quercetin conjugates with one or two glucuronyl, methyl, or sulfate groups attached (Table 2). The use of full scan tandem mass spectrometric fragmentation patterns to identify these quercetin metabolites has been discussed in detail by Mullen et al. (2003). In the present study, eight new radiolabeled metabolites were detected (peaks 1, 2, 7, 14, 15, 17, 23, and 27) along with 19 metabolites identified previously (Mullen et al., 2002). Peaks 1, 2, and 4 were not identified because they did not yield recognizable mass spectra. Peak 27 was identified as isorhamnetin (3′-O-methylquercetin) on the basis of its coelution with an authentic standard and matching UV diode array spectra.
Classes of Quercetin Metabolites Detected.Ingested [2-14C]-Quercetin-4′-glucoside. [2-14C]Quercetin-4′-glucoside (peak 18) was present in the GI tract at 0.5 and 1 h after the meal (Table 3). No intact [2-14C]quercetin-4′-glucoside was detected in the kidneys and plasma; however, the liver contained a very small amount at 0.5 h.
Quercetin. Thirty minutes after ingestion of [2-14C]quercetin-4′-glucoside, the aglycone quercetin (peak 24) represented 25% of the radioactivity in the GI tract. At later time points, the levels of free quercetin in the GI tract declined to <8%. No free quercetin was detected in plasma, but small amounts were present in liver and kidneys at some time points (Table 3).
Monoglucuronides. Seven different radiolabeled compounds (peak 12, 15, 19, 20, 21, 22, and 23) were identified as glucuronide conjugates of quercetin and methyl-quercetin. Their different retention times indicated different positioning of the methyl and/or glucuronide groups on the quercetin molecule, changing the polarity and the elution behavior of that molecule. On the basis of cochromatography with a reference compound and matching mass spectra, peak 12 was identified as quercetin-3-glucuronide, which has been previously identified in rat plasma 1 h after ingestion of [2-14C]quercetin-4′-glucoside (Mullen et al., 2002). Monoglucuronides were present in all organs, but only at 0.5, 1, and 2 h after the ingestion of [2-14C]quercetin-4′-glucoside. After 5 h, monoglucuronides were found only in the liver in relatively small amounts (Table 3).
Diglucuronides. Nine of the radiolabeled metabolites (peak 3, 5, 6, 7, 8, 9, 10, 11, and 14) were identified as diglucuronide conjugates of quercetin and methyl-quercetin. Due to the fragmentation pattern of these compounds (Table 2) it was assumed that the two glucuronyl moieties were attached at different positions on the flavonol ring (Mullen et al., 2002). Diglucuronides formed the major group of metabolites in liver and kidneys at all time points, and in the GI tract at 5 h. In plasma, the quantity of diglucuronides was highest at 0.5 h and decreased thereafter (Table 3).
Sulfates. Five different radiolabeled compounds contained a sulfate moiety. Peaks 13, 16, and 17 were identified as glucuronated and sulfated quercetin or methyl-quercetin. In plasma, the quantity of glucuronated sulfates increased over time, and at 5 h, 76% of all metabolites present in plasma were glucuronated sulfates. Peak 26 was a quercetin sulfate and peak 25 was a methyl-quercetin sulfate. Both of these sulfated metabolites were found only in the intestine at 0.5, 1, and 2 h (Table 3).
Unidentified Metabolites. Metabolite peaks 1, 2, and 4 were not identified, as they did not yield recognizable mass spectra. All three metabolites had comparatively short HPLC retention times, indicating that they are relatively polar molecules. Metabolite peak 4 occurs in small quantities in intestine, liver, and kidney. Metabolites 1 and 2 occur only in the intestine 5 h after the meal.
Metabolites in Tissues and Plasma.GI Tract. Thirty minutes after feeding, more than one-half of the ingested radioactivity had undergone metabolic modification. At this early stage, monoglucuronides, diglucuronides, and sulfated quercetin and/or methyl-quercetin conjugates were detected in the GI tract together with free quercetin and its methylated derivative isorhamnetin. One and 2 h after the meal, monoglucuronides formed a major portion of the metabolites. Five hours after the meal the monoglucuronides had disappeared and diglucuronides, in particular peak 10, formed the major class of metabolites (Table 3).
Plasma. Neither quercetin nor the ingested quercetin-4′-glucoside was detected in plasma at any time point (Table 3). In plasma, the amount of diglucuronides decreased over the time course of the experiment and monoglucuronides were present only at 0.5 and 1 h after the meal. However, the amount of glucuronidated sulfates increased steadily and 5 h after the meal glucuronated sulfates form the major group of metabolites in plasma in the form of peak 13, a sulfated quercetin glucuronide, and peak 16 a methylated, sulfated quercetin glucuronide. This latter compound was the predominant metabolite in plasma at 5 h, representing 62% of the radioactivity in plasma at that time point.
Liver and Kidneys. In both liver and kidney, the most abundant group of metabolites were diglucuronides followed by monoglucuronides and sulfated glucuronides. The amount of diglucuronides increased steadily over the time of the experiment. Small amounts of free quercetin were detected in the liver and in kidney. In the liver, a small amount of quercetin-4′-glucoside was present at 0.5 h (Table 3).
Discussion
The biological fate of dietary quercetin was the focus of the present study. To obtain unambiguous data on the dynamics of quercetin absorption, metabolite formation, and distribution throughout body tissues and plasma, radiolabeled quercetin-4′-glucoside, a major quercetin conjugate in onion, was synthesized and fed to rats. This article contains new data on body distribution and biotransformation of [2-14C]quercetin-4′-glucoside in rats at 0.5, 2, and 5 h postadministration, and includes previously reported data on body distribution and metabolism at 1 h postadministration (Mullen et al., 2002).
Absorption. The administered dose of 3 mg of [2-14C]quercetin-4′-glucoside to rats corresponds with the quercetin-4′-glucoside content in 250 g of lightly fried onions consumed by a 70-kg human subject (Tsushida and Suzuki, 1995; Crozier et al., 2000). Around 6% of the administered radioactivity was absorbed into internal organs and circulating blood, with a peak plasma concentration of 14 μM (2.9% of the administered dose) at 0.5 h. This is in keeping with recent findings in human trials where peak plasma concentrations ranged from 0.74 to 7.6 μM after oral ingestion of a dietary relevant dose of onion or quercetin-4′-glucoside (Hollman et al., 1997; Olthof et al., 2000; Graefe et al., 2001). In these studies, between 1.4 and 6.4% of the ingested quercetin dose was found in urine. These results suggest that the relative bioavailability of quercetin from a single dietary relevant dose (i.e., 1–10 mg quercetin/kg body weight) of quercetin-4′-glucoside supplement or from onion ranges from 1 to 7%.
In contrast, two studies with ileostomy patients (Hollman et al., 1995; Walle et al., 2000) found that more than 50% of the ingested quercetin (in fried onions) disappeared between oral ingestion and ileostomy effluent collection. Because only 5% of this was due to degradation in either the stomach or ileal fluid, it was concluded that the “missing quercetin” must have been absorbed (Hollman et al., 1995). Our data show that up to 44% of the quercetin metabolites in the GI tract of the rats were methylated. The methylation of quercetin, presumably to isorhamnetin, or tamarixetin, and its further conjugation with glucuronic acid/sulfate groups has been described in a number of human and animal studies (Hollman et al., 1997; Manach et al., 1997, 1999; Moon et al., 2000; Graefe et al., 2001). However, in the ileostomy studies (Hollman et al., 1995; Walle et al., 2000), ileal fluid was analyzed for the presence of only quercetin, quercetin glucosides, and quercetin conjugates. Metabolic transformation of quercetin to methyl-quercetin and its conjugates might therefore provide an explanation for the “high disappearance” of quercetin in the ileostomy model. Thus, the disappearance of quercetin from the ileostomy fluid may overestimate its absorption into the body and quercetin bioavailability may be substantially (∼44%) lower than suggested by the studies of Hollman et al. (1995) and Walle et al. (2000).
High absorption of quercetin has also been reported by Walle et al. (2001) on the basis of a study where human volunteers ingested [4-14C]quercetin aglycone. The absorption of radiolabeled quercetin was estimated to be 36 to 53% of the ingested dose (100 mg) by calculating the “area under the curve” of the total radioactivity present in plasma over a period of 72 h. During this period, 4.2% of the orally ingested radioactivity was excreted in urine, whereas 52% was exhaled as 14CO2. It was assumed that the exhaled 14CO2 originated from the portion of [14C]quercetin that was not absorbed from the small intestine but underwent bacterial degradation in the colon. The excretion of radioactivity in urine returned to baseline within 24 h, but the pharmacokinetic profile of total 14C in plasma showed that sizable amounts of 14C were still present 24 to 72 h after ingestion of [14C]quercetin. Paradoxically, most supplementation studies with quercetin-rich foods have shown that plasma quercetin levels return to baseline values within 24 h (Aziz et al., 1998; Crozier et al., 2000; Erlund et al., 2000; Graefe et al., 2001). Therefore, radioactivity detected in plasma after 24 h by Walle et al. (2001) is likely to include 14CO2, en route to the lungs for exhalation, together with other radiolabeled products such as hydroxyphenylacetic acids (Aura et al., 2002) and related colonic breakdown products of [14C]quercetin. Consequently, we believe that the area under the curve calculations, based on plasma radioactivity levels, used by Walle et al. (2001) to estimate quercetin absorption from the small intestine at 36 to 53%, is probably a substantial over-estimate. Urinary excretion of 4.2% of the ingested radioactivity may better reflect the true absorption of the quercetin supplement.
Site of Metabolite Formation. The catalytic activity of drug-metabolizing enzymes in the small intestine is generally lower than the corresponding values in the liver (Lin et al., 1999). Therefore, it is usually assumed that the liver is the major site for metabolism of xenobiotics such as quercetin (Griffiths, 1982; Hackett, 1986). However, from our results we conclude that most of the metabolites must have been formed in the GI tract. This conclusion is based on following observations: 1) all ingested radiolabeled quercetin-4′-glucoside had undergone metabolic changes within 2 h; and 2) at all time points, most of the radiolabeled compounds were present in the GI tract (86–93% of administered dose). If the liver were the main site of quercetin metabolism, and the metabolites were excreted into the intestine via the bile, unless there was an extremely rapid rate of turnover, the liver should contain much higher levels of radioactivity, at least at one time point. In practice, the liver contained no more than 2% of the ingested radioactivity at any time point. The hypothesis, that quercetin conjugation takes place predominantly in the GI tract, is supported by Crespy et al. (2003) who perfused rat intestine in situ with quercetin and collected intestinal eluent and bile separately. Bile contained only 10% of the quercetin conjugates; intestinal eluent contained 90% of the quercetin conjugates, indicating that the majority of the metabolism had occurred in the gut not in the liver. Preliminary data by Cermak et al. (2003) supports our conclusion, because it was reported that portal blood of pigs contained exclusively quercetin metabolites and that quercetin was metabolically transformed before reaching the liver.
A similar mechanism of presystemic elimination of xenobiotics through phase II metabolism and immediate excretion of the metabolites into the lumen of the gut has been described for a range of other drugs (Suzuki and Sugiyama, 2000), and Walgren et al. (2000) demonstrated the efflux of quercetin-4′-glucoside across human intestinal Caco-2 cell monolayers by the apical multidrug resistance-associated protein-2.
In the current study, 0.5 h after [2-14C]quercetin-4′-glucoside intake, a greater number of metabolites were present in the liver, kidneys, and in plasma in comparison with the GI tract (Table 3). This suggests that after metabolism and absorption from the GI tract, further methylation and glucuronidation occur in the liver and kidneys, in agreement with previous reports (Moon et al., 2000; Day et al., 2000; O'Leary et al., 2003).
Groups of Metabolites. In most previous studies the presence of quercetin metabolites in human and animal plasma was shown indirectly, by releasing free quercetin and/or methyl-quercetin from glucuronyl and sulfate moieties via enzymatic or acid hydrolysis before HPLC analysis (Manach et al., 1999, 1998; Crespy et al., 1999). In the present study using a radiolabeled parent compound and LC-MS-MS analysis with on line radioactivity detection, seven quercetin monoglucuronides and nine quercetin diglucuronides were detected in tissues and plasma together with three glucuronated sulfates and two quercetin sulfates. Glucuronidation of quercetin may occur at different and multiple hydroxyl groups on the quercetin molecule (Day et al., 2000). Isomers of quercetin glucuronide have identical molecular mass, and fragmentation patterns; however, they can be separated by HPLC. Assuming that all five hydroxyl groups on the quercetin molecule are possible binding sites for glucuronyl units and that hydroxyl groups 3′ and 4′ are potential loci for methylation, eight isoforms of methylated quercetin monoglucuronides and five isoforms of quercetin monoglucuronides could theoretically occur. In agreement with our data, Day et al. (2000) and O'Leary et al. (2003) identified seven quercetin monoglucuronides (three methylated and four unmethylated isoforms) produced by human liver cell free extracts and human HepG2 cells. Oliveira et al. (2002) and Oliveira and Watson (2000) reported the formation of four different isomers of quercetin monoglucuronide by rat hepatocytes in suspension and human glucuronyl transferase (UGT) microsomes. Thus, four different laboratories report the same number of isoforms for quercetin mono-glucuronides despite the use of different extraction techniques, chromatographic conditions, and equipment, supporting the hypothesis that UGTs selectively conjugate specific hydroxyl groups, most likely at positions 3, 7, 3′, and 4′ on the quercetin molecule (Day et al., 2000; Boersma et al., 2002). Boersma et al. (2002), however, also report significant differences of the conjugation rate and conjugation position of UDP glucuronosyltransferases derived from different organs (liver, intestine) and species (human, rat). Individual metabolite isomers may have different biological activities (Yamamoto et al., 1999; Day et al., 2000); therefore, species-dependent variation in metabolite formation may result in different biological effects.
In the present study, the main metabolic end products, 5 h after quercetin-4′-glucoside ingestion, were diglucuronides in liver, kidneys, and the GI tract. In plasma, the main group of metabolites was glucuronated sulfates. Manach et al. (1997) also reported that the main circulating forms of quercetin in rat plasma were sulfated and methylated quercetin glucuronides. Our results clearly show that the metabolite profile in plasma does not reflect the metabolite profile in liver, kidneys, and the GI tract.
Unidentified Metabolites. It has been reported that the microflora in the cecum or colon can form oxidative breakdown products of quercetin (Aura et al., 2002). Five hours after the meal, 21% of the recovered radioactivity from the GI tract was present in form of metabolite peaks 1 and 2. The identity of these two peaks could not be determined; however, it is possible that these two compounds were oxidative breakdown products of quercetin formed by colonic microflora. The present study measured the quercetin absorption and metabolism in the small intestine. Oxidative degradation of quercetin into hydroxyphenylacetic acids and other products would be expected to occur in the cecum and colon of the rat at time points around 5 h after the meal (Brown et al., 1988).
In conclusion, this study supports the view that ∼6% of dietary quercetin is absorbed from the GI tract, at least in rats. Absorbed quercetin occurs in the body almost exclusively metabolized with one or more methyl, sulfate or glucuronide groups attached. From our results, we hypothesize that most of the ingested dietary quercetin is metabolized directly in the GI tract. The retention of this flavonoid and its metabolites in the GI tract suggests that the gut epithelium may be one of the main end users of their antioxidant and anticancer properties. Animal studies have reported an inverse relationship between quercetin intake and the development of colon cancer (Deschner et al., 1991; Yang et al., 2000). The potential bioactivity of these metabolites needs to be determined. In addition, as this study was stopped at 5 h, more research is needed to determine whether these are the final metabolic products or whether subsequent exposure to the colonic microflora results in further metabolism.
Acknowledgments
We thank Alison Sutcliffe, University of Glasgow, for assistance with the development of the extraction and purification procedures and Dr Gary Williamson, Food Research Institute, Norwich, for generously supplying a sample of quercetin-3-glucuronide.
Footnotes
-
This study was supported by a University of Glasgow Fleck Postgraduate Fellowship (to S.T.C.) and the New Zealand Crop Research Institute, Christchurch. G.G.D. received financial support from the Scottish Executive Environmental and Rural Affairs Department. The HPLC-MS-MS used in the study was purchased with a Biotechnology and Biological Sciences Research Council grant (to A.C. and J. R. Coggins).
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.104.002691.
-
ABBREVIATIONS: GI, gastrointestinal tract; HPLC, high-performance liquid chromatography; RC, radiocounting; MS-MS, tandem mass spectrometry.
- Received November 2, 2004.
- Accepted April 12, 2005.
- The American Society for Pharmacology and Experimental Therapeutics