| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
[Departments of1 Nutrition and Food Science-CeRTA, and2 Internal Medicine, Hospital Clínic, Institut d’Investigació Biomèdica August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain;
aaddress correspondence to this author at: Department of Nutrition and Food Science, University of Barcelona, Av/Joan XXIII s/n. 08028, Barcelona, Spain; fax 34-93-403-59-31, e-mail lamuela{at}ub.edu]
Abstract
Background: Plant polyphenols have been studied largely because of the possibility that they might underlie the protective effects afforded by fruit and vegetable intake against cancer and other chronic diseases. Measurement of polyphenol content excreted in urine as an indicator of polyphenol consumption may offer a routine screening method that could be used for these pathologies.
Methods: Thirty-six healthy volunteers each received 2 interventions, one with a polyphenol-rich food (cocoa beverage) and one with a polyphenol-free food (milk) as a control, in a randomized cross-over design with 1-week intervals. The total polyphenol content excreted in urine during the 6 h after consumption of the test meals was measured by a modified Folin-Ciocalteu assay after sample cleanup by solid-phase extraction.
Results: The mean (SD) concentrations of polyphenols excreted in the urine 6 h after consumption of the test meals differed significantly: 140.95 (49.27) mg catechin/g of creatinine after the polyphenol-rich meal vs 90.43 (46.07) mg catechin/g of creatinine after the control meal (P <0.05).
Conclusions: This method allows analysis of a large number of samples per day, which is ideal for use in epidemiologic studies and may enable estimation of polyphenol consumption and determination of their possible role in preventing of certain pathologies, such as cancer, cardiovascular and degenerative diseases.
The idea that the health benefits associated with consumption of fruits, vegetables, tea, cocoa, and red wine are probably linked to polyphenol content has been supported by several recent studies (1)(2)(3)(4). The Folin–Ciocalteu (F-C) assay has for many years been used to measure of total phenolics in natural products (5)(6). Since introduction of the improvements recommended by Singleton and Rossi (7), reduction of phenols has also become more specific (8)(9). Nonetheless, there are limited reports describing use of this procedure for biological samples (10)(10)(11)(12)(13).
A wide range of water-soluble compounds are typically present in urine, although other substances, such as proteins, glucose, erythrocytes, and ketones bodies, can also be found when the body’s processes are not operating efficiently (14). The F-C assay is affected by several interfering substances, such as sugars, aromatic amines, sulfur dioxide, ascorbic acid, organic acids, and Fe(II), as well as nonphenolic organic substances that react with the F-C reagent (15)(16). We report here the use of a solid-phase extraction (SPE) procedure to remove such water-soluble compounds from urine samples. Combining this SPE with the Singleton and Rossi F-C assay (7), with certain modifications, provides an effective technique for quantifying the total polyphenols excreted in urine; these results can then be related to polyphenol intake. To study this we selected cocoa powder, a food rich in polyphenols. Cocoa and its derived products contain a diverse mixture of flavonoids, such as anthocyanins, flavonols (quercetin and its glycosides), and flavan-3-ols (epicatechin, catechin, and related procyanidin oligomers) (17).
Reagents were obtained from the following sources: methanol (HPLC grade) from Scharlau; F-C reagent from Panreac, formic acid, caffeic acid, (+)-catechin, gallic acid, and quercetin from Sigma; picric acid solution and creatinine from Fluka, Biochemika; and tyrosol and naringin from Extrasynthese. All chemicals used were of analytical or chromatographic grade. The water was purified in a MilliQ water purification system. The composition of the cocoa power was determined as described by Andrés-Lacueva et al. (18); the flavonoid concentrations for the cocoa powder portion used in this study were 56.4 mg of (–)-epicatechin, 51 mg of procyanidin B2, 16.8 mg of catechin, and 4 mg of flavonols, including isoquercitrin, quercetin, quercetin-glucoside, and quercetin-arabinoside. Synthetic urine was prepared as described by Miró-Casas et al. (19) to avoid any possible interference generated by the matrices with F-C reagent. A series of (+)-catechin calibrators with concentrations of 1, 2, 4, 6, 8, 10, and 12 mg/L was prepared in this synthetic urine. Calibrator preparation and sample processing were performed in a darkened room with a red safety light to avoid oxidation of analytes.
This randomized, crossover trial included 36 healthy adult volunteers (16 women and 20 men; age range, 18–49 years); all were nonsmokers with no history of heart disease or homeostatic disorders. The study was carried out in accordance with the Helsinki Declaration of 1975, as revised in 1996, and the protocol was approved by the Institutional Review Board of the Hospital Clinic, Barcelona. The test meal used was a polyphenol-rich food (PRF) in the form of a cocoa beverage containing 40 g of cocoa powder (Nutrexpa, Spain) dissolved in 250 mL of whole milk. As a control food (CF), 250 mL of whole milk was administered.
All volunteers consumed the 2 test meals 1 week apart in a crossover trial. Participants were instructed to abstain from vitamin supplements, medicines, alcoholic beverages, and any polyphenol-containing foods for at least 24 h before and during the test day and to fast at least 8 h before consuming the test food. Urine samples were collected in sterilized 1.5-L bottles. One sample was collected just before ingestion of the cocoa beverage; the second sample comprised all urine voided for 6 h after ingestion of either the PRF or CF. The urine samples were acidified with HCl to preserve the phenolic compounds before storage at –80 °C.
For SPE cleanup, 1 mL of the (+)-catechin calibrator or acidified patient urine was applied to an activated Waters Oasis HLB 3-mL (60 mg) cartridge. The cartridge was washed with 2 mL of 1.5 mol/L formic acid and 2 mL of water–methanol (95:5 by volume). The polyphenols were eluted with 1 mL of methanol containing 1 mL/L formic acid.
The F-C method described by Singleton and Rossi (7) was used with some modifications. Briefly, 200 µL of each methanolic fraction eluted after SPE was mixed in a 4.5-mL cell with 140 µL of F-C reagent diluted in 2.4 mL of water and 420 µL of sodium carbonate (200 g/L). The mixture was then incubated for 1 h at room temperature. After the reaction period, 910 µL of water was added to the cell and mixed, and the absorbance at 765 nm was measured in the same cell.
We used the Jaffe alkaline picrate method (20) to determine the urinary creatinine concentration. Briefly, 40 µL of urine was mixed in a 4.5-mL spectrophotometer cell with 800 µL of aqueous picric acid solution (10 mL/L) and 60 µL of sodium hydroxide (100 mg/L). After shaking, the mixture was left in the dark for 15 min at room temperature. After this reaction period, 3.1 mL of water was added to the cell, and the absorbance at 500 nm was measured in the same cell.
In the absence of disease, creatinine concentrations in serum and urine are usually very stable and can be used to estimate the urinary excretion of substances with only spot urine samples (21)(22)(23)(24). Therefore, in this study, we expressed total urinary polyphenol excretion per grams of creatinine excreted in the urine rather than per milliliters of urine.
To evaluate the linearity of the F-C assay, we prepared calibration curves with (+)-catechin added to blank synthetic urine. The F-C method was linear over the working range between 1 and 14 mg/L. Least-squares regression analysis gave the following results for the (+)-catechin calibration curve: mean (SD), slope, 0.094 (0.009); y-intercept, 0.064 (0.02); r2 = 0.998 (0.001). The limit of detection (0.9 mg/L) and the limit of quantification (2.1 mg/L) were calculated by repeated injections of diluted solutions of (+)-catechin in urine blanks, applying the Long–Wineford formula (25). Assay precision was determined by replicate analyses of samples containing known amounts of (+)-catechin prepared in blank urine by SPE (n = 5). The precision data obtained by analysis of calibration curves are shown in Table 1
. The criteria for acceptance (26) of precision were used at all concentrations. The measured concentrations (as a percentage of the expected value) of catechin and 5 common dietary polyphenols at concentrations ranging from 2 to 12 mg/L [mean (relative standard deviation); n = 3] were as follows: catechin, 93.7 (6.6)%; caffeic acid, 94.92 (7.6)%; tyrosol, 97.27 (3.27)%; quercetin, 78.46 (9.82)%; naringin, 95.65 (7.42)%; and gallic acid, 89.02 (0.23)%.
|
We tested for interference from sugar (glucose; 2 mg/L), iron [Fe(II); 1 mg/L], organic acids (oxalic, citric, and tartaric; 100 mg/L), amino acids (phenylalanine, tyrosine, glutamine, and arginine; 1 mg/L), vitamins (ascorbic acid and folic acid; 100 mg/L), and hippuric acid (10 mg/L). None of these substances reacted with the F-C reagent after the cleanup procedure.
We used liquid chromatography–tandem mass spectrometry (27) to confirm that the increment in polyphenols excreted in urine after ingestion of cocoa beverage, as measured by the F-C method, was attributable to cocoa polyphenol metabolites. Because (–)-epicatechin was the main polyphenol monomer in the cocoa powder ingested by the volunteers, we calculated the total concentration of quantified (–)-epicatechin metabolites excreted in urine 6 h after PRF intake. The mean (SD) concentration of (–)-epicatechin metabolites in urine from all volunteers was 279.39 (214.60) µg (–)-epicatechin/g of creatinine. Linear regression analyses showed that the liquid chromatography–tandem mass spectrometry values were significantly correlated with the urinary excretion of total polyphenols, as determined by the F-C assay (r = 0.813; P <0.001).
We found no significant differences in baseline urinary polyphenol concentrations between the CF and PRF groups (Fig. 1
). Statistical analysis (Kolmogorov–Levene and paired Student t-test) showed a significant increase (P <0.05) in polyphenol concentrations in the urine collected during the 6 h after ingestion of cocoa beverage [from 72.49 (17.6) to 140.95 (49.27) mg catechin/g of creatinine] in the PRF group, indicating that the total polyphenols measured in urine by the F-C assay can be related to polyphenol intake. We observed no significant changes in the CF group [increase from 79.27 (2.3) to 90.43 (46.07) mg catechin/g of creatinine]. Total urinary polyphenol concentrations increased up to 94% after ingestion of the PRF compared with an increase of only 14% after ingestion of the CF.
|
Duthie et al. (12) determined total polyphenols in urine 4 h after the ingestion of red wine, 12 year-old malt whisky (matured in an oak cask), or newly distilled whisky. They observed significantly more total phenols in the urine of volunteers who had consumed wine containing 125 mg of polyphenols [gallic equivalents (GAE)] as determined by F-C assay, 32 (3) GAE, in mg/L] compared with the aged whisky. However, even this lower urinary phenol concentration after the consumption of aged whisky containing 18 mg of polyphenols [22 (1) GAE, in mg/L] was significantly greater than that detected in volunteers who ingested the phenolic-free distilled whisky. In our study, we found values 4-fold higher than those obtained by Duthie et al. (12) in the urine of all volunteers 6 h after ingestion of a PRF containing 920 mg of polyphenols [(+)-catechin equivalents], as determined by F-C assay.
Krogholm et al. (28) also measured the total flavonoids excreted in urine by liquid chromatography–mass spectrometry to determine whether the flavonoid concentration in urine might reflect the intake of fruits and vegetables. They concluded that the total urinary excretion of flavonoids in 24 h may be used as a biomarker for fruit and vegetable intake. We have arrived at similar conclusions, finding that the total polyphenol concentrations in urine can be correlated to the polyphenol intake. Our method is less expensive and easier to perform than the method described by Krogholm et al. (28) and does not require liquid chromatography–mass spectrometry instrumentation. Further studies are required to determine whether this method is sensitive enough to detect the differences between a fruit- and vegetable-rich diet.
Acknowledgments
We are grateful for the financial support by the Spanish Government [Ministry of Education and Science (AGL 2005-0559/ALI and 2004-08378-C02-01/02) and Ministry of Health (CG03/140)]. We also are grateful to the Ramon y Cajal program from the Ministry of Education and Science and European Social Fund and to the University of Barcelona (Recerca i Docència grant). We thank all of the volunteers involved in this study as well as Nutrexpa S.A. for supplying the cocoa powder.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
