JFS H: Health, Nutrition, and Food. H: Health, Nutrition, & Food. Introduction Anthocyanins (ACN), derivatives of the 2-phenylbenzopyrylium

Transcription

1 JFS H: Health, Nutrition, and Viscous Matrix Influences Absorption and Excretion but Not Metabolism of Blackcurrant Anthocyanins in Rats MICHAELA C. WALTON,WOUTER H. HENDRIKS,ANNE M. BROOMFIELD, AND TONY K. MCGHIE ABSTRACT: The aim of the present study was to investigate the effect of a simultaneous intake of food and anthocyanins (ACNs) on ACN absorption, metabolism, and excretion. Blackcurrant ACNs (BcACNs) were dissolved in water with or without the addition of oatmeal and orally administered to rats, providing approximately 25 mg total ACNs per kilogram BW. Blood, urine, digesta, and tissue samples of the stomach, jejunum, and colon were subsequently collected at.25,.5, 1, 2, 3, 7, and 24 h. Identification and quantification of ACNs were carried out by Reversed phase-high-performance liquid chromatography (RP- HPLC) and liquid chromatography-mass spectrometry (LC-MS). Four major ACNs were present in the blackcurrant extract: delphinidin 3-O-glucoside, delphinidin 3-O-rutinoside, cyanidin 3-O-glucoside, and cyanidin 3-O-rutinoside. In plasma, the 4 ACNs of blackcurrant were identified and quantified. The time to reach maximal total ACN plasma concentration (C max BcACN/water =.37 ±.7 μmol/l; C max BcACN/oatmeal =.2 ±.5 μmol/l) occurred faster after BcACN/water (t max =.25 h), than after BcACN/oatmeal administration (t max = 1. h). In digesta and tissue samples, the 4 original blackcurrant ACNs were detected. The relative concentration of rutinosides in the digesta increased during their passage through the gastrointestinal tract, while the glucosides decreased. Maximum ACN excretion in urine occurred later after BcACN/oatmeal than after BcACN/water administration (3 compared with 2 h). The 4 original ACNs of blackcurrant in their unchanged form, as well as several metabolites, were identified in the urine samples of both groups. The simultaneous intake of food affects ACN absorption and excretion in the urine, but not metabolism. Keywords: absorption, anthocyanins, blackcurrants, gastrointestinal tract, metabolism Introduction Anthocyanins (ACN), derivatives of the 2-phenylbenzopyrylium (flavylium) cation, are a group of flavonoid compounds that are mainly responsible for the red, blue, and purple pigments in many fruits and fruit juices (Mazza and Miniati 1993). Anthocyanins are bioactive compounds and are of substantial nutritional interest as daily intakes are 18 to 215 mg in the United States (Kuhnau 1976). This is much higher than the intake (23 mg/d) of most other flavonoids, such as quercetin, kaempferol, myricetin, apigenin, and luteolin (Hertog and others 1993). Berry fruits, in particular, are rich dietary sources of ACNs, and some can contribute 1s mg in a single serving (McGhie and others 23). For example, blackcurrants (Ribes nigrum L.) contain as much as 732 mg/1 g fresh fruit (Koeppen and Herrmann 1977). The berries of blackcurrant contain 4 major ACNs, delphinidin 3-O-glucoside (D3G), delphinidin 3-O-rutinoside (D3R), cyanidin 3-O-glucoside (C3G), and cyanidin 3-O-rutinoside (C3R) (Figure 1) (Slimestad and Solheim 22) and blackcurrant extracts have been widely used to investigate ACN absorption in humans and animals (Matsumoto and others 21; Netzel and others 21; Nielsen and others 23; Ichiyanagi and others 25b; Wu and others 25). MS Submitted 6/18/28, Accepted 9/21/28. Authors Walton and Broomfield are with Inst. of, Nutrition, and Human Health, Massey Univ., Private Bag , Palmerston North, New Zealand. Author Hendriks is with Animal Nutrition Group, Dept. of Animal Sciences, Wageningen Univ., Wageningen, The Netherlands. Author McGhie is with The Horticulture and Research Inst. of New Zealand, Private Bag 11 3, Palmerston North, New Zealand. Direct inquiries to author McGhie ( tmcghie@hortresearch.co.nz). There is great interest in exploring the health benefits of ACNs found in fruits and vegetables because of their proposed healthrelated effects, and the number of studies investigating ACN bioavailability has increased over the last few years. Anthocyanins act as antioxidants (Wang and others 1997) both in vitro (Tsuda and others 1994, 1996a,b; Frankel and others 1995; Rice-Evans and others 1995; Satue-Gracia and others 1997; Wang and others 1997; Youdim and others 2; Landrault and others 21; Proteggente and others 22) and in vivo (Tsuda and others 1998, 1999a, 22; Janssen and others 2; Ramirez-Tortosa and others 21; Mazza and others 22). However, the majority of studies focusing on ACN absorption have shown that the bioavailability of ACNs is quite low (Cao and Prior 1999; Bub and others 21; Netzel and others 21; Felgines and others 22, 23; Mazza and others 22; Wu and others 22). So far, most of these studies have only administered purified ACNs or berry fruit extracts dissolved in water to fasting subjects. A few studies administered the ACNs to humans as part of a meal consisting of sugar, bread, and butter (Felgines and others 25), or with a high fat meal (Mazza and others 22). However, those studies did not measure the effect of the food matrix on ACN absorption. Nielsen and others (23) investigated the effect of a blackcurrant juice compared with an aqueous citric acid mix on ACN absorption in rabbits, as well as the additional ingestion of a rice cake on ACN absorption in humans. ACNs were slightly better absorbed when administered as juice, but no influence of the intake of a rice cake was observed. One study has investigated the codigestion of an ACN extract with commonly combined food stuffs, such as bread, breakfast cereal, ice cream, and cooked minced beef using an in vitro digestion model (McDougall and others 25). Apparent H22 JOURNAL OF FOOD SCIENCE Vol. 74, Nr. 1, 29 C 28 Institute of Technologists R doi: /j x Further reproduction without permission is prohibited

2 matrix and anthocyanin absorption... ACN absorption was unaffected, or slightly increased (minced beef), by co-incubation with foodstuffs. Thus, there are insufficient data about the effects of the food matrix on ACN absorption and metabolism in vivo. The latter is important for maximizing bioabsorption and any potential health related effects. As ACN bioavailability has been shown to be extremely low, ways to increase their absorption could assist in enhancing possible health benefits. The aim of this study was to investigate the effect of the simultaneous ingestion of ACNs in a food matrix on absorption and metabolism of ACNs. Rats were orally administered either blackcurrant ACN extract in water or in water and oatmeal. Blood, urine, as well as digesta and tissue samples of the stomach, jejunum, and colon were collected after several time intervals, and analyzed for ACN content. Materials and Methods Anthocyanins A commercial blackcurrant powdered concentrate (Currantex 3 R, total ACN content 32.9%, w/w) was kindly provided by Just the Berries Ltd., Palmerston North, New Zealand. The ACN composition of the blackcurrant concentrate was 15.7% D3G, 37.7% D3R, 7.1% C3G, and 34.4% C3R. All other nutrients, reagents, and chemicals used were purchased from commercial sources. Animals and diets Male Sprague-Dawley rats (n = 7, approximately 3 g) were bred and raised at the Small Animal Production Unit, Massey Univ., Palmerston North, New Zealand. They were housed in groups of 4 in polycarbonate rodent cages and kept in a room with controlled temperature (21 ± 1 C), humidity (55% ± 5%), and lighting (12 h light and dark cycles with dawn and dusk transitional periods). The animals had free access to tap water and a standard rodent chow (Matuschek and others 26), which was prepared at the Processing Unit, Massey Univ. One week before the experiment, rats were individually housed for adaptation under the same conditions. The experimental protocol was approved and followed the procedures set out by the Massey Univ. Animal Ethics Committee of Massey Univ. (protocol 4/59), New Zealand (Anonymous 23). Study design After the adaptation period, food was withdrawn, but water was continued ad libitum overnight (approximately 15 h). The following morning, the rats were randomly assigned to 2 groups containing 7 time-based subgroups (.25,.5, 1, 2, 3, 7, 24 h), each consisting of 5 rats. One group received the blackcurrant powder dissolved in.1 M citric acid buffer (ph 3.6) (BcACN/water) while the other group received blackcurrant powder dissolved in.1 M citric acid buffer containing 6.7% (w/w) finely ground oatmeal (Uncle Toby s Milk Oaties, Goodman Fielder Ltd. Auckland, New Zealand) (BcACN/oatmeal). Both treatments were prepared to provide 25 mg total ACNs per kilogram BW, and administered to the rats by stomach intubation (2.15 ±.2 ml/rat), using an infant polyvinyl chloride feeding tube (W/X-ray line, Fg 8; Unomedical, Australia). After administration, rats were placed into individual metabolic cages (Tecniplast, Italy) with free access to tap water until sampling. Treatment and sampling were appropriately staggered to ensure the correct sampling time for each subgroup and sufficient time for the collection of plasma, tissue, and digesta. All samples were stored at 2 C until analyzed. Sampling At the respective time points, animals were anaesthetized with isofluorane, containing 3% oxygen, and blood samples obtained by cardiac puncture into 1 ml evacuated glass tubes (Vacutainer, Becton Dickinson, Franklin Lakes, N.J., U.S.A.). The blood samples were kept on ice until they were centrifuged at 4165 g for 2 min at 4 C for plasma collection. Aliquots of plasma (1 ml) were acidified with.2 ml of 5% trifluoroacetic acid and stored at 2 Cuntil analysis. Within 1 min of euthanasia after blood sampling, the abdomen of the animals was opened by a midline incision, and the stomach and intestinal tract dissected out. The digesta of the entire stomach, and a subsample of the jejunum and colon (approximately 4 cm each) were collected into glass tubes and stored at 2 Cuntil analysis. The emptied tissue samples were thoroughly washed with cold tap water to remove any ACNs that adhered to the surface of the mucosa, collected into microfuge tubes, and stored at 2 C until analysis. Figure Chemical structures of the 4 major ACNs in blackcurrants: delphinidin-3-glucoside (D3G), delphinidin-3-rutinoside (D3R), cyanidin-3-glucoside (C3G), and cyanidin-3-rutinoside (C3R). Vol. 74, Nr. 1, 29 JOURNAL OF FOOD SCIENCE H23

3 matrix and anthocyanin absorption... Urine samples were collected at 2, 3, 7, and 24 h using metabolic cages (Tecniplast, Italy), and the volumes recorded. Aliquots (1 ml) were acidified with.2 ml of 5% trifluoroacetic acid and stored at 2 C until analysis. Anthocyanin concentration of treatments Frozen samples of the treatments were thawed at RT, diluted with 5% formic acid in H 2 O, and a 2 μl portion analyzed together with the plasma and urine samples by high-performance liquid chromatography (HPLC). Anthocyanin extraction from plasma Anthocyanins were extracted from acidified plasma as previously described (Walton and others 26). After evaporation to dryness, the sample residues were dissolved in 3 μl 5% formic acid in H 2 O, centrifuged for 1 min at 3 g at RT and 2 μl of the supernatant used for HPLC analysis. The recoveries for ACNs were found to be 94.9% ± 2.8% using authentic C3G (Extrasynthese, Genay, France). Anthocyanin extraction from urine Frozen, acidified urine samples were thawed quickly in a water bath at 37 C and centrifuged for 5 min at 1 g at RT. The supernatants were transferred into HPLC vials for analysis ( urine, 2 μl). Anthocyanins bound to proteins were extracted by washing the remaining pellet 5 times with 5 μl 1% TCA. A portion (1 ml) of the combined wash-supernatant (2.5 ml) was transferred into HPLC vials for analysis ( wash-urine, 2 μl). The ACN concentration was calculated as a sum of the urine and the wash-urine. The recovery of ACNs with this method was tested with spiking experiments using C3G and was found to be 91.2% ± 3.%. Urinary creatinine excretion was measured using a commercial diagnostic kit (Roche Diagnostic, New Zealand Ltd.; kit-no ). The samples were analyzed on a Flexor E clinical chemistry analyzer (Vital Scientific, Dieren, The Netherlands). RP HPLC with photodiode array detection as previously described (Matuschek and others 26). The sample injection volume was 2 μl for all samples. D3G-, D3R-, C3G-, and C3R-concentrations were calculated as C3G-equivalents, using an authentic standard of C3G (Extrasynthese, Genay, France) with known concentration. Liquid chromatography-mass spectrometry (LC-MS) analysis LC-MS analysis was performed as previously described by Cooney and others (24), using a Surveyor TM HPLC and PDA detector coupled to a LCQ Deco ion trap mass spectrometer fitted with an ESI interface (ThermoQuest, Finnigan, San Jose, Calif., U.S.A.). Solvents were (A) 5:3:92 (v/v/v) acetonitrile:formic acid: H 2 O and (B) acetonitrile +.1% formic acid and the flow rate was 2 μl/min. The initial mobile phase, 1% A, was ramped linearly to 83% A at 17 min, 8% A at 2 min, 7% A at 26 min, 5% A at 28.5 min, 5% A at 32 min and held for 3 min before resetting to the original conditions. Sample injection volume was 2 μl. Detection was by absorbance at 53 nm. MS data were acquired in the positive mode using a datadependent LC-MSn method with dynamic exclusion enabled and a repeat count of two. The electroface interspray voltage, capillary temperature, sheath gas pressure, and auxiliary gas were set at 27 V, 3 C, 45 psi, and 1 psi, respectively. Statistical analysis Data are presented as means ± SEM. Statistical analysis was carried out using SAS system for Windows, version 8 (SAS Inst. Inc., Cary, N.C., U.S.A.). The residuals of each analysis were tested for normality. Where not normally distributed, data were corrected by natural log transformation. Area under the plasma compared with time curve (AUC 7h ) was calculated using the linear trapezoidal rule. The significance of differences was assessed by one-way ANOVA for comparison of individual means. Differences with P.5 were considered significant. Anthocyanin extraction from digesta Frozen digesta samples were blended twice with 6 volumes of a solution of 5/5/1 methanol/h 2 O/formic acid using an Ultra- Turrax basic homogenizer. Samples were centrifuged at 57 g for 1 min at RT, the supernatants combined and a 1 ml portion filtered (MILLEX-GP Filter Unit; 22 μm pore size; Millipore, Billerica, Mass., U.S.A.). The prepared samples were diluted 1:1 with the methanol/h 2 O/formic acid (5/5/1), transferred into HPLCvials, and stored at 2 C until HPLC-analysis (2 μl). Recovery of C3G-spiked digesta samples was 97.3% ± 2.3%. Anthocyanin extraction from tissues Frozen tissue samples were manually cut into small pieces and blended twice with 6.6 (stomach), 15 (jejunum), and 8.6 (colon) (w/v) of 5% formic acid/h 2 O using an Ultra-Turrax basic homogenizer. Samples were then centrifuged at 57 g for 8 min at 4 C, the resulting supernatants combined and centrifuged again at 5 g for 1 min at RT. A 2-mL portion of the SN was evaporated under N 2 (< 35 C), redissolved with 1 ml 5% formic acid/h 2 O, and sonicated for 3 s. Finally, the samples were passed through a membrane filter (MILLEX-GP Filter Unit; 22 μm pore size; Millipore) into HPLC-vials, and stored at 2 C until HPLC-analysis (2 μl). Recovery of C3G-spiked tissue samples was 97.9% ± 3.4%. HPLC analysis Anthocyanin concentrations of the treatments, as well as in plasma, urine, digesta, and tissue samples were determined by Results ACN concentrations in plasma The 2 different treatments showed a different pattern of ACNs in plasma over time (Figure 2). The maximum total ACN concentration (Figure 2C) was reached sooner after BcACN/water administration (t max =.25 h, t 1/2 abs <.8 h; P <.5) than after BcACN/oatmeal administration (t max = 1. h, t 1/2 abs =.43 ±.21 h) (Table 1). Furthermore, the maximum total ACN concentration in plasma was significantly higher after BcACN/water administration (.37 ±.7 μmol/l; P <.5) than after BcACN/oatmeal administration (.2 ±.5 μmol/l) (Table 1). The total amount of total ACNs absorbed between and 7 h(auc 7h ) was significantly higher after BcACN/water administration (AUC 7h : 1. ±.11 μmol h/l, P <.5), than after BcACN/oatmeal administration (AUC 7h :.63 ±.9 μmol h/l, Table 1). The individual ACN concentrations in rat plasma for 7 h after administration of the BcACN/water and BcACN/oatmeal treatments are shown in Figure 2. After administration of BcACN/water, all 4 individual ACNs reached maximum plasma concentration after.25 h (Figure 2A, Table 1), whereas after BcACN/oatmeal administration, the maximum concentration was reached between.5 and 1 h (Figure 2B, Table 1). Furthermore, the absorption half-live (t 1/2abs ) for both rutinosides was longer after BcACN/oatmeal administration (D3R:.36 ±.26 h, NS; C3R:.5 ±.24 h, P <.5) than after BcACN/water administration (D3R:.6 ±.8 h; C3R: H24 JOURNAL OF FOOD SCIENCE Vol. 74, Nr. 1, 29

4 matrix and anthocyanin absorption... <.8 h). The total amount of individual ACNs absorbed between administration and 7 h (AUC-7h) was smaller after administrating BC/oatmeal, than BcACN/water (Table 1). However, the maximum concentrations of the individual ACNs were not significantly different between the 2 treatments (Table 1). After BcACN/water administration, there was another increase in the plasma concentration after 1 h for the rutinosides D3R and C3R (Figure 2A), which was not observed after BcACN/oatmeal administration. The concentration of ACNs in the plasma had returned to baseline after 7 h. ACN concentrations in digesta and gut tissue The HPLC chromatograms of digesta and tissue samples of the stomach, the jejunum, and the colon showed only the 4 original Plasma concentration [μmol/l] A B Dp-glu Cy-glu Dp-rut Cy-rut BC/water BC/oatmeal ACNs of BC; no additional peaks that would indicate the presence of ACN metabolites were observed (data not shown). The ratios of total ACN concentration found in the tissue samples (microgram per gram wet tissue) to total ACN concentration present in the corresponding digesta samples microgram digesta as-is) over the entire experimental period ( to 24 h) are shown in Figure 3. For both treatments, the jejunum showed a higher ratio (P <.1) than the stomach and the colon. Figure 4 presents the total ACN concentration in tissue samples of the stomach, the jejunum and the colon after BcACN/water or BcACN/oatmeal administration over time. After BcACN/water administration, a high mean concentration of ACNs was found in the stomach tissue after.25 h, which gradually decreased over time (Figure 4A). After BcACN/oatmeal administration however, there was an initial increase, with the highest mean concentration recorded at.5 h, before the ACN concentration also gradually decreased (Figure 4B). The jejunal tissue samples contained the highest mean ACN concentration at 1 h for both the BcACN/water (Figure 4A) and BcACN/oatmeal treatment (Figure 4B). Anthocyanins were also detected in colon tissues of the rats in both treatments. The maximum mean concentration was reached at 2 h for the BcACN/water, and at 3 h for the BcACN/oatmeal treatment (Figure 4A, 4B). The ratio of individual blackcurrant ACNs expressed as a percentage of the total ACN content in the treatments, digesta and tissue samples are shown in Figure 5A and 5B, respectively. The percentage of both glucosides (D3G, C3G) in the digesta decreased numerically (NS) during their passage through the gastrointestinal tract (GIT). The percentage of the rutinoside C3R in the digesta remained similar throughout the GIT, while the percentage of the rutinoside D3R was increased in the jejunum (P <.5) and colon (P <.1) digesta (Figure 5A). Similar to the digesta samples, the percentage of both glucosides decreased (D3G: P <.1, C3G: NS) in the tissue samples during the passage of digesta through the GIT (Figure 5B). However, the percentage of the rutinoside D3R in the various tissue samples of the GIT was not different (NS), while the relative amount of C3R increased in the stomach, jejunum (P <.5), and colon (P <.1) tissues compared with amount in the treatment C BC/water BC/oatmeal total ACNs Time [h] Figure ACN concentrations in rat plasma. (A) profile of individual ACNs in rat plasma after BcACN/water administration; (B) profile of individual ACNs in rat plasma after BcACN/oatmeal administration; and (C) total ACNs in rat plasma. Mean ± SEM, n = 5. Significant difference between the treatments at.25 h after administration, P <.1. Data were only analysed between and 3 h, as the concentrations reached baseline values from 7 h onwards. ACN metabolism The HPLC chromatogram (Figure 6) of a urine sample taken 2 h after administration of BcACN/water to a rat, showed the 4 original ACNs of blackcurrant (peak 1 to 4), and 4 additional metabolites (peak 5 to 8). HPLC chromatograms of urine samples taken at other time points as well as after administration of BcACN/oatmeal showed similar profiles (data not shown). Additional peak identification was performed by LC-MS analysis, and the RP-HPLC and LC-MS results are shown in Table 2. The identities of the 4 original ACNs of blackcurrant were confirmed by LC-MS analysis. Peaks 5 to 8 were identified as the methylated forms of the 4 original ACNs, methylated D3G, methylated D3R, peonidin-3-glucoside, and peonidin-3-rutinoside. In addition, traces of metabolites tentatively identified as diglucuronidated forms of delphinidin, cyanidin, and petunidin by their molecular and aglycon ion combinations were detected. The concentration of ACNs excreted per unit of urinary creatinine between 2 and 24 h is shown in Figure 7. Sufficient urine samples were only collectable from animals sacrificed between 2 and 24 h after administration of the treatments. The maximum urinary ACN excretion per unit creatinine after BcACN/oatmeal administration occurred later than the maximum after the BcACN/water administration (t max = 3 compared with 2 h). The mean concentration of ACNs excreted at 2 h was higher Vol. 74, Nr. 1, 29 JOURNAL OF FOOD SCIENCE H25

5 matrix and anthocyanin absorption... (P <.1) after administration of BcACN/water. In addition, the maximum urinary concentration of ACNs excreted was lower after administration of BcACN/oatmeal than after BcACN/water (28.8 ± 17.6 compared with 44.3 ± 2.8 ug total ACNs per milligram creatinine; P =.6). The 24 h ACN recovery (percent of the initial dose) of the total ACNs in the urine of the rats receiving the BcACN/water and BcACN/oatmeal treatments was.3% ±.2% and.2% ±.1%, respectively. Discussion Anumber of studies have been conducted to investigate ACN absorption in rats, pigs, and humans. However, most of these studies administered purified ACNs or berry fruit extracts dissolved in water (Matsumoto and others 21; Felgines and others 23; McGhie and others 23; Cooney and others 24; Ichiyanagi and others 24, 25b; Wu and others 24; Kay and others 25; Passamonti and others 25), thereby not taking into account possible interactions of ACNs with other dietary components. With the exception of 1 study, none of the previous studies investigating ACN bioavailability has compared the effect of different coadministered foods on ACN bioavailability in vivo. In this study, rats were given either a blackcurrant concentrate dissolved in acidified water (BcACN/water) or the same mixture containing additional oatmeal as a food source (BcACN/oatmeal). The latter was chosen to represent a typical breakfast type meal that could be administered to rats via stomach intubation. ACN absorption and excretion In the plasma samples of the rats in both treatment groups, we detected the 4 ACNs of blackcurrant that were present in the original concentrate. The main difference observed between the BcACN/water and BcACN/oatmeal treatments was the longer time to reach maximum concentration (t max ) for plasma ACNs with the BcACN/oatmeal compared with the BcACN/water treatment (Table 1, Figure 2A, 2B). The reason for the delayed absorption of ACNs from the BcACN/oatmeal treatment might be the higher Table Pharmacokinetic parameters of anthocyanins in rat plasma after a single oral dose of approximately 25 mg total anthocyanin per kilogram BW. A C max (μmol/l) A t max (h) B AUC 7h (μmol h/l) C t 1/2abs (h) D BC/water BC/oatmeal BC/water BC/oatmeal BC/water BC/oatmeal BC/water BC/oatmeal D3G.2 ±.1.1 ± ±.2.1 ±. NA NA D3R.9 ±.3.5 ± ±.4.19 ±.4 a.6 ±.8.36 ±.26 C3G.2 ±.1.2 ± ±..2 ±.1 NA NA C3R.23 ±.4.15 ± ±.8.42 ±.6 <.8.5 ±.24 a Total.37 ±.7.2 ±.5 a ± ±.9 a <.8.43 ±.21 a (dose corrected) b (.37) (.27) (1.) (.85) A Values are means ± SEM, n = 5. B Maximal plasma concentration. C Time to reach C max. D Area under the plasma compared with time curve. E Absorption half-live. a Significant difference between treatments, P <.5. b C max and AUC values divided by normalized dose. NA = not applicable A stomach jejunum colon 12 8 Ratio of total ACNs in tissue/digesta s j c s j c BC/water BC/oatmeal Figure Ratio of total ACN concentration in tissue samples (μg/g wet tissue) to the total ACN concentration in digesta samples (μg/g digesta as-is) of the stomach (s), the jejunum (j), and the colon (c) after administration of BcACN/water or BcACN/oatmeal over 24 h. Mean ± SEM, n = 26 to 29 (stomach), n = 13 to 15 (jejunum), n = 13 (colon). Ratios could only be calculated when ACN was detected by HPLC in the tissue. Significant difference within treatments between the intestinal regions, P <.1. Total ACNs [μg/g tissue] B Time [h] Figure Total ACN concentration in tissue samples of the stomach, the jejunum, and the colon after administration of BcACN/water (A), or BcACN/oatmeal (B). Mean ± SEM, n = 5. H26 JOURNAL OF FOOD SCIENCE Vol. 74, Nr. 1, 29

6 matrix and anthocyanin absorption... viscosity of the BcACN/oatmeal and the longer transit time of the chyme through the GIT, or the potential binding of ACNs to other food components. In a previous study, we showed that ACNs are maximally absorbed from the jejunum in mice (Matuschek and others 26), which is consistent with the view that the more viscous BcACN/oatmeal treatment would have taken longer to reach, and transit, the jejunum than the BcACN/water treatment. Interestingly, for the BcACN/water treatment we observed a second small but not significant increase in plasma concentrations for the rutinosides (D3R and C3R) 1 h after administration (Figure 2A). An earlier study also identified 2 absorption maxima, at.25 Percentage of individual ACNs [% of total] A treatment stomach jejunum colon B D3G D3R C3G C3R treatment stomach jejunum colon * * * Figure Percentage of individual BcACNs in digesta (A) and tissue (B) samples over 24 h. Mean ± SEM, Values considered only when ACN was detected by HPLC in digesta or tissue. Sample numbers, for digesta (A) n = 3 (treatment), n = 55 to 56 (stomach), n = 3 to 31 (jejunum), n = 28 to 29 (colon); for tissues (B) n = 3 (treatment), n = 36 to 34 (stomach), n = 38 to 44 (jejunum), n = 25 to 32 (colon). Significant difference of the individual ACNs in the different intestinal regions (within the respective sample), P <.5; P <.1; P <.1. Absorbance [AU] minutes Figure HPLC chromatogram of a rat urine sample, taken 2 h after administration of BC/water, and showing the original ACNs of BC (peak 1 to 4) and 4 additional metabolites (peak 5 to 8). Detection is at 52 nm. 4 6 ** 7 8 and 1 h in rat plasma after oral administration of 1 mg D3G per kilogram BW (Ichiyanagi and others 24). Ichiyanagi and others (25b) suggested that the 2nd increase in plasma concentrations of these compounds could be the result of the enterohepatic cycle, where absorbed ACN are transported to the hepatocytes, and subsequently excreted with the bile into the duodenum, where a re-absorption of the ACNs in the jejunum could take place. Interestingly, this 2nd maxima of plasma ACNs was not observed in the rats administered the BcACN/oatmeal. It is possible that a slower transit and longer exposure time of ACNs through the GIT would increase absorption and possible health-related benefits. However, a lower mean total ACN concentration in the plasma (.23 ±.53 compared with.369 ±.72 μmol/l), and a lower total ACN urinary excretion between and 7 h (AUC to 7 h:.63 ±.9 compared with 1. ±.11 μmol h/l) after BcACN/oatmeal administration compared with BcACN/water administration was observed (Table 1) and would seem to contradict this hypothesis. Examination of the ACN concentrations in each of the treatments used in this study showed that the BcACN/oatmeal contained only 74% of the BcACN/water mixture even though the treatments were designed to contain equivalent amounts of BcACN. Thus, the animals in the BcACN/oatmeal treatment received 24 ± 4.7 mg total ACNs per kilogram BW compared with 276 ± 5.2 mg total ACNs per kilogram BW for BcACN/water treatment. When the C max and AUC ( 7h) values for total anthocyanin were corrected for the actual differences in dose amounts, both C max and AUC ( 7h) were lower for the BcACN/oatmeal treatment compared with the BcACN/water treatment. The results of this study show that oatmeal slowed ACN absorption, and probably reduced the total absorption of anthocyanins from the gut compared with the consumption of BcACN in water. ACN in digesta and gut tissue To gain a better understanding of the fate of individual anthocyanins as they transit the gut, we measured ACN concentration in digesta and the corresponding gut tissue. Consistent with previous studies, we found that the relative concentrations of the ACN components changed during transit of the GIT. We observed a decrease in the relative concentration of glucosides, whereas the relative concentrations of rutinosides increased during their transit through the GIT. The variation of the different ACNs in different segments of the GIT from pigs was recently Table Peak identification of urine samples by RP-HPLC and LC-MS. RP-HPLC a LC-MS b Compound peak m/z Original ACNs D3G 1 465/33 D3R 2 611/33 C3G 3 449/287 C3R 4 595/287 Methylated forms methylated D3G 5 479/317 methylated D3R 6 625/317 peonidin-3-glucoside 7 463/31 peonidin-3-rutinoside 8 69/31 Diglucuronidated forms delphinidin diglucuronide ND 655/33 cyanidin diglucuronide ND 639/287 petunidin diglucuronide ND 669/317 peonidin diglucuronide ND ND a Reversed phase-hplc. b Liquid chromatography-mass spectrometry. ND = not detected. Vol. 74, Nr. 1, 29 JOURNAL OF FOOD SCIENCE H27

7 matrix and anthocyanin absorption... reported by Wu and others (26), who found that except for C3G, the profile of 4 other major ACNs in black raspberry was very similar to that in the berry. The instability of glucosides in the gut content has previously been shown by He and others (25), and has been suggested to be the result of lactase phloridzin hydrolase activity, a mammalianβ-glucosidase present in the brush border of the small intestine, which is also responsible for flavonoid deglycosylation in sheep and humans (Day and others 1998, 2). Evidence suggests that the in vivo stability of rutinoside ACNs is greater than that of glucoside ACNs, as the rutinoside moiety appears to reduce the metabolism of ACNs (Wu and others 25). The rapid appearance of ACNs in plasma following consumption can also be explained by potential absorption of ACNs in the stomach (Talavera and others 23). In the present study, we found substantial concentrations of ACNs in the stomach tissue, which would support this view. However, the ratio of the ACN concentration ( to 24 h) of the tissue to the digesta samples of the stomach and the jejunum indicates that the absorption by the jejunum was significantly more efficient (Figure 3). However, it should be noted that determination of the absorption site by calculating this ratio does not take into account the movement of digesta in the GIT, and the method needs to be further validated in additional in vitro systems. Furthermore, the maximum concentration of ACNs in jejunal tissue coincided with the maximum total ACN plasma concentration, supporting the view that the jejunum plays an important role in ACN absorption. Anthocyanins, which are not absorbed in the small intestine, appear further down the GIT. The colon digesta samples collected between 2 and 7 h in the present study showed a dark purple color and contained ACNs (BcACN/water: 1.17 ±.23 mg/g digesta; BC/oatmeal:.7 ±.2 mg/g digesta). This confirms the observation made by He and others (25), who investigated ACNs in intestinal contents after feeding rats a berry-enriched diet for 14 d, and reported an intense coloration of the fecal and caecal contents. Surprisingly, we also detected ACNs in the colon tissues, which would indicate that ACNs are absorbed by colonocites. This is in contrast to our previous in vitro study, where ACNs were found not to be absorbed by mouse colon tissue (Matuschek and others 26). The present study, however, used a relatively high dosage rate of ACNs (approximately 25 mg total ACNs per kilogram BW) in comparisonwiththeearlierin vitro study. Excretion [μg total ACNs/mg creatinine] BC/water Time [h] BC/oatmeal Figure ACN excretion in urine. Mean ± SEM, n = 1to5. Significant difference of urinary ACN excretion between the 2 treatments at 2 h, P <.1. ACN metabolism No difference was observed in the plasma profile of the 4 ACNs measured between the two treatments, indicating that the food matrix had no effect on the metabolism of ACNs. The present results show that ACNs are absorbed in their intact form as glycosides, as has been shown in previous rat and human studies (Miyazawa and others 1999; Tsuda and others 1999b; Bub and others 21; Cao and others 21; Matsumoto and others 21; Mazza and others 22). More recently, the metabolism of ACNs has been described to include methylation and glucuronidation, and the respective methylated and glucuronidated forms have been reported in the digestive area of rats, including the stomach, jejunum, liver, and kidney, as well as in plasma and urine of humans and rats (Wu and others 22; Felgines and others 23; Kay and others 24; Talavera and others 25). In the present study, we only detected metabolites of the blackcurrant ACNs in the urine samples, but not in plasma, digesta, or tissue samples. Besides the 4 major ACNs of blackcurrant, several additional peaks were detected in urine samples, after administration of either BcACN/water or BC/oatmeal. LC-MS analysis identified the additional peaks as the methylated and possibly diglucuronidated forms of the original four ACNs. A number of studies have reported the methylation and glucuronidation of ACNs in rats, pigs, and humans (Wu and others 24, 25; Felgines and others 25; Ichiyanagi and others 25a; Talavera and others 25). It has been shown that the methylation of D3G occurs at the 4 OH position in rats (Ichiyanagi and others 24), therefore the two methylated forms of D3G and D3R in the present study are probably the 4 -Omethyl-D3G and 4 -O-methyl-D3R metabolites. A previous study on blackcurrant ACN absorption and metabolism (Wu and others 25) presented a similar HPLC-chromatogram of pig urine after administration by gastric intubation. Compared with our study, however, the authors identified isopeonidin metabolites and reported the monoglucuronides of cyanidin, isopeonidin, and peonidin, whereas we detected traces of compounds that we have tentatively identified as diglucuronides. These differences could indicate a species specific metabolism of ACNs. However, monoglucuronides of C3G have also been reported in the urine of rats (Ichiyanagi and others 25a). The results of the present study indicate that the addition of a food matrix such as oatmeal does not affect metabolism of ACNs. The low ACN urinary excretion in the present study (.3% BcACN/water;.2% BcACN/oatmeal) confirm the previously reported low bioavailability of ACNs when assessed by urinary excretion. Most previous studies on ACN bioavailability in rats and humans reported ACN recoveries in urine between.4% and.11% (Bub and others 21; Cao and others 21; Matsumoto and others 21; Wu and others 22). Three more recent studies by Wu and others reported recoveries in pigs of.67%,.73%, and.88%, respectively (Wu and others 24, 25, 26). Conclusions This study demonstrates that a simultaneous intake of a food source affects ACN absorption and urinary excretion. ACNs were absorbed and excreted faster, when administered in an acidified water solution than in an acidified oatmeal solution. The administration of ACNs in acidified water resulted in a higher ACN concentration in blood plasma, compared with when oatmeal was simultaneously ingested. However, the differences in the amount of ACN consumed may account for this finding. The additional food matrix does not appear to have an effect on ACN metabolism. In addition, the results of the present study support our previous findings that ACNs are mainly absorbed from the jejunum. H28 JOURNAL OF FOOD SCIENCE Vol. 74, Nr. 1, 29

8 matrix and anthocyanin absorption... Acknowledgments This study was funded by the New Zealand Foundation for Research Science and Technology, grant number C6X7. The authors would like to thank Just the Berries Ltd. (Palmerston North, New Zealand) for providing the blackcurrant powdered concentrate. We also thank Dr. Janine Cooney (HortResearch, Hamilton, New Zealand) for the LC-MS analysis, and the late Dr. Phil Pearce (Massey Univ., Palmerston North, New Zealand) for the creatinine analysis of the rat urine samples. References Anonymous. 23. Code of ethical conduct for the use of live animals for teaching and research. Revised edition. Palmerston North, New Zealand: Massey Univ. Bub A, Watzl B, Heeb D, Rechkemmer G, Briviba K. 21. Malvidin-3-glucoside bioavailability in humans after ingestion of red wine, dealcoholized red wine and red grape juice. Eur J Nutr (3): Cao G, Prior RL Anthocyanins are detected in human plasma after oral administration of an elderberry extract. Clin Chem 45(4): Cao G, Muccitelli HU, Sanchez-Moreno C, Prior RL. 21. Anthocyanins are absorbed in glycated forms in elderly women: a pharmacokinetic study. Am J Clin Nutr 73(5):92 6. Cooney JM, Jensen DJ, McGhie TK. 24. Lc-ms identification of anthocyanins in boysenberry extract and anthocyanin metabolites in human urine following dosing. J Sci Agric 84(3): Day AJ, DuPont MS, Ridley S, Rhodes M, Rhodes MJ, Morgan MR, Williamson G Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and liver beta-glucosidase activity. FEBS Lett 436(1):71 5. DayAJ,CanadaFJ,DiazJC,KroonPA,MclauchlanR,FauldsCB,PlumbGW,Morgan MR, Williamson G. 2. Dietary flavonoid and isoflavone glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett 468(2 3): Felgines C, Texier O, Besson C, Fraisse D, Lamaison J-L, Remesy C. 22. Blackberry anthocyanins are slightly bioavailable in rats. J Nutr 132(6): Felgines C, Talavera S, Gonthier MP, Texier O, Scalbert A, Lamaison JL, Remesy C. 23. Strawberry anthocyanins are recovered in urine as glucuro- and sulfoconjugates in humans. J Nutr 133(5): Felgines C, Talavera S, Texier O, Gil-Izquierdo A, Lamaison JL, Remesy C. 25. Blackberry anthocyanins are mainly recovered from urine as methylated and glucuronidated conjugates in humans. J Agric Chem 53(2): Frankel EN, Waterhouse AL, Teissedre PL Principal phenolic phytochemicals in selected california wines and their antioxidant activity in inhibiting oxidation of human low-density lipoproteins. J Agric Chem 43(4):89 4. He J, Magnuson BA, Giusti MM. 25. Analysis of anthocyanins in rat intestinal contents impact of anthocyanin chemical structure on fecal excretion. J Agric Chem 53(8): Hertog MGL, Hollman PCH, Katan MB, Kromhout D Intake of potentially anticarcinogenic flavonoids and their determinants in adults in The Netherlands. Nutr Cancer 2(1):21 9. Ichiyanagi T, Rahman MM, Kashiwada Y, Ikeshiro Y, Shida Y, Hatano Y, Matsumoto H, Hirayama M, Tsuda T, Konishi T. 24. Absorption and metabolism of delphinidin 3-o-[beta]-glucopyranoside in rats. Free Radic Biol Med 36(7):93 7. Ichiyanagi T, Shida Y, Rahman MM, Hatano Y, Konishi T. 25a. Extended glucuronidation is another major path of cyanidin 3-o-beta-d-glucopyranoside metabolism in rats. J Agric Chem 53(18): Ichiyanagi T, Shida Y, Rahman MM, Hatano Y, Matsumoto H, Hirayama M, Konishi T. 25b. Metabolic pathway of cyanidin 3-o-beta-d-glucopyranoside in rats. J Agric Chem 53(1): Janssen M, Netzel M, Strasse G, Kesenheimer B, Herbst M, Carle E, Bitsch I, Boehm V, Bitsch R. 2. Increased plasmatic antioxidant capacity in humans after ingestion of blackcurrant and elderberry juice. Polyphenols commun Kay CD, Mazza G, Holub BJ, Wang J. 24. Anthocyanin metabolites in human urine and serum. Br J Nutr 91(6): Kay CD, Mazza G, Holub BJ. 25. Anthocyanins exist in the circulation primarily as metabolites in adult men. J Nutr 135(11): Koeppen BH, Herrmann K Flavonoid glycosides and hydroxycinnamic acid esters of blackcurrants (ribes nigrum). Z Lebensm Unters Forsch 164: Kuhnau J The flavonoids. A class of semi-essential food components: their role in human nutrition. World Rev Nutr Diet 24: Landrault N, Poucheret P, Ravel P, Gasc F, Cros G, Teissedre PL. 21. Antioxidant capacities and phenolics levels of french wines from different varieties and vintages. J Agric Chem 49(7): Matsumoto H, Inaba H, Kishi M, Tominaga S, Hirayama M, Tsuda T. 21. Orally administered delphinidin 3-rutinoside and cyanidin 3-rutinoside are directly absorbed in rats and humans and appear in the blood as the intact forms. J Agric Chem 49(3): Matuschek MC, Hendriks WH, McGhie TK, Reynolds GW. 26. The jejunum is the main site of absorption for anthocyanins in mice. J Nutr Biochem 17(1):31 6. Mazza G, Miniati E Anthocyanins in fruits, vegetables, and grains. Boca Raton, Fla.: CRC Press. Mazza G, Kay CD, Cottrell T, Holub BJ. 22. Absorption of anthocyanins from blueberries and serum antioxidant status in human subjects. J Agric Chem 5(26): McDougall GJ, Dobson P, Smith P, Blake A, Stewart D. 25. Assessing potential bioavailability of raspberry anthocyanins using an in vitro digestion system. J Agric Chem 53(15): McGhie TK, Ainge GD, Barnett LE, Cooney JM, Jensen DJ. 23. Anthocyanin glycosides from berry fruit are absorbed and excreted unmetabolized by both humans and rats. J Agric Chem 51(16): Miyazawa T, Nakagawa K, Kudo M, Muraishi K, Someya K Direct intestinal absorption of red fruit anthocyanins, cyanidin-3-glucoside and cyanidin-3,5- diglucoside, into rats and humans. J Agric Chem 47(3): Netzel M, Strass G, Janssen M, Bitsch I, Bitsch R. 21. Bioactive anthocyanins detected in human urine after ingestion of blackcurrant juice. J Environ Pathol Toxicol Oncol 2(2): Nielsen ILF, Dragsted LO, Ravn-Haren G, Freese R, Rasmussen SE. 23. Absorption and excretion of blackcurrant anthocyanins in humans and watanabe heritable hyperlipidemic rabbits. J Agric Chem 51(9): Passamonti S, Vrhovsek U, Vanzo A, Mattivi F. 25. Fast access of some grape pigments to the brain. J Agric Chem 53(18): Proteggente AR, Pannala AS, Paganga G, Van Buren L, Wagner E, Wiseman S, Van De Put F, Dacombe C, Rice-Evans CA. 22. The antioxidant activity of regularly consumed fruit and vegetables reflects their phenolic and vitamin c composition. Free Radic Res 36(2): Ramirez-Tortosa C, Andersen OM, Gardner PT, Morrice PC, Wood SG, Duthie SJ, Collins AR, Duthie GG. 21. Anthocyanin-rich extract decreases indices of lipid peroxidation and DNA damage in vitamin e-depleted rats. Free Radic Biol Med 31(9): Rice-Evans CA, Miller NJ, Bolwell PG, Bramley PM, Pridham JB The relative antioxidant activities of plant-derived polyphenolic flavonoids. Free Radic Res 22(4): Satue-Gracia MT, Heinonen M, Frankel EN Anthocyanins as antioxidants on human low-density lipoprotein and lecithin-liposome systems. J Agric Chem 45(9): Slimestad R, Solheim H. 22. Anthocyanins from black currants (ribes nigrum l.). J Agric Chem 5: Talavera S, Felgines C, Texier O, Besson C, Lamaison J-L, Rémésy C. 23. Anthocyanins are efficiently absorbed from the stomach in anesthetized rats. J Nutr 133(12): Talavera S, Felgines C, Texier O, Besson C, Gil-Izquierdo A, Lamaison JL, Remesy C. 25. Anthocyanin metabolism in rats and their distribution to digestive area, kidney, and brain. J Agric Chem 53(1): Tsuda T, Watanabe M, Ohshima K, Norinobu S, Choi SW, Kawakishi S, Osawa T Antioxidative activity of the anthocyanin pigments cyanidin 3-o-beta-d-glucoside and cyanidin. J Agric Chem 42(11):27 1. Tsuda T, Ohshima K, Kawakishi S, Osawa T. 1996a. Oxidation products of cyanidin 3-o-beta-d-glucoside with a free radical initiator. Lipids 31(12): Tsuda T, Shiga K, Ohshima K, Kawakishi S, Osawa T. 1996b. Inhibition of lipid peroxidation and the active oxygen radical scavenging effect of anthocyanin pigments isolated from phaseolus vulgaris l. Biochem Pharmacol 52(7): Tsuda T, Horio F, Osawa T Dietary cyanidin 3-o-beta-d-glucoside increases ex vivo oxidation resistance of serum in rats. Lipids 33(6): Tsuda T, Horio F, Kitoh J, Osawa T. 1999a. Protective effects of dietary cyanidin 3-obeta-d-glucoside on liver ischemia-reperfusion injury in rats. Arch Biochem Biophys 368(2): Tsuda T, Horio F, Osawa T. 1999b. Absorption and metabolism of cyanidin 3-o-betad-glucoside in rats. FEBS Lett 449(2 3): Tsuda T, Horio F, Kato Y, Osawa T. 22. Cyanidin 3-o-beta-d-glucoside attenuates the hepatic ischemia-reperfusion injury through a decrease in the neutrophil chemoattractant production in rats. J Nutr Sci Vitaminol 48(2): Walton MC, Lentle RG, Reynolds GW, Kruger MC, McGhie TK. 26. Anthocyanin absorption and antioxidant status in pigs. J Agric Chem 54(2):79 6. Wang H, Cao GH, Prior RL Oxygen radical absorbing capacity of anthocyanins. J Agric Chem 45(2):34 9. Wu X, Cao G, Prior RL. 22. Absorption and metabolism of anthocyanins in elderly women after consumption of elderberry or blueberry. J Nutr 132(7): Wu X, Pittman HE, 3rd, Prior RL. 24. Pelargonidin is absorbed and metabolized differently than cyanidin after marionberry consumption in pigs. J Nutr 134(1): Wu X, Pittman HE, 3rd, McKay S, Prior R. 25. Aglycones and sugar moieties alter anthocyanin absorption and metabolism after berry consumption in weanling pigs. J Nutr 135: Wu X, Pittman HE, 3rd, Prior RL. 26. Fate of anthocyanins and antioxidant capacity in contents of the gastrointestinal tract of weanling pigs following black raspberry consumption. J Agric Chem 54: Youdim KA, Martin A, Joseph JA. 2. Incorporation of the elderberry anthocyanins by endothelial cells increases protection against oxidative stress. Free Radic Biol Med 29(1):51 6. Vol. 74, Nr. 1, 29 JOURNAL OF FOOD SCIENCE H29