Dietary intake of 337 polyphenols in French adults1–3

Jara Pe´rez-Jime´nez, Le´opold Fezeu, Mathilde Touvier, Nathalie Arnault, Claudine Manach, Serge Hercberg, Pilar Galan,
and Augustin Scalbert

ABSTRACT benes) or as esters (phenolic acids esterified to polyols such as

Background: Epidemiologic studies have suggested an association quinic acid) (4).

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

between polyphenol intake and health. These studies have been The bioavailability and biological properties of dietary poly-
limited to 40 flavonoid and lignan aglycones. phenols vary to a great extent and depend on their chemical
Objective: We estimated intakes of all known individual polyphe- structure (5, 6). The number and position of hydroxyl groups and
nols in the French cohort SUpplémentation en VItamines et Minér- the structure of the heterocycle in flavonoids are important
aux AntioXydants (SU.VI.MAX) by using the recently developed variables that influence their biological properties. Glycosylation
database Phenol-Explorer, which contains content values for 502 and acylation are also important variables that influence ab-
polyphenols in 452 foods. sorption in the gut and bioavailability (7, 8).
Design: A total of 4942 men and women, who were aged 45–60 y For these reasons, it is important to precisely know the intake of
and who had completed at least six 24-h dietary records, partici- individual polyphenols and to relate their intakes in populations to
pated in this study. Foods documented in 24-h dietary records and health and disease outcomes. However, in most cohort studies,
the Phenol-Explorer database were matched, and intakes of all in- a limited number of polyphenols (eg, daidzein, genistein, secoi-
dividual polyphenols were calculated. solariciresinol, matairesinol, and quercetin) have been considered
Results: A total of 337 polyphenols were consumed by SU.VI. and their intake levels determined (9–11). Other authors have used
MAX subjects, including 258 polyphenols consumed by at least the US Department of Agriculture database that contains food-
one-half of the population and 98 polyphenols consumed in an composition data for 38 flavonoid aglycones (12–14) sometimes
amount >1 mg/d. Mean total polyphenol intake was estimated at together with specific composition data for phenolic acids (15).
1193 6 510 mg/d (or 820 6 335 mg/d when expressed as aglycone
These data have been used to calculate intakes and unravel new
equivalents), with hydroxycinnamic acid esters and proanthocyani-
associations between intakes and disease outcomes (16–19).
dins being the most largely consumed polyphenols. These values
The new Phenol-Explorer database (www.phenol-explorer.eu)
may have been underestimated because of insufficient data or lack
contains food-composition data for all known polyphenols
of accurate data on the content in foods for proanthocyanidins and
(flavonoids, phenolic acids, lignans, stilbenes, and other minor
thearubigins. Nonalcoholic beverages and fruit were the most im-
polyphenols) in foods (3). Moreover, it includes data on gly-
portant contributors to polyphenol intakes.
Conclusions: The current study provides intake data for all indi-
cosides and esters. It contains data on a total of 502 polyphenols
vidual polyphenols known to be present in the diet of a cohort. This in 452 foods. The Phenol-Explorer database is freely accessible
information will be essential to characterize the health effects of on the Web (www.phenol-explorer.eu). We report the first ap-
individual phenolic compounds that differ widely in their bioavail- plication of the Phenol-Explorer data to estimate dietary intakes
ability and physiologic properties. The SU.VI.MAX study was reg- of these 502 polyphenols in a well-established nutritional cohort
istered at clinicaltrials.gov as NCT00272428. Am J Clin Nutr
1
2011;93:1220–8. From the Clermont Université, Université d’Auvergne, Unité de
Nutrition Humaine, Clermont-Ferrand, France (JP-J, CM, and AS); the In-
stitut National de la Recherche Agronomique (INRA), Unite Mixte de Re-
INTRODUCTION cherche 1019, Unite de Nutrition Humaine, Centre de Recherche en Nutrition
Humaine Auvergne, Clermont-Ferrand, France (JP-J, CM, and AS); the In-
A number of clinical trials and cohort studies have suggested stitut National de la Sante de de la Recherche Medicale U557, INRA U1125,
a role of dietary polyphenols in the prevention of several major Conservatoire National des Arts et Metiers, Université de Paris 13, Bobigny,
chronic diseases such as cardiovascular diseases, cancers, di- France (LF, MT, NA, SH, and PG); the Centre de Recherche en Nutrition
abetes, neurodegenerative diseases, and osteoporosis (1, 2). In Humaine Ile-de-France, Unité de Recherche en Epidémiologie Nutritionnelle,
many of these studies, only a limited number of polyphenols have Bobigny, France (LF, MT, NA, SH, and PG); and the Département de Santé
been considered that are far from representing the large diversity Publique, Hôpital Avicenne, Bobigny, France (SH).
2
of compounds observed in the diet. Nevertheless, .500 different Supported by Unilever, Danone, and Nestlé.
3
Address correspondence to A Scalbert, International Agency for Re-
molecules are known in foods, from low-molecular-weight
search on Cancer (IARC), Nutrition and Metabolism Section, Biomarkers
phenolic acids to highly polymerized proanthocyanidins (3, 4). Group, 150 cours Albert Thomas, F-69372 Lyon cedex 08, France. E-mail:
Dietary polyphenols belong to 4 main classes of flavonoids, [email protected].
phenolic acids, stilbenes, and lignans, that are largely present in Received October 27, 2010. Accepted for March 22, 2011.
a glycosidic form (glycosides of flavonoids, lignans, and stil- First published online April 13, 2011; doi: 10.3945/ajcn.110.007096.

1220 Am J Clin Nutr 2011;93:1220–8. Printed in USA. Ó 2011 American Society for Nutrition
 DIETARY INTAKE OF POLYPHENOLS IN FRENCH ADULTS 1221
of French adults called the SUpplémentation en VItamines et the Phenol-Explorer database. These food entries included some
Minéraux AntioXydants (SU.VI.MAX) study. To our knowledge, spirits such as tequila, coconut milk, honey, some breakfast
this study provides the most comprehensive data on polyphenol cereals, and certain minor oils such as walnut oil, maize oil, or
intake so far published. grape seed oil. The consumption or polyphenol content of these
foods was low, and therefore, the contribution of these foods to
polyphenol intake was considered insignificant. In contrast, some
SUBJECTS AND METHODS
SU.VI.MAX food entries could have corresponded to several
Study population entries in the Phenol-Explorer database. For example, the SU.VI.
MAX item olive oil could have corresponded to either extra-
Subjects were participants of the SU.VI.MAX study, which
virgin, virgin, or refined olive oil in Phenol-Explorer database.
was a randomized double-blind, placebo-controlled, primary
The polyphenol content in the SU.VI.MAX food was weighted
prevention trial that was undertaken to determine whether sup-
according to their respective consumptions in the French pop-
plementation with antioxidant vitamins and minerals at nutri-
ulation. For mixed foods made of polyphenol-containing
tional doses could reduce incidences of cancers, ischemic heart
ingredients (eg, cocoa products) and for recipes, polyphenol

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

disease, and overall mortality. The rationale, design, and methods
contents were calculated on the basis of contents of the ingredient
of the study have been described in detail elsewhere (20, 21). In
(eg, cocoa) or food component and their polyphenol composition.
brief, 13,017 eligible subjects (women aged 35–60 y and men
aged 45–60 y) were included in 1994 to be followed for 7.5 y.
Participants were volunteers recruited by using mass-media Estimation of polyphenol intakes in the SU.VI.MAX cohort
campaigns and were free of chronic diseases and apparently
healthy at baseline. The SU.VI.MAX study was approved by the An advanced search was carried out in the Phenol-Explorer
ethical committee for studies with human subjects (CCPPRB no. database (www.phenol-explorer.eu/contents) to retrieve mean
706) of Paris-Cochin and the Commission Nationale de content values for all polyphenols contained in the SU.VI.MAX
l’Informatique et des Libertés (no. 334641), which advocated that foods. The total polyphenol content was considered as the sum of
all medical information was confidential and anonymous. The all individual polyphenols as determined by chromatography,
study was also registered at clinicaltrials.gov as NCT00272428. except for proanthocyanidins for which content data obtained by
normal phase HPLC were used (4). For foods that contained
polyphenols linked to the food matrix that were only solubilized
Assessment of food intakes and quantified after basic or acid hydrolysis, content values
Dietary data were collected by using the Minitel Telematic obtained by chromatography after hydrolysis were used (4); in
Network, which was widely used as an adjunct to the telephone in particular, this concerned lignans in several foods, ellagic acid in
France at the beginning of the SU.VI.MAX study. A tiny central walnuts, and hydroxycinnamic acids in cereals, white beans, and
processing unit that contained specialized software enabled olives. Similar advanced searches in Phenol-Explorer–retrieved
subjects to fill out the computerized dietary record. Participants total antioxidant contents were determined by the Folin assay
were invited to provide a 24-h dietary record every 2 mo from with gallic acid, catechin, and caffeic acid used as standards (23).
1995 to 1996. At enrollment, subjects received a scheduled The polyphenol intake was also calculated as aglycone equiv-
calendar for recording of 24-h dietary data. The 24-h records alents by taking into account all possible forms of glycosides and
were randomly distributed for 2 weekend days and 4 weekdays esters by removing, for each individual polyphenol, the contri-
per year so that each day of the week and all seasons were covered bution to molecular weight of the nonphenolic part of the
to account for the individual variability in intake. An instruction molecule (eg, sugar from flavonoid glycosides or quinic acid
manual for the codification of foods and beverages, which in- from chlorogenic acids).
cluded photographs to facilitate the estimation of portion sizes, The following data that were missing in Phenol-Explorer were
was mailed to each subject. This manual included validated obtained by extrapolation. The hydroxycinnamic acid content in
photographs of .250 foods (which corresponded to 1000 ge- breakfast cereals was obtained from values in wheat flour. Some
neric foods) that were represented in 3 different portion sizes. missing data for orange fruit were extrapolated from orange juice
Along with the 2 intermediate and 2 extreme quantities, there data. For some cooked foods consumed in the SU.VI.MAX
were 7 choices of amounts. Photos of portion sizes were pre- cohort and absent in the Phenol-Explorer database (a number of
viously validated by using 780 subjects in a pilot study (22). vegetables, cereals, and seeds), yield factors (24) were applied to
(See supplemental Table 1 under “Supplemental data” in the take into account the gain or loss of water during cooking.
online issue for the average consumption of main foods that Several products that contained refined wheat flour were in-
contributed to polyphenol intake.) cluded in the list of foods (eg, different kinds of breads and
biscuits). Polyphenol contents in these foods were estimated from
their wheat-flour contents. Although the polyphenol intake from
Correspondence between food items in dietary records and
each of these foods was calculated, they are presented together as
in the Phenol-Explorer database
refined wheat flour products.
SU.VI.MAX dietary records included 736 food items (ex-
cluding recipes). The Phenol-Explorer database contains data on
the content of 502 polyphenols in 452 foods (3, 4). All animal Statistical analyses
foods that contain no or only traces of plant polyphenols were Analyses focused on the 4942 subjects who were aged 45–60 y
excluded. Certain food entries that may have contained poly- and had completed at least six 24-h dietary records during the first
phenols were present in SU.VI.MAX dietary records but not in 2 y of follow-up (1994–1995). The numbers of included dietary
1222 PÉREZ-JIMÉNEZ ET AL

records were chosen to be balanced between winter and summer TABLE 2

to more precisely take into account the seasonal variability in Total polyphenol intakes in the SUpplémentation en VItamines et
food intake. These subjects did not differ from the total pop- Minéraux AntioXydants (SU.VI.MAX) cohort according to
sociodemographic characteristics1
ulation with respect to major characteristics. Data are presented
as means (6SDs) or medians for continuous variables and fre- Polyphenol intake
quencies or percentages for categorical variables. The general
characteristics of the study population are shown in Table 1. The n Mean 6 SD Median
mean and median intakes of all individual polyphenols, poly- mg/d
phenol subclasses, and total polyphenols in the SU.VI.MAX Total population 4942 1193 6 510 1123
cohort were determined for the whole study population as well Sex
as according to different sociodemographic characteristics such Male 2596 1270 6 514 1213
as sex, age, occupation, and educational level. Differences in Female 2346 1108 6 491 1038
P — 0.001 —
intakes between groups were tested by using Student’s t test for
Age class
independent samples or by analyses of covariance, and differ- 1189 6 513

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

45–49 y 1994 1114
ences below the probability level (P , 0.05) were considered 50–54 y 1544 1191 6 502 1130
significant. Statistical analyses were performed with SAS soft- 55–60 y 1404 1201 6 510 1137
ware (version 9.2; SAS Institute Inc, Cary, NC). P — 0.80 —
Educational level
Elementary school 1128 1154 6 508 1094
RESULTS Secondary school 1961 1178 6 509 1104
University 1853 1233 6 509 1163
Total intakes of polyphenols P — 0.001 —
Among the 736 food items considered in the SU.VI.MAX Occupational level
dietary records, 232 food items were shown to contain poly- Upper management 1443 1243 6 508 1169
Middle management, employee 2540 1175 6 504 1101
phenols according to the Phenol-Explorer database. A total of
Farmer, self-employed 234 1264 6 580 1189
337 polyphenols were shown in these foods. One hundred thirty Laborer 219 1230 6 498 1177
polyphenols of the Phenol-Explorer database were absent from Unemployed 244 1087 6 474 1016
foods of the SU.VI.MAX dietary records (see supplemental P — 0.001 —
Table 2 under “Supplemental data” in the online issue). 1
Comparisons across categories were performed by using Student’s t
Total polyphenol intake in the SU.VI.MAX cohort was de- test with equal variance for sex and ANOVA for age class, educational level,
termined as the sum of the intakes of all individual polyphenols and occupational level.
(Table 2). Mean and median polyphenol intakes for the whole
population were 1193 and 1123 mg polyphenols/d, re- jects with a higher educational level (university) showed a sig-
spectively. The majority of dietary polyphenols in foods were nificantly (P = 0.001) higher polyphenol intake (1154 6 508
glycosides (80% of 277 flavonoids) and esters (50% of 108 compared with 1233 6 509 mg/d) than did subjects with a lower
phenolic acids) (4), and therefore, these intake values included educational level (primary school) mainly because of higher
intakes of sugars and polyols linked to aglycones. The total intakes of flavonols, proanthocyandins, catechins, and thea-
intake of aglycones was also determined. Mean (6SD) and flavins.
median intakes of polyphenol aglycones were 820 6 335 and The contribution of different food groups to the total poly-
531 mg/d, respectively phenol intake as well as the main food contributors in each food
Polyphenol intake was also calculated according to sex, age, group are shown in Table 3. Nonalcoholic beverages and fruit
educational level, and occupation. Intakes were higher in men were the main polyphenol providers with respective con-
than in women (Table 2). Higher intakes of anthocyanins, tributions of 658 and 206 mg/d (390 and 162 mg/d as agly-
proanthocyandins, and stilbenes were observed in men than in cones). Alcoholic beverages, cocoa products, and vegetables
women (70 6 52 compared with 43 6 36 mg/d, 260 6 178 provided ’100 mg/d each; cereals provided ’50 mg/d, and
compared with 191 6 136 mg/d, and 7 6 6 compared with 2 6 seeds and oils each provided ,8 mg/d. Seasonings, which in-
3 mg/d, respectively; all P , 0.001). In contrast, women had cluded vinegar, herbs, and spices, had only a minor contribution
a higher intake of catechins (114 6 133 compared with 86 6 of ,1 mg/d. Within each food group, the few main food con-
93 mg/d) and theaflavins (16 6 23 compared with 8 6 16 mg/d) tributors were identified as coffee and tea for nonalcoholic
than did men (all P , 0.001). Age, within the limited range beverages and red wine for alcoholic beverages (Table 3). For
tested, had no significant influence on polyphenol intakes. Sub- other food groups and, in particular, fruit and vegetables,
polyphenols were more widely distributed over several food
TABLE 1 sources. Flavonoids and phenolic acid were the 2 main classes of
General characteristics of the study population (n = 4942) polyphenols consumed with intakes of 506 and 639 mg/d, re-
spectively (Table 3). Fruit was the main sources of flavonoids
Characteristics Mean 6 SD Minimum Maximum
(35%), whereas nonalcoholic beverages were the main sources
Age (y) 51.5 6 4.4 45 60 of phenolic acids (nearly 80%).
Weight (kg) 69.2 6 13.1 37 146 Polyphenol intakes were highly dependent on food preferences
BMI (kg/m2) 24.5 6 3.6 15.8 45.3 and, more particularly, on preferences for their main dietary
Energy intake (kcal/d) 2166 6 608 582 5067 sources. Consumers of coffee or red wine had systematically
 DIETARY INTAKE OF POLYPHENOLS IN FRENCH ADULTS 1223
TABLE 3
Contributions of different food groups to polyphenol intakes in the 4942 participants in the SUpplémentation en VItamines et Minéraux AntioXydants
(SU.VI.MAX) cohort and main food sources
Polyphenol intake1

Phenolic Total polyphenols Main food contributors (% contribution to

Food class Total polyphenols Flavonoids acids as aglycones polyphenol intake in the food group)2

mg/d per person

Nonalcoholic beverages 658 6 426 114 6 74 524 6 339 390 6 258 Coffee (79), tea (17), orange juice (2), pummelo juice (0.2)
Fruit 206 6 151 172 6 130 32 6 23 162 6 119 Apples (45), strawberries (15), plums (8), cherries (8)
Alcoholic beverages 99 6 136 73 6 100 19 6 26 83 6 114 Red wine (94), white wine (2), rosé wine (2), blond beer (1)
Cocoa products 90 6 132 90 6 132 16 1 91 6 133 Dark chocolate (73), cocoa powder (23), milk chocolate (3)
Vegetables3 81 6 42 26 6 13 51 6 26 52 6 27 Potatoes (30), green chicory (19), onions (14), spinach (9)
Cereals 46 6 18 28 6 11 76 3 32 6 12 Refined wheat-flour products (66), whole-grain wheat-flour

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

products (26), breakfast cereals (7), rice (2)
Seeds 8 6 18 3 6 7 5 6 11 8 6 18 Walnuts (37), hazelnuts (27), white beans (15), chestnuts (14)
Oils 4 6 2 0.1 6 0.05 0.02 6 0.01 2 6 1 Extra-virgin olive oil (99), rapeseed oil (0.7)
Seasonings 0.4 6 0.3 0.06 6 0.04 0.2 6 0.1 0.4 6 0.3 Vinegar (90), parsley (4), ginger (4), curry (1)
Total foods 1193 6 510 506 6 219 639 6 273 820 6 335 Coffee (44), tea (9), apples (6), red wine (6)
1
Values are means 6 SDs.
2
The first 4 main food sources in each food group are given; a lower number of foods indicates the absence of other food sources that contained
polyphenols in the food group considered.
3
Including tubers.

a higher mean total polyphenol intake than did nonconsumers. Intakes of different polyphenol classes
Coffee drinkers (92% of the total population) had a total poly- Polyphenol intake in the SU.VI.MAX cohort was also cal-
phenol intake of 1224 6 471 mg/d, and coffee accounted for 44% culated for each polyphenol class (Table 4). Hydroxycinnamic
of this intake. This intake was significantly higher (P , 0.001) than acids were the most largely consumed polyphenols, with an
that measured in noncoffee drinkers (807 6 343 mg/d). For red- intake close to 600 mg/d. Proanthocyanidins were the second
wine consumers (75% of the total population), total polyphenol most largely consumed polyphenols (227 mg/d), and this intake
intake was 1242 6 460 mg/d, with 11.1% of the intake coming could have actually been higher if more data on the content of
from red wine, which was an amount that was also significantly proanthocyanidin oligomers were available (4). Catechins and
higher (P , 0.001) than that observed in nonconsumers (1042 6 anthocyanins were the 2 next classes for polyphenol intake (99
474 mg/d). In contrast, no difference in total polyphenol intake was and 57 mg/d, respectively) and were followed in decreasing
observed for tea between consumers (52% of the total population) order by flavonols, hydroxybenzoic acids, flavones, flavanones,
and nonconsumers. theaflavins, and dihydroflavonol. The other polyphenols included

TABLE 4
Intakes of main polyphenol classes and subclasses in the 4942 participants in the SUpplémentation en VItamines et Minéraux AntioXydants (SU.VI.MAX)
cohort and main food sources
Polyphenol intake2
Main food contributors (% contribution to
Polyphenol subclass1 Total in the subclass Total as aglycones intake of the polyphenol subclass)3

mg/d per person

Hydroxycinnamic acids 599 6 426 316 6 225 Coffee (83), potatoes (4), apples (2), green chicory (2)
Proanthocyanidins 227 6 162 227 6 162 Apples (31), cocoa products (18), red wine (12), strawberries (8)
Catechins 99 6 116 87 6 102 Tea (71), red wine (10), apples (6), cocoa products (5)
Anthocyanins 57 6 47 35 6 29 Red wine (41), cherries (23), strawberries (20), black grapes (6)
Flavonols 51 6 28 34 6 19 Tea (23), onions (23), spinach (13), red wine (12)
Hydroxybenzoic acids 41 6 39 40 6 38 Tea (49), red wine (15), green chicory (9), walnuts (8)
Flavones 33 6 17 18 6 9 Refined wheat-flour products (64), whole-grain wheat-flour
products (20), oranges (6), orange juice (5)
Flavanones 26 6 29 13 6 14 Oranges (50), orange juice (44), red wine (3), pummelo juice (2)
Theaflavins 12 6 21 9 6 16 Tea (100)
Dihydroflavonols 7 6 9 5 6 5 Red wine (96), white wine (2), rosé wine (1)
Other polyphenols 41 6 20 28 6 14 Coffee (21), refined wheat-flour products (18), whole-grain
wheat-flour products (16), extra-virgin olive oil (14)
1
Only polyphenol classes or subclasses with a daily mean intake .5 mg/d per person.
2
Values are means 6 SDs.
3
Four main food sources for each polyphenol; a lower number indicates the absence of other food sources.
1224 PÉREZ-JIMÉNEZ ET AL

tyrosols (14.7 6 5.9 mg/d), alkylphenols (11.3 6 15.4 mg/d), hydroxycinnamic acids (55 compounds), flavonols (54 com-
stilbenes (4.8 6 5.2 mg/d), dihydrochalcones (3.7 6 3.6 mg/d), pounds), and anthocyanins (45 compounds), which all showed
alkylmethoxyphenols (2.8 6 2.2 mg/d), methoxybenzaldehydes a large diversity of hydroxylation, methylation, glycosylation, or
(0.89 6 1.06 mg/d), methoxyphenols (0.37 6 0.32 mg/d), lignans esterification patterns. The number of aglycones was also high
(0.4 6 0.2 mg/d), hydroxycoumarins (0.1 6 0.16 mg/d), for proanthocyanidins (because of their polymeric nature) and
hydroxybenzaldehydes (0.09 6 0.09 mg/d), furanocoumarins for hydroxybenzoic acids, hydroxycinnamic acids, flavanones,
(0.04 6 0.13 mg/d), methoxyphenylpropenes (0.01 6 0.07 mg/d), and flavones. The distribution of some of these polyphenols may
isoflavonoids (0.01 6 0.247 mg/d), and curcuminoids (0.003 6 have been limited to only one food consumed by a minor fraction
0.029 mg/d). When expressed as aglycone equivalents, the rela- of the cohort (eg, the [6]-gingerol characteristic of ginger).
tive contribution of hydroxycinnamic acids was reduced from The numbers of more commonly consumed polyphenols are
50% to 40% and that of proanthocyanidins increased from 19% to also shown in Table 5. A total of 258 compounds (76%) out of the
29%, whereas the relative contributions of the other polyphenol 337 consumed polyphenols had median intake values higher than
subclasses remained similar. zero and corresponded to polyphenols consumed by at least one-
The main food contributors to the intake of each polyphenol half of cohort members. These commonly consumed polyphenols

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

class or subclass are also shown in Table 4. Hydroxycinnamic again included large numbers of hydroxycinnamic acids, fla-
acids arose mainly from coffee but also from potatoes, apples, vonols, and anthocyanins.
and green chicory. Potatoes, vegetables, and fruit together pro- The intake of all individual polyphenols consumed in the
vide ’100 mg/d. Tea and red wine were the main sources of SU.VI.MAX cohort was determined (see supplemental Tables 3
hydroxybenzoic acids, with contributions of 20 and 6 mg/d, and 4 under “Supplemental data” in the online issue). Ninety-
respectively, mainly as gallic acid. Sources of proanthocyanidins eight polyphenols were consumed in amounts .1 mg/d The 25
were fruit, cocoa products, and red wine. Apart from these same first compounds in the list were all consumed in amounts greater
sources, the major source of catechins was tea. Anthocyanins than 8 mg/d (Table 6). Nine of the 25 most ingested polyphenols
arose from red wine and red fruit; flavonols arose from tea, in the SU.VI.MAX cohort were phenolic acids, including 7
onions, some green vegetables, and wine; and flavanones arose hydroxycinnamic acids that largely originated from coffee. This
from citrus fruit. Main sources for other polyphenols were red list also included 6 catechins that principally originated from
wine and olive oil for tyrosols and cereals for alkylphenols. tea, red wine, and cocoa products. The remaining compounds
among these 25 polyphenols were 5 proanthocyanidins that
Intakes of individual polyphenols came from tea, fruit, wine, and cocoa products, 2 anthocyanins
A total of 337 polyphenols that corresponded to 155 different that came from red wine, red fruit, and black olives, 2 flavones
polyphenol aglycones were found in the various foods consumed that mainly came from wheat products, and one flavanone,
by the SU.VI.MAX cohort (Table 5). The larger groups were hesperidin, that mainly came from oranges and orange juice.

TABLE 5
Number of individual polyphenols consumed by the 4942 participants in the SUpplémentation en VItamines et Minéraux
AntioXydants (SU.VI.MAX) cohort
All polyphenols consumed Polyphenols commonly consumed1

Aglycones, Aglycones,
glycosides, and esters Aglycones2 glycosides, and esters Aglycones2

Flavonoids
Anthocyanins 45 9 36 9
Chalcones 1 1 0 0
Dihydrochalcones 3 1 3 1
Dihydroflavonols 2 2 2 2
Catechins 9 4 8 4
Theaflavins 4 1 4 1
Proanthocyanidins 16 16 12 12
Flavanones 15 8 11 4
Flavones 19 7 15 7
Flavonols 54 13 38 7
Isoflavonoids 13 4 0 0
Phenolic acids
Hydroxybenzoic acids 28 17 22 12
Hydroxycinnamic acids 55 12 50 8
Hydroxyphenylacetic acids 7 7 4 2
Stilbenes 7 4 7 4
Lignans 7 7 5 4
Other polyphenols 52 42 41 30
Total 337 155 258 107
1
Median intake value .0.
2
The number of aglycones consumed as such or in the form of glycosides and esters.
 DIETARY INTAKE OF POLYPHENOLS IN FRENCH ADULTS 1225
Together, the 25 most-consumed polyphenols accounted for termined in the French SU.VI.MAX cohort. The remaining
68% of the total polyphenol intake. compounds were either not consumed (eg, anethole in fennel tea
Many other polyphenols were consumed in amounts less than or star anise) or present in foods that were not documented in the
8 mg polyphenols/d (see supplemental Table 3 under “Supple- dietary questionnaires (eg, eugenol in cloves). Most of these
mental data” in the online issue). The most largely consumed polyphenols were observed in minor foods, were not consumed,
flavonol, tyrosol, theaflavin, alkylresorcinol, and stilbene were or were consumed in very low amounts (see supplemental Table
quercetin 3,4#-O-diglucoside, tyrosol, theaflavin 3#-O-gallate, 2 under “Supplemental data” in the online issue).
5-heneicosylresorcinol, piceatannol 3-O-glucoside, and iso- Most previous studies on polyphenol intake were focused on
lariciresinol, respectively. specific classes of polyphenols, flavonoids (16, 28–38), lignans
Intakes of all polyphenol aglycones were also determined (see (10), or stilbenes (39). The most comprehensive data were
supplemental Table 4 under “Supplemental data” in the online obtained in a Finnish cohort (15), which included intake values
issue). Fifty-five polyphenol aglycones were consumed in for flavonoids, phenolic acids, and lignans. However, only
amounts greater than 1 mg /d. The 10 aglycones with the highest aglycones were considered in these studies because of the lack of
mean intakes were caffeic acid (264 6 108 mg /d), ferulic acid data on the content of glycosides and esters in common food-

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

(39 6 16 mg/d), gallic acid (35 6 21 mg/d), (2)-epicatechin composition databases.
(30 6 18 mg/d), (2)-epigallocatechin (23 6 29 mg/d), pro- Intake values for the different classes of polyphenols varied
cyanidin dimer B2 (22 6 13 mg/d), quercetin (20 6 5 mg/d), widely between studies (eg, 4–121 mg/d for catechin monomers,
(+)-catechin (20 6 13 mg/d), apigenin (17 6 6 mg/d), and 3–47 mg/d for anthocyanins, 20–78 mg/d for flavonones, and
procyanidin dimer B1 (15 6 12 mg/d). 5–25 mg/d for flavonols) (9, 40). Such large differences could be
explained by various dietary habits in the different populations.
However, a number of technical issues could also explain these
DISCUSSION differences. First, the presence or absence of data for a given
Intakes of 337 phenolic compounds out of the 502 polyphenols polyphenol in food-composition tables may have introduced
documented in the Phenol-Explorer database have been de- a bias in the comparison of intakes between studies. For example,

TABLE 6
List of the 25 most-consumed individual polyphenols in the 4942 participants of the SUpplémentation en VItamines et Minéraux AntioXydants
(SU.VI.MAX) cohort and their main food sources
Main food contributors (% contribution
Polyphenol Polyphenol subclass Polyphenol intake1 to intake of the polyphenol)2

mg/d per person

5-Caffeoylquinic acid Hydroxycinnamic acids 216 6 142 Coffee (76), potatoes (10), apples (4), artichokes (3)
3-Caffeoylquinic acid Hydroxycinnamic acids 141 6 117 Coffee (90), plums (3), cherries (2), prunes (1)
4-Caffeoylquinic acid Hydroxycinnamic acids 131 6 104 Coffee (97), tea (0.8), prunes (0.3), tomatoes (0.2)
5-Feruloylquinic acid Hydroxycinnamic acids 27 6 23 Coffee (99), carrots (0.3)
(2)-Epicatechin Catechins 24 6 15 Tea (28), apples (24), cocoa products (17), red wine (15)
Procyanidin dimer B2 Proanthocyanidins 22 6 13 Apples (48), red wine (23), tea (14), cocoa products (10)
4-Feruloylquinic acid Hydroxycinnamic acids 20 6 17 Coffee (99), carrots (0.1)
(+)-Catechin Catechins 17 6 12 Red wine (41), tea (15), cocoa products (10), peaches (6)
Hesperidin Flavanones 17 6 20 Oranges (53), orange juice (46), lemon juice (0.6)
(2)-Epigallocatechin 3-O-gallate Catechins 16 6 29 Tea (100)
Apigenin galactoside-arabinoside Flavone 16 6 10 Refined wheat flour products (75), whole-grain wheat flour
products (25)
Procyanidin dimer B1 Proanthocyanidins 15 6 12 Peaches (30), tea (29), red wine (26), cocoa products (5)
Malvidin 3-O-glucoside Anthocyanins 14 6 16 Red wine (85), black grapes (14), red raspberries (0.06)
Procyanidin dimer B3 Proanthocyanidins 14 6 15 Red wine (85), tea (6), peaches (4), strawberries (1)
(+)-Gallocatechin Catechins 14 6 24 Tea (99), red wine (0.4), strawberries (0.04), lentils (0.01)
5-O-Galloylquinic acid Hydroxybenzoic acids 13 6 24 Tea (100)
Gallic acid Hydroxybenzoic acids 12 6 11 Tea (44), red wine (29), green chicory (17), chestnuts (4)
(2)-Epigallocatechin Catechins 12 6 22 Tea (99), red wine (0.3), almonds (0.2), broad bean seeds (0.1)
Procyanidin dimer B4 Proanthocyanidins 12 6 12 Red wine (72), tea (25), plums (0.07), strawberries (0.029)
Cyanidin 3-O-rutinoside Anthocyanins 11 6 21 Cherries (82), plums (15), black currants (1), black olives (1)
Apigenin arabinoside-glucoside Flavones 11 6 6 Refined wheat-flour products (78), whole-grain wheat-flour
products (22)
3-Feruloylquinic acid Hydroxycinnamic acids 9.9 6 8.3 Coffee (98), plums (0.8), carrots (0.5), cherries (0.3)
(2)-Epicatechin 3-O-gallate Catechins 9.9 6 15.7 Tea (92), red wine (6), black grapes (0.6), strawberries (0.4)
Procyanidin trimer T2 Proanthocyanidins 8.8 6 10.7 Red wine (100)
Ferulic acid Hydroxycinnamic acids 8.3 6 4.1 Refined wheat-flour products (78), cocoa products (11), white
beans (4), red wine (0.8)
1
Values are means 6 SDs.
2
Four main food sources for each polyphenol; a lower number indicates the absence of other food sources.
1226 PÉREZ-JIMÉNEZ ET AL

thearubigins, which are oxidation products of catechins, were not and Folin values were collated for a large number of foods in the
included in the Phenol-Explorer database because of their ill- Phenol-Explorer database (3) and were used to calculate the total
defined structure and the lack of accurate methods for their es- polyphenol intake in the SU.VI.MAX. cohort. An intake of 1998
timation, whereas thearubigins were included in an Australian mg polyphenols/d was obtained (data not shown), which was
study (41). Thearubigin content values used in the Australian a value that was much higher than the sum of individual poly-
study (41) were extracted from the US Department of Agriculture phenols (820 6 335 mg/d) expressed as aglycone equivalents
database and were obtained by an unspecific spectrophotometric (the aglycone is the phenolic fraction of the polyphenol mole-
method (13). These values were not included in the Phenol- cule that reduces the Folin reagent) (23). This difference was
Explorer database, which resulted in a much lower flavanol intake explained by the presence in foods of nonphenolic food con-
in the current study (87 mg/d) than in the Australian study stituents such as ascorbic acid that also responded to the Folin
(454 mg flavanols/d). Second, content values may differ from one assay and led to an overestimation of the polyphenol content in
food-composition table to another (23). Proanthocyanidin intake foods (23). The difference could also be explained by the lack of
estimated in the Finnish cohort (116 mg/d) was lower than that in suitable methods to estimate complex phenolic compounds, such
the SU.VI.MAX cohort (227 mg/d), and this may have been as thearubigins and proanthocyanidins, that resulted in an un-

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

because of the much lower proanthocyanidin content value in derestimation of the total polyphenol content in the current
dark chocolate used in the first study (,500 mg/100 g) than the work. Due to interference with nonphenolic reducing com-
value used in the present study (1490 mg/100 g) (3, 15). pounds, the Folin assay cannot be considered as a suitable
Thus, the comparison of polyphenol intakes in different method for the total polyphenol estimation.
populations appeared to be difficult because of the large het- The main food contributors to polyphenol intakes were de-
erogeneity of food-composition data. The use of the same food- termined in the SU.VI.MAX. cohort, and the data largely con-
composition database or of harmonized food-composition data is firmed those obtained in other Western populations (15, 34).
highly desirable. The systematic publication of the polyphenol Three beverages (coffee, tea, and red wine) accounted for 44%,
food-composition data used in epidemiologic studies should also 9%, and 6%, respectively, of the total polyphenol intake, and
be encouraged to facilitate the comparison of polyphenol intake fruit, cocoa products, vegetables, and cereals accounted for 17%,
data between studies. All food-composition data used in the 8%, 7%, and 4%, respectively, of the total polyphenol intake
current work were available on the Phenol-Explorer website (Table 3). Oils and seasonings provided only a minor fraction of
(www.phenol-explorer.eu) together with all original data used to the total polyphenols consumed.
build the database (3). In conclusion, for the first time to our knowledge, the current
Thus, only a crude comparison of intakes of polyphenol study provided information on intakes in a cohort of 337 in-
aglycones with values previously published was possible. The dividual polyphenols with a level of detail that was not achieved
total polyphenol aglycone intake of 820 mg/d observed in the before. However, the study also had certain limitations. As
present study was close to the intake obtained in the Finnish mentioned, food-content data for some relatively abundant
cohort (752 mg/d) (15). Phenolic acids were the main poly- phenolic compounds were missing (thearubigins) or insufficient
phenols consumed in both studies, and their intake was 2 times (proanthocyanidins) (4, 46). For some major polyphenol dietary
higher in the Finnish cohort than in the French cohort (608 sources, composition data are still scarce and intake values might
compared with 356 6 228 mg/d, respectively, expressed as be revised as more data become available. This was particularly
aglycone equivalents), which was a difference that was con- evident for coffee for which only 4 literature sources have been
sistent with the twice higher consumption of coffee in Finland used to calculate intakes of phenolic acids and these calculations
(42). Proanthocyanidins were the second most abundant pol- did not take into account the different brewing recipes. Data on
yphenols in both cohorts (116 and 227 mg /d in the Finnish cooked foods were also lacking in the Phenol-Explorer database
and French cohorts, respectively). Other flavonoid classes (a new module is under construction). Taking into account losses
were consumed in amounts that varied between 5 and 87 mg/d, in polyphenol content during cooking will be particularly im-
depending on the class and cohort. Some polyphenol classes portant for vegetables and may result in lower estimates of their
were hardly consumed, with intakes ,1 mg polyphenols/d, contribution to polyphenol intakes.
because of their low contents in foods (lignans and stilbenes) The level of detail on dietary intakes of polyphenols achieved
or to the low consumption of their main dietary sources in the present study will be essential to study associations be-
(isoflavones). Although these particular polyphenols exhibit tween the intake of individual polyphenols with health and
some interesting biological properties, their low daily in- disease outcomes. Focus should clearly be given on individual
takes in the French and other Western populations raises polyphenol compounds rather that polyphenol classes or sub-
questions about their significance in terms of health and dis- classes to take into account subtle structural differences between
ease prevention. individual compounds within a class or subclass that can result in
Total polyphenol intake was influenced by sex with a higher major differences in the bioavailability and biological effects (8,
intake of polyphenols in men than in women. A similar difference 47). With the development of omics approaches in human bi-
according to sex was previously reported (15, 34) and was largely ology, the complexity of human organism has been commonly
explained by higher consumptions of coffee and red wine in men. described and integrated into complex models. In contrast, the
Age within the limited range studied (45–60 y) had no significant chemical complexity of foods and the human diet have rarely
influence on intakes. been considered (48, 49). An appraisal of the diet in all its
Total polyphenol intake was calculated in several studies by chemical complexity should be essential to understand the
using the Folin colorimetric assay (43–45). This assay has often complex interactions that link foods to human health. This work
been used to estimate total polyphenol content in foods (25–27), illustrates how the food complexity could be characterized to
 DIETARY INTAKE OF POLYPHENOLS IN FRENCH ADULTS 1227
better define the exposure to food bioactives and, ultimately, the to phytoestrogen intake derived from an improved database. Am J Clin
effects on health. Nutr 2010;91:440–8.
19. Nöthlings U, Murphy SP, Wilkens LR, Henderson BE, Kolonel LN.
The authors’ responsibilities were as follows—AS, PG, and SH: designed, Flavonols and pancreatic cancer risk: the multiethnic cohort study. Am
planned, and monitored the study; JP-J and NA: matched the foods in the J Epidemiol 2007;166:924–31.
food-frequency questionnaire to the foods in the Phenol-Explorer database; 20. Hercberg S, Preziosi P, Briancon S, et al. A primary prevention trial
using nutritional doses of antioxidant vitamins and minerals in car-
MT, NA, and LF: performed statistical analyses; JP-J, LF, and AS: interpreted
diovascular diseases and cancers in a general population: the SU.VI.
data; JP-J and AS: wrote the first draft of the manuscript; and all authors: MAX study–design, methods, and participant characteristics. Control
contributed to the writing of the manuscript and approved the final version Clin Trials 1998;19:336–51.
of the manuscript. The funders of the study did not participate in the design 21. Hercberg S, Galan P, Preziosi P, et al. The SU.VI.MAX study–a ran-
and implementation of the study or the data analyses and interpretation. None domized, placebo-controlled trial of the health effects of antioxidant
of the authors had a personal or financial conflict of interest. vitamins and minerals. Arch Intern Med 2004;164:2335–42.
22. Le Moullec N, Deheeger M, Preziosi P, et al. Validation of photo-
graphic document used to estimate the amounts of foods eaten by
subjects in the Suvimax study. Cahiers de Nutrition et de Dietetique
REFERENCES 1996;31:158–64.

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023

1. Arts ICW, Hollman PCH. Polyphenols and disease risk in epidemio- 23. Pérez-Jiménez J, Neveu V, Vos F, Scalbert A. Identification of the 100
logic studies. Am J Clin Nutr 2005;81:317S–25S. richest dietary sources of polyphenols: an application of the Pheno-
2. Scalbert A, Manach C, Morand C, Remesy C, Jimenez L. Dietary l-Explorer database. Eur J Clin Nutr 2010;64 (suppl 3):s112–20.
polyphenols and the prevention of diseases. Crit Rev Food Sci Nutr 24. Bognar A. Tables on weight yield of food and retention factors of food
2005;45:287–306. constituents for the calculation of nutrient composition of cooked foods
3. Neveu V, Pérez-Jiménez J, Vos F, et al. Phenol-Explorer: an online (dishes). Karlsruhe, Germany: Berichte der Bundesforschungsanstalt
comprehensive database on polyphenol contents in foods. Database für Ernährung, 2002. Available from: http://www.eurofir.org/?q=node/9
(Oxford) 2010;2010:bap024. (cited 5 July 2009).
4. Perez-Jimenez J, Neveu V, Vos F, Scalbert A. Systematic analysis of 25. Vinson JA, Hao Y, Su XH, Zubik L. Phenol antioxidant quantity and
the content of 502 polyphenols in 452 foods and beverages: an appli- quality in foods: vegetables. J Agric Food Chem 1998;46:3630–4.
cation of the Phenol-Explorer database. J Agric Food Chem 2010;58: 26. Wu X, Beecher GR, Holden JM, Haytowitz DB, Gebhardt SE, Prior
4959–69. RL. Lipophilic and hydrophilic antioxidant capacities of common
5. Manach C, Williamson G, Morand C, Scalbert A, Remesy C. Bio- foods in the United States. J Agric Food Chem 2004;52:4026–37.
availability and bioefficacy of polyphenols in humans. I. Review of 97 27. Scalbert A. Quantitative methods for the estimation of tannins in
bioavailability studies. Am J Clin Nutr 2005;81:230S–42S. plant-tissues. In: Hemingway RW, Laks PE, eds. Plant polyphenols:
6. Loke WM, Proudfoot JM, Stewart S, et al. Metabolic transformation synthesis, properties, significance. New York, NY: Plenum Press, 1992:
has a profound effect on anti-inflammatory activity of flavonoids such 259–80.
as quercetin: lack of association between antioxidant and lipoxygenase 28. Arai Y, Watanabe S, Kimira M, Shimoi K, Mochizuki R, Kinae N.
inhibitory activity. Biochem Pharmacol 2008;75:1045–53. Dietary intakes of flavonols, flavones and isoflavones by Japanese
7. Lafay S, Morand C, Manach C, Besson C, Scalbert A. Absorption and women and the inverse correlation between quercetin intake and
metabolism of caffeic acid and chlorogenic acid in the small intestine plasma LDL cholesterol concentration. J Nutr 2000;130:2243–50.
of rats. Br J Nutr 2006;96:39–46. 29. Boker LK, Van Der Schouw YT, De Kleijn MJ, Jacques PF, Grobbee
8. Nielsen ILF, Chee WSS, Poulsen L, et al. Bioavailability is improved DE, Peeters PH. Intake of dietary phytoestrogens by Dutch women.
by enzymatic modification of the citrus flavonoid hesperidin in hu- J Nutr 2002;132:1319–28.
mans: a randomized, double-blind, crossover trial. J Nutr 2006;136: 30. Geleijnse JM, Launer LJ, Van Der Kuip DA, Hofman A, Witteman JC.
404–8. Inverse association of tea and flavonoid intakes with incident myo-
9. Erdman JW Jr, Balentine D, Arab L, et al. Flavonoids and heart health: cardial infarction: the Rotterdam Study. Am J Clin Nutr 2002;75:
proceedings of the ILSI North America Flavonoids Workshop, May 880–6.
31-June 1, 2005, Washington, DC. J Nutr 2007;137:718S–37S. 31. Knekt P, Kumpulainen J, Järvinen R, et al. Flavonoid intake and risk of
10. Nurmi T, Mursu J, Peñalvo JL, Poulsen HE, Voutilainen S. Dietary chronic diseases. Am J Clin Nutr 2002;76:560–8.
intake and urinary excretion of lignans in Finnish men. Br J Nutr 2010; 32. Gu L, Kelm MA, Hammerstone JF, et al. Concentrations of proan-
103:677–85. thocyanidins in common foods and estimations of normal consump-
11. Peterson J, Dwyer J, Adlercreutz H, Scalbert A, Jacques P, Mccullough tion. J Nutr 2004;134:613–7.
M. Dietary Lignans: physiology and potential for cardiovascular dis- 33. Wu X, Beecher GR, Holden JM, Haytowitz DB, Gebhardt SE, Prior
ease risk reduction. Nutr Rev 2010;68:571–603. RL. Concentrations of anthocyanins in common foods in the United
12. USDA. Nutrient Data Laboratory. USDA database for the proantho- States and estimation of normal consumption. J Agric Food Chem
cyanidin content of selected foods. 2004. Available from: http://www. 2006;54:4069–75.
ars.usda.gov/nutrientdata (cited on 9 March 2009). 34. Chun OK, Sang JC, Song WO. Estimated dietary flavonoid intake and
13. USDA. Nutrient Data Laboratory. USDA database for the flavonoid major food sources of U.S. adults. J Nutr 2007;137:1244–52.
content of selected foods. Release 2.1. 2007. Available from: http: 35. Dilis V, Vasilopoulou E, Trichopoulou A. The flavone, flavonol and
//www.ars.usda.gov/nutrientdata (cited on 9 March 2009). flavan-3-ol content of the Greek traditional diet. Food Chem 2007;105:
14. USDA. Nutrient Data Laboratory. USDA database for the isoflavone 812–21.
content of selected foods. Release 2.0. 2008. Available from: http: 36. Somerset SM, Johannot L. Dietary flavonoid sources in Australian
//www.ars.usda.gov/nutrientdata (cited on 9 March 2009). adults. Nutr Cancer 2008;60:442–9.
15. Ovaskainen ML, Torronen R, Koponen JM, et al. Dietary intake and 37. Dilis V, Trichopoulou A. Antioxidant intakes and food sources in
major food sources of polyphenols in Finnish adults. J Nutr 2008;138: Greek adults. J Nutr 2010;140:1274–9.
562–6. 38. Zamora-Ros R, Andres-Lacueva C, Lamuela-Raventós RM, et al. Es-
16. Mink PJ, Scrafford CG, Barraj LM, et al. Flavonoid intake and car- timation of dietary sources and flavonoid intake in a Spanish adult
diovascular disease mortality: a prospective study in postmenopausal population (EPIC-Spain). J Am Diet Assoc 2010;110:390–8.
women. Am J Clin Nutr 2007;85:895–909. 39. Zamora-Ros R, Andres-Lacueva C, Lamuela-Raventós RM, et al.
17. Mursu J, Voutilainen S, Nurmi T, Tuomainen TP, Kurl S, Salonen JT. Concentrations of resveratrol and derivatives in foods and estimation of
Flavonoid intake and the risk of ischaemic stroke and CVD mortality in dietary intake in a Spanish population: European Prospective In-
middle-aged Finnish men: the Kuopio Ischaemic Heart Disease Risk vestigation into Cancer and Nutrition (EPIC)-Spain cohort. Br J Nutr
Factor Study. Br J Nutr 2008;100:890–5. 2008;100:188–96.
18. Ward HA, Kuhnle GGC, Mulligan AA, Lentjes MAH, Luben RN, 40. Hollman PA, Cassidy A, Comte B, et al. Antioxidant activity of pol-
Khaw KT. Breast, colorectal, and prostate cancer risk in the European yphenols and cardiovascular health: application of the PASSCLAIM
Prospective Investigation into Cancer and Nutrition-Norfolk in relation criteria. J Nutr (in press).
1228 PÉREZ-JIMÉNEZ ET AL
41. Johannot L, Somerset SM. Age-related variations in flavonoid intake 46. Tarascou I, Souquet JM, Mazauric JP, et al. The hidden face of food
and sources in the Australian population. Public Health Nutr 2006;9: phenolic composition. Arch Biochem Biophys 2010;501:16–22.
1045–54. 47. Zou S, Sinclair J, Wilson MA, et al. Comparative approaches to fa-
42. Bonita JS, Mandarano M, Shuta D, Vinson J. Coffee and cardiovascular cilitate the discovery of prolongevity interventions: effects of toco-
disease: in vitro, cellular, animal, and human studies. Pharmacol Res pherols on lifespan of three invertebrate species. Mech Ageing Dev
2007;55:187–98. 2007;128:222–6.
43. Reddy MB, Hurrell RF, Cook JD. Estimation of nonheme-iron bio- 48. Lemay DG, Zikovic AM, German JB. Building the bridges to bio-
availability from meal composition. Am J Clin Nutr 2000;71:937–43. informatics in nutrition research. Am J Clin Nutr 2007;86:1261–9.
44. Vinson JA, Su X, Zubik L, Bose P. Phenol antioxidant quantity and 49. van Ommen B, Bouwman J, Dragsted LO, et al. Challenges of mo-
quality in foods: Fruits. J Agric Food Chem 2001;49:5315–21. lecular nutrition research 6: the nutritional phenotype database to store,
45. Saura-Calixto F, Serrano J, Goñi I. Intake and bioaccessibility of total share and evaluate nutritional systems biology studies. Genes Nutr
polyphenols in a whole diet. Food Chem 2007;101:492–501. 2010;5:189–203.

Downloaded from https://academic.oup.com/ajcn/article/93/6/1220/4597718 by guest on 03 January 2023