An antiinflammatory dietary mix modulates inflammation and oxidative

and metabolic stress in overweight men: a nutrigenomics approach14

Gertruud CM Bakker, Marjan J van Erk, Linette Pellis, Suzan Wopereis, Carina M Rubingh, Nicole HP Cnubben,
Teake Kooistra, Ben van Ommen, and Henk FJ Hendriks

INTRODUCTION

The prevalence of overweight is increasing globally and has

become a serious public health problem. Overweight increases
the risk of chronic metabolic diseases such as type 2 diabetes and
cardiovascular disease (1, 2). Low-grade inflammatory status in
overweight persons has been proposed as one of the mediating
processes in metabolic disease development, such as cardiovascular diseases (CVDs) (3, 4) and diabetes (5). Several studies
support a link between oxidative stress and inflammation in
atherogenesis (6, 7).
Adipose tissue is crucial for the inflammatory status associated
with obesity, primarily because of macrophage infiltration (8).
Adipocytes secrete both pro- and antiinflammatory adipokines,

1044

including proinflammatory tumor necrosis factor-a (TNF-a),

interleukin-6 (IL-6), and the antiinflammatory adiponectin (9
11). Reduced adiponectin (12, 13) and increased C-reactive
protein (CRP) (5, 14) concentrations are associated with CVD
and type 2 diabetes. The established inflammatory marker CRP
originates from the liver.
A reduction in inflammatory status may prevent the occurrence
of disorders and diseases related to overweight. Many food
compounds have been reported to have antiinflammatory and/or
antioxidant effects in various populations (15). The Mediterranean diet contains several of these compounds and (16) has been
associated with a reduction of CVD and type 2 diabetes (17).
Expressed in nutrients, this means high contents of antioxidant
polyphenols, vitamins, long-chain unsaturated fatty acids, and
carotenoids.
In the present study we aimed to investigate the antiinflammatory effects induced by nutritional intervention in overweight men with mildly increased CRP concentrations.
Therefore, we selected potentially effective compounds aiming to
cover a wide range of actions in the reduction of inflammation.
We selected polyphenols from green tea (epigallocatechin gallate) and red wine (resveratrol), which reportedly have inhibitory
effects on nuclear transcription factor jB (NF-jB) and activator
protein 1 (AP-1) (18). Resveratrol and omega-3 (n23) fatty
acids are known for their inhibitory effects on cyclooxygenase 2.
Antiinflammatory effects of n23 fatty acids have been shown by
reduced plasma concentrations of CRP, TNF-a, and IL-6 (19).
Similarly, vitamin E intakes reduced plasma CRP and IL-6
concentrations in diabetic subjects (20). In addition, vitamin E
as well as tomato juice increased the lag time of LDL oxidation
(20). Hence, the antiinflammatory dietary mix (AIDM) consisted of fish oil, green tea extract, resveratrol, vitamin E, vitamin C, and tomato extract.
The effects of these dietary compounds were studied by using
a nutrigenomics approach by large-scale profiling of genes,
1

From TNO Quality of Life, Business Unit Biosciences, Zeist (GCMB,

MJvE, LP, SW, CMR, NHPC, BvO, and HFJH) and Leiden (TK), Netherlands.
2
GCMB, MJvE, and LP contributed equally to this work.
3
Supported by TNO Quality of Life. TNO is a Dutch acronym for Netherlands Organization for Applied Research.
4
Address correspondence to GCM Bakker, TNO Quality of Life, PO Box
360, 3700 AJ Zeist, Netherlands. E-mail: gertruud.bakker@tno.nl.
Received October 16, 2009. Accepted for publication January 22, 2010.
First published online February 24, 2010; doi: 10.3945/ajcn.2009.28822.

Am J Clin Nutr 2010;91:104459. Printed in USA. 2010 American Society for Nutrition
Supplemental Material can be found at:
http://ajcn.nutrition.org/content/suppl/2010/03/19/ajcn.2009.
28822.DC1.html

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

ABSTRACT
Background: Low-grade chronic inflammation in overweight subjects is thought to play an important role in disease development.
Objective: It was hypothesized that specific dietary components are
able to reduce low-grade inflammation as well as metabolic and
oxidative stress.
Design: Dietary products [resveratrol, green tea extract, a-tocopherol, vitamin C, n23 (omega-3) polyunsaturated fatty acids, and
tomato extract] selected for their evidence-based antiinflammatory
properties were combined and given as supplements to 36 healthy
overweight men with mildly elevated plasma C-reactive protein
concentrations in a double-blind, placebo-controlled, crossover
study with treatment periods of 5 wk. Inflammatory and oxidative
stress defense markers were quantified in plasma and urine. Furthermore, 120 plasma proteins, 274 plasma metabolites (lipids, free
fatty acids, and polar compounds), and the transcriptomes of peripheral blood mononuclear cells and adipose tissue were quantified.
Results: Plasma adiponectin concentrations increased by 7%,
whereas C-reactive protein (principal inflammation marker) was
unchanged. However, a multitude of subtle changes were detected
by an integrated analysis of the omics data, which indicated modulated inflammation of adipose tissue, improved endothelial function, affected oxidative stress, and increased liver fatty acid
oxidation.
Conclusion: An intervention with selected dietary products affected
inflammatory processes, oxidative stress, and metabolism in humans, as shown by large-scale profiling of genes, proteins, and
metabolites in plasma, urine, and adipose tissue. This trial was
registered at clinical trials.gov as NCT00655798.
Am J Clin
Nutr 2010;91:104459.

1045

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

proteins, and metabolites in blood, urine, and fat tissue. We report

the effects on markers of inflammation, oxidation, and metabolism by integrating the results from a multiplatform approach.
SUBJECTS AND METHODS

Study design

Subjects
Male subjects were selected from the pool of volunteers of
TNO Quality of Life (Zeist, Netherlands) and recruited by
advertisements and flyers. The men selected were generally
healthy with a body mass index (BMI; in kg/m2) between 25.5
and 35.0 and a low-grade inflammation determined on the basis
of a CRP concentration of 110 mg/L. Subjects were excluded if
they had a fasting blood glucose concentration of .6.9 mmol/L,
fasting cholesterol concentration of .8 mmol/L, blood hemoglobin concentration of ,8 mmol/L, or high blood pressure [age
,55 y: diastolic blood pressure (DBP) .100 mm Hg or systolic
blood pressure (SBP) .160 mm Hg; age 5559 y: DBP .90
mm Hg or SBP .140 mm Hg]. Those with acute inflammation,
as assessed on the basis of white blood cell count, were excluded. Furthermore, subjects with chronic diseases related to
inflammation, using medication against blood clotting, with
a medical history that might affect the study outcome, or using
antiinflammatory medication were excluded. Additional exclusion criteria were as follows: food allergy, smoking, unexplained
weight loss, consumption of .28 units of alcohol/wk, following
a slimming diet, use of food supplements, and use of probioticcontaining products. Forty-three men were eligible and 42 were
included, of whom 6 were appointed as a reserve and remained
in the study until day 34. The baseline characteristics of the 36
men who completed the study are given in Table 1.
Study products
The study products included AIDM capsules, 2 yogurts
containing different probiotic strains, and a placebo treatment.
This article focuses on the effects of AIDM; the effects of the
yogurt treatments will be described in a separate paper (MJ van

Treatment

Mean

Minimum

Maximum

Age (y)
BMI (kg/m2)
Diastolic blood pressure (mm Hg)
Systolic blood pressure (mm Hg)
Waist-hip ratio
Glucose (mmol/L)
Cholesterol, total (mmol/L)
HDL cholesterol (mmol/L)
LDL cholesterol (mmol/L)
Total:HDL cholesterol
Triglycerides (mmol/L)
Insulin (mU/L)
CRP (lg/L)
Fibrinogen (mg/dL)

46
29.7
84
128
0.97
5.9
5.9
1.24
3.8
4.8
1.9
10.8
2.5
297.1

22
25.6
64
112
0.89
5.2
3.4
0.91
1.9
3.0
0.4
3.3
1.0
205.7

58
34.7
101
155
1.08
6.7
8.0
1.78
5.6
6.5
3.8
30.6
8.1
385.8

Data are included only for subjects (n = 36) who completed the study.
CRP, C-reactive protein.

Erk, S Wopereis, GCM Bakker, et al, unpublished observations,

2010).
The subjects consumed 2 hard capsules and 2 soft capsules
daily at breakfast with 200 mL plain yogurt (Chr Hansen,
Hrsholm, Denmark) and at the evening meal. All doses were
based on reported effects on inflammatory markers in humans
and on tolerable levels (Dutch Health Council) and generally
exceeded the regular recommended daily nutritional dose.
Capsules consisted of a hard gelatin capsule (Microz Food
Supplements, Geleen, Netherlands) containing 6.3 mg resveratrol
(21) (daily dose equivalent to 4 L red wine), tomato extract
containing 3.75 mg lycopene (20) (daily dose equivalent to 500
mL tomato juice), 94.5 mg green tea extract (40% epigallocatechin gallate; daily dose equivalent to 300 mL green tea)
(22, 23), 90.7 mg a-tocopherol (6) [Recommended Daily Intake
(RDI): 18 IU; upper tolerable level: 1000 IU], and 125 mg vitamin C (upper tolerable level: 2000 mg/d) and a soft gel capsule
containing 1200 mg cold water fish oil composed of 380 mg
eicosapentaenoic acid (EPA), 260 mg docosahexaenoic acid
(DHA) (24), and 60 mg other n23 polyunsaturated fatty acids
(Omega-3 700; Solgar Vitamin and Herb, Leonia, NJ) (RDI:
200 mg/d). Placebo hard and soft gel capsules contained 365 mg
microcrystalline cellulose (Microz Food Supplements) and 1360
mg soy lecithin (soya lecithin; Solgar Vitamin and Herb),
respectively.
Compliance was 99%, as assessed by evaluation of completed
intake diaries and returned capsules. Compliance was supported
by increased plasma concentrations of vitamin E after treatment
with AIDM (mean increase: 83%; range: 42134%).

Study procedures
The subjects were requested to maintain their habitual lifestyle
and diet throughout the study, the use of nonsteroidal antiinflammatory agents was not allowed, and paracetamol use was
limited. Furthermore, the subjects were not allowed to consume
alcohol or to perform additional physical exercise on the last 2 d
at the end of each treatment period. Study products provision and
compliance checks were on a weekly basis.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

The study was a randomized, double-blind, placebo-controlled, crossover trial with 4 treatment periods of 5 wk each.
Group size was based on power calculations by using a power of
80% and a 2-sided test with an a of 0.05 on the variables CRP
and adiponectin. To show a difference of 1.4 mg/L for CRP, we
needed a sample size of 32 subjects, which would enable detection of a difference of 0.4 mg/L for adiponectin. The study
started with 36 subjects, allowing for some dropout.
The study was performed according to the International
Conference on Harmonization of Technical Requirements for
registration of Pharmaceuticals for Human use; guidelines for
Good Clinical Practice; the Helsinki Declaration of 1975, as
revised in 2000; and the Dutch Regulations on Medical Research
involving Human Subjects (WMO, 1999 as revised in 2000). The
study protocol had been approved by the independent Medical
Ethics Committee METOPP located in Tilburg, the Netherlands.
The study was conducted between December 2006 and June
2007.

TABLE 1
Demographic characteristics and laboratory values of subjects at inclusion1

1046

BAKKER ET AL

The subjects were instructed to fast overnight(from 2200)before

each blood collection. On the last day of each treatment period,
a fat-load test was performed. On the day before this test, the
subjects started to collect their urine for a period of 24 h. On the next
morning, the subjects consumed a standardized light breakfast
(1597 kJ, 8.6% of energy as protein, 17.0% of energy as fat, and
72.4% of energy as carbohydrates) provided by TNO. After 4 h,
they received a fat-load test consisting of a 500-mL milkshake
containing high-fat dairy products supplying 46.1 g fat, of which
27.1 g was saturated fat (total energy content: 2945 kJ per serving). Within 6 h after consumption of the milkshake, 7 blood
samples were collected at 0, 30, 60, 120, 180, 240, and 360 min.
After dinner, a biopsy sample was obtained by a physician, who
took subcutaneous fat from the abdomen for gene profiling.
Adverse events and medication use

Blood and urine samples

EDTA-blood was used for the white blood cell count, differentiation, and CRP analysis. Sodium-citrate blood was used for the
fibrinogen analyses and stored for additional analyses later on.
At week 4 of each treatment period, peripheral blood
mononuclear cells (PBMCs) were isolated from EDTA-blood by
using Leucosep tubes containing Ficoll Paque (Greiner, Alphen
a/d Rijn, Netherlands) diluted with saline. Samples were centrifuged for 30 min at 800 g at room temperature and PBMCs
were collected, washed 2 times with saline, and divided over 4
aliquots, each containing 1 107 cells. Samples were stored
for inhibitor-jB (I-jB), NF-jB, and AP-1 analyses and for
isolation of RNA for gene profiling.
At week 5 of each treatment period, serum and EDTA-blood was
collected for clinical chemistry tests and the measurement of a range
of inflammatory and antioxidant markers, respectively. Serum
clinical chemistry tests included the measurement of c-glutamyltransferase, alanine aminotransferase, aspartate aminotransferase,
alkaline phosphatase, glucose, insulin, total bilirubin, albumin,
creatinine, urea, total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, CRP, and fibrinogen were analyzed by using
immunoturbidimetric techniques. Total cholesterol, HDL cholesterol, and triglycerides were measured with enzymatic techniques
(Boehringer-Mannheim, Mannheim, Germany) on an Olympus
AU400 clinical chemistry analyzer (Olympus-Diagnostica Europe,
Hamburg, Germany). LDL cholesterol was calculated according to
Friedewald et al (25). Hematology variables were measured by
using routine clinical techniques with an ADVIA 120 (Siemens
Health Care Diagnostics, Deerfield, IL).
Inflammatory markers consisted of TNF-a, IL-6, IL-10, IL-12,
IL-1b, IL-2, IL-4, IL-5, IL-7, IL-8, IL-13, interferon-c, and
granulocyte-macrophage colony stimulating factor, all measured

Lipidomics and metabolomics

The LC-MS methods used for the measurement of plasma
lipids and FFAs and the GC-MS method used for the measurement of a broad range of metabolites were identical to the
methods reported by Wopereis et al (27). The samples were
analyzed in randomized order. For GC-MS, a total of 530 plasma
samples were analyzed in 19 different batches. It was ensured that
samples for one subject were analyzed within the same batch.
Data for each subject were corrected for the recovery of the
internal standard for injection. The performance of the applied
metabolic profiling platforms was assessed through the frequent
analysis of the quality control sample, and method performance
was carefully monitored by using multiple internal standards (5
10 depending on the method, including analogs and 2H- and 13Clabeled metabolites), as described previously (28). Batches were
only accepted if the relative SD (RSD) of the peak area ratio for all
internal standards was ,20%. Metabolites were only accepted if
the RSD was ,20%, unless large differences between treatment
groups were observed. Finally, the LC-MS FFA data set contained
21 metabolites: the LC-MS lipids data set consisted of 108 metabolites, and the GC-MS data set consisted of 145 metabolites
(see Supplementary Table 2 under Supplemental data in the
online issue). Also, 29 different metabolite ratios and sums were
calculated (see Supplementary Table 3 under Supplemental
data in the online issue). Metabolites were annotated by using an
in-house metabolite database containing retention time information, MS spectra (electron impact ionization for GC-MS
data), MS/MS spectra (LC-MS), and accurate mass data (LC-MS)
of reference substances. The confidence of identification was
100%, unless indicated otherwise. Accurate MS and MS/MS data
of reference substances and metabolites in the study samples were
acquired by using Thermo LTQ-FT and Thermo LTQ-Orbitrap
instruments (Thermo Fisher Scientific, Waltham, MA).
Proteomics
Plasma samples were sent to RulesBasedMedicine Inc (Austin,
TX) for measurement of expression levels of 124 proteins

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Adverse events and medication use were classified according

to ICD-10. No serious adverse events (AEs) occurred. In total, 80
and 100 AEs were reported by 28 of the 36 subjects when
consuming the control mix and AIDM, respectively. The categories, intensity, and number of AEs were similar for both
treatments. The predominant AEs were diarrhea and change in
bowel habits. The most commonly used medications were overthe-counter drugs, of which antiinflammatory drugs, such as
products based on acetylsalicylic acid, were used incidentally.

in plasma. The analyses were performed by using multiplex kits

(high-sensitivity human cytokine premixed Lincoplex kit; Linco
Research, St Charles, MO). In addition, plasma IL-17 and total
adiponectin were analyzed by using enzyme-linked immunosorbent assay kits (Quantikine kit; R&D systems, Minneapolis,
MN). The antioxidant markers 3-nitrotyrosine and troloxequivalent antioxidant activity (TEAC) were measured as described previously (26). Indomethacin was added before plasma
separation for prostaglandin E2 analyses (27).
Blood samples for liquid chromatographymass spectrometry
(LC-MS) analysis of lipids and free fatty acids (FFAs) were
collected after an overnight fast at the end of the 5-wk dietary
intervention. Blood samples for the analysis of GC-MS and
protein profiling were taken before and after the high fat load at
regular intervals (0, 30, 60, 120, 180, 240, and 360 min.). These
blood samples were collected into lithium-heparin tubes, and
plasma was separated within 30 min at 4C and stored at 270C
until analyzed. Aliquots of 24-h urine sample were used for the
measurement of creatinine, 8-isoprostaglandin F2-a, and 8-hydroxy-2-deoxyguanosine (26).

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

(HumanMAP). Data were available for 33 subjects. The so-called

80% rule (29) was applied to retain only those proteins for which

80% of the values were above the detection limit for
1 of the
2 treatment groups, resulting in retention of 79 of the 124 variables. Values below the detection limit that remained in the
truncated data set were replaced by a value of half of the detection limit. Values for remaining samples that were not measurable on the standard curve for a specific protein were set at
0.1 times the detection limit for that protein.

1047

biological interpretation and network construction if the P value

was ,0.05 together with .20% up-regulation in
10 subjects
or if the P value was ,0.05 together with .20% down-regulation in
10 subjects. Genomatix software tools (Genomatix
GmbH, Munich, Germany) were used for transcription factor
and biological network analysis. The microarray data have been
submitted to ArrayExpress (http://www.ebi.ac.uk/microarray-as/
ae/; accession number E-TABM-871).
Statistical analysis

Transcriptomics

xm 
ym 2 zm=
ym 3 100

where Xm is the mean treatment effect for metabolite m in the

difference between AIDM and placebo (%), Ym is the mean
intensity for metabolite m for placebo treatment, and Zm is the
mean intensity for metabolite m for AIDM treatment. Gene
expression data analysis is described in the paragraph above.
Biological interpretation
Functional analysis of the data were performed by using Ingenuity Pathway Analysis version 8.0 (Ingenuity Systems Inc,
Redwood City, CA). This tool allows for simultaneous analysis of
clinical chemistry, gene, protein, and metabolite data. To prevent
domination of enrichment analysis by the gene expression data
set, the analysis was performed both with and without the gene
expression data.
Further information on the function and biological role of the
variables was derived from MetaCore version 5.2 (GeneGo, St
Joseph, MI), EntrezGene (http://www.ncbi.nlm.nih.gov/sites/
entrez?db=gene), nutritional metabolomics database for metabolites (http://wiki.nugo.org/index.php/Nutritional_Metabolomics_
Database) and for proteins (http://wiki.nugo.org/index.php/

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Total RNA was extracted from PBMCs and adipose tissue

samples by using RNAzol (Campro Scientific, Veenendaal,
Netherlands) and glass beads according to the manufacturers
instructions. RNA from adipose tissue samples was subjected to
a clean-up step with the use of NucleoSpin columns (MachereyNagel, Germany).
The isolated RNA samples were sent to ServiceXS BV
(Leiden, Netherlands) where they were processed according to
Affymetrix protocols. In brief, RNA concentrations were measured by absorbency at 260 nm, and quality and integrity were
verified with the RNA 6000 Nano assay on the Agilent 2100
Bioanalyzer (Agilent Technologies, Santa Clara, CA). In total,
134 RNA samples from PBMCs and 77 RNA samples from
adipose tissue were of sufficient quality and subjected to array
analysis.
The labeled cRNA, generated as described by Radonjic et al
(30), was further used for the hybridization to the NuGO
Affymetrix Human Genechip NuGO_Hs1a520180 (30). After an
automated process of washing and staining, absolute values of
expression were calculated from the scanned array by using
Affymetrix GCOS software.
Quality control and normalization of microarray data were
performed by using R/BioConductor packages through the
NuGO pipeline that is available as a Genepattern procedure at
http://nbx2.nugo.org (31). One array from PBMC samples and 4
arrays from the adipose tissue samples did not pass quality
control. As a result, data for 22 subjects were included in the
intraindividual comparison of AIDM compared with placebo
effects in adipose tissue and data for 32 subjects were included in
the intraindividual comparison of AIDM compared with placebo
effects in PBMCs.
Raw signal intensities were normalized by using the GCRMA
algorithm. The custom CDF file for NuGO_Hs1a520180 (based
on Entrez Gene, version 10.0.0; http://nugo-r.bioinformatics.nl/
NuGO_R.html) was used to re-annotate the probes to new probe
sets, remove poor quality probes, and derive unique signal values
for different probe sets representing the same gene (32). This
resulted in gene expression values for 16,242 genes with unique
identifiers.
Transcriptome data sets for PBMCs and adipose tissue were
analyzed separately. Genes were filtered on expression value .5
in
10 samples, resulting in a set of 10,812 genes for PBMCs
and in a set of 11,283 genes for adipose tissue. Expression data
were log transformed (base 2). Statistical analysis was performed in BRB ArrayTools (developed by Richard Simon and
Amy Peng Lam; http://linus.nci.nih.gov/BRB-ArrayTools.html).
In addition to a threshold for P values, a threshold was applied
to fold change induced by intervention. Genes were used for

All data were analyzed for treatment differences by using

analysis of variance (ANOVA). Data were first analyzed for
a carryover effect of treatments between periods. Because this
appeared not to be relevant, the ANOVA was simplified by not
including this effect in further analyses. If the ANOVA indicated
an overall treatment effect, comparisons between treatment
means of the variables after AIDM and placebo intervention were
performed by using a 2-sided (paired) Students t test. Data were
log transformed if necessary.
For LC-MS lipids and FFA data, a Tukey-Kramer adjustment
was used to correct for multiple comparisons between treatments.
The time curve of GC-MS and protein profiling data were analyzed on treatment differences by using a repeated-measures
ANOVA. If a significant interaction between time and treatment
was found, comparisons between treatment means of the variables were performed per time point by using a 2-sided (paired)
Students t test. If no significant interaction was found, main
treatment effects averaged over time were investigated. Bonferroni adjustment was used to correct for multiple testing for
GC-MS and protein profiling data.
In all statistical tests performed, the null hypothesis (no effect)
was rejected at the 0.05 level of probability. The SAS statistical
software package (versions 8.2 and 9.1; SAS Institute Inc, Cary,
NC) was used for univariate statistical analysis.
For each significant metabolite, the mean treatment effect
(expressed as change, in %) was calculated as follows:

1048

BAKKER ET AL

Category:Protein), Human Metabolome database (http://www.

hmdb.ca/), and PubMed (http://www.ncbi.nlm.nih.gov/sites/
entrez?db=PubMed&itool=toolbar).
RESULTS

This study aimed to evaluate the effects of an AIDM in overweight men with mildly elevated CRP concentrations. The effects
were characterized by measurement of established classic inflammation markers, such as CRP and adiponectin in combination
with profiling of gene expression, proteins, and metabolites.
Subjects maintained their habitual lifestyle, which was supported
by unchanged body weight (99.9 6 10.6 kg for the placebo and
99.8 610.5 kg for the AIDM treatment) and blood pressure (SBP:
124 612 mm Hg for the placebo and 123 612 mm Hg for the
AIDM treatment) after both treatment periods compared with
their corresponding inclusion values (Table 1).
Overview of changes induced by AIDM

TABLE 2
Overview of number of detected and significantly affected variables for
each measurement platform1
Number of variables

Metabolites, lipids platform

Metabolites, free fatty acids platform
Metabolites, GC-MS platform
Metabolites, ratios or sums
Proteins, multiplex analysis
Genes, transcriptome adipose tissue
Genes, transcriptome PBMCs
Clinical chemistry

Detected

Significantly
changed2

108
21
145
29
79
11,283
10,812
47

71
4
60
16
18
195
217
8

1
GC-MS, gas chromatographymass spectrometry; PBMCs, peripheral
blood mononuclear cells.
2
Changed as a result of the antiinflammatory dietary mix (P , 0.05,
ANOVA and/or paired Students t test as described in Subjects and Methods), including Tukey-Kramer adjustment for lipids and free fatty acid data
(n = 36 subjects), Bonferroni adjustment for GC-MS (n = 36 subjects) and
protein data (n = 33 subjects), and a threshold of a 20% increase or decrease in
10 subjects for transcriptome data (n = 32 subjects for PBMCs,
n = 22 subjects for adipose tissue). Clinical chemistry data were available
for 36 subjects.

Inflammation-related processes
The altered metabolites, proteins, and genes associated with
inflammation, with mean percentage change in response to the
AIDM, are shown in Table 3. The metabolites, proteins, and
genes are grouped in subprocesses, which are discussed below.
Eicosanoid-related inflammation
Metabolic profiling showed that the omega-3omega-6 eicosanoid precursor ratio in plasma (EPA compared with AA; for
details, see Supplementary Table 3 under Supplemental data
in the online issue) was reduced, which indicated that the
amounts of antiinflammatory eicosanoid precursors increased
relative to the amounts of proinflammatory eicosanoid precursor
(Table 3). In addition, many genes involved in prostaglandin
metabolism were up-regulated in adipose tissue, namely PTGIS
(prostaglandin I2 synthase), PGDS (prostaglandin D2 synthase),
HPGD (hydroxyprostaglandin dehydrogenase), and PTGFRN
(prostaglandin F2 receptor negative regulator).
Inflammatory mediators and signaling molecules
The plasma adiponectin concentration increased and the
plasma prolactin concentration decreased in response to the
AIDM. The antiinflammatory genes IL10RA (interleukin 10
receptor alpha) and SOCS3 (suppressor of cytokine signaling 3)
were up-regulated in adipose tissue in response to the AIDM.
The plasma concentrations of the proinflammatory cytokine
IL18 and chemokine (C-C motif) ligand 22 (CCL22), which may
play a role in the trafficking of activated T lymphocytes to inflammatory sites, decreased. Also, the concentration of b2-microglobulin protein and the b-chain of major histocompatibility
complex class I molecules decreased in plasma. Expression of
the inflammatory gene IGHD (immunoglobulin heavy constant
delta) increased in PBMCs in response to AIDM intervention,
whereas gene expression of MYOM1 (myomesin 1), IGHV3-47
(immunoglobulin heavy variable 347), and IL12A (interleukin
12A) decreased.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

CRP and adiponectin were the main classic inflammation

markers measured in this study. CRP concentrations did not
change in response to the AIDM, whereas adiponectin concentrations increased significantly from 6.03 6 2.06 mg/L after
placebo intervention to 6.48 6 2.57 mg/L after AIDM intervention (P , 0.05). The effects of the AIDM were investigated in more detail by large-scale analysis of gene
expression, proteins, and metabolites in blood, urine, and adipose tissue biopsy samples. An overview of the numbers of
measured variables that responded to the AIDM are shown in
Table 2. The supplementary tables show a detailed overview of
all measured variables and the changes induced by the AIDM
(see Supplementary Table 1 under Supplemental data in the
online issue: multiplex analysis of proteins in plasma; see
Supplementary Table 2 under Supplemental data in the online

issue: analysis of metabolites in plasma; see Supplementary

Table 3 under Supplemental data in the online issue: plasma
metabolite ratios and sums; see Supplementary Table 4 under
Supplemental data in the online issue: significantly changed
gene expressions in peripheral blood mononuclear cells; see
Supplementary Table 5 under Supplemental data in the online
issue: significantly changed gene expressions in adipose tissue
biopsies; see Supplementary Table 6 under Supplemental data in
the online issue: clinical chemistry variables in plasma and urine).
The complete set of significantly changed variables (Table 2)
was subjected to functional enrichment analysis. The top 10
enriched biofunctions, shown in Figure 1A, confirm an effect of
the AIDM on inflammation (inflammatory response, antigen
presentation). Furthermore, the AIDM resulted in responses related to lipid metabolism and metabolic disease. The highest
scoring network in the analysis without the gene expression data
(Figure 1B) illustrates the effects of the AIDM on inflammation
(immune response), oxidative stress (production of reactive oxygen species), and lipid metabolism (quantity of lipid). The network indicates a central role for NF-jB in the effects of AIDM. A
more detailed biological interpretation of the data focuses on these
3 processes: inflammation, oxidation, and metabolism.

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

1049

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

FIGURE 1. A: Top 10 enriched functions within a combined set of changed clinical chemistry markers, proteins, and metabolites. Shown is the number of changed
variables in each biofunction in this analysis (white bars) and in the analysis including gene expression data (gray bars). Corresponding P values for the biofunctions
ranged from 7.9 10212 to 2.5 1023 (Ingenuity Pathway Analysis version 8.0; Ingenuity Systems Inc, Redwood City, CA). B: Top scoring network based on the
functions and/or diseases that are most significant to the network objects (Ingenuity Pathway Analysis) in the analysis of the combined set of changed clinical chemistry
markers, proteins, and metabolites. The biofunctions immune response, production of reactive oxygen species, and quantity of lipid are overlaid onto the network,
showing which variables (nodes) are directly involved in these processes. Color coding of the nodes corresponds to the direction of responses (up-regulation shown
in red and down-regulation shown in green). ADIPOQ, adiponectin; AFP, a-fetoprotein; APOA1, apolipoprotein A-I; APOC3, apolipoprotein C-III; CCL22, C-C
motif chemokine ligand 22; CSF1, colony stimulating factor 1; F7, coagulation factor VII; FTH1, ferritin heavy polypeptide 1; ICAM1, intercellular adhesion
molecule 1; IL1, interleukin-1; IL18, interleukin-18; IL12, interleukin-12; MPO, myeloperoxidase; TNFRSF1B, tumor necrosis factor receptor superfamily member
1B; VCAM1, vascular cell adhesion molecule 1; ifngamma, interferon c; MHC Class II, major histocompatibility class II.

Plaque formation and coagulation

Plasma factor VII and TNF RII concentrations decreased,
which are related to plaque formation and coagulation. In
PBMCs, gene expression of VEGFB (growth factor for endothelial cells) increased, whereas that of PEAR1 (platelet endo-

thelial aggregation receptor 1) decreased. Also, in adipose

tissue, the expression of genes involved in plaque formation and
coagulation changed. Expression of several platelet-related
genes were down-regulated, including PF4 (platelet factor 4),
NAPG (N-ethylmaleimide-sensitive factor attachment protein

1050

BAKKER ET AL

TABLE 3
Significantly altered metabolites, proteins, and genes in response to the
antiinflammatory dietary mix (AIDM) associated with inflammation1
Name

Tissue2

Change3

Supplementary
table4

TABLE 3 (Continued )

Name

Tissue2

Change3

Supplementary
table4

Platform

1.84
1.91
4.100
4.149
4.146
5.157
5.115
5.105
5.158
5.186
5.190
5.166
5.183
5.55

Pr
Pr
T
T
T
T
T
T
T
T
T
T
T
T

Platform
%

Pl

276

3.15

Pl

425

3.17

Pl
Pl
AT
AT
AT
AT

351
28
36
30
19
21

2.270
2.11
5.4
5.8
5.41
5.30

M
M
T
T
T
T

6
7.4
27
21
26
36
30
19
21
19
24
41
24
29
27
24
210
22
210
27
29

1.3
6.1
1.99
5.26
5.12
5.4
5.8
5.41
5.30
5.42
5.17
5.3
2.88
2.143
1.66
1.85
4.165
4.5
4.163
1.13
4.146

Pr
C
Pr
T
T
T
T
T
T
T
T
T
M
M
Pr
Pr
T
T
T
Pr
T

23
25
8
216
19
26
239
11
13
214
223
12
246

1.35
1.119
4.119
4.204
5.42
5.10
5.185
5.95
5.75
5.104
5.149
5.89
5.191

Pr
Pr
T
T
T
T
T
T
T
T
T
T
T

23
24
17
26
33

1.51
1.122
5.49
5.10
5.5

Pr
Pr
T
T
T

Pl
Pl
Pl
AT
AT
AT
AT
AT
AT
AT
AT
AT
Pl
Pl
Pl
Pl
PBMCs
PBMCs
PBMCs
Pl
PBMCs

Pl
Pl
PBMCs
PBMCs
AT
AT
AT
AT
AT
AT
AT
AT
AT
Pl
Pl
AT
AT
AT

(Continued)

Pl
Pl
PBMCs
PBMCs
PBMCs
AT
AT
AT
AT
AT
AT
AT
AT
AT

29
24
10
29
29
226
216
215
226
242
245
227
235
17

1
AT, adipose tissue; PBMCs, peripheral blood mononuclear cells; Pl,
plasma; M, metabolomics; Pr, proteomics; T, transcriptomics; C, clinical
data; PUFAs, polyunsaturated fatty acids; IL-18, interleukin-18; TNF RII,
tumor necrosis factor receptor type 2; ICAM-1, intercellular adhesion molecule 1; VCAM-1, vascular cell adhesion molecule 1; CSF-1, macrophage
colony stimulating factor 1; CCL-22, C-C motif chemokine 22 or monocyte
chemotactic protein 1. All variables were significantly changed by the AIDM
intervention (P , 0.05, ANOVA and/or paired Students t test as described
in Subjects and Methods), including Tukey-Kramer adjustment for lipids and
free fatty acid data (n = 36 subjects), Bonferroni adjustment for gas chromatographymass spectrometry (n = 36 subjects) and protein data (n = 33
subjects), and a threshold of a 20% increase or decrease in
10 subjects for
transcriptome data (n = 32 subjects for PBMCs, n = 22 subjects for adipose
tissue). Clinical chemistry data were available for 36 subjects.
2
Tissue refers to tissues in which a variable was measured.
3
Mean percentage change by AIDM.
4
Refers to supplementary tables; first digit corresponds to the supplementary table (1: proteins; 2: metabolites; 3: metabolite ratios or sums; 4: PBMC
genes; 5: adipose tissue genes; 6: clinical data), and the second number (after the
decimal point) corresponds to the row number in the supplementary table.

gamma), and PPBP (pro-platelet basic protein). Gene expression of a specific endothelial cell receptor (THBD), which ultimately reduces the amount of thrombin generated, was
increased, as was the gene expression of PROCR (protein C
receptor-endothelial). PROCR, PF4, and THBD are all part of
the protein C pathway controlling blood coagulation.
Endothelial function
Plasma concentrations of intercellular adhesion molecule 1
(ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) decreased. The gene expression of SMOC2 (SPARC related modular
calcium binding 2), which is able to stimulate endothelial cell
proliferation, increased in adipose tissue. Gene expression of
ANGPTL5, which belongs to the angiopoietin familyan important growth factor specific for vascular endotheliumalso
increased in adipose tissue.
Blood cell differentiation
Two plasma proteins, 4 PBMC genes and 9 adipose tissue
genes related to blood cell differentiation, responded to AIDM

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Eicosanoid-related
inflammation
Omega-3/omega-6
eicosanoid
precursor ratio
Balance omega-3/
omega-6 PUFAs
Eicosapentaenoic acid
Arachidonic acid
PTGIS
PGDS
HPGD
PTGFRN
Inflammatory mediators
and signaling
molecules
Adiponectin
Adiponectin
Prolactin
DPP4
TLR7
PTGIS
PGDS
HPGD
PTGFRN
SELP
IL10RA
SOCS3
Inosine
Uric acid
IL-18
CCL-22
MYOM1
IGHD
IGHV3-47
b2-Microglobulin
IL12A
Plaque formation and
coagulation
Factor VII
TNF RII
VEGFB
PEAR1
SELP
THBD
PF4
F10
PROCR
NAPG
FGF10
EDG2
PPBP
Endothelial function
ICAM-1
VCAM-1
ANGPTL5
THBD
SMOC2

Blood cell
differentiation
CSF-1
Myeloperoxidase
IL4R
CCL21
IL12A
IL8RB
IL7R
KLRG1
TESC
HBB
HBM
NKTR
ALAS2
IL15

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

intervention. The gene expression of CCL21, which inhibits

hemopoiesis and stimulates chemotaxis, decreased in PBMCs.
Gene expression of TESC (tescalcin), a key factor in megakaryocytic differentiation, was down-regulated in adipose tissue.
IL-15 is a cytokine that stimulates the proliferation of T lymphocytes, and its expression was up-regulated in adipose tissue.
IL4R gene expression was up-regulated in PBMC. IL-4R can
bind IL-4 to promote differentiation of T helper 2 cells and
regulate immunoglobulin E production. Genes such as IL12A,
IL7R, KLRG1 (killer cell lectin-like receptor subfamily G
member 1), and NKTR (natural killer triggering receptor, all
involved in activating T lymphocytes and natural killer cells,
were down-regulated.
Oxidative stressrelated processes

TABLE 4
Significantly altered metabolites, proteins, and genes in response to the
antiinflammatory dietary mix (AIDM) associated with oxidative stress1
Name

Tissue2 Change3

Myeloperoxidase
2,3,4-Trihydroxybutanoic
acid
8-Isoprostaglandin F2-a
Indole-3-propionic acid
Uric acid
Vitamin E
Ferritin
LACTB
HEPH
NDUFS1
SLC25A27

Pl
Pl

%
24
24

U
29.8
Pl
28
Pl
29
Pl
83
Pl
211
PBMCs
12
AT
19
AT
222
AT
216

Supplementary
table4
Platform
1.91
2.108

Pr
M

6.31
2.58
2.34
2.145
1.37
4.55
5.39
5.145
5.120

C
M
M
M
Pr
T
T
T
T

1
AT, adipose tissue; PBMCs, peripheral blood mononuclear cells; Pl,
plasma; M, metabolomics; Pr, proteomics; T, transcriptomics; C, clinical
data; U, urine. All variables were significantly changed by the AIDM intervention (P , 0.05, ANOVA and/or paired Students t test as described in
Subjects and Methods), including Tukey-Kramer adjustment for lipids and
free fatty acid data (n = 36 subjects), Bonferroni adjustment for gas chromatographymass spectrometry (n = 36 subjects) and protein data (n = 33
subjects), and a threshold of a 20% increase or decrease in
10 subjects for
transcriptome data (n = 32 subjects for PBMCs, n = 22 subjects for adipose
tissue). Clinical chemistry data were available for 36 subjects.
2
Tissue refers to tissues in which variables were measured.
3
Mean percentage change by AIDM.
4
Refers to supplementary tables; first digit corresponds to the supplementary table (1: proteins; 2: metabolites; 3: metabolite ratios or sums; 4: PBMC
genes; 5: adipose tissue genes; 6: clinical data), and the second number (after the
decimal point) corresponds to the row number in the supplementary table.

[NADH dehydrogenase (ubiquinone) Fe-S protein 1] and

SLC25A27 (solute carrier family 25 member 27, also known as
UCP4), were down-regulated in adipose tissue in response to
AIDM. NDUFS1 is the core subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase, and it is a ubiquinone iron-sulfur complex. SLC25A27 (UCP4), a member of
mitochondrial uncoupling proteins, separates oxidative phosphorylation from ATP synthesis, whereby energy is dissipated as heat.
Metabolism-related processes
The AIDM had a profound effect on metabolism, specifically
on lipoprotein metabolism. Concentrations of total cholesterol,
LDL cholesterol, and HDL cholesterol increased and total triglyceride concentrations decreased in response to the AIDM. All
affected metabolites, proteins, and genes involved in metabolism
with mean percentage change in response to the AIDM are shown
in Table 5. Metabolites, proteins, and genes were grouped into
different subprocesses, which are discussed below.
Lipoprotein metabolism
The AIDM intervention clearly affected lipoprotein metabolism at various levels (Table 5). In plasma, total triglycerides
decreased by 5% and total diglycerides by 16%. Metabolic
profiling (lipidomics) indicated significant changes, mainly occurring in medium and long chain triglycerides (50 or more
C-atoms). Within this group of triglycerides; those with 04
double bonds decreased and those with
5 double bonds increased in response to the AIDM. Total plasma cholesterol increased by 5%, due to both plasma HDL (+5%) and LDL (+9%)
cholesterol. The ratio of total cholesterol to HDL cholesterol
remained unchanged. Of the increased individual cholesterol
esters (Table 5), 85% could be attributed to cholesterol esters
containing EPA (20:5) and 13% to DHA (C22:6) (see Supplementary Table 2 under Supplemental data in the online issue).
Furthermore, the major protein component of HDL cholesterol,
apolipoprotein A-I (apo A-I), increased, whereas the protein
component for VLDL, apolipoprotein C-III (apo C-III), decreased. In adipose tissue, expression of genes involved in lipid
transport were up-regulated, including cholesteryl ester transfer
protein and fatty acid binding protein 3. In PBMCs, gene expression of LDL receptor and LRP5L (LDL receptorrelated
protein 5) and LRP12 were down-regulated, whereas gene expression of fatty acid transporter 1 (SLC27A1) was up-regulated.
Expression of glycerol kinase, which phosphorylates glycerol,
and diacylglycerol O-acyltransferase homolog 1 (DGAT1), which
catalyzes the conversion of diglycerides to triglycerides, were
increased in PBMCs.
Elongation and desaturation of fatty acids
Steroyl-CoA desaturase activity decreased in response to the
AIDM, as indicated by the ratios of monounsaturated to saturated fat
fromlipidssynthesizedinlivercontainingasinglefattyacid(ratiosof
16:0 and 18:0, see Supplementary Table 3 under Supplemental
data in the online issue) and by down-regulated expression of the
gene (SCD) in PBMCs. In addition, plasma d6 desaturase activity,
as assessed by metabolite ratio (see Supplementary Table 3 under
Supplemental data in the online issue), increased.
Plasma total elongase activity decreased and d6 desaturase
activity increased in plasma, whereas gene expression of ELOVL7

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Concentrations of the oxidative stress marker 8-iso prostaglandin F2-a were reduced in urine, as were uric acid concentrations in plasma (Table 4). Plasma myeloperoxidase
concentrations increased by 24%. A similar increase was observed for 2,3,4-triihydroxybutanoic acid (a degradation product
of vitamin C) and indole-3-propionic acid. Elevated vitamin E
concentrations are related to the presence of vitamin E in the
AIDM. The plasma ferritin concentration was decreased and
adipose tissue gene expression of HEPH (hephaestin) was increased, which are both involved in iron homeostasis. Expression of 2 genes encoding for mitochondrial proteins, NDUFS1

1051

1052

BAKKER ET AL

TABLE 5
Significantly altered metabolites, proteins, and genes in response to the antiinflammatory dietary mix (AIDM) associated with metabolic stress1
Name

Tissue2

Change3

Supplementary
table4

Platform

%
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
PBMCs
PBMCs
PBMCs
PBMCs
PBMCs
PBMCs
AT
AT
AT
AT
AT
AT
AT

24.8
26
21
23 to 213
28 to 215
24 to 217
28 to 219
254
48 to 96
108 to 178
47 to 185
50 to 595
48 to 2857
11
13 to 469
28 to 219
7 to 9
4 to 16
7 to 13
8 to 13
5 to 15
3 to 15
216
29 to 217
7
4.9
8.8
5.1
25
8
28
83
10
17
13
239
28
8
41
221
219
17
23
213
219

6.37
3.24
2.214
2.218 to 2.222
2.225 to 2.228
2.232 to 2.235
2.239 to 2.241
2.201
2.223 to 2.224
2.230 to 2.231
2.237 to 2.238
2.243 to 2.247
2.248 to 2.254
3.28
2.188, 2.195, 2.198 to 2.199
2.189, 2.194, 2.196
3.29, 3.6
2.175, 2.178 to 2.179, 2.181 to 2.185
2.97 to 2.99, 2.102
2.109, 2.110
2.115 to 2.118
2.120 to 2.126
3.27
2.171 to 2.174
2.144
6.33
6.35
6.34
2.140
1.9
1.10
2.145
4.93
4.18
4.47
4.217
4.2
4.126
5.2
5.143
5.134
5.50
5.21
5.101
5.132

C
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
C
C
C
M
Pr
Pr
M
T
T
T
T
T
T
T
T
T
T
T
T
T

Pl
PBMCs

27 to 219
219

3.11 to 3.13
4.209

M
T

Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl
Pl

23
25 to 28
26 to 27
23
28
24
23
23
210

3.4
2.30, 2.39, 2.40
2.55, 2.56
2.13
2.44
2.27
2.53
2.72
2.14

M
M
M
M
M
M
M
M
M
(Continued)

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Lipoprotein metabolism
Total triglycerides
Total triglycerides
48:1
50:050:4
52:152:4
54:154:4
56:256:4
40:1
50:550:6
52:652:7
54:654:7
56:656:10
58:658:12
Total cholesterol esters
16:0, 18:4, 20:5, 22:6
16:1, 18:3, 20:3
Total sphingomyelins
d18:1/14:0, 16:1, 18:0, 22:024:2
d16:1/16:0, 18:0, 20:0, 24:1
d17:1/16:0, 18:0
d18:2/16:0, 18:0, 20:0, 24:0, 24:1
d18:1/16:0, 17:0, 18:0, 22:0, 23:0, 24:0, 24:1
Total diglycerides
36:136:4
Cholesterol
Total cholesterol
LDL cholesterol
HDL cholesterol
18:1 Glycerol ester
Apo A-I
Apo C-III
Vitamin E
DGAT1
GK
EDG4
LDLR
RSAD2
SLC27A1
CETP
DGKI
CEPT1
CYP4B1
FABP3
APOL2
PAPSS2
PPARa-related metabolism
Stearoyl-CoA desaturase-1 activity
SCD
Amino acid metabolism
Total amino acids
Branched-chain amino acids
Branched-chain amino acid derivatives
Alanine
b-Alanine
Proline
Phenylalanine
Threonine
Sarcosine
Energy metabolism

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

1053

TABLE 5 (Continued )

Tissue2

Change3

Supplementary
table4

Platform

Citric acid
a-Ketoglutaric acid
Total organic acids
PANK1
DHTKD1
L2HGDH
HCAL1
DLST
ACACB
ACOX1
Elongation and desaturation of fatty acids
d6 Desaturase activity
Total elongase activity
ELOVL7
Glycemic control
1,5 Anhydro-D-glucitol
GLP-1, total
NAGPA
DPP4
SLC37A4
SNX15
IGF2
Adiponectin
Adiponectin
Sex hormonebinding globulin
Thyroxine-binding globulin
Endocrine factors
GLP-1, total
Prolactin
Sex hormonebinding globulin
Thyroxine-binding globulin
Other
a-Fetoprotein
b2-Microglobulin
Alkaline phosphatase
Creatinine

Pl
Pl
Pl
PBMCs
PBMCs
PBMCs
PBMCs
AT
AT
AT

%
27
24
24
10
9
215
210
227
224
217

2.25
2.104
3.5
4.102
4.107
4.202
4.164
5.165
5.150
5.124

M
M
M
T
T
T
T
T
T
T

Pl
Pl
AT

11
225
18

3.16
3.22
5.43

M
M
T

Pl
Pl
PBMCs
AT
AT
AT
AT
Pl
Pl
Pl
Pl

23
210
13
21
213
15
225
7.4
6
7
4

2.68
1.44
4.54
5.26
5.100
5.67
5.155
6.1
1.3
1.108
1.116

M
Pr
T
T
T
T
T
C
Pr
Pr
Pr

Pl
Pl
Pl
Pl

210
27
7
4

1.44
1.99
1.108
1.116

Pr
Pr
Pr
Pr

Pl
Pl
Pl
Pl

5
27
25.4
23.1

1.6
1.13
6.39
6.42

Pr
Pr
C
C

AT, adipose tissue; PBMCs, peripheral blood mononuclear cells; Pl, plasma; M, metabolomics; Pr, proteomics; T, transcriptomics; C, clinical data;
GLP-1, glucagon-like peptide 1; Apo, apolipoprotein; PPARc, peroxisome proliferator-activated receptor c. All variables were significantly changed by the
AIDM intervention (P , 0.05, ANOVA and/or paired Students t test as described in Subjects and Methods), including Tukey-Kramer adjustment for lipids
and free fatty acid data (n = 36 subjects), Bonferroni adjustment for gas chromatographymass spectrometry (n = 36 subjects) and protein data (n = 33
subjects), and a threshold of a 20% increase or decrease in
10 subjects for transcriptome data (n = 32 subjects for PBMCs, n = 22 subjects for adipose tissue).
Clinical chemistry data were available for 36 subjects.
2
Tissue refers to tissues in which variables were measured
3
Mean percentage change by AIDM.
4
Refers to supplementary tables; first digit corresponds to the supplementary table (1: proteins; 2: metabolites; 3: metabolite ratios or sums; 4: PBMC
genes; 5: adipose tissue genes; 6: clinical data), and the second number corresponds to the row number in the supplementary table.

(elongation of long-chain fatty acids family member 7) was upregulated in adipose tissue (Table 5).
Amino acid metabolism
Plasma concentrations of total amino acids, branched-chain
amino acids, and several individual amino acids and organic
acids decreased, including alanine, proline, and phenylalanine
(Table 5).
Energy metabolism
Gene expression of ACACB (acetyl-coenzyme A carboxylase
b, which is suggested to control fatty acid oxidation), ACOX1

(acyl-coenzyme A oxidase 1 palmitoyl, the first and rate-limiting enzyme in the peroxisomal fatty acid b-oxidation pathway),
and DLST (dihydrolipoamide succinyltransferase, which catalyzes the oxidative decarboxylation of a-keto acids) all decreased in response to the AIDM in adipose tissue (Table 5). In
line with decreased gene expression of DLST, the plasma
a-ketoglutaric acid concentration decreased, whereas gene expression of L2HGDH (L-2-hydroxyglutarate dehydrogenase) in
PBMCs, an enzyme that oxidizes L-2-hydroxyglutarate to
a-ketoglutarate, increased. Gene expression of PANK1 (pantothenate kinase 1), an essential regulatory enzyme in CoA biosynthesis, was up-regulated in PBMCs.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

Name

1054

BAKKER ET AL

Glycemic control

DISCUSSION

Plasma 1,5-anhydro-D-glucitol and total glucagon-like peptide

1 decreased, which indicated a change in glycemic control
(Table 5). Moreover, gene expression of the transporter
SLC37A4 was down-regulated in adipose tissue. This protein
plays a central role in homeostatic regulation of blood glucose
concentrations by transporting glucose-6-phosphate from the
cytoplasm to the lumen of the endoplasmic reticulum.

This study showed that supplementation with food components with evidence-based antiinflammatory properties modulated inflammation and oxidation and altered the metabolism
status of healthy overweight subjects. These effects were detected
by a nutrigenomics approach consisting of large-scale analysis of
gene expression, proteins, and metabolites followed by integrated
biological interpretation of these data generated on multiple
omics platforms.
The main focus of the study was on inflammation, with adiponectin and CRP as established readouts. The plasma concentration
of the antiinflammatory marker adiponectin increased in response
to the AIDM. In contrast with unchanged concentrations of CRP,
we detected numerous subtle changes in other markers related to
inflammation, which together provided evidence of inflammatory
modulation. In a similar study using a mild antiinflammatory

Endocrine factors
Plasma concentrations of the hormones prolactin and glucagonlike peptide 1 decreased, whereas plasma concentrations of sex
hormonebinding globulin and thyroxine-binding globulin increased (Table 5). Concentrations of testosterone and sex hormonebinding globulin were found to be lower in obese men (33).

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

FIGURE 2. Summary of the main effects of the antiinflammatory dietary mix (AIDM) on inflammation-related processes. The matrix in which the
variables were measured [plasma, peripheral blood mononuclear cells (PBMC), adipose tissue, and urine] and/or tissue from which a variable is likely to
originate are shown. The changed variables are indicated in the tissue where they were measured by a colored border (red indicates up-regulation and green
indicates down-regulation in response to the AIDM). The variables without a colored border represent variables in their likely tissue of origin. The different
colors for variables refer to the different subprocesses. Tissues are represented in italic type, and an AIDM component causing a direct effect is illustrated by
an orange capsule. PUFA, polyunsaturated fatty acids; PPARc, peroxisome proliferatoractivated receptor c; F7, coagulation factor VII; PRL, prolactin; B2M,
b2-microglobulin; VCAM1, vascular cell adhesion molecule 1; ICAM1, intercellular adhesion molecule 1; AA, arachidonic acid; EPA, eicosapentaenoic acid;
MPO, myeloperoxidase; CSF1, macrophage colony stimulating factor 1; CCL22, C-C motif chemokine 22 or monocyte chemotactic protein 1.

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

FIGURE 3. Network displaying the interactions, cocitations, and

peroxisome proliferatoractivated receptor (PPARc) and nuclear
transcription factor jB (NFjB) binding sites for prolactin, adiponectin,
and adipose tissue genes based on analysis in Biblisphere (Genomatix
Software GmbH, Munich, Germany). The red arrows indicate inhibition,
and the green arrows indicate activation. The blue rectangles indicate that
the promoter region of the gene contains a binding site for NFjB promoter,
and the purple rectangles indicate that the promoter region of the gene
contains a binding site for PPARc promoter. The gray block arrows
indicate up-regulation (when pointing upward) or down-regulation (when
pointing downward) in response to the antiinflammatory dietary mix
intervention.

reported for resveratrol (49), fish oil (36, 50), and vitamin E
(51).
The data suggest a beneficial effect of AIDM on endothelial
function and plaque formation. Lower concentrations of ICAM-1
and VCAM-1 indicate improvements in endothelial function,
which may diminish atherosclerotic vascular disease risk in
obesity (52). Interestingly, supplementation of AIDM did result
in decreased atherosclerosis in an atherosclerosis-prone mouse
model (R Kleemann and T Kooistra, personal communication,
2008). Gene expression data in both human PBMCs and adipose
tissue suggest altered blood cell differentiation, with indications
of a lowering of natural killer cells, but not of macrophages (53,
54). Resveratrol (49, 55), lycopene (56), green tea catechins
(57), and vitamin E (51) have been shown to ameliorate endothelial health and to reduce vascular inflammation and platelet
aggregation.
AIDM also affected oxidative stress markers (summarized in
Figure 4). Such an effect was anticipated, because antioxidant
effects have been reported for multiple components of the
AIDM: resveratrol (58), green tea catechins (59), lycopene (60),
vitamin E, and vitamin C. Potential reduction of oxidative stress
by the AIDM is supported by decreased urinary concentrations
of 8-isoprostaglandin F2-a, decreased plasma concentrations of
uric acid, increased plasma concentrations of vitamin E, and
increased plasma concentrations of the antioxidant indole-3propionic acid. Indole-3-propionic acid is synthesized by gut
microbiota (61), which indicates an effect of AIDM in the
gastrointestinal tract. The data also indicated an antioxidant
effect in adipose tissue: myeloperoxide expression and 2,3,4trihydroxybutanoic acid concentration both increased. Myeloperoxide may play a role in vitamin C degeneration into
2,3,4-trihydroxybutanoic acid (62).
In addition, lipid metabolism was affected by AIDM (summarized in Figure 5), eg, plasma triglycerides decreased. The
changes were most likely attributable to fish oil including EPA
and DHA. These fatty acids activate PPARa (63), increase hepatic fatty acid oxidation, and reduce triglyceride synthesis,
resulting in decreased plasma triglyceride concentrations (64).
The green tea catechins in AIDM may have contributed as well
through the inhibition of intestinal lipid absorption (65). Furthermore, PPARa activation is known to affect liver apolipoprotein synthesis; it increases apo A-I and decreases apo A-III
(66). Both changes may have contributed to the observed increases in HDL and LDL cholesterol. Apo A-I is an important
component of HDL, and a reduction in apo C-III may result
in the conversion of VLDL into smaller intermediate-densitylipoprotein and LDL particles (66). Lipoprotein changes observed in this study together with changes in other established
PPARa targets, such as d6 desaturase activity and SCD1 expression and activity (66), correspond to an activation of
PPARa. Earlier, we proposed that metabolic stress, together
with inflammatory and oxidative stress, is one of the overarching processes whereby disbalance can lead to a range of
diseases and homeostasis can be maintained by nutrition (67).
In obesity, metabolic stress can be defined as a disturbance of
metabolic homeostasis as a result of caloric excess. Therefore,
modulation of the overarching processes may be relevant for
maintaining or ameliorating health.
Interestingly, some of the effects observed in our study may
suggest effects of the AIDM in organs other than those directly

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

drug, CRP concentrations in overweight men also remained

unchanged (34).
An effect on inflammation was anticipated because the food
components were selected based on their antiinflammatory
properties (6, 3539). Previously, we showed that a short-term
intervention changed the expression of inflammatory genes in
adipose tissue (40). In the current study, most of the effects on
inflammation (summarized in Figure 2) were also observed in
adipose tissue, which is the critical site of the inflammatory state in
obesity (8). Our analyses did not suggest an antiinflammatory
effect in the liver. However, we did not analyze liver tissue because
samples are not easily collected, in contrast with PBMCs.
NF-jB and peroxisome proliferatoractivated receptors
(PPARs) are key regulators of obesity-associated inflammation
(8, 4143). Transcription factor analysis of the adipocyte-related
changes identified inflammatory markers, such as adiponectin
and prolactin, as having binding sites for NF-jB and/or PPARc
in their promoter region (Figure 3). This suggests that NF-jB
and PPARc are the regulators of antiinflammatory effects of
AIDM in adipose tissue. Interestingly, the adiponectin gene was
also found to have a binding site for transcription factor HNF4a,
which we identified previously as a relevant transcription factor
in adipose tissue (40). Adiponectin is produced by adipose tissue
and prolactin by both brain and adipose tissue (44, 45). AIDM
intervention increased adiponectin concentrations, whereas
prolactin concentrations decreased. This corresponds to the
notion that prolactin may reduce adiponectin production and
secretion in adipose tissue (4648).
Plasma fatty acid ratios (specifically a decreased eicosanoid
precursor ratio of n23 to n26) and gene expression in adipose
tissue indicated an AIDM effect on eicosanoids, namely an inhibition of the synthesis of inflammation-related eicosanoids.
Plasma prostaglandin E2 concentrations, however, did not
change. The former observation is consistent with an antiinflammatory activity through inhibition of eicosanoid synthesis

1055

1056

BAKKER ET AL

analyzed, such as the brain. For instance, concentrations of

prolactin, which is mainly synthesized in the brain, decrease after
AIDM intervention. Prolactin is a well-established metabolic
regulator that affects body weight regulation, pancreatic islet
development, and other processes. Possible effects on the brain may
support the association between increased intake of essential fatty
acids (especially EPA and DHA), vitamin E, and vitamin Call
of which are present in the AIDMand improved mental health
(6871). This findings may provide interesting leads for future
research (72).
Several limitations of this study need to be addressed. In this
study, we used a specific combination of compounds in an effort
to better mimic a real life situation, because a healthy diet
consists of a mix of components. Choices for these compounds
were mainly based on their individual antiinflammatory activities; a more optimal combination may exist. The mild effects
observed may be attributed to the length of the intervention
period. Also, it should be noted that, although we applied
multiple platforms for large-scale screening of genes, proteins,
and metabolites, it does not necessarily mean that we analyzed all
variables of interest. This was because of limitations in the

sampling of specific tissues and limitations in the analysis

techniques, mainly on the protein and metabolite concentrations.
Despite these limitations, our approach allowed for the detection of multiple subtle health effects of a mix of dietary
components in relatively healthy overweight subjects. Furthermore, the changes in concentrations of genes, proteins, and
metabolite induced by the AIDM appeared to be consistent. This
may prove to be an additional advantage of nutrigenomics
technologies in human intervention studies, namely enabling the
high-sensitivity detection of multiple physiologic changes.
In contrast with the accepted biomarkers, the application of
nutrigenomics techniques for large-scale profiling of genes,
proteins, and metabolites showed that the AIDM was able to
influence processes of inflammation, oxidative stress, and metabolism in humans. The use of comprehensive techniques such
as metabolic, protein, and gene profiling strongly facilitated the
accurate and detailed quantification and description of the molecular processes involved.
We express our gratitude to the volunteers who participated in the study, to
the staff of the Metabolic Research Unit, to the laboratories who assisted in the

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

FIGURE 4. Summary of the main effects of the antiinflammatory dietary mix (AIDM) on oxidation-related processes. The matrix in which the variables
were measured [plasma, peripheral blood mononuclear cells (PBMC), adipose tissue, and urine] and/or tissue from which a variable is likely to originate are
shown. The changed variables are indicated in the tissue where they were measured by a colored border (red indicates up-regulation and green indicates downregulation in response to the AIDM). The different colors for variables refer to the different subprocesses. Tissues are represented by italic type, and an AIDM
component causing a direct effect is illustrated by an orange capsule. MPO, myeloperoxidase.

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

1057

organization of the study, to groups that contributed to the measurements

and data analysis, to Trinette van Vliet and Robert Kleemann for their valuable contribution to designing the experiment, and to Peter Gillies (research fellow at DuPont Applied BioSciences) for critical reading of
the manuscript.
The authors responsibilities were as followsHFJH and TK: designed the
experiment; GCMB and MJvE: organized and managed the conduct of the
study; GCMB, MJvE, LP, SW, and CMR: compiled the data; HFJH, TK,
and BvO: supervised the final compilation of the manuscript; LP and SW:
designed the tables and figures; and GCMB, NHPC, and BvO: coordinated
the complete project. All authors approved the manuscript, took part in
the discussion of the results, and were involved in writing the manuscript.
All authors are employed by TNO Quality of Life, and none of them reported
a conflict of interest related to this article. TNO is a member of the European
Nutrigenomics Organization (www.nugo.org).

3.

4.
5.
6.
7.
8.
9.

REFERENCES
1. Fontana L, Klein S. Aging, adiposity, and calorie restriction. JAMA
2007;297:98694.
2. Ginsberg HN, Maccallum PR. The obesity, metabolic syndrome, and
type 2 diabetes mellitus pandemic: Part I. Increased cardiovascular

10.

11.

disease risk and the importance of atherogenic dyslipidemia in persons

with the metabolic syndrome and type 2 diabetes mellitus. J Cardiometab Syndr 2009;4:1139.
Folsom AR, Aleksic N, Catellier D, Juneja HS, Wu KK. C-reactive
protein and incident coronary heart disease in the Atherosclerosis Risk
In Communities (ARIC) study. Am Heart J 2002;144:2338.
Saadeddin SM, Habbab MA, Ferns GA. Markers of inflammation and
coronary artery disease. Med Sci Monit 2002;8:RA512.
Hu FB, Meigs JB, Li TY, Rifai N, Manson JE. Inflammatory markers and
risk of developing type 2 diabetes in women. Diabetes 2004;53:693700.
Singh U, Devaraj S, Jialal I. Vitamin E, oxidative stress, and inflammation. Annu Rev Nutr 2005;25:15174.
Ross R. Atherosclerosisan inflammatory disease. N Engl J Med 1999;
340:11526.
Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006;
444:8607.
Ronti T, Lupattelli G, Mannarino E. The endocrine function of adipose
tissue: an update. Clin Endocrinol (Oxf) 2006;64:35565.
Trayhurn P, Beattie JH. Physiological role of adipose tissue: white adipose tissue as an endocrine and secretory organ. Proc Nutr Soc 2001;60:
32939.
Wang P, Mariman E, Renes J, Keijer J. The secretory function of adipocytes
in the physiology of white adipose tissue. J Cell Physiol 2008;216:313.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

FIGURE 5. Summary of the main effects of the antiinflammatory dietary mix (AIDM) on metabolism-related processes. The matrix in which the variables
were measured [plasma, peripheral blood mononuclear cells (PBMC), adipose tissue, and urine] and/or tissue from which a variable is likely to originate are
shown. The changed variables are indicated in the tissue where they were measured by a colored border (red indicates up-regulation and green indicates downregulation in response to the AIDM). The different colors for variables refer to the different subprocesses. Tissues are represented by italic type, and an AIDM
component causing a direct effect is illustrated by an orange capsule. PPARc, peroxisome proliferatoractivated receptor c; GLP1, glucagon-like peptide 1;
TG, triglyceride; DG, diglyceride; SHBG, sex hormonebinding globulin; BCAA, branched-chain amino acids; PUFA, polyunsaturated fatty acids; ALP,
alkaline phosphatase; TBG, thyroxine binding globulin; AFP, a-fetoprotein; PRL, prolactin.

1058

BAKKER ET AL
38. Esposito K, Maiorino M, Giugliano D. Which diet is best for diabetes?
Diabetologia 2009;52:9889.
39. Heber D, Lu QY. Overview of mechanisms of action of lycopene. Exp
Biol Med 2002;227:9203.
40. van Erk MJ, Pasman WJ, Wortelboer HM, van Ommen B, Hendriks HF.
Short-term fatty acid intervention elicits differential gene expression
responses in adipose tissue from lean and overweight men. Genes Nutr
2008;3:12737.
41. Carlsen H, Haugen F, Zadelaar S, et al. Diet-induced obesity increases
NF-kappaB signaling in reporter mice. Genes Nutr 2009;4:21522.
42. Moller DE, Berger JP. Role of PPARs in the regulation of obesity-related
insulin sensitivity and inflammation. Int J Obes Relat Metab Disord
2003;27(suppl 3):S1721.
43. Radonjic M, de Haan JR, van Erk MJ, et al. Genome-wide mRNA expression analysis of hepatic adaptation to high-fat diets reveals switch
from an inflammatory to steatotic transcriptional program. PLoS One
2009;4:e6646.
44. McFarland-Mancini M, Hugo E, Loftus J, Ben-Jonathan N. Induction of
prolactin expression and release in human preadipocytes by cAMP activating ligands. Biochem Biophys Res Commun 2006;344:916.
45. Hugo ER, Borcherding DC, Gersin KS, Loftus J, Ben-Jonathan N.
Prolactin release by adipose explants, primary adipocytes, and LS14
adipocytes. J Clin Endocrinol Metab 2008;93:400612.
46. Nilsson L, Binart N, Bohlooly Y, et al. Prolactin and growth hormone
regulate adiponectin secretion and receptor expression in adipose tissue.
Biochem Biophys Res Commun 2005;331:11206.
47. Nanbu-Wakao R, Fujitani Y, Masuho Y, Muramatu MA, Wakao H.
Prolactin enhances CCAAT enhancer-binding protein b (C/EBPb) and
peroxisome proliferator-activated receptor c (PPARc) messenger RNA
expression and stimulates adipogenic conversion of NIH-3T3 cells. Mol
Endocrinol 2000;14:30716.
48. Swarbrick MM, Havel PJ. Physiological, pharmacological, and nutritional regulation of circulating adiponectin concentrations in humans.
Metab Syndr Relat Disord 2008;6:87102.
49. Pace-Asciak CR, Hahn S, Diamandis EP, Soleas G, Goldberg DM. The
red wine phenolics trans-resveratrol and quercetin block human platelet
aggregation and eicosanoid synthesis: implications for protection against
coronary heart disease. Clin Chim Acta 1995;235:20719.
50. Browning LM. n23 polyunsaturated fatty acids, inflammation and
obesity-related disease. Proc Nutr Soc 2003;62:44753.
51. Reiter E, Jiang Q, Christen S. Anti-inflammatory properties of a and c
tocopherol. Mol Aspects Med 2007;28:66891.
52. Ohman MK, Wright AP, Wickenheiser KJ, Luo W, Eitzman DT. Visceral
adipose tissue and atherosclerosis. Curr Vasc Pharmacol 2009;7:16979.
53. Schenk S, Saberi M, Olefsky JM. Insulin sensitivity: modulation by
nutrients and inflammation. J Clin Invest 2008;118:29923002.
54. Breitling R. Robust signaling networks of the adipose secretome. Trends
Endocrinol Metab 2009;20:17.
55. Bertelli AAE, Giovannini L, Giannessi D, et al. Antiplatelet activity of
synthetic and natural resveratrol in red wine. Int J Tissue React 1995;17:
13.
56. Hung CF, Huang TF, Chen BH, Shieh JM, Wu PH, Wu WB. Lycopene
inhibits TNF-a-induced endothelial ICAM-1 expression and monocyte-endothelial adhesion. Eur J Pharmacol 2008;586:27582.
57. Babu PV, Liu D. Green tea catechins and cardiovascular health: an
update. Curr Med Chem 2008;15:184050.
58. Baur JA, Sinclair DA. Therapeutic potential of resveratrol: the in vivo
evidence. Nat Rev Drug Discov 2006;5:493506.
59. Serafini M, Ghiselli A, Ferro-Luzzi A. In vivo antioxidant effect of
green and black tea in man. Eur J Clin Nutr 1996;50:2832.
60. Rao AV, Rao LG. Carotenoids and human health. Pharmacol Res 2007;
55:20716.
61. Wikoff WR, Anfora AT, Liu J, et al. Metabolomics analysis reveals large
effects of gut microflora on mammalian blood metabolites. Proc Natl
Acad Sci USA 2009;106:3698703.
62. Jahn M, Baynes JW, Spiteller G. The reaction of hyaluronic acid and its
monomers, glucuronic acid and N-acetylglucosamine, with reactive
oxygen species. Carbohydr Res 1999;321:22834.
63. Kersten S, Desvergne B, Wahli W. Roles of PPARS in health and disease. Nature 2000;405:4214.
64. Kersten S. Peroxisome proliferator activated receptors and lipoprotein
metabolism. PPAR Res 2008;2008:132960.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016

12. Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB.
Plasma adiponectin levels and risk of myocardial infarction in men.
JAMA 2004;291:17307.
13. Schulze MB, Rimm EB, Shai I, Rifai N, Hu FB. Relationship between
adiponectin and glycemic control, blood lipids, and inflammatory
markers in men with type 2 diabetes. Diabetes Care 2004;27:16807.
14. Danesh J, Wheeler JG, Hirschfield GM, et al. C-reactive protein and
other circulating markers of inflammation in the prediction of coronary
heart disease. N Engl J Med 2004;350:138797.
15. Puglisi MJ, Fernandez ML. Modulation of C-reactive protein, tumor
necrosis factor-a, and adiponectin by diet, exercise, and weight loss.
J Nutr 2008;138:22936.
16. Trichopoulou A, Bamia C, Trichopoulos D. Anatomy of health effects of
Mediterranean diet: Greek EPIC prospective cohort study. BMJ 2009;
338:b2337.
17. Esposito K, Marfella R, Ciotola M, et al. Effect of a Mediterraneanstyle diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 2004;
292:14406.
18. Shimizu M, Weinstein IB. Modulation of signal transduction by tea
catechins and related phytochemicals. Mutat Res 2005;591:14760.
19. Simopoulos AP. Omega-3 fatty acids in inflammation and autoimmune
diseases. J Am Coll Nutr 2002;21:495505.
20. Upritchard JE, Sutherland WHF, Mann JI. Effect of supplementation
with tomato juice, vitamin E, and vitamin C on LDL oxidation and
products of inflammatory activity in type 2 diabetes. Diabetes Care
2000;23:7338.
21. Goldberg DM, Yan J, Soleas GJ. Absorption of three wine-related polyphenols in three different matrices by healthy subjects. Clin Biochem 2003;
36:7987.
22. Williamson G, Manach C. Bioavailability and bioefficacy of polyphenols in humans. II. Review of 93 intervention studies. Am J Clin
Nutr 2005;81(suppl):230S42S.
23. Fukino Y, Shimbo M, Aoki N, Okubo T, Iso H. Randomized controlled
trial for an effect of green tea consumption on insulin resistance and
inflammation markers. J Nutr Sci Vitaminol (Tokyo) 2005;51:33542.
24. Kelley DS, Hubbard NE, Erickson KL. Regulation of human immune
and inflammatory responses by dietary fatty acids. Adv Food Nutr Res
2005;50:10138.
25. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of
the preparative ultracentrifuge. Clin Chem 1972;18:499502.
26. Van Den Berg R, van Vliet T, Broekmans WMR, et al. A vegetable/fruit
concentrate with high antioxidant capacity has no effect on biomarkers
of antioxidant status in male smokers. J Nutr 2001;131:171422.
27. Wopereis S, Rubingh CM, van Erk MJ, et al. Metabolic profiling of the
response to an oral glucose tolerance test detects subtle metabolic
changes. PLoS One 2009;4:e4525.
28. van der Kloet FM, Bobeldijk I, Verheij ER, Jellema RH. Analytical error
reduction using single point calibration for accurate and precise metabolomic phenotyping. J Proteome Res 2009;8:513241.
29. Bijlsma S, Bobeldijk I, Verheij ER, et al. Large-scale human metabolomics studies: a strategy for data (pre-) processing and validation.
Anal Chem 2006;78:56774.
30. Radonjic M, van Erk MJ, Pasman WJ, Wortelboer HM, Hendriks HF,
van Ommen B. Effect of body fat distribution on the transcription response to dietary fat interventions. Genes Nutr 2009;4:1439.
31. De Groot PJ, Reiff C, Mayer C, Muller M. NuGO contributions to
GenePattern. Genes Nutr 2008;3:1436.
32. Dai M, Wang P, Boyd AD, et al. Evolving gene/transcript definitions
significantly alter the interpretation of GeneChip data. Nucleic Acids
Res 2005;33:e175.
33. Mah PM, Wittert GA. Obesity and testicular function. Mol Cell Endocrinol 2010;316:1806.
34. van Erk MJ, Wopereis S, Rubingh CM, et al. Insight in modulation of
inflammation by gene, protein and metabolite profiling in overweight
males: a human intervention study. BMC Med Genom (in press).
35. Butt MS, Sultan MT. Green tea: natures defense against malignancies.
Crit Rev Food Sci Nutr 2009;49:46373.
36. Calder PC. n23 Polyunsaturated fatty acids, inflammation, and inflammatory diseases. Am J Clin Nutr 2006;83(suppl):1505S19S.
37. Curin Y, Andriantsitohaina R. Polyphenols as potential therapeutical
agents against cardiovascular diseases. Pharmacol Rep 2005;57:
97107.

DIETARY MIX AND INFLAMMATORY STATUS IN OVERWEIGHT MEN

65. Koo SI, Noh SK. Green tea as inhibitor of the intestinal absorption of lipids:
potential mechanism for its lipid-lowering effect. J Nutr Biochem 2007;18:
17983.
66. Duval C, Muller M, Kersten S. PPARalpha and dyslipidemia. Biochim
Biophys Acta 2007;1771:96171.
67. van Ommen B, Keijer J, Heil SG, Kaput J. Challenging homeostasis to
define biomarkers for nutrition related health. Mol Nutr Food Res 2009;
53:795804.
68. Bourre JM. Effects of nutrients (in food) on the structure and function of
the nervous system: update on dietary requirements for brain. Part 1:
micronutrients. J Nutr Health Aging 2006;10:37785.

1059

69. Bourre JM. Effects of nutrients (in food) on the structure and function of
the nervous system: update on dietary requirements for brain. Part 2:
macronutrients. J Nutr Health Aging 2006;10:38699.
70. Riediger ND, Othman RA, Suh M, Moghadasian MH. A systemic review of the roles of n23 fatty acids in health and disease. J Am Diet
Assoc 2009;109:66879.
71. Kaplan BJ, Crawford SG, Field CJ, Simpson JS. Vitamins, minerals, and
mood. Psychol Bull 2007;133:74760.
72. Simon GE, Von Korff M, Saunders K, et al. Association between obesity
and psychiatric disorders in the US adult population. Arch Gen Psychiatry 2006;63:82430.

Downloaded from ajcn.nutrition.org by guest on July 14, 2016