Bioavailability and pharmacokinetics of caffeoylquinic acids and flavonoids after oral administration of Artichoke leaf extracts in humans.
Extracts from artichoke leaves are traditionally used in the treatment of dyspeptic and hepatic disorders. Various potential pharmacodynamic effects have been observed in vitro for mono- and dicaffeoylquinic acids (e.g. chlorogenic acid, cynarin), caffeic acid and flavonoids (e.g. luteolin-7-O-glucoside) which are the main phenolic constituents of artichoke leaf extract (ALE). However, in vivo not only the genuine extract constituents but also their metabolites may contribute to efficacy. Therefore, the evaluation of systemic availability of potential bioactive plant constituents is a major prerequisite for the interpretation of in vitro pharmacological testing. In order to get more detailed information about absorption, metabolism and disposition of ALE, two different extracts were administered to 14 healthy volunteers in a crossover study. Each subject received doses of both extracts. Extract A administered dose: caffeoylquinic acids equivalent to 107.0 mg caffeic acid and luteolin glycosides equivalent to 14.4 mg luteolin. Extract B administered dose: caffeoylquinic acids equivalent to 153.8 mg caffeic acid and luteolin glycosides equivalent to 35.2 mg luteolin. Urine and plasma analysis were performed by a validated HPLC method using 12-channel coulometric array detection. In human plasma or urine none of the genuine target extract constituents could be detected. However, caffeic acid (CA), its methylated derivates ferulic acid (FA) and isoferulic acid (IFA) and the hydrogenation products dihydrocaffeic acid (DHCA) and dihydroferulic acid (DHFA) were identified as metabolites derived from caffeoylquinic acids. Except of DHFA all of these compounds were present as sulfates or glucuronides. Peak plasma concentrations of total CA, FA and IFA were reached within 1 h and declined over 24 h showing almost biphasic profiles. In contrast maximum concentrations for total DHCA and DHFA were observed only after 6-7 h, indicating two different metabolic pathways for caffeoylquinic acids. Luteolin administered as glucoside was recovered from plasma and urine only as sulfate or glucuronide but neither in form of genuine glucosides nor as free luteolin. Peak plasma concentrations were reached rapidly within 0.5 h. The elimination showed a biphasic profile.
[c] 2004 Elsevier GmbH. All rights reserved.
Keywords: Artichoke; Caffeoylquinic acids; Luteolin-7-O-glucoside; Bioavailability; Clinical trial
Extracts from artichoke leaves are traditionally used in the treatment of dyspeptic and hepatic disorders. Efficacy has been demonstrated in several clinical studies (Wegener and Fintelmann, 1999). High-dosed artichoke leaf extract (ALE) caused a considerable increase in choleresis, most likely due to an increased bile production into the duodenum of healthy volunteers (Kirchhoff et al., 1994). Thus, promotion of bile secretion might be responsible for the successful treatment of the dyspetic syndrome. In addition a recent study showed a reduction of the symptoms of the irritable bowel syndrome (IBS) after the treatment with ALE (Walker et al., 2001). Among the choleretic effect a positive influence on serum levels of cholesterol and LDL-cholesterol were reported in patients with hyperlipoproteinemia after intake of ALE (Fintelmann, 1996; Englisch et al., 2000). Thus, the administration of ALE may reduce the risk of coronary heart diseases and artheriosclerosis. In different in vitro assays artichoke leaf extract has shown an increase in bile production (Gebhardt, 2001), the inhibition of hepatocellular cholesterol biosynthesis (Gebhardt, 1995, 1998, 2002), the inhibition of LDL-oxidation (Brown and Rice-Evans, 1998) and antioxidant activities (Zapolska-Downar et al., 2002). Luteolin-7-O-glucoside as well as cynarin (1,5-dicaffeoylquinic acid) and several other polyphenolic extract constituents such as chlorogenic acid (5-caffeoylquinic acid) and caffeic acid (CA) (Fig. 1) have been reported to have antioxidant capacity (Gebhardt and Fausel, 1997; Brown and Rice-Evans, 1998). In contrast inhibition of cholesterol biosynthesis and biliary secretion were observed mainly by the aglycon luteolin and only in a lesser extent by luteolin-7-O-glucoside while cynarin and chlorogenic acid were almost ineffective (Gebhardt, 1998, 2001). However, in vivo not only the genuine extract constituents but also their metabolites may contribute to efficacy. Therefore, the evaluation of systemic availability of potential bioactive plant constituents is a major prerequisite for the interpretation of in vitro pharmacological testing results and in vivo effects.
Regarding caffeoylquinic acids and luteolin-7-O-glucoside as main constituents of ALE only few studies have been published concerning the absorption, metabolism and distribution of these compounds. Ferulic acid (FA), dihydroferulic acid (DHFA), isoferulic acid (IFA) and vanillic acid were reported as main urinary metabolites from caffeoylquinic acids in humans after oral administration of ALE and coffee, respectively. All of them, with the exception of DHFA, have been found not free but conjugated to sulfuric or glucuronic acid. No caffeoyl or feruloyl quinic acid derivatives as genuine components or their conjugates were detected (Rechner et al., 2001a). Nardini et al. and DuPont et al. determined high levels of conjugated CA in human plasma, but no chlorogenic acid, after coffee and apple cider consumption, respectively (DuPont et al., 2002; Nardini et al., 2002). These results were in good agreement with previous studies in rats, which demonstrated that after ingestion of chlorogenic acid, conjugated CA and FA, but not intact chlorogenic acid, were found in plasma and urine (Booth et al., 1957; Choudhury et al., 1999; Azuma et al., 2000). Regarding the flavonoids, bioavailability and metabolism of luteolin-7-O-glucoside has been mainly studied in animals. Deglucosylation to luteolin (LUT) and glucuronidation during the passing across the gut was observed (Shimoi et al., 1998).
In order to get more detailed information about absorption, metabolism and disposition of the constituents from ALE, two extracts with different patterns of phenolic compounds were administered to 14 healthy volunteers in a two-way crossover study. One of the extracts was a commercial available herbal remedy the other one derived from the first extract by extraction with aliphatic alcohols.
Subjects and methods
The study was carried out at the Institute of Clinical Pharmacology, University of Rostock, Germany. Fourteen healthy volunteers, seven men and seven women, were recruited for the study. Subjects were determined to be healthy on the basis of medical history, physical examination, electrocardiogram, routine urine-, clinical chemical- and hematological screening. All volunteers had no evidence of hepatitis B-, hepatitis C- or HIV-infection. None of the volunteers took drugs or medications, except hormonal contraceptives. Mean age was 25.1 [+ or -] 3.9 years (mean [+ or -] SD), and mean body mass index was 21.7 [+ or -] 2.3 kg [m.sup.-2] (mean [+ or -] SD). Participants were excluded if they had any significant medical history, a history of acute infection within the last 4 weeks before admission, use of an investigational drug within 2 month before participation, use of prescription or "over the counter" drugs (including herbal medicines) within 4 weeks before enrollment. All subjects gave written informed consent before being enrolled. The study protocol was approved by the ethic committee of the University of Rostock and was performed according to national law and ICH Good Clinical Practice.
Study design and supplements
Each subject received, according to a randomized two-way crossover design, the following treatments separated by a 10-days wash out interval: 2.4 g ALE A (Lichtwer Pharma AG, Berlin, Germany) containing 3.67% monocaffeoylquinic acid, 2.53% dicaffeolyquinic acid, 0.03% caffeic acid (providing a total amount of 107.0 mg caffeic acid equivalents) and 0.94% flavonoids (providing 14.4 mg luteolin equivalents); 0.625 g ALE B (Lichtwer Pharma AG, Berlin, Germany) containing 8.91% monocaffeoylquinic acid, 19.47% dicaffeolyquinic acid, 0.56% caffeic acid (providing a total amount of 153.8 mg caffeic acid equivalents) and 8.83% flavonoids (providing 35.2 mg luteolin equivalents). Each extract was suspended in a formulation of a 0.5% hydroxypropyl-methylcellulose gel (Methocel[R]K4M Premium EP, Dow Germany, Inc., Schwalbach, Germany) containing 20% ethanol (m/m). Subjects got a plant material free diet by avoiding all fruits and vegetables two days prior medication and during the 24 h sample collecting period. After the run-in phase of two days subjects received one of the two supplements on day 3 and 17 and were only allowed to drink water and to eat standardized meals according to the plant material free diet.
[FIGURE 1 OMITTED]
Collection of blood and urine samples
Venous blood samples (9 ml per blood sample) were collected into EDTA tubes (S-Monovetten[R]) once before subjects were administered the medication and 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24 h after administration. Blood was centrifuged for 10 min at 4700 X g at 4 [degrees]C. The supernatant was transferred into reaction cups in aliquots of 0.5 ml and 20 [micro]l acetic acid 0.58 M were added to each aliquot for stabilization. The samples were stored at -80 [degrees]C prior to analysis. Urine was collected in plastic bottles over 24 h after administration. An aliquot of 50 ml was mixed with 1 g ascorbic acid as antioxidant and stored at -80 [degrees]C prior to analysis.
Sample analysis was performed for the target compounds CA, dihydrocaffeic acid (DHCA), FA, DHFA, IFA and LUT. In order to assay the free metabolites beside their conjugates each plasma sample was prepared with and without enzymatic hydrolysis by sulfatase/glucuronidase of conjugated metabolites prior to HPLC CoulArray analysis (Wittemer and Veit, 2003). For urine analysis 0.1 M NaOH compared to 0.58 M acetic acid used in the plasma method was necessary to adjust to pH 5, because ascorbic acid was added before the samples were frozen. In addition, after evaporation the residual was redissolved in 250 [micro]l methanol/water 1:1 (v/v) instead of 150 [micro]l methanol/water 1:1 (v/v) for the plasma analysis, due to higher concentrations in urine samples.
Validation of the methods was performed according to the FDA Draft Guidance for Industry No. 2578 (Bioanalytical Method Validation for Human Studies). For both plasma and urine analytical method validation data were found to be within an acceptable range (Wittemer and Veit, 2003). The lower limit of quantification for plasma analysis was defined at 2.1 ng [ml.sup.-1] (CA), 2.0 ng [ml.sup.-1] (DHCA), 2.2 ng [ml.sup.-1] (FA), 2.1 ng [ml.sup.-1] (DHFA), 1.1 ng [ml.sup.-1] (IFA) and 2.1 ng [ml.sup.-1] (LUT) and for urine analysis at 28.8 ng [ml.sup.-1] (CA), 53.6 ng [ml.sup.-1] (DHCA), 110.9 ng [ml.sup.-1] (FA) and 52.2 ng [ml.sup.-1] (DHFA, IFA, LUT). At the limit of quantification (LOQ) within-day precision of the plasma and urine assay was for all target compounds <18.5% (coefficient of variation) and <11.5% at higher concentrations. Accuracy was <18.5% (relative error) at the LOQ and <9.5% at higher concentrations for plasma and urine. Calibration curves were obtained repeatedly during sample analysis by spiking blank plasma and urine samples with reference compounds at six different concentrations in duplicates, followed by extraction and HPLC analysis. The correlation coefficient for all calibration curves was >0.99.
In spite of a diet free of plant material, analysis of plasma samples showed measurable concentrations for CA, FA, DHFA and LUT before intake of the extracts. These baseline levels were observed again 24 h after administration and were subtracted from each time point prior to pharmacokinetic data analysis. Pharmacokinetic parameters were determined by noncompartmental analysis using Kinetica[R] 2000 vers. 4.01 (InnaPhase Corp., Philadelphia). The maximum observed plasma concentration [C.sub.max] and the time to reach [C.sub.max] ([t.sub.max]) were determined directly from the data. The AU[C.sub.0[right arrow]last] was calculated using the log-linear method, trapezoidal when [C.sub.n] > [C.sub.n-1]. AU[C.sub.tot] was computed with automatic estimation of [k.sub.el] by linear regression on the logarithmic transformation of the last data points of the curve. The slope of the curve was equal to [k.sub.el]/2.303. The apparent terminal half-life was calculated by the following equation: [t.sub.1/2] = [log.sub.(2)]/[k.sub.el].
Kinetica[R] 2000 vers. 4.01 (InnaPhase Corp., Philadelphia) was used for statistical analysis. Data are given as mean with corresponding standard deviation. To achieve a better approximation to a normal distribution data were transformed logarithmically. Differences among [C.sub.max], [t.sub.max], AU[C.sub.tot], [t.sub.1/2] and Ut after administration of the two treatments were tested for significance by a two-way analysis of variance (ANOVA) with a significance level of p < 0.05. Due to the different amounts of caffeoylquinic acids and flavonoids in the study medications statistical analysis of [C.sub.max], AU[C.sub.tot] and Ut were performed with dose-adjusted values. Calculations for CA, DHCA, FA, DHFA based on the total amount of caffeoylquinic acids expressed as caffeic acid equivalents (extract A 107.0 mg; extract B 153.8 mg). Calculation for LUT based on the total amounts of flavonoids expressed as luteolin equivalents (extract A 14.4 mg; extract B 35.2 mg).
After administration of extract A as well as extract B caffeoylquinic acids were not present in human plasma or urine. However, after [beta]-glucuronidase treatment CA, its O-methylated products FA and IFA and the hydrogenation products DHCA and DHFA could be identified. Prior to enzymatic hydrolysis only DHFA was detected in the plasma samples. Thus, phase-II-conjugates (sulfates or glucuronides) of CA, DHCA, FA, DHFA and IFA as well as nonconjugated DHFA could be found as metabolites of caffeoylquinic acids after ingestion of ALE. However, the presence of nonconjugated CA, DHCA and FA in plasma cannot be completely excluded since some CA, DHCA and FA was detected in urine in the nonconjugated form (Fig. 2). Most likely the concentration of these compounds in plasma was below the limit of detection.
Luteolin-7-O-glucoside was not present in human plasma or urine after administration of extracts A and B as well. Also, no free LUT could be detected in plasma or urine prior to enzymatic hydrolysis of conjugates. However, after [beta]-glucuronidase treatment LUT was identified. This indicates that after ingestion of ALE LUT was in plasma only available as phase-II-metabolites (sulfates or glucuronides).
The plasma concentration curves of nonconjugated DHFA and total DHFA after enzymatic hydrolysis of conjugates showed parallel profiles (Fig. 3). For the other compounds the nonconjugated form could not be detected. Hence, the quantification of these compounds was based on their total concentrations after hydrolysis of conjugates.
The time courses of total CA, DHCA, FA, DHFA and IFA concentrations found in human plasma after administration of extract A are shown in Fig. 4. Peak plasma concentrations of 6.51 [+ or -] 1.89, 8.89 [+ or -] 1.66 and 7.89 [+ or -] 2.17 ng [ml.sup.-1] for CA, FA and IFA, respectively, were reached within 1 h (Table 1). The concentration versus time profiles of all three compounds were biphasic, most prominent for FA, with a rapid distribution phase and a slower terminal elimination phase lasting up to 24 h after administration. DHCA and DHFA showed almost three-fold higher peak plasma concentrations of 21.30 [+ or -] 12.35 and 27.56 [+ or -] 13.79 ng [ml.sup.-1], respectively. Maximum concentrations were reached only within 6-7 h (Table 1).
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
The time courses of total CA, DHCA, FA, DHFA and IFA concentrations found in human plasma after administration of extract B are shown in Fig. 4. The plasma concentration versus time profiles of all compounds run almost parallel to that of extract A. Due to the 1.4-fold higher intake of caffeic acid equivalents the maximum peak concentrations for CA, FA, IFA, DHCA and DHFA were increased to 8.04 [+ or -] 2.64, 15.37 [+ or -] 3.85, 10.51 [+ or -] 2.43, 29.53 [+ or -] 12.40 and 39.72 [+ or -] 15.51 ng [ml.sup.-1], respectively (Table 1). Considering the different amounts of caffeic acid equivalents administered, the pharmacokinetic parameters did not differ significantly (p < 0.05) between the two treatments.
The elimination half-lives of CA, DHCA, DHFA and IFA were between 2 and 3 h after administration of both treatments. Due to the more prominent biphasic elimination profile of FA the elimination half-life was 5.5-6.5 h, i.e. more than two-fold higher.
[FIGURE 4 OMITTED]
The time courses of total LUT concentrations found in human plasma after administration of extract A and B are shown in Fig. 4. The plasma concentration versus time profiles of total LUT after intake of extracts A and B run parallel. Considering the 2.4-fold higher amount of luteolin equivalents in extract B compared to extract A the pharmacokinetic parameters as given in Table 1 did not differ significantly (p < 0.05). LUT was rapidly absorbed from extract A as well as from extract B. Peak plasma concentrations of 59.07 [+ or -] 32.82 and 156.5 [+ or -] 92.25 ng [ml.sup.-1], respectively, were reached within 30-40 min. The plasma concentration versus time profile was biphasic, subdivided into a distribution phase and a slow elimination phase. Elimination half-life was calculated to be 2-3 h for both treatments.
Cumulatively renally excreted amounts of free and conjugated CA, DHCA, FA, DHFA and IFA in 24-h after administration of extracts A and B were 4.74 [+ or -] 1.82% and 4.01 [+ or -] 1.45% of caffeoylquinic acids intake, respectively. For both treatments the amounts of caffeoylquinic acids recovered as total DHFA, DHCA and FA were much higher than the amounts recovered as total CA and IFA (Table 2).
After administration of extracts A and B 1.72 [+ or -] 1.01% and 1.99 [+ or -] 1.50% of luteolin-7-O-glucoside intake, respectively, were recovered as LUT conjugates in the 24-h urine (Table 2).
The data obtained for the renal clearance are presented in Table 2. They differed among the various compounds, however the difference between the two treatments was not significant (p < 0.05), which is in agreement with the assumption that identical metabolites are formed out off the two extracts.
Caffeoylquinic acids were only present in human plasma and urine in form of conjugated and nonconjugated hydroxycinnamates. However, the conjugates (sulfates or glucuronides) could not be further identified in this study. The analytical methods were developed primarily for the parent compounds, and the conjugates presumably coeluted with matrix compounds.
So far little has been known about the metabolites of caffeoylquinic acids in human plasma since in several previous studies metabolites of caffeoylquinic acids have only been determined in urine (Booth et al., 1957; Rechner et al., 2001a, b). The only two studies with data of metabolites from caffeoylquinic acids present in human plasma failed to detect other compounds than conjugated and nonconjugated CA (DuPont et al., 2002; Nardini et al., 2002). However, the results of this study demonstrate that after ingestion of ALE conjugates of DHCA, FA, DHFA and IFA and nonconjugated DHFA are formed as metabolites from caffeoylquinic acids. These metabolites were identical independently of the ALE administered. Thus, the different pattern of caffeoylquinic acids in the two extracts seems to have no influence on the formation of metabolites. This assumptions is also supported by the fact that the elimination half-life and the renal clearance for each of the metabolites did not differ significantly (p < 0.05) between extracts A and B.
After intake of both extracts considerable plasma concentrations of total CA, FA and IFA could already be detected after 15 min and maximum concentrations were reached within 1 h. This confirms previous observations made by DuPont et al. and Nardini et al. After oral application of caffeoylquinic acids in forms of apple cider and coffee maximum concentrations of total CA were reached within 1 h as well (DuPont et al., 2002; Nardini et al., 2002). In contrast, peak plasma concentrations of total DHCA and DHFA occurred after 6-7 h. Comparable data for DHCA and DHFA are lacking. However, the results correspond well with urinary data obtained by Rechner et al. In urine, total DHFA was detected 8-12 h after coffee consumption (Rechner et al., 2001b).
The disposition of the caffeoylquinic acids metabolites behaved in a dose-proportional manner. Differences in the pharmacokinetic parameters between the two extracts were due to the different amounts of total caffeic acid equivalents administered. Dose-adjusted bioavailability and pharmacokinetic parameters of total CA, DHCA, FA, DHFA and IFA did not differ significantly (p < 0.05) after administration of extracts A and B.
The mechanism and site of absorption of caffeoylquinic acids is still unclear. The rapid increase of plasma concentrations in the case of CA, FA and IFA conjugates suggests that some absorption takes place in the upper gut. To form these metabolites from caffeoylquinic acids the ester bound has to be cleaved. The absence of caffeoylquinic acids in human plasma indicates that the cleavage occurs somewhere before or at least immediately after entering the systemic circulation. However, there is no published evidence for enzymatic hydrolysis of chlorogenic acid by intestine, liver or plasma extracts (Plumb et al., 1999; Andreasen et al., 2001). Moreover, chlorogenic acid has been reported to be stable in digestive or intestinal juice (Takenaka et al., 2000; Olthof et al., 2001). Despite the missing evidence the [t.sub.max] values of 1 h for CA, FA and IFA conjugates and the absence of caffeoylquinic acids in human plasma suggest absorption and de-esterification of caffeoylquinic acids somewhere in the upper gut.
After the release of CA from caffeoylquinic acids conjugation with glucuronic acid presumably occurs in the enterocytes. In vitro experiments in rats showed that 63% of CA were present as conjugates in the serosal fluid after perfusion of the small intestine (Spencer et al., 1999). After entering the systemic circulation CA conjugates are most likely methylated during the first liver passage as maximum concentrations of the methylation products FA and IFA were obtained after 1 h. Methylation of CA in the liver is supported by in vitro experiments using catechol-O-methyltransferase isolated from rat hepatocytes (Scheline, 1991; Moridani et al., 2002). The higher amounts of total FA compared to total IFA in plasma and urine might be explained by the greater susceptibility of the 4-O-methylether formation towards subsequent O-demethylation (Nielsen et al., 1998).
The [t.sub.max] values of 6-7 h for total DHCA and DHFA suggest that beside the uptake in the upper gut, the colon is another a bsorption site of caffeoylquinic acids yielding different metabolites. This results are in agreement with in vitro experiments, showing esterase activities for bacteria of the colonic microflora for chlorogenic acid (Plumb et al., 1999; Andreasen et al., 2001).
The lack of an extended second maximum around 6 h in the concentration time profile of total CA as well as the presence of DHCA and DHFA in plasma indicate, that the released CA is metabolized by the colonic microflora prior to absorption. CA metabolism has been observed for different isolated bacteria from human feces. DHCA acid was formed as well as several other unspecific compounds (Peppercorn and Goldman, 1971). According to the formation of FA from CA by catechol-O-methyltransferases DHFA most likely derives from DHCA by methylation in the liver. This model is supported by in vitro studies published by Moridani et al. using rat hepatocytes. Moridani et al. also described the dehydrogenation of DHFA to form FA, which might be a possible explanation for the more prominent biphasic elimination profile of total FA concentration (Moridani et al., 2002). A model of the metabolic pathways of caffeoylquinic acids after intake of ALE, which is in agreement with our findings is presented in Fig. 5.
After application of both extracts LUT was present in human plasma only in conjugated form. This confirms previous studies by Shimoi et al. in rats. After oral administration of luteolin-7-O-glucoside to rats LUT was mainly present as monoglucuronide in plasma (Shimoi et al., 1998). The analytical procedure used in this study was not able to identify the conjugates. However, the presence of luteolin-7-O-glucuronide could be excluded by comparison of chromatograms from plasma samples and a reference compound.
[FIGURE 5 OMITTED]
After intake of both extracts considerable plasma concentrations of total LUT could already be detected after 15 min and peak concentrations were reached after 30-40 min. As this was the first study on pharmacokinetics of metabolites from luteolin-7-O-glucoside comparable data are lacking. However, the concentration time profile of total LUT is similar to those of other flavonoids after oral intake of their respective glucosides (Graefe et al., 2001).
Like the disposition of caffeoylquinic acid metabolites the disposition of LUT conjugates behaved in a dose proportional manner. Differences in the pharmacokinetic parameters between the two extracts were due to the different amounts of total luteolin equivalents administered. Dose adjusted the bioavailability and pharmacokinetic parameters of total LUT did not differ significantly (p < 0.05) after administration of extracts A and B.
The mechanism of absorption of luteolin-7-O-glucoside is still unclear. The fast increase in plasma concentration suggests the upper gut as site of absorption. The genuine glucoside is most likely hydrolyzed during absorption. This assumption is supported by in vitro experiments showing that the structurally related apigenin-7-O-glucoside is hydrolyzed by a cell free extract of human small intestine (Day et al., 1998). Recently, it has been reported, that enzymes that are able to hydrolyze flavonolglucosides are located in the cells (cytosolic beta-glucosidase, CBG) and on the apical membrane (lactase-phlorizin hydrolase, LPH) (Nemeth et al., 2003). Thus, flavonolglucosides may be cleaved by LPH, after which the aglycon may diffuse passively into the cell (Day et al., 2000). Alternatively, flavonols may enter the cell as an intact glucoside by sodium-dependent glucose transporter (SGTL1) (Arts et al., 2002) and then be cleaved inside the cell by CBG. Probably both mechanism are involved in the absorption of flavonolglucosides. However, it is questionable if this mechanism can also be assumed for luteolin-7-O-glucoside.
Due to the complete lack of free LUT in the plasma, LUT seems to be quickly metabolized after absorption. Thus conjugation has likely to occur primarily in the enterocytes. In vitro experiments demonstrated, that 74% of LUT were conjugated to glucuronic acid after incubation with microsomal samples from human intestine. Most common binding sites of the molecule were the hydroxyl groups in the 3'- and 4'-position (51%;44%) whereas the 7-O-glucuronide was only formed to 5% (Boersma et al., 2002). This might explain the lack of luteolin-7-O-glucuronide in plasma seen in this study.
As human glucuronidase present in various tissue is capable of hydrolyzing conjugates (Paigen, 1989; Sperker et al., 1997; O'Leary et al., 2003), the released aglycones together with detected metabolites are putatively biological active at the target sites. Considering the concentrations found in this study this should be taken in account for the assessment of in vitro data.
Table 1. Pharmacokinetic parameters of total CA, DHCA, FA, DHFA, IFA and LUT absorption and elimination in human plasma after single oral administration of artichoke leaf extracts A and B in a crossover study Parameter CA DHCA Extract A (107.0 mg CA; 14.4 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 6.51[+ or -]1.89 21.30[+ or -]12.35 [t.sub.max] (h) 0.83[+ or -]0.30 6.28[+ or -]1.24 AU[C.sub.0-last] (ng 19.60[+ or -]5.35 85.23[+ or -]58.15 [ml.sup.-1]*h) AU[C.sub.tot] (ng 23.02[+ or -]7.15 114.9[+ or -]79.88 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.28[+ or -]0.13 0.29[+ or -]0.16 [t.sub.1/2] (h) 3.08[+ or -]1.53 3.10[+ or -]1.73 Extract B (153.8 mg CA; 35.3 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 8.04[+ or -]2.64 29.53[+ or -]12.40 [t.sub.max] (h) 0.94[+ or -]0.47 5.77[+ or -]1.69 AU[C.sub.0-last] (ng 22.38[+ or -]9.54 124.8[+ or -]62.57 [ml.sup.-1]*h) AU[C.sub.tot] (ng 26.32[+ or -]10.69 146.7[+ or -]62.93 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.37[+ or -]0.20 0.38[+ or -]0.18 [t.sub.1/2] (h) 2.69[+ or -]1.80 2.52[+ or -]2.14 Parameter FA DHFA Extract A (107.0 mg CA; 14.4 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 8.89[+ or -]1.66 27.56[+ or -]13.79 [t.sub.max] (h) 0.77[+ or -]0.26 6.34[+ or -]1.05 AU[C.sub.0-last] (ng 47.34[+ or -]9.22 119.8[+ or -]62.91 [ml.sup.-1]*h) AU[C.sub.tot] (ng 68.86[+ or -]22.09 147.3[+ or -]95.33 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.14[+ or -]0.07 0.32[+ or -]0.25 [t.sub.1/2] (h) 6.35[+ or -]2.95 2.91[+ or -]1.35 Extract B (153.8 mg CA; 35.3 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 15.37[+ or -]3.85 39.72[+ or -]15.51 [t.sub.max] (h) 0.98[+ or -]0.35 6.21[+ or -]1.28 AU[C.sub.0-last] (ng 77.07[+ or -]26.54 180.0[+ or -]88.14 [ml.sup.-1]*h) AU[C.sub.tot] (ng 105.7[+ or -]54.87 205.6[+ or -]98.12 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.17[+ or -]0.08 0.36[+ or -]0.19 [t.sub.1/2] (h) 5.23[+ or -]2.73 2.48[+ or -]1.35 Parameter IFA LUT Extract A (107.0 mg CA; 14.4 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 7.89[+ or -]2.17 59.07[+ or -]32.82 [t.sub.max] (h) 0.83[+ or -]0.26 0.36[+ or -]0.18 AU[C.sub.0-last] (ng 34.94[+ or -]21.92 159.0[+ or -]73.87 [ml.sup.-1]*h) AU[C.sub.tot] (ng 41.19[+ or -]25.96 168.6[+ or -]76.85 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.28[+ or -]0.19 0.32[+ or -]0.13 [t.sub.1/2] (h) 3.46[+ or -]1.99 2.50[+ or -]0.85 Extract B (153.8 mg CA; 35.3 mg LUT) [C.sub.max] (ng [ml.sup.-1]) 10.51[+ or -]2.43 156.5[+ or -]92.29 [t.sub.max] (h) 1.10[+ or -]0.43 0.46[+ or -]0.18 AU[C.sub.0-last] (ng 41.58[+ or -]17.10 464.8[+ or -]183.1 [ml.sup.-1]*h) AU[C.sub.tot] (ng 47.37[+ or -]24.04 499.6[+ or -]194.0 [ml.sup.-1]*h) [k.sub.el] ([h.sup.-1]) 0.36[+ or -]0.18 0.37[+ or -]0.24 [t.sub.1/2] (h) 2.69[+ or -]2.07 2.45[+ or -]1.14 [C.sub.max], peak plasma concentration; [t.sub.max], time to reach [C.sub.max]; AU[C.sub.0-last], area under the curve from 0 to the last sampling time, AU[C.sub.tot], area under the curve from 0 to infinity; [k.sub.el], elimination constant; [t.sub.1/2], elimination half-life. Data are expressed as mean [+ or -] SD (n = 14). Table 2. Renale clearance of total CA, DHCA, FA, DHFA, IFA and LUT and elimination of total CA, DHCA, FA, DHFA, IFA and LUT expressed as percentage of CA and LUT intake, respectively, within 24 h after single oral administration of the two artichoke leaf extracts Parameter CA DHCA Extract A (107.0 mg CA; 14.4 mg LUT) Recovery (%) 0.21[+ or -]0.10 1.61[+ or -]1.11 Renal clearance (1 [h.sup.-1]) 10.73[+ or -]6.12 17.02[+ or -]7.58 Extract B (153.8 mg CA; 35.3 mg LUT) Recovery (%) 0.23[+ or -]0.10 1.29[+ or -]0.78 Renal clearance (1 [h.sup.-1]) 13.30[+ or -]3.96 16.76[+ or -]12.16 Parameter FA DHFA Extract A (107.0 mg CA; 14.4 mg LUT) Recovery (%) 1.14[+ or -]0.32 1.62[+ or -]0.76 Renal clearance (1 [h.sup.-1]) 21.08[+ or -]12.64 16.24[+ or -]12.08 Extract B (153.8 mg CA; 35.3 mg LUT) Recovery (%) 0.94[+ or -]0.36 1.38[+ or -]0.79 Renal clearance (1 [h.sup.-1]) 17.53[+ or -]11.44 10.70[+ or -]5.23 Parameter IFA Total Extract A (107.0 mg CA; 14.4 mg LUT) Recovery (%) 0.16[+ or -]0.09 4.74[+ or -]1.82 Renal clearance (1 [h.sup.-1]) 5.39[+ or -]4.76 -- Extract B (153.8 mg CA; 35.3 mg LUT) Recovery (%) 0.17[+ or -]0.07 4.01[+ or -]1.45 Renal clearance (1 [h.sup.-1]) 6.09[+ or -]3.63 -- Parameter LUT Extract A (107.0 mg CA; 14.4 mg LUT) Recovery (%) 1.72[+ or -]1.04 Renal clearance (1 [h.sup.-1]) 1.58[+ or -]0.92 Extract B (153.8 mg CA; 35.3 mg LUT) Recovery (%) 1.99[+ or -]1.50 Renal clearance (1 [h.sup.-1]) 1.39[+ or -]0.71 Data are expressed as mean [+ or -] SD (n = 14).
The authors thank the Zentralinstitut Arzneimittelforschung GmbH for technical and Lichtwer Pharma AG for financial support.
Received 13 June 2003; accepted 24 November 2003
Andreasen, M.F., Kroon, P.A., Williamson, G., Garcia-Conesa, M.T., 2001. Esterase activity able to hydrolyze dietary antioxidant hydroxycinnamates is distributed along the intestine of mammals. J. Agric. Food Chem. 49 (11), 5679-5684.
Arts, I.C., Sesink, A.L., Hollman, P.C., 2002. Quercetin-3-glucoside is transported by the glucose carrier SGLT1 across the brush border membrane of rat small intestine. J. Nutr. 132(9), 2823 (author reply 2824).
Azuma, K., Ippoushi, K., Nakayama, M., Ito, H., Higashio, H., Terao, J., 2000. Absorption of chlorogenic acid and caffeic acid in rats after oral administration. J. Agric. Food Chem. 48 (11), 5496-5500.
Boersma, G.M., van der Woude, H., Bogaards, J., Boern, S., Vervoort, J., Cnubben, N.H.P., van Iersel, M.L.P.S., van Bladeren, P.J., Rietjens, I.M.C.M., 2002. Regioselectivity of phase II metabolism of luteolin and quercetin by UDP-glucuronosyl tranferases. Chem. Res. Toxicol. 15, 662-670.
Booth, A.N., Emerson, O.H., Jones, F.T., DeEds, F., 1957. Urinary metabolites of caffeic and chlorogenic Acids. J. Biol. Chem. 51-59.
Brown, J.E., Rice-Evans, C.A., 1998. Luteolin-rich artichoke extract protects low density lipoprotein from oxidation in vitro. Free Radic. Res. 29 (3), 247-255.
Choudhury, R., Srai, S.K., Debnam, E., Rice-Evans, C.A., 1999. Urinary excretion of hydroxycinnamates and flavonoids after oral and intravenous administration. Free Radic. Biol. Med. 27 (3-4), 278-286.
Day, A.J., DuPont, M.S., Ridley, S., Rhodes, M., Rhodes, M.J., Morgan, M.R., Williamson, G., 1998. Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and liver [beta]-glucosidase activity. FEBS Lett. 436, 71-75.
Day, A.J., Canada, F.J., Diaz, J.C., Kroon, P.A., McLauchlan, W.R., Faulds, C.B., Plumb, G.W., Morgan, M.R., Williamson, G., 2000. Dietary flavonoid and isoflavone glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett. 468, 166-170.
DuPont, M.S., Bennett, R.N., Mellon, F.A., Williamson, G., 2002. Polyphenols from alcoholic apple cider are absorbed, metabolized and excreted by humans. J. Nutr. 132 (2), 172-175.
Englisch, W., Beckers, C., Unkauf, M., Ruepp, M., Zinserling, V., 2000. Efficacy of Artichoke dry extract in patients with hyperlipoproteinemia. Arzneimittelforschung 50 (3), 260-265.
Fintelmann, V., 1996. Antidyspeptische und lipidsenkende Wirkungen von Artischockenblatter Extrakt. Ergebnisse klinischer Untersuchungen zur Wirksamkeit und Vertraglichkeit von Hepar-SL forte an 553 Patienten. Z. Allg. Med. 72, 48-57.
Gebhardt, R., 1995. Artischockenextrakt: In-vitro-Nachweis einer Hemmwirkung auf die Cholesterinbiosynthese. Die Medizinische Welt 46 (7), 348-350.
Gebhardt, R., 1998. Inhibition of cholesterol biosynthesis in primary cultured rat hepatocytes by artichoke (Cynara scolymus L.) extracts. J. Pharmacol. Exp. Ther. 286 (3), 1122-1128.
Gebhardt, R., 2001. Anticholestatic activity of flavonoids from artichoke (Cynara scolymus L.) and of their metabolites. Med. Sci. Monit. 7 (Suppl. 1), 316-320.
Gebhardt, R., 2002. Inhibition of cholesterol biosynthesis in HepG2 cells by artichoke extracts is reinforced by glucosidase pretreatment. Phytother. Res. 16 (4), 368-372.
Gebhardt, R., Fausel, M., 1997. Antioxidant and hepatoprotective effects of artichoke extracts and constituents in cultured rat hepatocytes. Toxicology in vitro 11, 669-672.
Graefe, E.U., Wittig, J., Mueller, S., Riethling, A.K., Uehleke, B., Drewelow, B., Pforte, H., Jacobasch, G., Derendorf, H., Veit, M., 2001. Pharmacokinetics and bioavailability of quercetin glycosides in humans. J. Clin. Pharmacol. 41 (5), 492-499.
Kirchhoff, R., Beckers, C.H., Kirchhoff, G.M., Trinczek-Gartner, H., Petrowicz, O., Reimann, H.-J., 1994. Increase in choleresis by means of artichoke extract. Results of a randomised placebo-controlled double-blind study. Phytomedicine 1, 107-115.
Moridani, M.Y., Scobie, H., O'Brien, P.J., 2002. Metabolism of caffeic acid by isolated rat hepatocytes and subcellular fractions. Toxicol. Lett. 133 (2-3), 141-151.
Nardini, M., Cirillo, E., Natella, F., Scaccini, C., 2002. Absorption of phenolic acids in humans after coffee consumption. J. Agric. Food Chem. 50 (20), 5735-5741.
Nemeth, K., Plumb, G.W., Berrin, J.G., Juge, N., Jacob, R., Naim, H.Y., Williamson, G., Swallow, D.M., Kroon, P.A., 2003. Deglycosylation by small intestinal epithelial cell beta-glucosidases is a critical step in the absorption and metabolism of dietary flavonoid glycosides in humans. Eur. J. Nutr. 42 (1), 29-42.
Nielsen, S.E., Breinholt, V., Justesen, U., Cornett, C., Dragsted, L.O., 1998. In vitro biotransformation of flavonoids by rat liver microsomes. Xenobiotica 28 (4), 389-401.
O'Leary, K.A., Day, A.J., Needs, P.W., Mellon, F.A., O'Brien, N.M., Williamson, G., 2003. Metabolism of quercetin-7- and quercetin-3-glucuronides by an in vitro hepatic model: the role of human [beta]-glucuronidase, sulfotransferase, catechol-O-methyltransferase and multiresistant protein 2 (MRP2) in flavonoid metabolism. Biochem. Pharmacol. 65, 479-491.
Olthof, M.R., Hollman, P.C., Katan, M.B., 2001. Chlorogenic acid and caffeic acid are absorbed in humans. J. Nutr. 131 (1), 66-71.
Paigen, K., 1989. Mammalian beta-glucuronidase: genetics, molecular biology, and cell biology. Prog. Nucl. Acid Res. Mol. Biol. 37, 155-205.
Peppercorn, M.A., Goldman, P., 1971. Caffeic acid metabolism by bacteria of the human gastrointestinal tract. J. Bacteriol. 108 (3), 996-1000.
Plumb, G.W., Garcia-Conesa, M.T., Kroon, P.A., Rhodes, M., Ridley, S., Williamson, G., 1999. Metabolism of chlorogenic acid by human plasma, liver, intestine and gut microflora. J. Sci. Food Agric. 79, 390-392.
Rechner, A.R., Pannala, A.S., Rice-Evans, C.A., 2001a. Caffeic acid derivatives in artichoke extract are metabolised to phenolic acids in vivo. Free Radic. Res. 35 (2), 195-202.
Rechner, A.R., Spencer, J.P., Kuhnle, G., Hahn, U., Rice-Evans, C.A., 2001b. Novel biomarkers of the metabolism of caffeic acid derivatives in vivo. Free Radic. Biol. Med. 30 (11), 1213-1222.
Scheline, R.R., 1991. CRC Handbook of Mammalian Metabolism of Plant Compounds. CRC Press, Boca Raton.
Shimoi, K., Okada, H., Furugori, M., Goda, T., Takase, S., Suzuki, M., Hara, Y., Yamamoto, H., Kinae, N., 1998. Intestinal absorption of luteolin and luteolin 7-O-beta-glucoside in rats and humans. FEBS Lett. 438 (3), 220-224.
Spencer, J.P., Chowrimootoo, G., Choudhury, R., Debnam, E.S., Srai, S.K., Rice-Evans, C., 1999. The small intestine can both absorb and glucuronidate luminal flavonoids. FEBS Lett. 458 (2), 224-230.
Sperker, B., Murdter, T.E., Schick, M., Eckhardt, K., Bosslet, K., Kroemer, H.K., 1997. Inter-individual variability in expression and activity of human beta-glucuronidase in liver and kidney: consequences for drug metabolism. J. Pharmacol. Exp. Ther. 28, 914-920.
Takenaka, M., Nagata, T., Yoshida, M., 2000. Stability and bioavailability of antioxidants in garland (Chrysanthemum coronarium L.). Biosci. Biotechnol. Biochem. 64 (12), 2689-2691.
Walker, A.F., Middleton, R.W., Petrowicz, O., 2001. Artichoke leaf extract reduces symptoms of irritable bowel syndrome in a post-marketing surveillance study. Phytother. Res. 15 (1), 58-61.
Wegener, T., Fintelmann, V., 1999. Pharmacological properties and therapeutic profile of artichoke (Cynara scolymus L.). Wien Med. Wochenschr. 149 (8-10), 241-247.
Wittemer, S.M., Veit, M., 2003. Validated method for the determination of six metabolites derived from artichoke leaf extract in human plasma by high-performance-liquid chromatography-coulometric-array detection. J. Chromatogr. B Biomed. Sci. Appl. 793, 367-375.
Zapolska-Downar, D., Zapolski-Downar, A., Naruszewicz, M., Siennicka, A., Krasnodebska, B., Koldziej, B., 2002. Protective properties of artichoke (Cynara scolymus) against oxidative stress induced in cultured endothelial cells and monocytes. Life Sci. 71 (24), 2897-2908.
S.M. Wittemer (a), M. Ploch (a), T. Windeck (a), S.C. Muller (b), B. Drewelow (b), H. Derendorf (c), M. Veit (d,*)
(a) Lichtwer Pharma AG, Berlin, Germany
(b) Institute of Clinical Pharmacology, University of Rostock, Germany
(c) College of Pharmacy, University of Florida, Gainesville, USA
(d) LAT GmbH Dr. Tittel, Am Haag 4, Grafelfing 82166, Germany
*Corresponding author. Tel.: +49-89-858967120; fax: +49-89-858967111
E-mail address: email@example.com (M. Veit).