Nutrient–Gene Interactions

Epicatechin and a Cocoa Polyphenolic Extract Modulate Gene Expression

in Human Caco-2 Cells1
Véronique Noé,2 Silvia Peñuelas,3 Rosa M. Lamuela-Raventós,* Joan Permanyer,*
Carlos J. Ciudad, and Maria Izquierdo-Pulido*
Department of Biochemistry and Molecular Biology, School of Pharmacy, University of Barcelona, E-08028
Barcelona, Spain and *Department of Nutrition and Food Science, School of Pharmacy, University of
Barcelona, E-08028 Barcelona, Spain

ABSTRACT We performed a functional genomic analysis to study the effect of epicatechin and polyphenolic
cocoa extract in the human colon adenocarcinoma cell line Caco-2. The specific Human Hematology/Immunology
cDNA arrays by Clontech, containing 406 genes in duplicate, were used. The differentially expressed genes were
classified according to their level of expression, calculated as the ratio of the value obtained after each treatment
relative to control cells, with a statistical significance of P ⬍ 0.05 (upregulated: ratio ⬎ 1.5; downregulated: ratio
⬍ 0.6). Treatment with epicatechin decreased the expression of 21 genes and upregulated 24 genes. Upon

Downloaded from jn.nutrition.org by on September 13, 2010

incubation with the cocoa polyphenolic extract, 24 genes were underexpressed and 28 were overexpressed. The
changes in expression for ferritin heavy polypeptide 1 (FTH1), mitogen-activated protein kinase kinase 1 (MAPKK1),
signal transducer and activator of transcription 1 (STAT1), and topoisomerase 1 upon incubation with epicatechin,
and for myeloid leukemia factor 2 (MLF2), CCAAT/enhancer binding protein gamma (C/EBPG), MAPKK1, ATP-
binding cassette, subfamily c member 1 (MRP1), STAT1, topoisomerase 1, and x-ray repair complementing
defective repair 1 (XRCC1) upon incubation with the cocoa polyphenolic extract were validated by RT-PCR.
Changes in the messenger RNA levels for MAPKK1, STAT1, MRP1, and topoisomerase 1 upon incubation with
either epicatechin or cocoa extract were further confirmed at the protein level by Western blotting. The changes in
the expression of STAT1, MAPKK1, MRP1, and FTH1 genes, which are involved in the cellular response to
oxidative stress, are in agreement with the antioxidant properties of cocoa flavonoids. In addition, the changes in
the expression of C/EBPG, topoisomerase 1, MLF2, and XRCC1 suggest novel mechanisms of action of flavonoids
at the molecular level. J. Nutr. 134: 2509 –2516, 2004.

KEY WORDS: ● cDNA arrays ● cocoa ● epicatechin ● gene expression ● antioxidant

Flavonoids are naturally occurring compounds found in
fruits, vegetables, and nuts with potential health benefits in
reducing the incidence of various chronic illnesses, including
cancer, stroke, and coronary heart disease (1–3). By virtue of
their chemical nature, flavonoids can act as antioxidants and
may help to maintain the body’s natural defenses against a
variety of diseases associated with oxidative stress, such as
inflammation, and proliferative and cardiovascular diseases
(4,5). Waterhouse et al. (6) suggested for the first time that the
intake of cocoa and chocolate could contribute to a large
proportion of the dietary antioxidants. Since then, several
reports have been published demonstrating that cocoa and its
products can be sources of flavonoids (7–9).
The primary flavonoids in cocoa and chocolate are the
flavan-3-ols, epicatechin and catechin (monomeric units),
and polymers of these, the proanthocyanidins, also termed
1
This work was supported by grants FBG-301666 from Nutrexpa S.A., CDTI
(P-02– 0277) and PROFIT (FIT-060000 –2002-99).
2
To whom correspondence should be addressed. E-mail: vnoe@ub.edu.
3
Recipient of a predoctoral fellowship from the Ministerio de Ciencia y
Tecnologı́a, Spain.
procyanidins (9). Depending on the method used in its
production, cocoa powder can contain as much as 10%
flavonoids on a dry-weight basis (5). The apparent bioavail-
abilty of flavonoids in humans ranges from 1 to 26%,
depending on the chemical structure (e.g., quercetin, epi-
catechin, and soy isoflavones), and shows a large interindi-
vidual variability. Epicatechin from chocolate is rapidly
absorbed (10), and the increase in its plasma concentration
is dose dependent. Moreover, in addition to the monomeric
flavonoids catechin and epicatechin, dimeric procyanidins
have been detected in human plasma following consump-
tion of flavonoid-rich cocoa (11).
The only data available about the contribution of cocoa
and chocolate to the total flavonoid intake were reported by
Arts et al. (12) in a Dutch nutritional survey indicating that
cocoa products contribute in 20% of the dietary catechins
intake in this population. According to the food consumption
data of the Spanish Department of Agriculture, Fishery and
Food, in children and teenagers, cocoa products can play an
important role as a source of dietary flavonoids, because cocoa
powder consumption in our country is 6 – 8 kg of cocoa powder
per child per year (13). However, the benefit that this cocoa

0022-3166/04 $8.00 © 2004 American Society for Nutritional Sciences.

Manuscript received 16 April 2004. Initial review completed 24 May 2004. Revision accepted 11 July 2004.

2509
2510
consumption level may represent due to its antioxidant prop-
erties has not yet been determined.
Flavonoids have multiple actions, both in vitro and in vivo.
However, the exact mechanisms of action of the monomeric
flavan-3-ols, oligomeric procyanidins, and their metabolites in
vivo remain to be elucidated (5). In addition to the free-
radical scavenging activity displayed by flavonoids, it is likely
that these compounds could interact with intracellular signal-
ing mechanisms to elicit a variety of biologic effects in the
vascular and the immune systems.
Food components play a role in influencing, directly or
indirectly, the expression of genes encoding proteins involved
in energy metabolism, cell growth, and cell differentiation.
Arrays represent a new powerful technology for high through-
put screening of differential expression and create an exciting
new field for nutrition and health care. According to Muller &
Kersten (14), nutrigenomics attempts to study the genome-
wide influences of nutrition and patterns of gene expression,
protein expression, and metabolite production in response to
particular compounds can be viewed as “dietary signatures.”
Several reports of the effects of food components that result in
marked increases and suppression in the expression of multiple
genes can be found in the literature [reviewed in (15)] In
addition, functional genomics enable the comparison of the
action of food components at the molecular level and provide
tools to generate new hypotheses on their mechanism of
action.
The objective of the present work was to perform a func-
tional genomic analysis to study the effects of epicatechin, the
main cocoa flavonoid, and cocoa powder extract on gene
expression in human cells. We report several genes whose
differential expression provides more information of the anti-
oxidant properties of these compounds and suggests new
mechanisms of action of flavonoids at the molecular level.
MATERIALS AND METHODS
Cocoa powder phenolic extract. Natural Forastero cocoa powder
from Malaysia was used for this study. Phenols were extracted from
10 g of cocoa as described in (16). The final residue was dissolved in
dimethyl sulfoxide (DMSO).4 The total phenolic content was deter-
mined in the extract according to the Folin-Ciocalteu method (17)
and is expressed in grams per liter of catechin equivalents.
Epicatechin. (⫺)-Epicatechin was purchased from Sigma. Stan-
dard solutions were prepared in DMSO at a concentration of 10
mmol/L in dimmed light and were stored at ⫺20°C.
Cell culture. The human colon adenocarcinoma cell line Caco-2
was used throughout the study as an experimental model, because it
serves as a common in vitro model for estimating the fraction ab-
sorbed of compounds via the intestinal tract (18). Cells were grown
in F-12 medium (Gibco) supplemented with 7% (v:v) fetal bovine
serum (Gibco), 100,000 U/L sodium penicillin G, and 0.1 g/L strep-
tomycin, and were maintained at 37°C in a humidified atmosphere
containing 5% CO2.
To investigate the effects of epicatechin and cocoa extract on
gene expression, the concentrations of these compounds used in this
study were pharmacological in magnitude, based on dietary intakes
reported in (14,15). Caco-2 cells (5 ⫻ 106) were incubated for 24 h
4
Abbreviations used: APRT, adenosyl phosphorribosyl transferase; BAT2,
HLA-B-associated transcript 2; C/EBP, CAAT/enhancer-binding protein;
C/EBPG, CCAAT/enhancer binding protein gamma; DMSO, dimethyl sulfoxide;
FTH1, ferritin heavy polypeptide 1; GSH, glutathione; GTP, green tea polyphenol;
IL, interleukin; MAPK, mitogen-activated protein kinase; MAPKK1, MAPK kinase
1; MLF, myeloid leukemia factor; mRNA, messenger RNA; MRP1, ATP-binding
cassette, subfamily c member 1; NPM, nucleophosmin; PARP-1, poly (ADPri-
bose) polymerase; STAT1, signal transducer and activator of transcription 1;
TOP1, DNA topoisomerase 1; XRCC1, x-ray repair complementing defective
repair 1.
NOÉ ET AL.
with either 100 ␮mol/L (29.03 g/L) epicatechin or 0.2155 g/L of
catechin equivalents of cocoa polyphenolic extract, both in DMSO
[final concentration of DMSO in the medium ⬍1% (v:v)]. The
concentration of epicatechin used in the incubations has been shown
not to be toxic for Caco-2 cells or to affect the cell cycle distribution
(19).
cDNA arrays. Caco-2 cells were harvested, and total RNA was
prepared by following the procedure recommended by Clontech. The
integrity of the RNA (2 ␮g) was assessed after agarose gel electro-
phoresis in the presence of formaldehyde. Gene expression was ana-
lyzed by hybridization to specific cDNA arrays (Human Hematology/
Immunology arrays from Clontech). This type of array was used
because it contained genes related with stress response. The hybrid-
ization procedure has been described elsewhere (20). Array data
analyses were performed by using the Atlas Image 2.01 software
(Clontech) and the GeneSpring 6.1 program (Silicon Genetics) (20).
The expression of each gene was reported as the ratio of the value
obtained after each treatment relative to control after the normal-
ization of the data (upregulated: ratio ⬎ 1.5; downregulated: ratio
⬍ 0.6). A cutoff of 1.5, representing a 50% overexpression or under-
expression compared with the control, was chosen, because small
changes in gene expression may represent important changes down-
stream of the differentially expressed genes. Lists of differentially
expressed genes with a P-value ⬍0.05 were generated by using data
Downloaded from jn.nutrition.org by on September 13, 2010
from 3 independent experiments for each condition. The genes in
these lists were further classified according to their function.
Quantitative RT-PCR. Levels of specific messenger RNAs
(mRNA) were assessed by quantitative RT-PCR with [␣-32P]-dATP
to produce a radioactive product that could be detected with great
sensitivity during the exponential phase of the reaction. The same
cDNA preparations were amplified with primers for the different
selected genes’ mRNAs. Adenosyl phosphorribosyl transferase
(APRT) mRNA was used as a control for both the reverse transcrip-
tion and the PCR reactions.
Total RNA was extracted from Caco-2 cells by using the Ultra-
spec RNA reagent (Biotecx) in accordance with the manufacturer’s
instructions. RT-PCR reactions were typically carried out as de-
scribed in Noé et al. (21). The primers used were as follows:
for mitogen-activated protein kinase kinase 1 (MAPKK1): 5⬘-
GCAACTCTCCGTACATCGTG-3⬘ and 5⬘-GCGACATGTAG-
GACCTTGTG-3⬘
for ferritin heavy polypeptide 1 (FTH1): 5⬘-GCGATGATGTGGCTT-
TGAAG-3⬘ and 5⬘-CAAGTTGGTCACGTGGTCAC-3⬘
for CCAAT/enhancer binding protein gamma (C/EBPG): 5⬘-CGCAG-
CAAAACAGCACTCCAG-3⬘ and 5⬘-GTACGTTGTCTGCA-
AGGTTG-3⬘
for ATP-binding cassette, subfamily c member 1 (MRP1): 5⬘-CTG-
GACTGATGACCCCATCG-3⬘ and 5⬘-CACCAATGACGTTG-
AACAGG-3⬘,
for HLA-B-associated transcript 2 (BAT2): 5⬘-CGTCTGGGCCCG-
GACCAAGC-3⬘ and 5⬘-GGCTCTACTAAGCGCATTGG-3⬘
for DNA topoisomerase 1 (TOP1): 5⬘-CAAGCAGCCCGAGGAT-
GATC-3⬘ and 5⬘-GCACTTTTCAGGTCTCTCCG-3⬘
for signal transducer and activator of transcription 1 (STAT1): 5⬘-
CTAGTGGAGTGGAAGCGGAG-3⬘ and 5⬘-CACCAACAGT-
CTCAACTTCAC-3⬘
for myeloid leukemia factor 2 (MLF2): 5⬘-GCTTTGGATATAGC-
CCCTTC-3⬘ and 5⬘-CCAATGGACATCTGCTCCAG-3⬘
for x-ray repair complementing defective repair 1 (XRCC1): 5⬘-CGCT-
GGGGAGCAAGACTATG-3⬘ and 5⬘-CAAATCCAACTTCC-
TCTTCC-3⬘
for APRT: 5⬘-GCAGCTGGTTGAGCAGCGGAT-3⬘ and 5⬘-AGA-
GTGGGGCCTGGCAGCTTC-3⬘
Preliminary experiments were carried out by using different num-
bers of cycles to determine the linear conditions of PCR amplification
for all the genes studied: 18 cycles for FTH1 (generating a fragment
of 338 bp), 20 cycles for TOP1 (333 bp); 22 cycles for C/EBPG (387
bp), BAT2 (325 bp), STAT1 (342 bp), and MLF2 (362 bp); 23 cycles
for APRT (270 bp), 24 cycles for MAPKK1 (333 bp) and MRP1 (316
bp); and 25 cycles for XRCC1 (517 bp). The quantification of the
intensity of the radioactive bands was carried out by phosphorimaging
with the ImageQuant software (Molecular Dynamics). The signals
corresponding to the APRT mRNA were used to normalize the
changes in the levels of mRNA for each particular case.
 GENE EXPRESSION MODULATION BY COCOA FLAVONOIDS
Western blot analyses. Whole extracts were obtained from con-
trol or treated Caco-2 cells, according to Kraus et al. (22). A total of
5 ␮L of the extract was used to determine protein concentration by
the Bradford assay [(23), Bio-Rad]. The extracts were frozen in liquid
N2 and stored at ⫺80°C.
Total extracts, 100 ␮g, were resolved on SDS-polyacrylamide gels
(24) and were transferred to polyvinylidene fluoride membranes (Im-
mobilon P, Millipore) by using a semidry electroblotter. The mem-
branes were probed either with anti-MEK-1 antibody C-18 at 1:50
dilution, anti-STAT1 p84/p91 antibody E-23 at 1:500 dilution, anti-
TOP1 antibody H-300 at 1:500 dilution, or anti-MRP1 antibody
H-70 at 1:100 dilution (all from Santa Cruz Biotechnology). Signals
were detected by secondary horseradish peroxidase-conjugated anti-
body (1:5000 dilution) and enhanced chemiluminiscence, as recom-
mended by the manufacturer (Amersham). Blots were reprobed with
antibodies against ␤-actin (Sigma) at a 1:2000 dilution or Oct-1
(C-21, Santa Cruz Biotechnology) at a 1:100 dilution to normalize
the results.
Statistical methods. Array data analyses were performed through
a parametric comparison by using all available error estimates as a
filter based on the variances estimated by the Cross-Gene Error
Model in GeneSpring 6.1 (Silicon Genetics). The Cross-Gene Error
Model performs a variance components analysis for the accurate
comparison of mean expression levels between experimental condi-
tions (25). In this model, separate estimates of 2 different kinds of
random variation are used to estimate the variability in gene expres-
sion measurements: 1) measurement variation, which comprises the
lowest level of variation, corresponding to the variation of the mea-
surement of a gene on a single chip around the true value that would
be achieved by a perfect measurement of the expression level of the
gene for that sample; and 2) sample-to-sample variation, which is
variation between replicates that represents biological or sampling
variability.
For the RT-PCR and Western blot analyses, values are expressed
as the mean ⫾ SE. Data were evaluated by unpaired Student’s t test,
and analysis was performed by using InStat v.3.0 software (GraphPad
Software). Differences with P-values ⬍ 0.05 were considered signif-
icant.
RESULTS
Differential gene expression analysis by using cDNA ar-
rays. Incubation of Caco-2 cells with epicatechin decreased
the expression of 21 genes and upregulated 24 other genes as
determined by cDNA arrays (Table 1). Upon treatment with
the cocoa polyphenolic extract, 24 genes were underexpressed
and 28 genes were overexpressed (Table 2). The genes that
were differentially expressed due to both treatments were
classified according to their function into several categories:
transcription; oncogenes and tumor suppressors; cell signaling
and extracellular communication proteins; intracellular trans-
ducers, effectors, and modulators; stress response proteins; cell
receptors and cell surface antigens; metabolism; apoptosis-
related proteins; DNA synthesis, recombination, and repair.
Validation of selected targets by RT-PCR. The RNA
data obtained in the screening performed by using the cDNA
arrays were validated by quantitative RT-PCR. The number of
PCR cycles was chosen within the exponential phase to give
reliable values for each cDNA product. In these conditions,
the correlation coefficient between the number of cycles and
the level of APRT mRNA detected, which was used as a
control to normalize the results, was 0.996 (data not shown).
For epicatechin, 3 genes were analyzed from the list of
upregulated genes: BAT2, C/EBPG, and FTH1; and 4 genes
were analyzed from the list of downregulated genes: MAPKK1,
MRP1, STAT1, and TOP1. For the polyphenolic cocoa ex-
tract, 2 genes were analyzed from the list of upregulated genes:
BAT2 and MLF2; and 7 genes were analyzed from the list of
downregulated genes, C/EBPG, FTH1, MAPKK1, MRP1,
STAT1, TOP1, and XRCC1.
2511
In epicatechin-treated cells, mRNA levels for FTH1 were
increased by 54.6% compared with control cells. We also
validated, by quantitative RT-PCR, the underexpression of
MAPKK1 (68.5%), STAT1 (41.4%), and TOP1 (26.4%)
upon incubation with epicatechin (Fig. 1). In cells treated
with cocoa extract, MLF2 mRNA was increased by 337%
compared with the control values. It was also demonstrated
that treatment with a cocoa extract caused a decrease in the
mRNA levels for C/EBPG of 82.1%, MAPKK1 of 36.8%,
MRP1 of 47.3%, STAT1 of 68.3%, TOP1 of 35.7%, and
XRCC1 of 13.8% (Fig. 2). All these changes were significant
(P ⬍ 0.05).
Western blot assays. Western blot analyses were per-
formed to investigate whether the changes at the RNA level
were translated into protein. Treatment with epicatechin de-
creased the protein levels of MAPKK1 by 32.02%, of STAT1
by 25.13%, and of TOP1 by 52.98%, in keeping with the
corresponding mRNA levels (Fig. 3A). Treatment with cocoa
extract decreased the protein levels of MRP1 by 52.35%, of
STAT1 by 14.4%, and of TOP1 by 56.38% in comparison
with the control levels (Fig. 3B). All these changes were
significant (P ⬍ 0.05). Thus, in all cases, the Western blot
Downloaded from jn.nutrition.org by on September 13, 2010
results confirmed that a decreased mRNA expression was
translated into lower levels of the encoded proteins.
DISCUSSION
In this report, we analyzed the gene expression profile of
human cells treated with either epicatechin or a cocoa poly-
phenolic extract. Epicatechin was chosen because it is the
main flavonoid in cocoa, and the purified cocoa polyphenolic
extract was used as representative of the wide flavonoid spec-
trum (mono- and oligomers) present in cocoa. By using cDNA
arrays, we evaluated the differential expression of several genes
involved in the immune response. The changes in mRNA
expression of several outlier genes observed in the arrays were
confirmed by RT-PCR and, at the protein level, by Western
blot analysis. The selection was made according to several
criteria: 1) genes that had already been reported to be modu-
lated by the antioxidant properties of flavonoids, such as
MRP1 and MAPKK1; 2) genes involved in oxidative stress
such as STAT1 and FTH1; 3) transcription factors (C/EBPG)
and enzymes involved in DNA repair and recombination
(XRCC1 and TOP1); and 4) other highly overexpressed genes
in the arrays such as BAT-2 and MLF2. Interestingly, the
differential expression of MRP1, MAPKK1, STAT1, and
FTH1 could provide a molecular explanation for the antioxi-
dant effect of flavonoids, because all of them are involved in
different pathways related to oxidative stress. In addition, the
changes in C/EPBG, TOP1, MLF2, and XRCC1 expression
could provide novel information about the mechanisms of
action of flavonoids.
MRP1 is a membrane protein that acts as a glutathione
(GSH) S-conjugate efflux pump by transporting drugs that are
conjugated with glutathione. Yamane et al. (26) described the
induction of MRP1 at the mRNA level with pro-oxidants and
its downregulation upon overexpression of antioxidant GSH.
The underexpression of MRP1 mRNA and protein upon in-
cubation with cocoa extract reported here could explain, at
the transcriptional level, the inhibition of MRP1-mediated
transport upon incubation with flavonoids observed by other
authors (27,28).
MAPKK1 corresponds to the dual specificity kinase that
phosphorylates extracellular regulatory kinase, ubiquitously
expressed and critically involved in the regulation of signaling
pathways involved in cell proliferation and differentiation. It
2512 NOÉ ET AL.

TABLE 1
Genes differentially expressed in Caco-2 cells upon incubation with epicatechin1

Ratio2 Gene name GenBank3 Functional classification4

Downregulated genes

0.116 MAPKK 1 L05624 Intracellular transducers/effectors

0.214 Interferon regulatory factor 4 U52682 Transcription
0.230 ATP/GTP-binding protein U73524 Oncogenes/tumor suppressors
0.235 Chemokine (C motif) receptor 1 L36149 Cell receptors
0.266 Heat shock 60kDa protein 1 M34664 Stress response proteins
0.270 C6.1A protein X64643 Oncogenes/tumor suppressors
0.323 Fms-related tyrosine kinase 3 U04806 Apoptosis associated proteins
0.339 STAT1 M97935 Transcription
0.390 DEK protein X64229 Oncogenes/tumor suppressors
0.414 v-yes-1 Yamaguchi sarcoma viral related homolog M16038 Intracellular transducers/effectors
0.436 Natural killer cell transcript 4 M59807 Immune system proteins
0.486 TOP1 J03250 DNA synthesis/recombination/repair
0.493 Tumor necrosis factor receptor superfamily member 8 M83554 Apoptosis associated proteins
0.501 Nucleophosmin M23613 Oncogenes/tumor suppressors
0.506 Ras-related C3 botulinum toxin substrate M29870 Oncogenes/tumor suppressors
0.519 Thymidine kinase 1, soluble K02581 Metabolism
0.522 Rho GTPase activating protein 4 X78817 Oncogenes/tumor suppressors

Downloaded from jn.nutrition.org by on September 13, 2010

0.534 IL-2 receptor alpha X01057 Cell receptors
0.535 Macrophage migration inhibitory factor M25639 Cell signaling
0.549 DNA topoisomerase II alpha J04088 Oncogenes/tumor suppressors
0.564 Splicing factor 1 D26120 Transcription

Upregulated genes

1.503 Zinc finger protein D45213 Transcription

1.542 Granulin M75161 Cell signaling
1.590 Transcription factor-7 X59869 Transcription
1.624 CD9 antigen M38690 Cell signaling
1.650 CD19 antigen M21097 Cell surface antigens
1.666 Intercellular adhesion molecule-1 J03132 Cell adhesion receptors/proteins
1.677 Soluble carrier family 20, member 1 L20859 Cell surface antigens
1.686 Activated p21cdc42Hs kinase L05624 Intracellular transducers/effectors
1.771 Integrin alpha L Y00796 Cell adhesion receptors/proteins
1.784 Notch homolog 1 M73980 Oncogenes/tumor suppressors
1.852 Annexin A2 D00017 Trafficking/targeting proteins
1.921 Ewing sarcoma breakpoint region 1 X66899 Oncogenes/tumor suppressors
1.942 IL-10 M57627 Cell signaling
2.022 CD81 antigen M33680 Cell surface antigens
2.133 Ribosomal protein L6 X69391 Transcription
2.264 B-cell translocation gene 1 X61123 Oncogenes/tumor suppressors
2.705 Homeo box D3 D11116 Transcription
2.800 FTH1 M97164 Metabolism
2.888 STAT6 U16031 Transcription
3.254 Activating transcription factor 4 D90209 Transcription
3.466 C/EBPG U20240 Transcription
3.578 Host cell factor C1 L20010 Transcription
5.583 STAT5B U47686 Transcription
15.480 BAT2 M33509 Immune system proteins

1 Genes that were upregulated (ratio ⬎ 1.5) or downregulated (ratio ⬍ 0.6) in epicatechin-treated cells in comparison with control cells, n ⫽ 3. P
⬍ 0.05.
2 Mean fold change in expression of each gene relative to the control.
3 GenBank accession number.
4 Functional categories to which the genes belong.

was reported that an increase in reactive oxygen species pro-
duction results in activation of MAPK pathways [reviewed in
(29)]. Modulation of the MAPKs pathway by flavonoids has
been reported by different authors (30 –32). In addition, stim-
ulation of MAPKs may be the potential signaling pathway
used by green tea polyphenols (GTP) to activate antioxidant-
response element-dependent genes (33,34) and could repre-
sent a mechanism for the protective effect attributed to GTP.
In our conditions, the observed decrease in MAPKK1 mRNA
and protein levels upon epicatechin incubation could be part
of the mechanism to explain the antioxidant properties of this
flavonoid.
Ferritin is a 24-subunit protein composed of 2 types of
subunits, H and L, and plays a central role in the maintenance
of the intracellular iron balance. Inflammatory cytokines, such
as tumor necrosis factor and interleukin-2, upregulate ferritin
synthesis, preferentially the H chains, thus determining an
increase of the catalytic sites and a reduction of cell iron
availability (35). Overexpression of the H subunit leads to a
reduced production of reactive oxygen species after exposure
 GENE EXPRESSION MODULATION BY COCOA FLAVONOIDS 2513

TABLE 2
Genes differentially expressed in Caco-2 cells upon incubation with cocoa extract1

Ratio2 Gene name GenBank3 Functional classification4

Downregulated genes

0.048 Heat shock 60 kDa protein 1 M34664 Stress response proteins

0.056 C/EBPG U20240 Transcription
0.143 STAT1 M97935 Transcription
0.176 Chemokine (C motif) ligand 1 U23772 Cell signaling
0.182 Interferon regulatory factor 4 U52682 Transcription
0.217 Transcription factor 3 M31523 Transcription
0.236 C6.1A protein X64643 Oncogenes/tumor suppressors
0.352 MRP1 L05628 Stress response proteins
0.353 ADP-ribosyltransferase M18112 Apoptosis associated proteins
0.373 NPM M23613 Oncogenes/tumor suppressors
0.378 TOP1 J03250 DNA synthesis/recombination/repair
0.397 Fms-related tyrosine kinase 3 ligand U04806 Apoptosis associated proteins
0.449 FTH1 M97164 Metabolism
0.450 Transcription factor-7 X59869 Transcription
0.454 Breakpoint cluster region X02596 Oncogenes/tumor suppressors
0.478 Thymidine kinase 1, soluble K02581 Metabolism
0.487 GSH S-transferase pi X15480 Stress response proteins

Downloaded from jn.nutrition.org by on September 13, 2010

0.494 XRCC1 M36089 DNA synthesis/recombination/repair
0.524 Ribosomal protein L6 X69391 Transcription
0.524 Rho-GTPase activating protein 4 X78817 Oncogenes/tumor suppressors
0.525 Ras-related C3 botulinum toxin substrate M64595 Oncogenes/tumor suppressors
0.530 DNA topoisomerase II alpha 170 kDa J04088 DNA synthesis/recombination/repair
0.552 CD79A antigen M74721 Cell surface antigens
0.553 MAPKK1 L05624 Intracellular transducers/effectors

Upregulated genes

1.581 POU domain class 3 transcription factor Z11933 Transcription

1.603 Soluble carrier family 20, member 1 L20859 Cell surface antigens
1.730 Granulin M75161 Cell signaling
1.832 Transcription factor-7 X59869 Transcription
1.850 CD81 antigen M33680 Cell surface antigens
1.882 STAT6 U16031 Transcription
1.954 Colony stimulating factor 2 receptor beta M59941 Cell surface antigens
2.197 Integrin alpha L (antigen CD11A) Y00796 Cell adhesion receptors/proteins
2.300 Integrin alpha 2B M34480 Cell surface antigens
2.490 Selectin L M25280 Cell surface antigens
2.538 Intercellular adhesion molecule 1 precursor (CD54) J03132 Cell adhesion receptors/proteins
2.560 B-cell growth factor 1, 12 kDa M15530 Cell signaling
2.720 Pre-B-cell leukemia transcription factor-2 X58842 Transcription
2.771 Activated p21cdc42Hs kinase L13738 Intracellular transducers/effectors
2.818 Nucleoporin 214kDa X64228 Oncogenes/tumor suppressors
2.911 Chemokine (C motif) receptor 1 L36149 Cell receptors
3.071 Host cell factor C1 L20010 Transcription
3.085 MLF2 U57342 Oncogenes/tumor suppressors
3.130 Zinc finger protein D45213 Transcription
3.217 BAT2 M33509 Immune system proteins
3.800 Interleukin-2 receptor alpha X01057 Cell receptors
4.136 Macrophage migration inhibitory factor M25639 Cell signaling
4.523 Homeo box D3 D11117 Transcription
4.619 Basigin L20471 Protein turnover
4.865 IL-10 M57627 Cell signaling
5.500 Interferon regulatory factor 2 X15949 Transcription
8.312 IL-1 receptor-associated kinase L76191 Intracellular transducers/effectors
8.647 V-ets erythroblastosis virus E26 oncogene homolog 1 J04101 Oncogenes/tumor suppressors

1 Genes that were upregulated (ratio ⬎ 1.5) or downregulated (ratio ⬍ 0.6) in cells treated with cocoa extract in comparison with control cells, n
⫽ 3. P ⬍ 0.05.
2 Mean fold change in expression of each gene relative to the control.
3 GenBank accession number.
4 Functional categories to which the genes belong.

to hydrogen peroxide, suggesting a role for ferritin in the reduced iron pool in the cell as a protective mechanism against
protection against oxidative damage (36). In this direction, an oxidant injury.
the overexpression of FTH1 mRNA observed upon epicat- STAT1 belongs to a family of transcription factors that
echin incubation could lead to the reduction of the reactive transduce a signal from a cytokine receptor upon activation by
2514 NOÉ ET AL.
FIGURE 1 Quantitation of FTH1, MAPKK1, STAT1, and TOP1
expression by RT-PCR in epicatechin-treated Caco-2 cells. A rep-
resentative autoradiogram of the PCR products for each mRNA
(panel A) and the quantification of the radioactive bands, expressed
as mRNA levels as percentage of the control (panel B), are shown.
Results represent the means ⫾ SE of 3 different experiments. As-
terisks indicate different from the control: *P ⬍ 0.05 and **P ⬍ 0.01.
FIGURE 2 Quantitation of MLF2, C/EBPG, MAPKK1, MRP1,
STAT1, TOP1, and XRCC1 by RT-PCR in Caco-2 cells treated with
cocoa extract. A representative autoradiogram of the PCR products for
each mRNA (panel A) and the quantification of the radioactive bands,
expressed as mRNA levels as percentage of the control (panel B) are
shown. Results represent the means ⫾ SE of 3 different experiments.
Asterisks indicate different from the control: *P ⬍ 0.05 and **P ⬍ 0.01.
Downloaded from jn.nutrition.org by on September 13, 2010
FIGURE 3 Quantitation of MAPKK-1, STAT1, and TOP1 protein
levels in control and epicatechin-treated Caco-2 cells (panel A) and
quantitation of STAT1, TOP1, and MRP1 protein levels in control
Caco-2 cells and cells treated with a cocoa extract (panel B). Results
represent the means ⫾ SE of 3 different experiments. Asterisks indicate
different from the control: *P ⬍ 0.05 and **P ⬍ 0.01.
phosphorylation of a specific tyrosine residue. STAT1 and
STAT3 are activated by the oxidative stress induced by oxi-
dized LDL, which can be prevented by the addition of vitamin
E or N-acetylcysteine, suggesting that activation of STAT
proteins can be regulated by the cellular redox status (37). The
underexpression of STAT1 observed upon cell incubation
with either epicatechin or cocoa polyphenolic extract could be
due to the antioxidant properties of these compounds and may
play a protective role in inflammation.
CAAT/enhancer-binding proteins (C/EBP) are a family of
leucine zipper transcription factors. The binding activity
and/or mRNA and protein levels of various C/EBPs are dif-
ferentially modulated in response to inflammatory stimuli and
recombinant cytokines. C/EBP-binding motifs have been
identified in the functional regulatory regions of genes encod-
ing the inflammatory cytokines, interleukines (IL) such as
IL-6, IL-1b, and tumor necrosis factor ␣, and other cytokines
such as IL-8 and IL-12 (38,39). The ability of cacao polyphe-
nols to modify the expression of different cytokines involved
in immune responses both at the level of transcription and
 GENE EXPRESSION MODULATION BY COCOA FLAVONOIDS
translation has been reported in several in vitro studies (40 –
43). In addition, cocoa flavonoids may have antiinflammatory
properties in vivo as suggested by Osakabe et al. (44). The
reduced expression of C/EBPG observed in cells treated with
cocoa polyphenols could lead to the underexpression of a
variety of genes containing C/EBP-binding motifs in their
promoter sequences that would reduce the inflammatory re-
sponse.
Human DNA topoisomerase I (TOP1) is a nuclear protein
that plays a key role in DNA replication, transcription, and
recombination. Because many neoplastic cells are character-
ized by high levels of TOP1 activity (45), this enzyme has
become one of the cellular targets for anticancer therapy, and
inhibitors for TOP1, e.g., camptothecin, have been used with
clinical applications (46). The underexpression observed in
TOP1 mRNA and protein levels upon incubation with either
epicatechin or cocoa extract could decrease the final activity
of TOP1 at the transcriptional level.
XRCC1 appears to operate as a molecular scaffold protein
that interacts with and stimulates enzymatic components of
single-strand break repair (47). XRCC1, through its BRCT I
domain, interacts with poly (ADPribose) polymerase
(PARP-1) (48). PARP-1 is required for assembly or stability of
XRCC1 nuclear foci after oxidative DNA damage, in agree-
ment with a model in which poly (ADPribose) synthesis serves
to recruit XRCC1 to sites of DNA strand breakage (48). Both
XRCC1 and PARP-1 are downregulated in Caco-2 cells upon
incubation with cocoa extract, suggesting that the presence of
these antioxidant flavonoids could prevent the oxidative DNA
damage that would trigger PARP-1 and XRCC1 overexpres-
sion.
Little is known about the promyelodysplasia/myeloid leu-
kemia factor 2 (MLF2). The MLF2 gene is highly related to
the previously identified MLF1 (49), an oncogene involved in
acute myeloid leukemia and myelodysplastic syndrome (50).
In myeloid leukemia cells, the translocation t(3;5) t(3;
5)(q25.1;q34) produces chimeric gene transcripts containing
5⬘ sequences encoding nucleophosmin (NPM), fused in-frame
to those of MLF1. MLF1 is normally located in the cytoplasm,
whereas NPM-MLF1 is targeted to the nucleus, indicating that
NPM trafficking signals direct MLF1 to an inappropriate cel-
lular compartment in leukemia cells (51). The final effect of
the overexpression of MLF2 observed upon cocoa extract
incubation remains unknown, although it is worth mentioning
that nucleophosmin mRNA is underexpressed upon incuba-
tion with either epicatechin or cocoa extract in the arrays.
In summary, by using cDNA arrays we determined the
changes in gene expression upon incubation with epicatechin
or cocoa extract. The differential expression of MRP1,
MAPKK1, STAT1, and FTH1 could explain the antioxidant
effect of flavonoids at the molecular level, because these genes
are involved in different oxidative pathways. In addition, the
changes observed in CEPBG, TOP1, MLF2, and XRCC1
expression could suggest novel mechanisms of action for these
compounds. In the future, array analyses could be used as a
screening for the study of the mechanism of action of food
components at the molecular level and could serve as a basis
for diet responsiveness.
LITERATURE CITED
1. Scalbert, A. & Williamson, G. (2000) Dietary intake and bioavaliability
of polyphenols. J. Nutr. 130: 2073S–2085S.
2. Geleijnse, J. M., Launer, L. J., Van der Kuip, D. A., Hofman, A. & Witte-
man, J. C. (2002) Inverse association of tea and flavonoid intakes with incident
myocardial infarction: the Rótterdam Study. Am. J. Clin. Nutr. 75: 880 – 886.
3. Arts, I.C.W., Hollman, P.C.H., Feskens, E.J.M., Bas Bueno de Mesquita,
2515
H. & Kromhout, D. (2001) Catechin intake might explain the inverse relation
between tea consumption and ischemic heart disease: the Zutphen Elderly Study.
Am. J. Clin. Nutr. 74: 227–232.
4. Nijveldt, R. J., van Nood, E., van Hoorn, D. E., Boelens, P. G., van Norren,
K., & van Leeuwen, P. A. (2001) Flavonoids: a review of probable mechanisms
of action and potential applications. Am. J. Clin. Nutr. 74: 418 – 425.
5. Steinberg, F. M., Bearden, M. M. & Keen, C. L. (2003) Cocoa and
chocolate flavonoids: implications for cardiovascular health. J. Am. Diet. Assoc.
103: 215–223.
6. Waterhouse, A. L., Shirley, J. R. & Donovan, J. L. (1996) Antioxidants
in chocolate. Lancet 348: 834.
7. Arts, I.C.W., van de Putte, B. & Hollman, P.C.H. (2000) Catechin
contents of foods commonly consumed in The Netherlands. 1. Fruits, vegetables,
staple foods, and processed foods. J. Agric. Food Chem. 48: 1746 –1751.
8. Hammerstone, J. F., Lazarus, S. A. & Schmitz, H. H. (2000) Procya-
nidin content and variation in some commonly consumed foods. J. Nutr. 130:
2086S–2092S.
9. Lamuela-Raventós, R. M., Andrés-Lacueva, C., Permanyer, J. & Izqui-
erdo-Pulido, M. (2001) More antioxidants in cocoa. J. Nutr. 131: 834.
10. Rein, R., Lotito, S., Holt, R. R., Keen, C. L., Schmitz, H. H. & Fraga, C. G.
(2000) Epicatechin in human plasma: in vivo determination and effect of choc-
olate consumption on plasma antioxidant capacity. J. Nutr. 130: 2109S–2114S.
11. Holt, R. R., Lazarus, S. A., Sullards, M. C., Zhu, Q. Y., Schramm, D. D.,
Hammerstone, J. F., Fraga, C. G., Schmitz, H. H. & Keen, C. L. (2002) Pro-
cyanidin dimer B2 [epicatechin- (4beta-8)-epicatechin] in human plasma after the
consumption of a flavanol-rich cocoa. Am. J. Clin. Nutr. 76: 798 – 804.
12. Arts, I.C.W., Hollman, P.C.H. & Kromhout, D. (1999) Chocolate as a
source of tea flavonoids. Lancet 354: 488.
13. Ministerio de Agricultura, Pesca y Alimentación (2002) La alimentación
Downloaded from jn.nutrition.org by on September 13, 2010
en España 2002. Ministerio de Agricultura, Pesca y Alimentación, Madrid, Spain.
14. Müller, M. & Kersten, S. (2003) Nutrigenomics: goals and strategies.
Nat. Rev. Genet. 4: 315–322.
15. van Ommen, B. & Stierum, R. (2002) Nutrigenomics: exploiting sys-
tems biology in the nutrition and health arena. Curr. Opin. Biotechnol. 13: 517–
521.
16. Andrés-Lacueva, C., Lamuela-Raventós, R. M., Jáuregui, O., Casals, I.,
Izquierdo-Pulido, M. & Permanyer, J. (2000) An LC method for the analysis of
cocoa phenolics. LC-GC Europe 12: 902–905.
17. Swain, T. & Hillis, W. E. (1969) The total phenolic constituents of
Prunus domestica. J. Sci. Food Agric. 46: 4292– 4296.
18. Artursson, P., Palm, K. & Luthman, K., (2001) Caco-2 monolayers in
experimental and theoretical predictions of drug transport. Adv. Drug Deliv. Rev.
46: 27– 43.
19. Salucci, M., Stivala, L. A., Maiani, G., Bugianesi, R. & Vannini, V. (2002)
Flavonoids uptake and their effect on cell cycle of human colon adenocarcinoma
cells (Caco2). Br. J. Cancer 86: 1645–1651.
20. Llaverias, G., Noe, V., Peñuelas, S., Vazquez-Carrera, M., Sanchez, R. M.,
Laguna, J. C., Ciudad, C. J. & Alegret, M. (2004) Atorvastatin reduces CD68,
FABP4, and HBP expression in oxLDL-treated human macrophages. Biochem.
Biophys. Res. Commun. 318: 265–274.
21. Noé, V., MacKenzie, S. & Ciudad, C. J. (2003) An intron is required for
dihydrofolate reductase protein stability. J. Biol. Chem. 278: 38292–33800.
22. Kraus, M. H., Popescu, N. C., Amsbaugh, S. C. & King, C. R. (1987)
Overexpression of the EGF receptor-related proto-oncogene erbB-2 in human
mammary tumor cell lines by different molecular mechanisms. EMBO J. 6: 605–
610.
23. Bradford, M. M (1976) A rapid and sensitive method for the quantita-
tion of microgram quantities of protein utilizing the principle of protein-dye bind-
ing. Anal. Biochem. 72: 248 –254.
24. Laemmli, U. K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227: 680 – 685.
25. Rocke, D. M. & Durbin, B. (2001) (2001) A model for measurement
error for gene expression analysis. J. Comput. Biol. 8: 557–570.
26. Yamane, Y., Furuichi, M., Song, R., Vani, N. T., Mulcahy, T. Ishikawa, T.
& Kuo, M. T. (1998) Expression of multidrug resistance protein/GS-X pump
and g-glutamylcysteine synthetase genes is regulated by oxidative stress. J. Biol.
Chem. 273: 31075–31085.
27. Nguyen, H., Zhang, S. & Morris, M. E. (2003) Effect of flavonoids on
MRP1-mediated transport in Panc-1 cells. J. Pharm. Sci. 92: 250 –257.
28. Wortelboer, H. M., Usta, M., van der Velde, A. E., Boersma, M. G.,
Spenkelink, B., van Zanden, J. J., Rietjens, I. M., van Bladeren, P. J. & Cnubben,
N.H.P. (2003) Interplay between MRP inhibition and metabolism of MRP
inhibitors: the case of curcumin. Chem. Res. Toxicol. 16: 1642–1651.
29. Torres, M. & Forman, H. J. (2003) Redox signaling and the MAP kinase
pathways. Biofactors 17: 287–296.
30. Kobuchi, H., Roy, S., Sen, C. K., Nguyen, H. G. & Packer, L. (1999)
Quercetin inhibits inducible ICAM-1 expression in human endothelial cells
through the JNK pathway. Am. J. Physiol. 277: 403– 411.
31. Kong, A.-N.T., Yu, R., Chen, C., Mandlekar, S. & Primiano, T. (2000)
Signal transduction events elicited by natural products: role of MAPK and
caspase pathways in homeostatic response and induction of apoptosis. Arch.
Pharm. Res. 23: 1–16.
32. Briviba, K., Pan, L. & Rechkemmer, G. (2002) Red wine polyphenols
inhibit the growth of colon carcinoma cells and modulate the activation pattern of
mitogen-activated protein kinases. J. Nutr. 132: 2814 –2818.
33. Yu, R., Jiao, J. J., Duh, J. L., Gudehithlu, K., Tan, T. H. & Kong, A. N.
2516 NOÉ ET AL.
(1997) Activation of mitogen-activated protein kinases by green tea polyphe-
nols: potential signaling pathways in the regulation of antioxidant-responsive
element-mediated phase II enzyme gene expression. Carcinogenesis 18: 451–
456.
34. Chen, C., Yu, R., Owuor, E. D. & Kong, A. N. (2000) Activation of
antioxidant-response element (ARE), mitogen-activated protein kinases (MAPKs)
and caspases by major green tea polyphenol components during cell survival and
death. Arch. Pharm. Res. 6: 605– 612.
35. Miller, L. L., Miller, S. C., Torti, S. V., Tsuji, Y. & Torti, F. M. (1991)
Iron-independent induction of ferritin H chain by tumor necrosis factor. Proc. Natl.
Acad. Sci. U.S.A. 88: 4946 – 4950.
36. Orino, L. K., Lehman, O., Tsuji, Y., Ayaki, H., Torti, S. V. & Torti, F. M.
(2001) Ferritin and the response to oxidative stress. Biochem. J. 357: 241–247.
37. Maziere, C., Alimardani, G., Dantin, F., Dubois, F., Conte, M. & Maziere,
J. (1999) Oxidized LDL activates STAT1 and STAT3 transcription factors:
possible involvement of reactive oxygen species. FEBS (Feb. Eur. Biochem. Soc.)
Lett. 448: 49 –52.
38. Shirakawa, F., Saito, K., Bonagura, C. A., Galson, D. L., Fenton, M. J.,
Webb, A. C. & Auron, P. E. (1993) The human prointerleukin 1 beta gene
requires DNA sequences both proximal and distal to the transcription start site for
tissue-specific induction. Mol. Cell. Biol. 13: 1332–1344.
39. Wedel, A., Sukski, G. & Ziegler-Heitbrock, H. W. (1996) CCAAT/
enhancer binding protein is involved in the expression of the tumour necrosis
factor gene in human monocytes. Cytokine 8: 335–341.
40. Mao, T. K., Van de Water, J., Keen, C. L., Schmitz, H. H. & Gershwin,
M. E. (2002) Effect of cocoa flavanols and their related oligomers on the
secretion of interleukin-5 in peripheral blood mononuclear cells. J. Med. Food 5:
17–22.
41. Schramm, D. D., Wang, J. F., Holt, R. R., Ensunsa, J. L., Gonsalves, J. L.,
Lazarus, S. A., Schmitz, H. H., German, J. B. & Keen, C. L. (2001) Chocolate
procyanidins decrease the leukotriene-prostacyclin ratio in humans and human
aortic endothelial cells. Am. J. Clin. Nutr. 73: 36 – 40.
42. Mao, T. K., Van de Water, J., Keen, C. L., Schmitz, H. H. & Gershwin,
M. E. (2000) Cocoa procyanidins and human cytokine transcription and se-
cretion. J. Clin. Nutr. 130: 2093S–2099S.
43. Sangboni, C., Suzuki, N. & Sakane, T. (1997) Polyphenols in choco-
late, which have antioxidant activity, modulate immune functions in humans in
vitro. Cell Immunol. 177: 129 –136.
44. Osakabe, N., Sanbongi, C., Yamagishi, M., Takizawa, T. & Osawa, T.
(1998) Effects of polyphenol substances derived from Theobroma cacao on
gastric mucosal lesion induced by ethanol. Biosci. Biotechnol. Biochem. 62:
1535–1538.
45. Husain, I., Mohler, J. L., Seigler, H. F. & Besterman, J. M. (1994)
Elevation of topoisomerase I messenger RNA, protein, and catalytic activity in
human tumors: demonstration of tumor-type specificity and implications for
cancer chemotherapy. Cancer Res. 54: 539 –546.
46. Takimoto, C. H., Wright, J. & Arbuck, S. G. (1998) Clinical applications
of the camptothecins. Biochim. Biophys. Acta 1400: 107–119.
47. Caldecott, K. W. (2001) Mammalian DNA single-strand break repair:
an x-ra(y)ted affair. Bioessays. 23: 447– 455.
48. Masson, M., Niedergang, C., Schreiber, V., Muller, S., Ménissier-de Mur-
cia, J. & de Murcia, G. (1998) XRCC1 is specifically associated with poly(ADP-
ribose) polymerase and negatively regulates its activity following DNA damage.
Mol. Cell. Biol. 18: 3563–3571.
49. Kuefer, M. U., Look, A. T., Williams, D. C., Valentine, V., Naeve, C. W.,
Behm, F. G., Mullersman, J. E., Yoneda-Kato, N., Montgomery, K., et al. (1996)
cDNA cloning, tissue distribution, and chromosomal localization of myelodyspla-
sia/myeloid leukemia factor 2 (MLF2). Genomics 35: 392–396.
50. Yoneda-Kato, N., Look, A. T., Kirstein, M. N., Valentine, M. B., Raimondi,
S. C., Cohen, K. J., Carroll, A. J. & Morris, S. W. (1996) The t (3;5)(q25.1;q34)
of myelodysplastic syndrome and acute myeloid leukemia produces a novel
Downloaded from jn.nutrition.org by on September 13, 2010
fusion gene, NPM-MLF1. Oncogene 12: 265–275.
51. Yoneda-Kato, N., Fukuhara, S. & Kato, J. Y. (1999) Apoptosis induced
by the myelodysplastic syndrome-associated NPM-MLF1 chimeric protein. On-
cogene 18: 3716 –3724.