Differential epigenomic and transcriptomic responses in subcutaneous
adipose tissue between low and high responders to caloric
restriction1–3
Luigi Bouchard, Re´mi Rabasa-Lhoret, May Faraj, Marie-E`ve Lavoie, Jonathan Mill, Louis Pe´russe, and Marie-Claude Vohl
ABSTRACT
Background: Caloric restriction is recommended for the treatment
of obesity, but it is generally characterized by large interindividual
variability in responses. The factors affecting the magnitude of
weight loss remain poorly understood. Epigenetic factors (ie, heritable but reversible changes to genomic function that regulate gene
expression independently of DNA sequence) may explain some of
the interindividual variability seen in weight-loss responses.
Objective: The objective was to determine whether epigenetics and
gene expression changes may play a role in weight-loss responsiveness.
Design: Overweight/obese postmenopausal women were recruited
for a standard 6-mo caloric restriction intervention. Abdominal subcutaneous adipose tissue biopsy samples were collected before (n =
14) and after (n = 14) intervention, and the epigenomic and transcriptomic profiles of the high and low responders to dieting, on the
basis of changes in percentage body fat, were compared by using
microarray analysis.
Results: Significant DNA methylation differences at 35 loci were
found between the high and low responders before dieting, with 3
regions showing differential methylation after intervention. Some of
these regions contained genes known to be involved in weight control and insulin secretion, whereas others were localized in known
imprinted genomic regions. Differences in gene expression profiles
were observed only after dieting, with 644 genes being differentially expressed between the 2 groups. These included genes likely
to be involved in metabolic pathways related to angiogenesis and
cerebellar long-term depression.
Conclusions: These data show that both DNA methylation and gene
expression are responsive to caloric restriction and provide new
insights about the molecular pathways involved in body weight loss
as well as methylation regulation during adulthood.
Am J Clin
Nutr 2010;91:309–20.
INTRODUCTION
In 2005, 1.1 billion adults and 10% of children were overweight or obese worldwide (1). Obesity is defined as an excessive
accumulation of fat resulting from a long-term imbalance between energy intake and expenditure (2). It has a dramatic effect
on an individual’s health, with musculoskeletal, pulmonary, and
psychosocial-related problems, and is associated with an increased risk of morbidity and mortality attributable to cardiovascular diseases, diabetes, and certain forms of cancer (1, 3).
Interestingly, only a moderate loss of initial body weight
provides significant metabolic improvements (4–6). However,
weight loss responses to caloric restriction show considerable
interindividual variability (7). Studies of genetically identical
monozygotic twins have been particularly useful in disentangling
the role of environmental and heritable factors in determining the
degree of weight loss. It has been shown that within-pair changes
in body fat variability after a caloric deficit is significantly lower
than between-pair variability, which suggests that genetic factors
have an important influence on an individual’s response to caloric
deficit (8, 9). However, the concordance between twin pairs was
not complete, which suggests that environmental factors or other
DNA sequence–independent mechanisms may be involved.
It has been suggested that monozygotic twin discordance for
complex traits such as body weight could be accounted for by
epigenetic factors (10, 11). Epigenetics refers to the heritable, but
reversible, regulation of various genomic functions, including
gene transcription, that are mediated principally through changes
in DNA methylation and chromatin structure (12). The epigenetic
regulation of cellular functions is a normal and essential process
in cell development and differentiation, and epigenetic factors are
subjected to reprogramming in response to both stochastic and
environmental stimuli (12). Such changes can be mitotically
1
From the Nutraceuticals and Functional Foods Institute (LB, LP, and MCV), the Department of Preventive Medicine (LP), and the Department of
Food Science and Nutrition (M-CV), Université Laval, Laval, Canada; the
Department of Medicine, Université de Montréal, Chicoutimi Hospital, Saguenay, Canada (LB); the Department of Nutrition (RR-L, MF, and M-EL),
the Montreal Diabetes Research Centre (RR-L and MF), Université de Montréal, Montreal, Canada; and the Institute of Psychiatry, SGDP Research
Centre, King’s College London, London, United Kingdom (JM).
2
Supported by the Canadian Institute of Health Research through the
MONET project (Montreal-Ottawa New Emerging Team; OHN-63279 and
MOP62976) and the Quebec New Emerging Team (OHN 63276). LB was
funded by the Laval University Merck Frosst/Canadian Institute of Health
Research Chair in Obesity and the Heart and Stroke Foundation of Canada/
Sanofi-Aventis research fellowship awards. RRL was supported by the Fonds
de la Recherche en Santé du Québec and held the chair for clinical research
J-A de Sève at IRCM (Montreal Institute for Clinical Research). MF was the
recipient of the Canadian Institute of Health Research New Investigator
Award. M-EL was supported by a scholarship from the Fonds de la Recherche en Santé du Québec.
3
Address correspondence to M-C Vohl, Lipid Research Center, 2705
Laurier Boulevard, (TR93), Québec City, PQ, Canada G1V 4G2. E-mail:
[email protected].
Received May 18, 2009. Accepted for publication November 2, 2009.
First published online November 25, 2009; doi: 10.3945/ajcn.2009.28085.
Am J Clin Nutr 2010;91:309–20. Printed in USA. Ó 2010 American Society for Nutrition
309
310
BOUCHARD ET AL
stable and enduring, producing long-term changes to gene expression, but can also be short-lived and rapidly reversed (13).
Indeed, such dynamic epigenetic changes have the potential to
offer a mechanism by which cellular metabolism can be rapidly
regulated independently of long-term, irreversible evolutionary
mutagenesis. Cytosine methylation (Cmet), occurring at position
5 of the cytosine pyrimidine ring in CpG dinucleotides, is the
best understood epigenetic modification. The methylation of
CpG sites disrupts the binding of transcription factors and attracts methyl-binding proteins that are associated with gene silencing and chromatin compaction (14). Because CpG
dinucleotide methylation is associated with the regulation of
gene expression, altered DNA methylation could explain interindividual phenotypic differences (15).
We report here the results of the first comprehensive analysis of
epigenomic and transcriptomic responses in subcutaneous adipose tissue after a caloric restriction intervention in overweight
and obese women. We found that although both DNA methylation and gene expression differences existed between the high
and low responders to dietary restriction after the intervention,
only epigenetic differences exist before dieting. These data
suggest that the epigenetic profile has the potential to differentiate
between good and poor responders to caloric restriction.
adipose tissue biopsy collections were preceded by a 4-wk
weight-stability period (within 62 kg), verified on a weekly
basis at our research unit. Also, the subjects were instructed not
to exercise and to eat a high-carbohydrate diet for the 3 d before
the biopsy procedure. Fasting baseline and postintervention
subcutaneous adipose tissue biopsy samples were obtained from
the periumbilical level at both sides of the body by needle biopsy under local anesthesia (20 mg xylocaine/mL) (18–20).
One hundred thirty-seven women were recruited and completed the 6-mo weight-loss program. Baseline and postintervention adipose tissue biopsy samples were available for 29
of these women. Consent for biopsy was an optional part of the
larger study. There was no other criteria to be met to be included
in the biopsy subsample. Fourteen women were further selected
for the current study based on their response to the caloric restriction. Women who lost 3% of their body fat were considered “high responders,” whereas those who lost ,3% of their
body fat were considered “low responders” to the caloric restriction. Low (n = 7) and high (n = 7) responders were matched
for baseline age, BMI, percentage body fat, resting blood
pressure, fasting blood lipids, glucose and insulin concentrations, and changes in fat-free mass (Tables 1 and 2).
Anthropometric and metabolic measurements
SUBJECTS AND METHODS
Study population and experimental design
The sample used in this study was a subsample of a larger
weight-loss study (n = 137) aimed at exploring the effect of
a caloric restriction intervention on body composition, energy
expenditure, insulin sensitivity, and metabolic, inflammatory,
hormonal, and psychosocial profiles in overweight and obese
postmenopausal women. The recruitment for the larger study
began in May 2003. The substudy presented here explored the
epigenomic and transcriptomic responses of 14 overweight and
obese postmenopausal women to a caloric restriction intervention. All individuals provided written informed consent
before their inclusion in the study, which was approved by the
Université de Montréal ethics committee.
The inclusion and exclusion criteria of the weight-loss study
were presented in detail previously (16, 17). Briefly, sedentary
overweight and obese postmenopausal women were included if
they had been weight stable for 3 mo before the study and were
not taking medications known to affect cardiovascular function
and/or metabolism. The subjects were nonsmokers, had a low-tomoderate alcohol consumption (,2 drinks/d), and were free of
diabetes and, uncontrolled thyroid, inflammatory, cardiovascular, or pituitary diseases.
The aim of the medically supervised weight-loss program was
to reduce body weight by 10% over 6 mo. To determine the levels
of caloric restriction, 500–800 kcal were subtracted from the
baseline resting metabolic rate (determined by indirect calorimetry) and then multiplied by a physical activity factor of 1.4,
which corresponds to a sedentary state. Dietary prescriptions
ranged from 1100 to 1800 kcal/d. The macronutrient composition
of the diets was standardized: 55%, 30%, and 15% of energy
intake from carbohydrates, fat, and protein, respectively (16). To
reduce the acute effect of weight changes on the outcomes
measured, baseline and postintervention measurements as well as
Standardized procedures were used to measure body weight,
height, waist girth, resting blood pressure, and blood lipid,
glucose, and insulin concentrations, as previously described (16,
17). Total energy expenditure was measured by using the doubly
labeled water technique (21), and resting metabolic rate was
measured by indirect calorimetry (16, 17). Total fat mass and fatfree mass were measured by dual-energy X-ray absorptiometry
(16, 17).
DNA and RNA extraction
Genomic DNA was extracted by using the Qiagen DNeasy
Blood and Tissue DNA purification kit (Qiagen, Mississauga,
Canada). Total RNA was prepared from ’50 mg of subcutaneous
adipose tissue biopsy samples and was concentrated with the
Qiagen RNeasy Lipid Tissue Minikit and Qiagen RNeasy
MinElute Cleanup Kit (Qiagen). The concentration (5–10 lg/50 mg
fat tissue) and integrity of purified total RNA was verified by
using an Agilent 2100 bioanalyzer with the RNA 6000 Nano
LabChip kit (Agilent Technologies, Palo Alto, CA).
Microarrays
Epigenomic profiling
Our epigenomic profiling of abdominal subcutaneous adipose
tissue used Human CpG-island 15K arrays obtained from the
University Health Network (Toronto, Canada). This array contains 14,923 CpG-island clones derived from the Human CpGisland 8.1K array (22) and a set of 6800 additional CpG-island
loci. Clones with internal repeat sequences, with multiple or
absent BLAST hits, mapped on mitochondrial chromosome or
those with .20% overlapping sequences (duplicates) were
excluded. Clones from Arabidopsis and the Stratagene’s
SpotReport Alien cDNA Array Validation System were also
spotted on the array and served as controls.
311
MOLECULAR PATHWAYS INVOLVED IN BODY WEIGHT LOSS
TABLE 1
Baseline anthropometric and metabolic characteristics of the low and high responders1
Low responders (n = 7)
Age (y)
BMI (kg/m2)
Fat mass (kg)
Body fat (%)
Fat-free mass (kg)
Fat-free mass (%)
Waist girth (cm)
DBP (mm Hg)
SBP (mm Hg)
Total cholesterol (mmol/L)
HDL cholesterol (mmol/L)
LDL cholesterol (mmol/L)
Triglycerides (mmol/L)
Glucose (mmol/L)
Insulin (mmol/L)2
Total energy expenditure (kcal/d)
Mean 6 SD
Range
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
53.04–64.24
26.53–38.03
28.04–48.84
40.30–50.30
35.73–57.22
49.70–59.70
90.50–136.00
61.00–79.00
95.00–146.00
4.76–6.27
1.13–2.04
2.93–4.08
0.49–1.76
3.93–6.57
5.98–21.55
2057–3301
57.71
32.25
38.21
45.91
44.82
54.09
105.96
72.43
115.43
5.71
1.59
3.49
1.37
5.22
14.30
2625
4.41
4.18
7.06
3.61
7.10
3.61
15.36
6.27
16.22
0.50
0.30
0.40
0.45
0.95
5.14
511
High responders (n = 7)
Mean 6 SD
Range
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
52.80–68.57
26.13–34.57
23.97–44.70
39.80–50.50
34.77–48.67
49.50–60.20
86.50–105.75
69.00–91.00
104.00–157.00
4.58–6.60
1.28–1.58
2.60–4.09
0.90–3.01
4.40–5.67
8.57–21.29
1832–2760
57.80
29.54
34.40
45.40
41.06
54.60
97.04
78.57
126.71
5.50
1.44
3.24
1.81
5.13
13.32
2317
5.30
2.78
6.64
4.30
5.31
4.30
7.87
7.46
19.73
0.78
0.11
0.61
0.77
0.48
5.83
337
1
DBP, diastolic blood pressure; SBP, systolic blood pressure. None of the differences between groups were statistically
significant, P , 0.05 (unpaired t test).
2
For the high responders, only 6 samples were available for fasting insulin measurements.
The unmethylated fraction of genomic DNA was enriched for
epigenomic profiling, as previously reported (23, 24). Briefly, 500 ng
genomic DNA was digested with the restriction enzyme HpaII
(New England Biolabs, Ipswich, MA). The DNA fragments
were then ligated with DNA adaptors (annealing products of
U-CG1A 5#-CGTGGAGACTGACTACCAGAT-3# and UCG1B
5#-AGTTACATCTGGTAGTCAGTCTCCA-3#) followed by further
digestion with McrBC restriction enzyme (New England Biolabs). HpaII cleaves DNA only when the restriction site is unmethylated, whereas McrBC cleaves methylated fragments only.
Therefore, both methylation-sensitive restriction enzymes contributed to enrich the unmethylated fraction of genomic DNA.
The remaining fragments were then amplified by polymerase
chain reaction (PCR) with U-CG1B primer in conditions such
that short fragments (,1.5 kb), and thus unmethylated DNA,
was preferentially amplified. For each PCR reaction, 80 ng digested genomic DNA were added to a reaction mixture (final
volume = 100 lL) containing 6.25 units Taq DNA polymerase
(New England Biolabs), 1 · thermopol buffer (New England
Biolabs), 2.5 mmol MgCl2/L, 125 lmol of each dNTP and 1.5
lmol U-CG1b primer. PCRs were run in triplicate as follow:
initial cycle at 72°C for 5 min and 95°C for 1 min, 25 cycles at
95°C for 40 s, 68°C for 2 min, 72°C for 20 s, and a final extension at 72°C for 5 min. The pooled triplicates provided
’10 lg enriched DNA for labeling and hybridization. We combined 2.5 lg of each DNA sample to produce a common
TABLE 2
Anthropometric and metabolic changes after the caloric restriction intervention1
Low responders
Mean 6 SD
Change
Change
Change
Change
Change
Change
Change
Change
Change
Change
Change
Change
Change
Change
1
BMI (kg/m )
fat mass (kg)
percentage body fat (%)
fat-free mass (kg)
percentage fat-free mass (%)
waist girth (cm)
DBP (mm Hg)
SBP (mm Hg)
total cholesterol (mmol/L)
HDL cholesterol (mmol/L)
LDL cholesterol (mmol/L)
triglycerides (mmol/L)
glucose (mmol/L)
insulin (mmol/L)5
20.91
21.81
21.07
20.55
1.07
24.46
0.29
4.71
20.07
20.10
20.03
0.13
0.03
21.89
6
6
6
6
6
6
6
6
6
6
6
6
6
6
0.92
1.84
1.19
1.07
1.19
3.44
7.54
9.07
0.52
0.22
0.55
0.41
0.56
2.79
Range
Mean 6 SD
22.03 to 0.74
23.69 to 0.77
22.70 to 1.00
21.95 to 1.09
21.00 to 2.70
28.25 to 2.00
216.00 to 6.00
27.00 to 17.00
21.08 to 0.31
20.30 to 0.34
20.93 to 0.48
20.28 to 0.95
21.10 to 0.53
25.45 to 1.29
22.78 6 1.51
27.34 6 3.053
26.10 6 2.864
0.26 6 2.27
6.10 6 2.864
27.79 6 5.15
24.86 6 8.63
25.57 6 9.11
20.12 6 0.58
20.01 6 0.10
0.03 6 0.69
20.29 6 0.62
20.16 6 0.32
21.93 6 4.59
Range
2
DBP, diastolic blood pressure; SBP, systolic blood pressure.
Significantly different from low responders (unpaired t test): 2P 0.05, 3P 0.01, 4P 0.00.
For the high responders, only 6 samples were available for fasting insulin measurements.
2–4
5
in
in
in
in
in
in
in
in
in
in
in
in
in
in
2
High responders
25.32 to 21.15
212.98 to 23.87
210.90 to 23.00
23.21 to 3.49
3.00 to 10.90
217.00 to 21.50
222.00 to 4.00
220.00 to 5.00
21.21 to 0.41
20.21 to 0.09
20.98 to 0.98
21.14 to 0.68
20.50 to 0.43
29.53 to 4.00
312
BOUCHARD ET AL
reference pool that was used to standardize between hybridizations. All samples were labeled with amino-allyl-dUTP, coupled to alexa dyes (alexa 647 for experimental samples and alexa
555 for the reference pool), following the Bioprime Plus Array
CGH Genomic labeling system protocol. The labeled probes were
then purified and hybridized (via incubation for 16 h at 42°C) on
Human CpG-island 15K arrays by using the Advalytix SlideBooster (Advalytix, Munich, Germany) and AdvaHybc hybridization solution. The arrays were scanned on an Agilent
G2565BA scanner, quantified with ArrayVision v.8.0 (Imaging
Research), and the data were normalized with MIDAS software
v.2.19 by using a subgrid intensity-based method. All 28 microarrays passed initial quality control measures and were suitable
for analysis.
CpG-island 15K microarray results validation
The Sequenom EpiTYPER (Sequenom Inc, San Diego, CA)
approach has been used to determine base-specific cytosine
methylation levels of loci showing differential methylation levels
using microarray (25). Briefly, the EpiTYPER assay combines
DNA sodium bisulfite conversion chemistry, PCR amplification
of target sequences, and base-specific cleavage. The cleavage
products are quantitatively analyzed by MALDI-TOF mass
spectrometry. The presence of methylated cytosine within the
original DNA sequence determines the cleavage pattern. The
EpiDESIGNER software has been used for PCR primer selection.
The loci showing the strongest methylation differences in lowand high-methylation groups (DNASE1L2 and TRIM14) were
selected for validation as well as the 2 regions with 2 proximal
probe sets with methylation differences (chr7p15 and chr10q26).
The PCR primers were as follows: DNASE1L2-2f-GTTTAGTAGTGTTTTGGGAGTTTGT and DNASE1L2-2r-CCTACCCACCACACCTATTAATCTA, DNASE1L2-6f-GGGTTTTTTTTATTTTTTAGGAAAG and DNASE1L2-6r-ACCACTTAAAAACCTCACTACTCCTC; TRIM14-4f-TTTTTGGGGTATTTTTGGTTTTTA and TRIM14-4r-CTTCCCATTTCTAATAAAACCACCT, TRIM14-8f-ATGTTTGGGTTGGTTTTTTTAATTT and
TRIM14-8r-CCCATCATCAAAACTACAATTTTCT; HOXA6-3fTGGTTTTTAGAAGTTTTTGTTTTTTTG and HOXA6-3r-CCAATCTCCTACCTAAACTAAACCC ; OAT-3f-TGGAATTGGTTTTATGTATAGGAGG and OAT-3r-AAAACACCAAATAACTCCCTACCTT, OAT-4f-TGGAATTGGTTTTATGTATAGGAGG and
OAT-4r-CAACCAAATTAATAATCAAAACACCA.
Transcriptomic profiling
Affymetrix HG U133 plus 2.0 GeneChip microarrays containing .47,000 transcripts were used to determine the transcriptome of subcutaneous adipose tissue by using the standard
manufacturer’s protocol (Affymetrix). Briefly, RNA was reverse
transcribed into cDNA, and in vitro transcription was performed
to generate biotin-labeled cRNA. Biotin-labeled cRNA was then
hybridized to microarrays and stained with streptavidin phycoerythrin. Arrays were scanned on a GeneArray scanner. One
microarray (high-responder group, postcaloric restriction sample) was discarded because of a low signal intensity likely resulting from poor RNA quality. In total, analyses were
performed by using data from 27 microarrays (low-responder
group: 7 arrays before and 7 after caloric restriction; highresponder group: 6 arrays before and 7 after caloric restriction).
Expression values were normalized by using the Robust Multiarray Average algorithm (26), as implemented in FlexArray
software (version 1.1) (27).
Affymetrix HG U133 plus 2.0 GeneChip microarray results
validation
Complementary DNA (cDNA) was generated from 40 ng total
RNA by using a random primer hexamer following the manufacturer’s protocol for Superscript II (Invitrogen, Carlsbad, CA).
Primers and TaqMan probes overlapping the first and second
exons of each of the selected genes were obtained from Applied
Biosystems (Table 3). Each sample was analyzed in duplicate
and calibrated to LRP10 and GAPDH genes (endogenous control; LRP10: Hs00204094_m1; GAPDH: Hs99999905_m1).
Relative quantification estimations were performed on an Applied Biosystems 7500 Real Time PCR System as recommended by the manufacturer (Applied Biosystems, Foster City,
CA), and the DDCT calculation method was used to evaluate
the mean fold expression differences (MFED) between the 2
groups (28).
Statistics
Sample characteristics
Baseline anthropometric and metabolic characteristics and
changes in these characteristics in the low and high responders to
caloric restriction were tested by using an unpaired t test. Correlations were tested by using Spearman rank correlation coefficients; t tests and correlations were performed by using SAS
software, version 9.1.3 (SAS Institute, Cary, NC).
Microarray-data analysis
Significance Analysis of Microarray (SAM) (29) was used to
determine significant differences in DNA methylation (via CpGisland arrays) and expression levels (via expression arrays) between the low- and high-responder groups before and after
dieting. A modified t test was applied to each probe set (log2
transformed), and the raw P values were converted to the false
discovery rate (FDR, q values) to correct for multiple testing
according to Benjamini and Hochberg (30). The same procedure
was applied to analyze the CpG-island validation results. For
gene expression arrays, the significance threshold was set to
detect at least a 1.20-fold change, with an FDR value 0.05.
Because the number of differentially methylated genes identified
was limited on the basis of the same criterion used for the
methylation experiment, a minimum fold-change of 1.15-fold
and an FDR ,0.10 were used instead to establish significant
differences between the groups. In Table 3, Table 4, and Table
5 and in supplementary Tables 1 and 2 (see “Supplemental data”
in the online issue), the low-responder group was the reference
for fold-change computation.
Power estimates for CpG-island and expression microarray
experiments
We assessed the power of the present study sample in SAM
using a test based on the permutation analysis of real data (31).
This analysis estimates the SD of each gene and the overall null
distribution of the genes. The results provide accurate estimates
313
MOLECULAR PATHWAYS INVOLVED IN BODY WEIGHT LOSS
TABLE 3
Significant differentially methylated loci (by false discovery rate) between low and high responders to the caloric restriction intervention1
Gene symbol
Before weight loss: hypomethylated (n = 3)
RAB3C
HYPK
DNASE1L2
Before weight loss: hypermethylated (n = 32)
KCNA3
LHX8
RTKN
WDR5B
CYTL1
PRDM8
C6orf120
CTTNBP2
SLC1A1
TRIM14
TRIM3
CLPB
ITGA7
AMDHD1
BX161511
NDRG4
CYB5A
LYL1
After weight loss: hypomethylated (n = 0)
None
After weight loss: hypermethylated (n = 3)
PLCL4
PRDM8
Probe set
Gene
symbol (5#)
Gene
symbol (3#)
Localization
Fold
change2
UHNhscpg0019055
UHNhscpg0007880
UHNhscpg0028273
FLJ33641
SERINC4
E4F1
PDE4D
MFAP1
DCI
chr5q12
chr15q15
chr16p16
21.28
21.23
21.32
UHNhscpg0027651
UHNhscpg0007328
UHNhscpg0028140
UHNhscpg0022257
UHNhscpg0022498
UHNhscpg0024207
UHNhscpg0024950
UHNhscpg0025147
UHNhscpg0018526
UHNhscpg0028283
UHNhscpg0014466
UHNhscpg0013650
UHNhscpg0004673
UHNhscpg0014893
UHNhscpg0025968
UHNhscpg0026046
UHNhscpg0026076
UHNhscpg0017442
UHNhscpg0022618
UHNhscpg0022695
UHNhscpg0022714
UHNhscpg0026822
UHNhscpg0019513
UHNhscpg0018627
UHNhscpg0023031
UHNhscpg0005471
UHNhscpg0023555
UHNhscpg0027656
UHNhscpg0023806
UHNhscpg0027654
UHNhscpg0004606
UHNhscpg0018857
KCNA2
C1orf171
AJ301580
ZNF691
MAP1LC3C
WDR54
C3orf28
MSX1
ANTXR2
THBS2
HOXA6
FKBP14
CFTR
KIAA1539
GLIS3
C9orf36
NANS
ZFP37
GALNAC4S-6
SPRN
HPX
CLPB
KIRREL3
METTL7B
CCDC38
RPS29
RASGRF1
FLJ13912
FOXL1
C18orf55
NFIX
C20orf74
CD53
SLC44A5
FAF1
SLC2A1
PLD5
ZNHIT4
KPNA1
STK32B
FGF5
PHF10
HOXA7
PLEKHA8
LSM8
UNC13B
C9orf68
MGC21881
CORO2A
SLC31A2
OAT
BC038300
ARFIP2
PDE2A
ETS1
BLOC1S1
HAL
RPL36AL
TMED3
NDRG4
FBXO31
CR749350
TRMT1
C20orf19
chr1p13
chr1p31
chr1p33
chr1p34
chr1q43
chr2p13
chr3q21
chr4p16
chr4q21
chr6q27
chr7p15
chr7p15
chr7q31
chr9p13
chr9p24
chr9q21
chr9q22
chr9q32
chr10q26
chr10q26
chr11p15
chr11q13
chr11q24
chr12q13
chr12q22
chr14q22
chr15q24
chr16q21
chr16q24
chr18q23
chr19p13
chr20p11
UHNhscpg0022169
UHNhscpg0018526
UHNhscpg0025297
PEX10
ANTXR2
MGC13034
PANK4
FGF5
BTF3
chr1p36
chr4q21
chr5q13
q Value
No. of samples
with consistent
differences in the
HWL group
,0.001
,0.001
,0.001
6
6
7
1.24
1.22
1.18
1.17
1.17
1.17
1.16
1.17
1.40
1.18
1.24
1.15
1.21
1.16
1.24
1.37
1.44
1.40
1.27
1.18
1.17
1.22
1.18
1.21
1.17
1.23
1.30
1.32
1.38
1.23
1.16
1.21
0.0602
,0.001
0.0802
0.0621
0.0845
0.0860
0.0802
0.0621
0.0602
0.0602
,0.001
0.0602
0.0621
0.0705
0.0602
0.0892
0.0507
0.0705
,0.001
0.0602
0.0507
0.0705
0.0602
0.0892
0.0802
0.0892
0.0892
0.0602
0.0802
0.0602
0.0621
,0.001
7
6
6
6
6
6
6
6
6
6
6
6
5
6
6
5
6
7
7
6
6
5
7
5
7
5
4
6
7
7
5
6
1.18
1.38
1.38
,0.001
,0.001
,0.001
7
6
6
1
Genes in the vicinity of those identified in the table [gene symbol, gene symbol (5#), and gene symbol (3#) columns] may also have been affected by the
identified DNA methylation changes. HWL, high-weight-loss group. A modified t test was applied as implemented in Significance Analysis of Microarray to
each probe set (log2 transformed), and the raw P values were converted to the false discovery rate (q values) to correct for multiple testing according to
Benjamini and Hochberg (30).
2
The low responders were the reference group for fold-change computation. The change reflects the mean methylation levels computed before and after
weight-loss treatment separately for each locus.
of the power (1-FDR) according to the number of samples. This
approach has the advantage that it is based on a real data set and
takes into account gene correlation. The mean difference in DNA
methylation or mRNA levels used for power computation was
1.4-fold, which we consider to be a reasonable estimate based on
the observed mean differences from the list of significant genes
obtained from SAM.
As estimated by the method proposed by Tibshirani (31), the
power to detect differences in DNA methylation level was .90%
before caloric restriction and .99% after caloric restriction (14
arrays each). For the expression arrays, the power to identify
genes with mean expression changes of 1.4-fold was
.60% before dieting (13 arrays) and .80% after dieting (14
arrays).
314
BOUCHARD ET AL
TABLE 4
Bisulfite-treated DNA analyses of 4 probe set sequences identified by using CpG-island microarray1
Probe set (gene)
UHNhscpg0028273 (DNASE1L2)
UHNhscpg0022618 (OAT)
UHNhscpg0026076 (TRIM14)
UHNhscpg0014466 (HOXA6)
1
2
No. of CpG
(HpaII sites) tested
CpG dinucleotide with
significant MFED
21.32
77 (3)
1.27
1.44
1.24
21 (9)
48 (3)
25 (6)
CpG 15
CpG 16
CpG 17
CpG 21
None
None
MFED array
MFED Sequenom
(FDR)
21.71
21.71
21.71
21.18
—
—
(0)
(0)
(0)
(0)
R2 (P value)
0.71 (0.005)
0.71 (0.005)
0.71 (0.005)
20.74 (0.003)
—
—
MFED, mean-fold expression difference; FDR, false discovery rate.
Spearman correlation coefficient between CpG island microarray and Sequenom (Sequenom Inc, San Diego, CA) results.
Pathway analysis
Pathway analyses allowed us to determine whether genes
found to be differentially expressed belong to predefined networks more than expected by chance alone and help to add
structure to the vast amount of data generated by microarrays.
The Database for Annotation, Visualization, and Integrated
Discovery (DAVID; http://david.abcc.ncifcrf.gov/home.jsp),
a web-based program (32), allowed us to classify our differentially expressed genes into shared biological categories or
chromosomal localization. DAVID computes Fisher’s exact
P values and their resulting FDR (q value).
RESULTS
Effects of caloric restriction
At baseline, the anthropometric and metabolic characteristics
were not significantly different between the low and high responders to caloric restriction (Table 1). By design, the high
responders had stronger overall changes in adiposity than did the
low responders (Table 2). On average, the 2 groups lost 1.81 61.84
and 7.34 63.05 kg (P , 0.01) of fat mass, respectively, whereas
changes in fat-free mass were not significantly different between
groups. Accordingly, changes in percentage body fat showed
very comparable trends, as compared with changes in fat mass.
In accordance with differences in fat mass loss between the
groups, leptin gene expression was down-regulated by 1.24-fold
(FDR = 0.0875) in the high-fat-loss group compared with the
nonresponsive women. No correlation between initial anthropometric and metabolic characteristics and changes in body
composition was observed (percentage body fat; Spearman
correlation coefficient = 0.30, P = 0.30).
As presented in Table 2, body fat loss in response to caloric
restriction was very heterogeneous among the participants,
ranging from a gain of 0.77 kg to a loss of 12.98 kg. There was no
significant difference between the 2 groups in the changes in
blood pressure, lipid profile, and glucose and insulin values after
the intervention.
were relatively hypermethylated in the high-responder group.
For 2 regions (chromosomes 7p15 and 10q26), supportive evidence for differential methylation was provided by 2 adjacent
probe sets located within a range of ,3 Mb. Interestingly, these
2 regions, in addition to a region nominated on chromosome
1p36, were previously identified as human imprinted loci (33).
The most biologically relevant genes identified were the potassium channel, voltage-gated, shaker-related subfamily,
member 3 (KCNA3; 1.24-fold), the glis family zinc finger protein 3 (GLIS3; 1.24-fold), the V-ets avian erythroblastosis virus
E26 oncogene homolog 1 (ETS; 1.18-fold), the nuclear factor
I/X (NFIX; 1.16-fold), and insulinoma-associated 1 (INSM1;
1.21-fold; directly flanking the C20orf74). Other genes located
in the vicinity (’0.5 Mb apart) of the differentially methylated
loci were also of early interest. Of these, corticotropin-releasing
hormone receptor 2 (CHRH2; 1.15-fold) was located at 0.6 Mb
of the PLEKHA8 gene (chromosome 7p15); enoyl-CoA hydratase, short-chain, 1, mitochondrial (ECHS1; 1.18-fold) was
found at 0.05 Mb of the SPRN gene (chromosome 10q26); and
cholecystokinin B receptor (CCKBR; 1.17-fold) was located at
0.2 Mb of the HPX gene (chromosome 11p15).
Epigenomic profile adaptation to caloric restriction
Probes near 3 loci (chromosomes 1p36, 4q21, and 5q13) were
found to be differentially methylated after caloric restriction
(Table 3). For each of these loci, postintervention DNA methylation levels were higher in the high-responder group than in the
low-responder group. Given the postulated role of ectodermalneural cortex 1 gene (ENC1) in obesity (34) and our observation
that it was also differentially expressed in the present study (see
below), the most attractive locus found to be differentially
methylated after caloric restriction was located at 5q13 (1.38fold)—the genomic region to which ENC1 belongs. ENC1 was
found at ’1 Mb from the probe set UHNhscpg0025297 on
chromosome 5q13 (Table 3).
The probe set UHNhscpg0018526, located on chromosome
4q21 (located in the vicinity of PRDM8), was the only one
to demonstrate differential methylation both before and after
intervention.
Epigenomic profiling (CpG-island profiling)
Adipose tissue epigenomic profiling before dieting
CpG-island microarray results validation
Before caloric restriction, 35 loci were differentially methylated (FDR , 0.10) between the groups (Figure 1 and Table 3).
Three loci were relatively hypomethylated in the high-responder
group compared with the low-responder group, whereas 32 loci
We further investigated DNA methylation across several loci
nominated from our epigenomic microarray analyses. A total of
4 CpG dinucleotides within 2 loci showed significant methylation
differences by using the bisulfite-based Sequenom EpiTYPER
315
MOLECULAR PATHWAYS INVOLVED IN BODY WEIGHT LOSS
TABLE 5
After-treatment significant differentially regulated genes (by false discovery rate) with 1.5-fold expression differences
between low and high responders to the caloric restriction intervention1
Gene symbol
UniGene ID
Fold change
PSPH
PHLDA2
SFRP4
CPEB4
hCG_2565
CLEC2B
ERBB4
TNMD
IGHM
ADAMTS18
NA
LHCGR
EGFL6
SFRP4
C9orf19
CCND2
PPP2R1B
PTGER3
USP36
PEMT
ANKRD30A
GCNT2
WASL
PALLD
CBL
PALLD
NA
KRT5
NA
PCDH17
NA
TCF4
QPRT
ENC1
NA
ABCC6
C9orf19
CNKSR3
NA
IQGAP1
GPLD1
KCMF1
CLCN4
TACSTD2
OSBPL11
HIST1H1C
NBN
NA
GPLD1
PDLIM4
GPLD1
NDUFS1
RASGRF2
TXN
NA
DHX9
HIST2H2A
NA
PKP1
Hs.512656
Hs.154036
Hs.658169
Hs.127126
Hs.144151
Hs.85201
Hs.390729
Hs.132957
NA
Hs.188746
Hs.670282
Hs.468490
Hs.12844
Hs.658169
Hs.493819
Hs.376071
Hs.584790
Hs.445000
Hs.464243
Hs.287717
Hs.373787
Hs.519884
Hs.143728
Hs.151220
Hs.504096
Hs.151220
Hs.658612
NA
Hs.650577
Hs.106511
Hs.585479
Hs.644653
Hs.513484
Hs.104925
Hs.134650
Hs.643018
Hs.493819
Hs.16064
Hs.660596
Hs.430551
Hs.591810
Hs.654968
Hs.495674
Hs.23582
Hs.477440
Hs.7644
Hs.492208
Hs.586365
Hs.591810
Hs.424312
Hs.591810
Hs.471207
Hs.162129
Hs.435136
Hs.363526
Hs.74578
Hs.530461
NA
Hs.497350
24.25
22.40
22.07
22.03
22.03
22.01
21.98
21.98
21.97
21.94
21.94
21.87
21.85
21.80
21.78
21.77
21.77
21.75
21.74
21.73
21.71
21.69
21.68
21.67
21.67
21.66
21.66
21.66
21.63
21.63
21.63
21.62
21.62
21.60
21.60
21.60
21.60
21.59
21.58
21.58
21.57
21.57
21.57
21.56
21.55
21.55
21.55
21.54
21.54
21.54
21.53
21.53
21.53
21.53
21.52
21.52
21.52
21.51
21.51
Probe set
205048_s_at
209803_s_at
204052_s_at
224828_at
237339_at
1556209_at
214053_at
220065_at
209374_s_at
230040_at
237356_at
207240_s_at
219454_at
204051_s_at
225604_s_at
200952_s_at
222351_at
210834_s_at
220370_s_at
207621_s_at
223864_at
230788_at
205809_s_at
200907_s_at
229010_at
200906_s_at
238752_at
201820_at
230064_at
205656_at
236297_at
212382_at
204044_at
201340_s_at
238755_at
214033_at
225602_at
227481_at
222877_at
213446_s_at
206265_s_at
222471_s_at
214769_at
202286_s_at
222586_s_at
209398_at
217299_s_at
228987_at
215554_at
211564_s_at
206264_at
236356_at
228109_at
216609_at
235696_at
212107_s_at
214290_s_at
240467_at
221854_at
Chromosomal location
chr7p15.2-p15.1
chr11p15.5
chr7p14.1
chr5q21
chr10p11.21
chr12p13-p12
chr2q33.3-q34
chrXq21.33-q23
chr14q32.33
chr16q23
chr16q23.1
chr2p21
chrXp22
chr7p14.1
chr9p13-p12
chr12p13
chr11q23.2
chr1p31.2
chr17q25.3
chr17p11.2
chr10p11.21
chr6p24.2
chr7q31.3
chr4q32.3
chr11q23.3
chr4q32.3
chr6p22.2
chr12q12-q13
NA
chr13q21.1
chr10p12.31
chr18q21.1
chr16p11.2
chr5q12-q13.3
chr11p15.2
chr16p13.1
chr9p13-p12
chr6q25.2
chr2q33.3
chr15q26.1
chr6p22.3-p22.2
chr2p11.2
chrXp22.3
chr1p32-p31
chr3q21
chr6p21.3
chr8q21
chr8q24.21
chr6p22.3-p22.2
chr5q31.1
chr6p22.3-p22.2
chr2q33-q34
chr5q13
chr9q31
chr8p12
chr1q25
chr1q21.2
NA
chr1q32
(Continued)
316
BOUCHARD ET AL
TABLE 5 (Continued )
Gene symbol
SYNC1
STK38L
TMEM20
TMEM182
GPR62
NPHP3
NA
CTSLL3
BCAP29
BCL2L11
LOC38983
HIF3A
NA
CETP
UniGene ID
Hs.696281
Hs.184523
Hs.632085
Hs.436203
Hs.232213
Hs.511991
Hs.673033
NA
Hs.303787
Hs.469658
Hs.632605
Hs.420830
NA
Hs.89538
Fold change
21.51
21.51
21.51
21.50
1.50
1.50
1.52
1.52
1.54
1.56
1.62
1.64
1.82
2.40
Probe set
221276_s_at
212565_at
239265_at
238867_at
1554559_at
235432_at
1557459_at
1563445_x_at
241640_at
1553096_s_at
227715_at
1555318_at
244181_at
206210_s_at
Chromosomal location
chr1p34.3-p33
chr12p11.23
chr10q23.33
chr2q12.1
chr3p21.1
chr3q22.1
chr11q23.1
chr9q22.1
chr7q22-q31
chr2q13
NA
chr19q13.32
chr5q13.1
chr16q21
1
NA, not available; ID, identification. A modified t test was applied as implemented in Significance Analysis of Microarray to
each probe set (log2 transformed), and the raw P values were converted to the false discovery rate (q values) to correct for multiple
testing according to Benjamini and Hochberg (30). The low responders were the reference group for fold-change computation.
analysis approach. Three of these CpG were located in the
DNASE1L2 probe set (CpG 15, 16, and 17; P = 0.002; FDR = 0%).
DNA methylation across all 3 CpG sites was highly correlated
with one other and positively correlated with the CpG-island
array results (Spearman r = 0.71, P = 0.005). Another CpG
dinucleotide with significant methylation differences between
the groups was located within the OAT probe set (CpG 21: P =
0.002 and FDR = 0%). The results correlated significantly with
the microarray results (Spearman r = 20.74, P = 0.003) (Table
4). The regions tested with bisulfite sequencing are somewhat
arbitrary, and it is likely that the specific critical CpG sites
causing differential enrichment (and thus microarray signal intensity) were not included in all the amplicons tested. However,
the observation that differential methylation was confirmed
within the sequence of 2 out of 4 probe sets tested suggests that
the microarray data reflect true differences. See Supplementary
Figure 1 (under “Supplemental data” in the online issue) for
Spearman correlation coefficients between CpG-island results
and corresponding HpaII-validated restriction sites.
Gene expression profiling
sponsive (24.25-fold). Only 5 other genes showed MFEDs .2,
and these were all found to be up-regulated: PHLDA2 (chromosome 11p15.5), SFRP4 (chromosome 7p14.1), and 3 other
hypothetical genes. Other genes of interest based on biological
function were those encoding phospholipase A2 group 6
(PLA2G6, 1.29-fold), phospholipase C beta 1 and beta 2
(PLCB1, 1.35-fold; PLCB2, 1.29-fold), retinoic acid receptor
gamma (RARg, 1.27-fold), SH2Badaptor protein 2 (SH2B2,
1.39-fold), nitric oxide synthase 1 (NOS1, 1.23-fold), sex hormone–binding globulin (SHBG, 1.29-fold), angiopoietin 2
(ANGPT2, 21.41-fold), tumor necrosis factor receptor–associated
factor 3 (TRAF3, 21.27-fold), oxysterol binding protein-like 11
(OSBPL11, 21.55-fold), glucocorticoid receptor DNA binding
factor 1 (GRLF1, 21.32-fold), and protein inhibitor of activated
STAT, 2 (PIAS2, 21.35-fold).
Interestingly, 5 differentially expressed loci were also found to
be differentially methylated. These loci were located on chromosomes 3q21 (KPNA1), 5q13 (ENC1), 6q27 (C6orf120), 11p15
(HPX), and 15q15 (HYPK). For these loci, DNA methylation
and gene expression were correlated for most of the probe sets
(Table 6).
Adipose tissue transcriptome before and after dieting
Expression microarray results validation
None of the tested genes showed differential adipose tissue
gene expression between the groups before caloric restriction.
However, after caloric restriction, a total of 334 transcripts were
up-regulated (1.2-fold to 2.39-fold), whereas 342 transcripts were
down-regulated (21.2-fold to 24.25-fold) in the high-responder
group as compared with the low-responder group (see supplementary Tables 1 and 2 under “Supplemental data” in the online
issue). The differentially expressed genes with 1.5-fold expression differences are shown in Table 5. These transcripts correspond to 327 and 317 differentially expressed genes in each
group, respectively. The gene encoding cholesteryl ester transfer
protein (CETP; NM_000078; chromosome 16q21) demonstrated
the highest increase in mRNA levels (2.40-fold), and, among
down-regulated genes, that encoding phosphoserine phosphatase
(PSPH; NM_004577; chromosome 7p15) was the most re-
A total of 11 genes showing 1.5-fold mRNA level differences were validated by real-time PCR (Table 7). Seven of these
FIGURE 1. Overview of the study design and of both methylation and
expression microarray results.
317
MOLECULAR PATHWAYS INVOLVED IN BODY WEIGHT LOSS
TABLE 6
Spearman correlation analysis between DNA methylation and expression levels for probe sets with significant differences
with both approaches
Gene
1
KPNA1
ENC1
C6orf120
HPX1
HYPK
1
Localization
Methylation
Expression levels
after intervention
3q21
5q13
6q27
11p15
15q15
Increased at baseline
Increased after treatment
Increased at baseline
Increased at baseline
Decreased at baseline
Down-regulated
Down-regulated
Down-regulated
Up-regulated
Down-regulated
R
P value
20.588
20.689
20.720
0.604
0.720
0.035
0.007
0.006
0.029
0.006
Two probe sets with significant expression level differences.
genes showed comparable MFEDs, with a strong correlation
between microarray and real-time PCR results. TXN also
showed very similar MFEDs with both methods, but the correlation between microarray and real-time PCR results was
modest. Overall, most of the genes tested were validated by RTPCR, which suggests that our microarray data represent true
expression changes.
Pathway analysis
DAVID software (32) was used to identify Gene Ontology
[GO, biological processes (BP)] terms, KEGG metabolic pathways, and chromosomal localizations in which a significant
proportion of differentially expressed genes could be found. We
report here only the most significant findings (P 0.05 and FDR
0.20). For up-regulated transcripts, none of the GO-BP terms
reached this significance level. However, the cerebellar longterm depression pathway was identified as the most significant
KEGG term (P = 0.02, FDR = 0.19). The differentially expressed genes falling into this category were as follows: PLCB1,
PLCB2, PLA2GVI, GNAO1, NOS1, and GUCY1B2. For the
down-regulated genes, protein transport, blood vessel development, and vasculature development GO-BP terms were
identified (P = 0.002, FDR = 0.04; P = 0.006, FDR = 0.10; and
P = 0.006, FDR = 0.10, respectively), but these genes did not
belong to any of the tested KEGG metabolic pathways. The
differentially expressed genes falling into the protein transport
category were as follows: TMEM48, EXOC2, RAB9A, C18orf55,
SRP19, PEX13, CHMP2B, TIMM17A, TLOC1, KPNA1, SCFD2,
RAB2A, SDAD1, PTPN11, NVTF2, RAB1A, RAB3IP, VPS37A,
and PAMC1. Those expressed in blood vessel development and
vasculature development were as follows: PDGFA, LAMA4,
ANGPT1, ANGPT2, C9orf105, and CHD7. Finally, up-regulated
genes were more likely to be located on chromosomes 3q21 (P =
0.007, FDR = 0.12) and 22q11.23 (P = 0.01, FDR = 0.17). These
genes were as follows: ABTB1, IFT122, and MYLK for the 3q21
region and C22orf15, DERL3, GGH, HS.648268, and
203877_AT for the latter region.
DISCUSSION
The aim of this study was to document genome-wide dynamic
adaptations to caloric restriction in terms of cytosine methylation
and transcriptional activity occurring in adipose tissue. Our
hypothesis was that failure to respond satisfactorily to caloric
restriction in terms of fat mass loss could be accounted for by
epigenomic and/or transcriptomic differences. At baseline, we
found that despite various loci being differentially methylated
between low and high responders to caloric restriction, none of
TABLE 7
Spearman correlations between real-time polymerase chain reaction (RT-PCR) and microarray measurements of selected
adipose tissue gene transcripts1
Genes
PSPH
SFRP4
SFRP4
PALLD
PALLD
PALLD
ABCC6
GPLD1
GPLD1
GPLD1
NDUFS1
TXN
BCL2L11
BCL2L11
OSBPL11
PCDH17
CETP
1
Probe set
RT-PCR assay
MFED array
MFED RT-PCR
205048_s_at
204052_s_at
204051_s_at
200907_s_at
200906_s_at
200897_s_at
214033_at
206265_s_at
215554_at
206264_at
236356_at
216609_at
1553096_s_at
1553088_a_at
222586_s_at
205656_at
206210_s_at
Hs00190154_m1
Hs00180066_m1
0.24
0.48
0.56
0.60
0.60
0.68
0.63
0.64
0.65
0.65
0.65
0.66
1.56
1.36
0.64
0.61
2.40
0.98
0.60
0.60
0.72
0.72
0.72
0.75
0.73
0.73
0.73
0.92
0.75
1.00
1.00
0.90
0.64
2.25
Hs00363101_m1
Hs01081201_m1
Hs00412832_m1
Hs00192297_m1
Hs00828652_m1
Hs00197982_m1
Hs00224361_m1
Hs00205457_m1
Hs00163942_m1
MFED, mean-fold expression difference.
R
20.07
0.92
0.94
0.79
0.77
0.74
0.77
0.73
0.91
0.90
0.10
0.46
0.37
0.03
0.65
0.64
0.95
P value
0.81
,0.0001
,0.0001
0.0008
0.001
0.003
0.001
0.003
,0.0001
,0.0001
0.73
0.099
0.19
0.91
0.01
0.01
,0.0001
318
BOUCHARD ET AL
the 47,000 transcripts tested was differentially expressed between
the groups. Recently, in a study designed to verify whether adipose tissue gene expression profiling can differentiate and
predict dietary responders, Mutch et al (35) also reported no
significant differences between the low and high dietary responders (at FDR 5%). Their conclusion was that transcriptomic data were, at best, only a weak predictor of weight
loss after caloric restriction. Our results and conclusions agree
with those reported by Mutch et al.
However, the observation of adipose tissue DNA methylation
differences characterizing each group at baseline suggests that
the epigenomic profile could represent a more promising approach to differentiate low from high responders to caloric restriction. Of the gene loci showing differential DNA methylation,
2 general categories emerged. KCNA3 and NFIX are representative of the first category and are associated with weight control. Knockout (KO) mice for both genes weigh significantly less
than wild-type animals and are resistant to weight gain (36, 37).
INSM1, GLIS3, and CCKBR are representative of the second
category and are associated with diabetes and/or insulin secretion. INSM1-KO mice have impaired pancreatic b cell development and therefore insulin production and secretion (38).
Defects in GLIS3 production are thought to be responsible for
permanent neonatal diabetes and congenital hypothyroidism in
human patients (39). Cholecystokinin (CCK) is a brain and gut
satiety peptide secreted in response to a meal that has the capacity to stimulate insulin secretion through the activation of its
receptors, CCKAR and CCKBR (40). Finally, the differentially
methylated region on chromosome 11p15 is located only 4.5 Mb
from the IGF2-H19 imprinted gene cluster, known to be associated with growth regulation (41, 42). The insulin-like growth
factor 2 gene (IGF2) is normally paternally expressed and has
been associated with eating disorders. Interestingly, a recent
study showed altered DNA methylation across IGF2 in individuals exposed to prenatal famine and low calorific intake
(43). Overall, even if only a small number of gene loci were
found differentially methylated, this approach identified several
strong candidate genes and proved to be potentially useful to
differentiate low from high dietary responders.
After caloric restriction, only 3 chromosomal regions were
differentially methylated between the groups. This strongly
suggests that epigenetic markers are responsive to dieting.
However, there is no obvious explanation for the modest number
of genes that were differentially methylated, but it is possible that
qualitative improvements in the diet, independently of fat mass
loss, may have made both groups more comparable in terms of
methylation patterns. In contrast, adipose tissue gene expression
profiles appear to be strongly responsive to caloric restriction
with a total of 644 genes showing differential transcript levels
after the intervention, compared with none at baseline. The
cholesterol ester transfer protein (CETP; chromosome 16q21) is
1 of the top 3 genes based on MFEDs between the groups and is
of biological relevance to obesity. CETP is well known for its
role in mediating the transfer of cholesterol ester (CE) and triglycerides among plasma lipoproteins (44–46) but has recently
been shown to be associated with lipid metabolism and storage
in adipose tissue (47). The results of a recent study suggest that
higher CETP concentrations in adipocytes facilitate lipid
transport to cellular sites, where access to hydrolytic enzymes is
improved and therefore their utilization in situation of caloric
restriction may be made easier (48). Thus, in our study, the
greater body fat loss in the high-responders group may be attributed, at least in part, to the increased adipose tissue CETP
expression. At the molecular level, no evidence of differential
DNA methylation was found for both probes located in the vicinity of the CETP gene. Some polymorphisms have been
shown to affect CETP plasma concentrations (49, 50) and
weight loss (51), but our study was not designed to test this
specific hypothesis. Further studies are therefore needed to explore the potential dysregulation of CETP cis and trans regulatory elements in weight loss.
Whereas the transcriptome is probably of limited use in differentiating dietary responders before an intervention, gene expression profiling after weight loss is more likely to be useful in
identifying genes and metabolic pathways that have to be activated or repressed to achieve a better fat mass loss. Our pathway
analysis results suggest that down regulation of genes involved in
angiogenesis characterizes high responders to caloric restriction.
Given that angiogenesis dysregulation has been associated with
obesity and its metabolic complications (52), achieving downregulation of these genes in the adipose tissue of obese individuals could offer a better prognosis in terms of fat mass loss
(53). On the other hand, the analyses showed that some of the upregulated genes in the high-responders group were associated
with cerebellar long-term depression pathway. In the cerebellum,
this pathway is associated with memory and learning, and its
long-term depression has been associated with expression
changes for many genes (54). The brain-derived neurotrophic
factor (BDNF) is one of these genes and has been associated
with obesity and weight control in numerous studies (55).
However, it was not found to be differentially expressed in this
study, and the role of this pathway in human adipose tissue remains to be determined. Finally, more up-regulated genes than
would be expected by chance alone were localized on chromosomes 3q21 and 22q11, which suggests that long-range DNA
modifications may have impaired gene expression at these loci.
Interestingly, differential methylation levels were also found for
probes in the chromosome 3q21 region. However, it is not
possible to determine whether the chromosome 3q21 locus
signal corresponds to a larger genomic region than the one
identified by the probe set contained on the microarrays used in
this study.
In this study, DNA methylation and gene expression were
correlated for a limited number of genes. Our data thus concur
with findings from the pilot Human Epigenome Project, which
suggest that the relation between the promoter methylation and
transcription is not necessarily straightforward. Whereas there
was a clear correlation between DNA methylation and gene
expression for some loci, this was not the case for most of the
genes (56). DNA methylation has several other genomic functions in addition to regulating gene expression, including protecting the genome against retroviral elements and playing a key
role in DNA replication, which should be taken into account
when interpreting the results.
A limitation of our study was that only a fraction of the genome
could be interrogated using restriction enzymes. Indeed, DNA
methylation at CpG dinucleotides can be assessed only if they are
located within McrBC [PuCmet(N40–3000)PuCmet] or HpaII (unmethylated CCGG) restriction sites. Therefore, it is plausible
that the remaining fraction of the genome harbors additional
MOLECULAR PATHWAYS INVOLVED IN BODY WEIGHT LOSS
regions showing differential methylation levels. Nevertheless,
this approach has proven to be effective (23) and more sensitive
than other strategies used to identify small DNA methylation
differences (57), such as those expected (44, 58). Moreover, it
has been shown that the methylation levels of neighboring cytosines correlate with one another, expanding the theoretical
genome coverage of this approach. Finally, although the current
sample size provided acceptable statistical power, the number of
samples used was still limited. Accordingly, these results need to
be validated in a larger study.
In summary, our results showed significant epigenomic and
transcriptomic differences in adipose tissue between the high and
low responders to caloric restriction. Furthermore, our data
showed that the epigenetic and transcriptomic profiles differentiating these groups were responsive to the intervention.
Whereas the results suggest that DNA methylation profiling
could be used to predict good responders to dieting, gene expression profiling may enable the identification of genes and
metabolic pathways that have the potential to improve fat mass
loss. Understanding adipose tissue molecular adaptation to caloric restriction may provide an opportunity to understand the
mechanisms put in place to protect the body’s energy reserves
and may thus help to tailor obesity prevention and treatment
interventions.
We express our gratitude to Zachary Kaminsky for his highly helpful advice while preparing DNA samples for epigenomic harm. We also acknowledge the contribution of the Gene Quantification core laboratory of the Centre
de Génomique de Québec, the Microarray Centre at the University Health
Network in Toronto (CpG-island arrays), the McGill University/Génome
Québec Innovation Center (expression arrays), and Bioserve Inc (Benjamin
Winterroth and Rama Modali).
The authors’ responsibilities were as follows—LB, RR-L, MF, LP, and MCV: designed the experiment; LB, MF, and M-EL: collected the data; LB:
performed the analyses and wrote the manuscript; and RR-L, MF, M-EL,
JM, LP, and M-CV: provided significant advice regarding the analyses and
interpretation of the data. All authors reviewed the manuscript. None of
the authors had a financial or personal interest in a company or an organization
that could benefit directly from this research.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
REFERENCES
1. Haslam DW, James WP. Obesity. Lancet 2005;366:1197–209.
2. Flatt JP. Dietary fat, carbohydrate balance, and weight maintenance:
effects of exercise. Am J Clin Nutr 1987;45:296–306.
3. Rodgers A, Ezzati M, Vander Hoorn S, Lopez AD, Lin RB, Murray CJ.
Distribution of major health risks: findings from the Global Burden of
Disease study. PLoS Med 2004;1:e27.
4. National Institutes of Health. Clinical guidelines on the identification,
evaluation, and treatment of overweight and obesity in adults—the evidence report. Obes Res 1998;6(suppl 2):51S–209S.
5. Katz DL, O’Connell M, Yeh MC, et al. Public health strategies for
preventing and controlling overweight and obesity in school and
worksite settings: a report on recommendations of the Task Force on
Community Preventive Services. MMWR Recomm Rep 2005;54:1–12.
6. Klein S, Burke LE, Bray GA, et al. Clinical implications of obesity with
specific focus on cardiovascular disease: a statement for professionals
from the American Heart Association Council on Nutrition, Physical
Activity, and Metabolism: endorsed by the American College of Cardiology Foundation. Circulation 2004;110:2952–67.
7. Hainer V, Zamrazilova H, Spalova J, et al. Role of hereditary factors in
weight loss and its maintenance. Physiol Res 2008;57:S1–15.
8. Bouchard C, Tremblay A, Després JP, et al. The response to exercise
with constant energy intake in identical twins. Obes Res 1994;2:400–10.
9. Hainer V, Stunkard AJ, Kunesova M, Parizkova J, Stich V, Allison DB.
Intrapair resemblance in very low calorie diet-induced weight loss in
27.
28.
29.
30.
31.
32.
33.
34.
35.
319
female obese identical twins. Int J Obes Relat Metab Disord 2000;24:
1051–7.
Petronis A. Human morbid genetics revisited: relevance of epigenetics.
Trends Genet 2001;17:142–6.
Petronis A. Epigenetics and twins: three variations on the theme. Trends
Genet 2006;22:347–50.
Henikoff S, Matzke MA. Exploring and explaining epigenetic effects.
Trends Genet 1997;13:293–5.
Kangaspeska S, Stride B, Metivier R, et al. Transient cyclical methylation of promoter DNA. Nature 2008;452:112–5.
Bird A. DNA methylation patterns and epigenetic memory. Genes Dev
2002;16:6–21.
Ling C, Del Guerra S, Lupi R, et al. Epigenetic regulation of PPARGC1A
in human type 2 diabetic islets and effect on insulin secretion. Diabetologia 2008;51:615–22.
Faraj M, Messier L, Bastard JP, et al. Apolipoprotein B: a predictor of
inflammatory status in postmenopausal overweight and obese women.
Diabetologia 2006;49:1637–46.
Karelis AD, Faraj M, Bastard JP, et al. The metabolically healthy but
obese individual presents a favorable inflammation profile. J Clin Endocrinol Metab 2005;90:4145–50.
Faraj M, Beauregard G, Loizon E, et al. Insulin regulation of gene expression and concentrations of white adipose tissue-derived proteins in vivo
in healthy men: relation to adiponutrin. J Endocrinol 2006;191:427–35.
Faraj M, Beauregard G, Tardif A, et al. Regulation of leptin, adiponectin
and acylation-stimulating protein by hyperinsulinaemia and hyperglycaemia in vivo in healthy lean young men. Diabetes Metab 2008;34:
334–42.
Meugnier E, Faraj M, Rome S, et al. Acute hyperglycemia induces
a global downregulation of gene expression in adipose tissue and skeletal muscle of healthy subjects. Diabetes 2007;56:992–9.
St-Onge M, Mignault D, Allison DB, Rabasa-Lhoret R. Evaluation of
a portable device to measure daily energy expenditure in free-living
adults. Am J Clin Nutr 2007;85:742–9.
Heisler LE, Torti D, Boutros PC, et al. CpG Island microarray probe
sequences derived from a physical library are representative of CpG
Islands annotated on the human genome. Nucleic Acids Res 2005;33:
2952–61.
Mill J, Tang T, Kaminsky Z, et al. Epigenomic profiling reveals
DNA-methylation changes associated with major psychosis. Am J Hum
Genet 2008;82:696–711.
Schumacher A, Kapranov P, Kaminsky Z, et al. Microarray-based DNA
methylation profiling: technology and applications. Nucleic Acids Res
2006;34:528–42.
Ehrich M, Nelson MR, Stanssens P, et al. Quantitative high-throughput
analysis of DNA methylation patterns by base-specific cleavage and
mass spectrometry. Proc Natl Acad Sci USA 2005;102:15785–90.
Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of
normalization methods for high density oligonucleotide array data based
on variance and bias. Bioinformatics 2003;19:185–93.
Blazejczyk M, Miron M, Nadon R. FlexArray: a statistical data analysis
software for gene expression microarrays. Montreal, Canada: Genome
Quebec, 2007.
Livak KJ, Schmittgen TD. Analysis of relative gene expression data
using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
Methods 2001;25:402–8.
Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays
applied to the ionizing radiation response. Proc Natl Acad Sci USA
2001;98:5116–21.
Benjamini Y, Hochberg Y. Controling the false discovery rate: a pratical and
powerfull approach to multiple testing. J R Stat Soc B 1995;57:289–300.
Tibshirani R. A simple method for assessing sample sizes in microarray
experiments. BMC Bioinformatics 2006;7:106.
Dennis G Jr, Sherman BT, Hosack DA, et al. DAVID: Database for
Annotation, Visualization, and Integrated Discovery. Genome Biol
2003;4:3.
Luedi PP, Hartemink AJ, Jirtle RL. Genome-wide prediction of imprinted murine genes. Genome Res 2005;15:875–84.
Zhao L, Gregoire F, Sul HS. Transient induction of ENC-1, a Kelch-related
actin-binding protein, is required for adipocyte differentiation. J Biol Chem
2000;275:16845–50.
Mutch DM, Temanni MR, Henegar C, et al. Adipose gene expression
prior to weight loss can differentiate and weakly predict dietary responders. PLoS One 2007;2:e1344.
320
BOUCHARD ET AL
36. Driller K, Pagenstecher A, Uhl M, et al. Nuclear factor I X deficiency
causes brain malformation and severe skeletal defects. Mol Cell Biol
2007;27:3855–67.
37. Xu J, Koni PA, Wang P, et al. The voltage-gated potassium channel
Kv1.3 regulates energy homeostasis and body weight. Hum Mol Genet
2003;12:551–9.
38. Gierl MS, Karoulias N, Wende H, Strehle M, Birchmeier C. The
zinc-finger factor Insm1 (IA-1) is essential for the development of
pancreatic beta cells and intestinal endocrine cells. Genes Dev 2006;20:
2465–78.
39. Taha D, Barbar M, Kanaan H, Williamson Balfe J. Neonatal diabetes
mellitus, congenital hypothyroidism, hepatic fibrosis, polycystic kidneys, and congenital glaucoma: a new autosomal recessive syndrome?
Am J Med Genet A 2003;122A:269–73.
40. Ahren B, Holst JJ, Efendic S. Antidiabetogenic action of cholecystokinin-8 in type 2 diabetes. J Clin Endocrinol Metab 2000;85:1043–8.
41. Smith AC, Choufani S, Ferreira JC, Weksberg R. Growth regulation,
imprinted genes, and chromosome 11p15.5. Pediatr Res 2007;61:43R–7R.
42. Waterland RA, Lin JR, Smith CA, Jirtle RL. Post-weaning diet affects
genomic imprinting at the insulin-like growth factor 2 (Igf2) locus. Hum
Mol Genet 2006;15:705–16.
43. Heijmans BT, Tobi EW, Stein AD, et al. Persistent epigenetic differences
associated with prenatal exposure to famine in humans. Proc Natl Acad
Sci USA 2008;105:17046–9.
44. Hesler CB, Swenson TL, Tall AR. Purification and characterization of
a human plasma cholesteryl ester transfer protein. J Biol Chem 1987;
262:2275–82.
45. Hesler CB, Tall AR, Swenson TL, Weech PK, Marcel YL, Milne RW.
Monoclonal antibodies to the Mr 74,000 cholesteryl ester transfer protein neutralize all of the cholesteryl ester and triglyceride transfer activities in human plasma. J Biol Chem 1988;263:5020–3.
46. Morton RE, Zilversmit DB. Purification and characterization of lipid
transfer protein(s) from human lipoprotein-deficient plasma. J Lipid Res
1982;23:1058–67.
47. Izem L, Morton RE. Possible role for intracellular cholesteryl ester
transfer protein in adipocyte lipid metabolism and storage. J Biol Chem
2007;282:21856–65.
48. Zhou H, Li Z, Hojjati MR, et al. Adipose tissue-specific CETP expression in mice: impact on plasma lipoprotein metabolism. J Lipid Res
2006;47:2011–9.
49. Lottenberg AM, Nunes VS, Nakandakare ER, et al. The human cholesteryl ester transfer protein I405V polymorphism is associated with
plasma cholesterol concentration and its reduction by dietary phytosterol
esters. J Nutr 2003;133:1800–5.
50. Sandhofer A, Tatarczyk T, Laimer M, et al. The Taq1B-variant in the
cholesteryl ester-transfer protein gene and the risk of metabolic syndrome. Obesity (Silver Spring) 2008;16:919–22.
51. Ruaño G, Windemuth A, Kocherla M, et al. Physiogenomic analysis of
weight loss induced by dietary carbohydrate restriction. Nutr Metab
(Lond) 2006;3:20.
52. Costa C, Incio J, Soares R. Angiogenesis and chronic inflammation:
cause or consequence? Angiogenesis 2007;10:149–66.
53. Rupnick MA, Panigrahy D, Zhang CY, et al. Adipose tissue mass can be
regulated through the vasculature. Proc Natl Acad Sci USA 2002;99:
10730–5.
54. Vigot R. [Cerebellar long-term depression: a mechanism for learning
and memory.] Med Sci (Paris) 2003;19:437–41 (in French).
55. Lebrun B, Bariohay B, Moyse E, Jean A. Brain-derived neurotrophic
factor (BDNF) and food intake regulation: a minireview. Auton Neurosci 2006;126-127:30–8.
56. Eckhardt F, Lewin J, Cortese R, et al. DNA methylation profiling of
human chromosomes 6, 20 and 22. Nat Genet 2006;38:1378–85.
57. Irizarry RA, Ladd-Acosta C, Carvalho B, et al. Comprehensive highthroughput arrays for relative methylation (CHARM). Genome Res
2008;18:780–90.
58. Tobi EW, Lumey LH, Talens RP, et al. DNA methylation differences
after exposure to prenatal famine are common and timing- and sexspecific. Hum Mol Genet 2009;18:4046–53.