Peter C. H. Hollman

Intake of potentially anticarcinogenic flavonoids and their determinants in adults in the Netherlands

Flavonoids are strong antioxidants that occur naturally in foods and can inhibit carcinogenesis in rodents. Accurate data on population-wide intakes of flavonoids are not available. Here, using data of the Dutch National Food Consumption Survey 1987-1988, we report the intake of the potentially anticarcinogenic flavonoids quercetin, kaempferol, myricetin, apigenin, and luteolin among 4,112 adults. The flavonoid content of vegetables, fruits, and beverages was determined by high-performance liquid chromatography. In all subjects, average intake of all flavonoids combined was 23 mg/day. The most important flavonoid was the flavonol quercetin (mean intake 16 mg/day). The most important sources of flavonoids were tea (48% of total intake), onions (29%), and apples (7%). Flavonoid intake did not vary between seasons; it was not correlated with total energy intake (r = 0.001), and it was only weakly correlated with the intake of vitamin A (retinol equivalents, r = 0.14), dietary fiber (r = 0.21), and vitamin C (r = 0.26). Our use of new analytic technology suggests that in the past flavonoid intake has been overestimated fivefold. However, on a milligram-per-day basis, the intake of the antioxidant flavonoids still exceeded that of the antioxidants beta-carotene and vitamin E. Thus flavonoids represent an important source of antioxidants in the human diet.

Dietary flavonoids: intake, health effects and bioavailability.

Flavonoids are polyphenolic compounds that occur ubiquitously in foods of plant origin. Over 4000 different flavonoids have been described. They may have beneficial health effects because of their antioxidant properties and their inhibitory role in various stages of tumour development in animal studies. An estimation of the total flavonoid intake is difficult, because only limited data on food contents are available. It is estimated that humans ingest a few hundreds of milligram per day. The average intake of the subclasses of flavonols and flavones in The Netherlands was 23 mg/day. The intake of flavonols and flavones was inversely associated with subsequent coronary heart disease in most but not all prospective epidemiological studies. A protective effect of flavonols on cancer was found in only one prospective study. Flavonoids present in foods were considered non-absorbable because they are bound to sugars as beta-glycosides. However, we found that human absorption of the quercetin glycosides from onions (52%) is far better than that of the pure aglycone (24%). Flavonol glycosides might contribute to the antioxidant defences of blood. Dietary flavonols and flavones probably do not explain the cancer-protective effect of vegetables and fruits; a protective effect against cardiovascular disease is not conclusive.

Relative bioavailability of the antioxidant flavonoid quercetin from various foods in man

Quercetin is a strong antioxidant and a major dietary flavonoid. Epidemiological studies suggest that consumption of quercetin protects against cardiovascular disease, but its absorption in man is controversial. We fed nine subjects a single large dose of onions, which contain glucose conjugates of quercetin, apples, which contain both glucose and non‐glucose quercetin glycosides, or pure quercetin‐3‐rutinoside, the major quercetin glycoside in tea. Plasma levels were then measured over 36 h. Bioavailability of quercetin from apples and of pure quercetin rutinoside was both 30% relative to onions. Peak levels were achieved less than 0.7 h after ingestion of onions, 2.5 h after apples and 9 h after the rutinoside. Half‐lives of elimination were 28 h for onions and 23 h for apples. We conclude that conjugation with glucose enhances absorption from the small gut. Because of the long half‐lives of elimination, repeated consumption of quercetin‐containing foods will cause accumulation of quercetin in blood.

Absorption, metabolism and health effects of dietary flavonoids in man

Flavonoids are polyphenolic compounds that occur ubiquitously in foods of plant origin. Over 4,000 different flavonoids have been described, and they are categorized into flavonols, flavones, catechins, flavanones, anthocyanidins and isoflavonoids. Flavonoids have a variety of biological effects in numerous mammalian cell systems, in vitro as well in vivo. Recently, much attention has been paid to their antioxidant properties and to their inhibitory role in various stages of tumour development in animal studies. Quercetin, the major representative of the flavonol subclass, is a strong antioxidant, and prevents oxidation of low density lipoproteins in vitro. Oxidized low density lipoproteins are atherogenic, and are considered to be a crucial intermediate in the formation of atherosclerotic plaques. This agrees with observations in epidemiological studies that the intake of flavonols and flavones was inversely associated with subsequent coronary heart disease. However, no effects of flavonols on cancer were found in these studies. The extent of absorption of flavonoids is an important unsolved problem in judging their many alleged health effects. Flavonoids present in foods were considered non-absorbable because they are bound to sugars as beta-glycosides. Only free flavonoids without a sugar molecule, the so-called aglycones, were thought to be able to pass through the gut wall. Hydrolysis only occurs in the colon by microorganisms, which at the same time degrade flavonoids. We performed a study to quantify absorption of various dietary forms of quercetin. To our surprise, the quercetin glycosides from onions were absorbed far better than the pure aglycone. Subsequent pharmacokinetic studies with dietary quercetin glycosides showed marked differences in absorption rate and bioavailability. Absorbed quercetin was eliminated only slowly from the blood. The metabolism of flavonoids has been studied frequently in various animals, but very few data in humans are available. Two major sites of flavonoid metabolism are the liver and the colonic flora. There is evidence for O-methylation, sulphation and glucuronidation of hydroxyl groups in the liver. Bacterial ring fission of flavonoids occurs in the colon. The subsequent degradation products, phenolic acids, can be absorbed and are found in urine of animals. Quantitative data on metabolism are scarce.

The sugar moiety is a major determinant of the absorption of dietary flavonoid glycosides in man

Flavonoids are antioxidants present in plant foods. They occur mainly as glycosides, i.e. linked with various sugars. It is uncertain to what extent dietary flavonoid glycosides are absorbed from the gut. We investigated how the nature of the sugar group affected absorption of one major flavonoid, quercetin. Quercetin linked with glucose, i.e. quercetin glucoside and quercetin linked with rutinose, i.e. quercetin rutinoside, both occur widely in foods. When we fed these compounds to nine volunteers, the peak concentration of quercetin (Cmax) in plasma was 20 times higher and was reached (Tmax) more than ten times faster after intake of the glucoside (Cmax = 3.5+/-0.6 microM (mean +/- SE); Tmax < 0.5 h) than after the rutinoside (Cmax = 0.18+/-0.04 microM; Tmax = 6.0+/-1.2 h). The bioavailability of the rutinoside was only 20% of that of the glucoside. We suggest that quercetin glucoside is actively absorbed from the small intestine, whereas quercetin rutinoside is absorbed from the colon after deglycosylation. Absorption of other food components might also be enhanced by attachment of a glucose group.

Absorption and disposition kinetics of the dietary antioxidant quercetin in man

Quercetin is a dietary antioxidant that prevents oxidation of low-density lipoproteins in vitro by scavenging to free oxygen radicals. Its intake was inversely associated with coronary heart mortality in Dutch elderly men. However, data on absorption of quercetin in man are scarce and contradictory. We studied the time course of the plasma quercetin concentration in two subjects after ingestion of fried onions containing quercetin glucosides equivalent to 64 mg of quercetin aglycone. Peak plasma levels of 196 ng/ml were reached after 2.9 h, with a half-life of absorption of 0.87 h. The half-life of the distribution phase was 3.8 h, and of the subsequent elimination phase 16.8 h. After 48 h the plasma concentration was about 10 ng/ml. We conclude that quercetin glucosides from onions are absorbed and are eliminated slowly throughout the day. Thus, the dietary antioxidant quercetin could increase the antioxidant capacity of blood plasma.

Flavonols, flavones and flavanols – nature, occurrence and dietary burden

Total flavonol and flavone contents of foods have been determined with validated state-of-the-art methods. Quercetin dominates, and flavonol levels found in vegetables and fruits are below 10 mg kg−1. However, high concentrations are found in onions (300 mg kg−1), kale (450 mg kg−1), broccoli (100 mg kg−1), beans (50 mg kg−1), apples (50 mg kg−1), blackcurrants (40 mg kg−1), and tea (30 mg l−1). The dietary intake of flavonols varies 10-fold between countries (6–60 mg day−1). Flavones are of minor importance in the diet. Tea, wine and fruits are the most important sources of flavanols, but there are gaps in our knowledge on flavanol levels of many foods. The absorption of dietary quercetin glycosides in humans ranges from 20 to 50%. The sugar moiety is an important determinant of the bioavailability of flavonols. The presence of a glucose moiety significantly enhances absorption. The extent of absorption of flavanols in humans seems similar to that of flavonols but has been little studied. Flavonols and flavanols are extensively metabolised, as only 1–2% of them are excreted with an intact flavonoid backbone. Hepatic biotransformations include glucuronidation and sulphatation of the phenolic hydroxyls and O-methylation of catechol groups. Bacteria of the colon cleave the C-ring of the flavonoid nucleus to phenolic acids which are subsequently absorbed. Apart from conjugates, virtually no metabolites have been characterised in humans. Absorption of flavanols is rather fast, with times to reach peak values between 0.5 and 4 h. Flavanols are rapidly excreted, with elimination half-lives of 1–6 h. Quercetin glycosides show rapid to slow absorption; peak values are reached between  < 0.5 and 9 h. The type of glycoside determines the rate of absorption. Excretion of quercetin glycosides is slow: elimination half-lives are 24 h, independent of the type of glycoside. Analytical data for flavanols in foods are needed. Tea, as an important dietary source, has to be studied. Research on the bioavailability of flavonols and flavanols has to be expanded. Attention is needed for the identification and quantification of their metabolites in body fluids. © 2000 Society of Chemical Industry

Lignan contents of Dutch plant foods: a database including lariciresinol, pinoresinol, secoisolariciresinol and matairesinol

Enterolignans (enterodiol and enterolactone) can potentially reduce the risk of certain cancers and cardiovascular diseases. Enterolignans are formed by the intestinal microflora after the consumption of plant lignans. Until recently, only secoisolariciresinol and matairesinol were considered enterolignan precursors, but now several new precursors have been identified, of which lariciresinol and pinoresinol have a high degree of conversion. Quantitative data on the contents in foods of these new enterolignan precursors are not available. Thus, the aim of this study was to compile a lignan database including all four major enterolignan precursors. Liquid chromatography-tandem mass spectrometry was used to quantify lariciresinol, pinoresinol, secoisolariciresinol and matairesinol in eighty-three solid foods and twenty-six beverages commonly consumed in The Netherlands. The richest source of lignans was flaxseed (301,129 microg/100 g), which contained mainly secoisolariciresinol. Also, lignan concentrations in sesame seeds (29,331 microg/100 g, mainly pinoresinol and lariciresinol) were relatively high. For grain products, which are known to be important sources of lignan, lignan concentrations ranged from 7 to 764 microg/100 g. However, many vegetables and fruits had similar concentrations, because of the contribution of lariciresinol and pinoresinol. Brassica vegetables contained unexpectedly high levels of lignans (185-2321 microg/100 g), mainly pinoresinol and lariciresinol. Lignan levels in beverages varied from 0 (cola) to 91 microg/100 ml (red wine). Only four of the 109 foods did not contain a measurable amount of lignans, and in most cases the amount of lariciresinol and pinoresinol was larger than that of secoisolariciresinol and matairesinol. Thus, available databases largely underestimate the amount of enterolignan precursors in foods.

The Biological Relevance of Direct Antioxidant Effects of Polyphenols for Cardiovascular Health in Humans Is Not Established

Human studies provide evidence for beneficial effects of polyphenol-rich foods on cardiovascular health. The antioxidant activity of polyphenols potentially explains these effects, but is the antioxidant activity a reliable predictor for these effects? An International Life Sciences Institute Europe working group addressed this question and explored the potential of antioxidant claims for polyphenols in relation to cardiovascular health by using the so-called Process for the Assessment of Scientific Support for Claims on Foods project criteria. In this process, analytical aspects of polyphenols, their occurrence in foods, dietary intake, and bioavailability were reviewed. Human studies on polyphenols and cardiovascular health were reviewed together with methods for biomarkers of oxidative damage and total antioxidant capacity (TAC). In retrospective studies, F2-isoprostanes and oxidized LDL, the most reliable biomarkers of lipid peroxidation, and measures for TAC showed the expected differences between cardiovascular disease patients and healthy controls, but prospective studies are lacking, and a causal relationship between these biomarkers and cardiovascular health could not be established. Therefore, the physiological relevance of a potential change in these biomarkers is unclear. We found limited evidence that some types of polyphenol-rich products modify these biomarkers in humans. A direct antioxidant effect of polyphenols in vivo is questionable, however, because concentrations in blood are low compared with other antioxidants and extensive metabolism following ingestion lowers their antioxidant activity. Therefore, the biological relevance of direct antioxidant effects of polyphenols for cardiovascular health could not be established. Overall, although some polyphenol-rich foods exert beneficial effects on some biomarkers of cardiovascular health, there is no evidence that this is caused by improvements in antioxidant function biomarkers (oxidative damage or antioxidant capacity).

Analysis and health effects of flavonoids.

Flavonoids are polyphenolic compounds that occur ubiquitously in foods of plant origin. Over 4000 different flavonoids have been described. Flavonoids have a variety of biological effects in numerous mammalian cell systems, in vitro as well as in vivo. Recently much attention has been paid to their antioxidant properties and to their inhibitory role in various stages of tumour development in animal studies. Hitherto, analytical research on flavonoids was mainly aimed at identification and not at quantification. As a consequence, no data were available for epidemiological investigations of flavonoid intake and chronic diseases. Flavonoids in foods are mostly linked to sugars, the so-called glycosides. As one parent compound or aglycone, e.g. quercetin, may be linked to a number of different sugars, quantification in foods is complex. Hydrolysis of the glycosides and subsequent determination of the parent aglycones simplifies this task. Following this approach we developed and validated an HPLC method and determined the flavonol and flavone content of vegetables, fruits and beverages commonly consumed. Subsequent epidemiological evaluation showed that the intake of flavonols and flavones was inversely associated with coronary heart disease in both a prospective cohort study and in a cross-cultural study. However, no relation with cancer risk could be established.