HAL Id: hal-01906543
https://hal.archives-ouvertes.fr/hal-01906543
Submitted on 26 May 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
Nutrimetabolomics: An Integrative Action for
Metabolomic Analyses in Human Nutritional Studies
M. M. Ulaszewska, C. H. Weinert, A. Trimigno, R. Portmann, C. Andres
Lacueva, R. Badertscher, L. Brennan, C. Brunius, A. Bub, F. Capozzi, et al.
To cite this version:
M. M. Ulaszewska, C. H. Weinert, A. Trimigno, R. Portmann, C. Andres Lacueva, et al.. Nu-
trimetabolomics: An Integrative Action for Metabolomic Analyses in Human Nutritional Studies.
Molecular Nutrition and Food Research, Wiley-VCH Verlag, 2019, 63 (1), �10.1002/mnfr.201800384�.
�hal-01906543�
R EVI E W Open Access
Food intake biomarkers for apple, pear, and
stone fruit
Marynka Ulaszewska
1†
, Natalia Vázquez-Manjarrez
2,3†
, Mar Garcia-Aloy
4,5
, Rafael Llorach
4,5
, Fulvio Mattivi
1,6
,
Lars O. Dragsted
3
, Giulia Praticò
3
and Claudine Manach
2*
Abstract
Fruit is a key component of a healthy diet. However, it is still not clear whether some classes of fruit may be more
beneficial than others and whether all individuals whatever their age, gender, health status, genotype, or gut
microbiota composition respond in the same way to fruit consumption. Such questions require further
observational and intervention studies in which the intake of a specific fruit can be precisely assessed at the
population and individual levels. Within the Food Biomarker Alliance Project (FoodBAll Project) under the Joint
Programming Initiative “A Healthy Diet for a Healthy Life”, an ambitious action was undertaken aiming at reviewing
existent literature in a systematic way to identify validated and promising biomarkers of intake for all major food
groups, including fruits. This paper belongs to a series of reviews following the same BFIRev protocol and is
focusing on biomarkers of pome and stone fruit intake. Selected candidate biomarkers extracted from the literature
search went through a validation process specifically developed for food intake biomarkers.
Keywords: Apple, Pear, Quince, Pome fruit, Stone fruit, Cherry, Plum, Prune, Apricot, Peach, Nectarine, Biomarkers, Intake
Background
Introduction
Fruit is an essential component of a healthy diet. In a
comparative risk assessment of global disease burden at-
tributable to 67 risk factors, diets low in fruit were esti-
mated to account for 30% of ischemic heart disease
disability-adjusted life years worldwide and ranked
among the five leading risk factors and as the first diet-
ary factor for global disease burden and mortality [1].
Large prospective cohort studies, increasingly supported
by well-designed randomized clinical trials, have conclu-
sively established the protective effects of high fruit in-
take regarding hypertension, cardiovascular disease, and
stroke, with some evidence of a dose-response relation-
ship [2–5]. High intake of fruit and vegetables (F&V)
have also been associated with prevention of other
chronic diseases such as several cancer types, obesity
and type 2 diabetes, or neurodegenerative disea ses, how-
ever, mixed results were reported, and the overall
evidence is more limited [6–9]. There is a strong need
for more research in the field to answer important pend-
ing questions and guide the development of more effi-
cient public health policies and healthy food production.
One major interrogation is whether the total quantity of
fruit consumed is the most important factor or whether
the intake of particular fruits or groups of fruit, or a high
diversity, matters . Some fruits exp ected to provide im-
portant amount of specific bioactive compounds in the
human diet, such as pomegranate, orange, or cranberry,
have received much interest in the last decades. Due to
the diversity of fruit composition regarding bioactives, it
is important to evaluate the specific health effects of the
individual fruits, to identify their protective constituents
and biological targets and eventually to determine the
most beneficial associations. This is particularly relevant
for the cancer-protective or the anti-obesity effects of
fruit, which were observed to differ for various types of
fruit [6, 10]. Another question concerns the
inter-individual variability in response to fruit consump-
tion. It is not clear whether everyone, regardless of age,
gender, lifestyle, gut microbiota composition, or geno-
type, responds similarly to fruit consumption and if
there is a risk associated with high fruit intake for some
* Correspondence: claudine.manach@inra.fr
†
Marynka Ulaszewska and Natalia Vazquez-Manjarrez contributed equally to
this work.
2
Human Nutrition Unit, Université Clermont Auvergne, INRA, F63000
Clermont-Ferrand, France
Full list of author information is available at the end of the article
© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.
Ulaszewska et al. Genes & Nutrition (2018) 13:29
https://doi.org/10.1186/s12263-018-0620-8
individuals. All these questions require further investiga-
tions in human intervention and large prospective stud-
ies in which an accurat e assessment of fruit intake can
be made for every subject, not only of total intake of
fruit but also for the intake of several classes of fruit and
individual fruits.
Fruit intake is traditionally assessed with dietary ques-
tionnaires. The usual consumpt ion of total F&V is som e-
times the only information inquired in Food Frequency
Questionnaires (FFQs), and even in the most detailed
FFQs classes of fruit (such as pome, citrus, drupes, ber-
ries , nuts….) may be distinguished, but rarely individual
species (apples, pears, oranges, grapefruits, etc.) and al-
most never down to the variety. Repeated 24-h recalls
are more precise but still biased by self-reporting in-
accuracy. The consumption of fruit and vegetables has
been shown to be particularly prone to overestimation
in dietary questionnaires, at least for a fraction of the
population [11]. Complementary to questionnaires, bio-
markers such as plasma vitamin C and plasma caroten-
oids have been widely used [5, 12]. However, as shown
in a systematic review and meta-analysis of 19 interven-
tion studies, these bioma rkers can reflect group-level dif-
ferences for assessing compliance to F&V interventions,
but are not accurate enough to precisely reflect
individual-level intakes [13]. These traditional bio-
markers give heterogeneous responses depending on the
type of F&V consumed and are affected by a range of in-
trinsic and environmental factors.
The Food Biomarker Alliance (FoodBAll), a project
funded by the Joint Programming Initiative, “A Healthy
Diet for a Healthy Life” (http://www.healthydietfor
healthylife.eu/), has undertaken a systematic evaluation
of traditional and newly discovered biomarkers of food
intake (BFIs). Guidelines were established for conducting
a systematic literature search dedicated to food intake
biomarkers [14] and for evaluating their level of valid-
ation using a set of consensus criteria [15]. The guide-
lines were applied for more than 140 foods from all
major food groups: fru it and vege tables, meats, fish and
other marine foods , dairy products, cereals and whole-
grains, alc oholic and non-alcoholic beverages, vegetable
oils, nuts, and spices and herbs (http: //foodmetabolom
e.org/wp3). The present article presents the results of
the in-depth exploration of possible biomarkers of intake
for important classes of fruit, the pome and stone fruit.
Methods
Selection of food groups
The most widely consumed pome and stone fruit were
inventoried [16]. For pome fruit, the apple (Malus
domestica Borkh.) and pear (Pyrus communis L.) were
selected, as well as quince (Cydonia oblonga Miller),
which is less frequently consumed as jams, marmalade,
jellies, or “pâte de fruit.” For stone fruit, sweet cherry (Pru-
nus avium L.), sour cherry (Prunus cerasus L.), plum and
prune (Prunus domestica L.), apricot (Prunus armeniaca
L.), peach (Prunus persica (L.) Batsch), and nectarine (Pru-
nus persica var. nucipersica (Borkh.) C.K.Schneid.) were
covered. Botanical genus and generic fruit group names
were also used in the search, as described below, to ensure
that no other important pome or stone fruits were missed.
Search for relevant BFI research papers
An extensive literature search was carried out to collect
all available information on the existing and new candi-
date BFIs for the selected fruits. The BFIRev protocol
(Food Intake Biomarker Reviews) described previously
was followed [14]. Briefly, a primary search was per-
formed in three databases, Scopus, Pub Med central, and
Web of Science with the name of the specific fruit and
its botanical genus, i.e., (pear OR pyrus*), (apple* OR
Malus domestica), (quince OR Cydonia oblonga), and
(plum OR peach OR nectarine OR cherry OR apricot
OR prunus OR drupe* OR stone fruit) along with the
common keywords: AND (urine OR plasma OR serum
OR excretion OR blood) AND (human* OR men OR
women OR patient* OR volunteer* OR participant*)
AND (biomarker* OR marker* OR metabolite* OR bio-
kinetics OR biotransformation OR pharmacokinetics OR
bioavailability OR ADME) AND (intake OR meal OR
diet OR ingestion OR administration OR consumption
OR eating OR drink*). Keywords were used in the fields
[Topic], [All fields], and [Article Title/Abstract/Key-
words] for Web of Science, PubMed, and Scopus, re-
spectively. All searches were carried out in March 2016
and updated in May 2017. Only papers in English lan-
guage were considered eligible, and no restriction on the
date of publication was applied. Articles showing results
of human intervention studies (randomized controlled
trials, acute, short-term or long-term studies) or obser-
vational studies (cohort, case-control, cross-sectional
studies) were considered eligible. After duplicate re-
moval, a first selection of pape rs was performed accord-
ing to relevance of abstract and title. Full texts were
obtained for the selected articles and further assessed for
eligibility according to their relevance in determining
BFIs for pome and stone fruit. Some of the publications
found in the reference list of the selected articles were
also included at this stage.
The process of selection of the articles identifying or
using potential biomarkers of intake is outlined in Fig. 1.
Identification and characterization of candidate BFIs
For each potential biomarker identified, a secondary
search allowed to retrieve relevant information to assess
the quality of the individual biomarkers, regarding their
specificity, their pharmacokinetics , dose-response
Ulaszewska et al. Genes & Nutrition (2018) 13:29 Page 2 of 16
relationship, the robustness, and reliability of their
method of analysis, in order to qualify their use as BFIs
according to the validating scheme established by
Dragsted et al. [15].
The name of the potential biomarkers and their syno-
nyms were queried in the previously mentioned data-
bases along with AND (biomarker* OR marker* OR
metabolite* OR biokinetics OR biotransformation OR
pharmacokinetics OR bioavailability OR ADME). Add-
itionally, the compounds were searched manually in the
online databases, HMDB (https://www.hmdb.ca), FooDB
(http://foodb.ca/), Phenol-Explorer (http://phenol-ex
plorer.eu/), Dictionary of Food Compounds (http://
dfc.chemnetbase.com/faces/chemical/ChemicalSearch.x h
tml), Duke’s phytochemical and ethnobotanical data-
bases (https://phytochem.nal.usda.gov/phytochem/sea
rch), eBASIS (http://ebasis.eurofir.org/Default.asp),
Knapsack (http://kanaya.naist.jp/knapsack_jsp/top.html),
and PhytoHub (http://phytohub.eu) to determine all the
possible dietary or metabolic origins of the candidate BFIs.
Specific and non-specific biomarkers were discussed in
the text, while only the most plausible candidate BFIs
for pome fruit have been reported in Table 1 along with
the information related to study designs and analytical
methods. The non-retained compounds for pome fruit
and stone fruit are listed in Tables 2 and 3, respectively,
along with the main reasons for exclusion and with the
corresponding references for an exhaustive presentation
of the results. The tables have been reviewed and agreed
upon by all authors and no additional markers were
suggested.
Application of validation criteria
According to Dragsted et al., a set of validation criteria
was applied to the candidate BFIs reported in Table 1,in
order to assess their current status of validation and to
identify the missing information for a full validation of
each of them [15]. The validation scheme is ba sed on
eight questions related to the analytical and biological
aspects: Q1: Is the marker compound plausible as a spe-
cific BFI for the food or food group (chemical/biological
plausibility)? Q2: Is there a dose-response relationship at
relevant intake levels of the targeted food (quantitative
aspect)? Q3: Is the biomarker kinetics described ad-
equately to make a wise choice of sample type, fre-
quency, and time window (time-response)? Q4: Has the
marker been shown to be robust after intake of complex
meals reflecting dietary habits of the targeted population
(robustness)? Q5: Has the marker been shown to com-
pare well with other markers or questionnaire data for
the same food/food group (reliability)? Q6: Is the marker
chemically and biologically stable during biospecimen
collection and storage, making measurements reliable
and feasible (stability)? Q7: Are analytical variability
(CV%), accuracy, sensitivity, and specificity known as ad-
equate for at least one reported analytical method
Fig. 1 Flow diagram of study selection according to the BFIRev procedure
Ulaszewska et al. Genes & Nutrition (2018) 13:29 Page 3 of 16
Table 1 List of studies reporting candidate biomarkers for pome fruit consumption
Dietary factor Study design Number of
subjects
Analytical
method
Biofluid Discriminating metabolites/
candidate biomarkers
Primary
reference(s)
Apple
Apple
(24-h recalls,
58.7 ± 113.5 g/day
+
)
Observational study 53
(31 women, 22 men)
HPLC-MS Urine
(spot and 24 h)
• Phloretin
• Othe r non-specific
polyphenol
metabolites
[40]
Apple
(FFQ and food diaries,
47 (3 –140)g/day
*
)
Observational study 119 (all women) HPLC-ESI-MS Urine
(24 h)
Plasma
• Phloretin
• Othe r non-specific
compound
[42]
Apple (24-h recall,
61 (0 –317) g/day
*
)
Observational study:
5 months of free
access to fruit
basket in
working place.
79 HPLC-ESI-MS Urine
(24 h)
• Phloretin
• Othe r polyphenols
for other fruits
[41]
Apple and pear
(24-h recalls,
228 ± 239 g/day
+
)
Observational study 481 Untargeted
HPLC-TOF-MS
Urine
(24 h)
• Phloretin
glucuronide
• Othe r non-specific
epicatechin
metabolites
[17]
40 g of lyophilized
apples:
polyphenol-rich
vs polyphenol-poor
apples
Double-blind,
randomized
cross-over trial,
4-week periods
30 (all men) LC-MS Morning spot urine • Phloretin [47]
25 g of unripe
apple processed
in powder
Randomized
cross-over study
(two 1-day
interventions: (1)
50 g OGTT and (2)
50 g OGTT+ 25 g
apple powder)
6 (all women) LC-MS Urine
(0 h, 0-2 h, 2–4h)
• Phloretin
• Phloretin
glucuronide
[45]
1 L cloudy apple juice Kinetics
intervention,
single dose
11
(healthy ileostomy
subjects)
HPLC-DAD;
HPLC-ESI-MS/MS
Ileostomy fluid
(0 h, 1 h, 2 h, 4 h,
6h,and 8h)
• Phloretin
2′-O-xyloglucoside
• Phloretin
2′-O-glucuronide
• Phloretin
• Othe r non-specific
polyphenol
metabolites
[46]
0.7 L of apple
smoothie
Single dose,
kinetic study
10
(healthy
ileostomy
persons)
HPLC-DAD and
HPLC-MS/MS
Ileostomy fluid
(0 h, 1 h, 2 h, 4 h,
6h,and 8h)
• Phloretin 2′-O-
xylogluco
side
• Phloretin 2′-O-
glucuronide
(and isomers)
• Phloretin
• Othe r non-specific
polyphenol
metabolites
[95]
1 kg of apple
(organic vs
conventional)
Randomized,
cross-over
single-dose study
(2 experimental
days, 1 per
intervention)
6 (all men) HPLC-MS Plasma
(0 h, 1 h, 2 h, 3 h,
4h,5h,6h,9h,
12 h, 24 h)
• Phloretin [43]
500 g/day apple for
4 weeks (organic
vs conventional
vs no apple)
Double-blind,
randomized
parallel study (3
interventions: (1)
organic apple, (2)
conventional apple,
(3) control)
43 (all men) HPLC-MS Plasma
(day 0 and day
28 24 h after last
intake of fruit)
• Phloretin [43]
Apple
(low: 200 ± 10 g,
medium: 400 ± 10 g
and high: 790 ± 10 g
consumption)
Acute parallel study
with three groups
30
(14 women,
16 men)
HPLC-ESI-QTrap Urine
(0 h, 0–2h,2–4h,
4–6h,6–8h,
8–10 h, 10–12 h,
12–14 h, 14–16 h,
48 h morning spot,
72 h morning spot,
• Phloretin
• Othe r non-specific
polyphenol
metabolites
[48]
Ulaszewska et al. Genes & Nutrition (2018) 13:29 Page 4 of 16
