Nutritional Epigenetics

Mihai D. Niculescu

Abstract plasticity and, by consequence, influence later health out-

comes, much of the research has been focused on epigenetic
Within the last two decades, significant progress has been mechanisms that could be responsible, in part, for the rela-
made in understanding the importance of epigenetic mecha- tionship between early nutrition and later development of
nisms in the regulation of gene expression as a consequence chronic conditions such as obesity and other metabolic dis-
of gene–environment interactions. Nutrition, among many eases (Gluckman et al. 2011).
other environmental factors, is a key player that can induce epi- Establishment and use of suitable animal models for nu-
genetic changes not only in the directly exposed organisms tritional epigenetics is an essential prerequisite for success-
but also in subsequent generations through the transgenera- ful research on establishing the role of nutrition in the
tional inheritance of epigenetic traits. This article aims to pro- epigenetic regulation of gene expression in normal and
vide insights into the usefulness of the mouse model for pathologic states and on transgenerational inheritance
epigenetic studies involving nutrition as well as the inherent (Niculescu and Lupu 2011). Because of the inherent ethical
limitations when compared with epigenetic phenomena in limitations of human studies, animal models are crucial for
humans. Mice are one of the most versatile models for nutri- deciphering the intimate molecular mechanisms that are re-
tion and epigenetic studies because of several features, such lated to the onset of chronic diseases. Because obesity and
as short life-span, relative low cost for generating samples, related conditions have increasingly become major factors
the existence of well-characterized genetically engineered affecting healthcare costs and loss in the quality of life, the
lines, the detailed sequencing of genomes, and the relative use of animal models for nutrition and epigenetic studies
similarity of their metabolic processes to human metabolism. also helps establish new paradigms that can be used to im-
However, several limitations have to be acknowledged, such prove existing prevention policies (Niculescu 2011).
as the different location of genes on the chromosomes (and Epigenetics refers to molecular events, other than
hence possibly different consequences of some epigenetic al- changes in DNA sequence, that establish heritable patterns
terations), differences in the epigenetic patterns established of gene expression and that are meiotically and mitotically
during late embryogenesis, and possible epigenetic differ- stable (Gravina and Vijg 2010; Skinner 2011). Such epigenetic
ences associated with cellular senescence caused by the dif- mechanisms include DNA methylation, histone modifica-
ferent structure of telomeres when compared with humans. tions, nucleosome positioning along DNA, and the modu-
All these aspects have to be carefully analyzed when decid- lation of gene expression by noncoding RNA (Mohr et al.
ing whether a mouse model should be considered for a study 2011; Niculescu and Lupu 2011; Skinner, 2011).
in nutrition and epigenetics. Consequently, the results ob-
tained from mouse studies should be carefully interpreted
regarding their relevance to humans. DNA Methylation

DNA methylation is a biologic process that consists of the

Introduction covalent addition of methyl groups to DNA (reviewed in
Cucu 2011). In eukaryotes, DNA methylation occurs mainly

T
he last two decades have registered significant prog- at the 5 position of the cytosine ring (5-methylcytosine) that
ress in the understanding of mechanisms involved in is followed by a guanine nucleotide (CpG sites), although
environment–gene interactions and the roles that the methyl groups can also be added to other nucleotides (Cucu
environment has in shaping phenotypic traits. Because nutrition 2011). First hypothesized in 1925 (Johnson and Coghill
was identified as a strong player that can alter development 1925), then confirmed beginning in 1948 (Hotchkiss 1948;
Wyatt 1950), cytosine methylation became the most widely
studied epigenetic modification and the first hallmark of
modern epigenetics (Comings 1972).
Mihai D. Niculescu, MD, PhD, is assistant professor in the Nutrition
Research Institute and Department of Nutrition at the University of North
DNA methylation is catalyzed by DNA methyltransferases
Carolina at Chapel Hill, Kannapolis, North Carolina. (DNMTs1). During cell division, the DNA methylation pattern
Address correspondence and reprint requests to Dr. Mihai D. Niculescu,
Nutrition Research Institute, 500 Laureate Way, Room 2104, Kannapolis, 1Abbreviation that appears ≥3x throughout the article: DNMT, DNA
NC 28081 or email Mihai_Niculescu@unc.edu. methyltransferase.

270 ILAR Journal

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 of the parent DNA strand serves as a template for the meth- several cell cycles (Chen and Riggs 2011). Recently, it has
ylation of the newly synthesized DNA (maintenance DNA been shown that the epigenetics of DNA is not confined to
methylation catalyzed by the maintenance methylase Dnmt1 DNA methylation but extends to the hydroxylation of the
[discussed in Bonifer and Cockerill 2011]). However, the methyl group attached to cytosine (Tahiliani et al. 2009).
methylation process can also occur at previously unmethylated This groundbreaking discovery provided the first plausible
CpG sites (de novo DNA methylation), with this reaction cata- mechanism that could contribute to the previously described
lyzed by Dnmt3a and Dnmt3b, and with the participation of active demethylation of DNA (Mohr et al. 2011) or, alterna-
Dnmt2 and Dnmt3l (Bonifer and Cockerill 2011). tively, a mechanism that could contribute to DNA methyla-
When DNA methylation occurs in promoter regions, it tion fidelity (Williams et al. 2011).
usually associates with gene underexpression and chroma-
tin compaction (Niculescu and Zeisel 2002), but there are
Histone Modifications
also instances when promoter hypermethylation prevents the
binding of inhibitory factors and thus allows for overex- Posttranslational modifications of histones, occurring espe-
pression (Mehedint et al. 2010). Importantly, specific DNA cially at their flexible tail regions, are another category of
methylation patterns contribute decisively to the establish- epigenetic events. Such modifications include, but are not
ment of specific cellular phenotypes, which become stable in limited to, methylation, acetylation, phosphorylation, ubiq-
differentiated cells (Bonifer and Cockerill 2011). uitination, and ADP ribosylation (Hake 2011). In concert with
Mouse models were essential for the discovery of ge- DNA methylation, histone modifications allow for the revers-
nomic imprinting, which allows genes to be expressed in a ible switch between chromatin compaction and relaxation,
parent-of-origin manner (imprinted genes) and thus contrib-
which dictates the degree of access for transcription factors
utes in part to the molecular basis for monoallelic expression
to promoter regions (Bonifer and Cockerill 2011; Hake
(Ferguson-Smith 2011). During embryogenesis and fetal
2011). For instance, methylation of histone H3 at its lysine
morphogenesis, most of the parental DNA methylation patterns
9 and 27 residues (K9 and K27) allows for chromatin com-
are erased by active and passive demethylation mechanisms
paction and, therefore, inhibition of gene expression, whereas
(with the exception of some imprinted regions), and new pat-
trimethylation of H3K4 allows for transcription activation
terns, which are specific for each cell lineage (Kaminen-
(Hake 2011).
Ahola et al. 2011), are established by de novo methylation
There is considerable debate on whether the epigenetic
(Figure 1).
states of histones can be inherited during cell division or
In opposition to DNA methylation, which is always an whether there is a de novo reestablishment of histone epi-
active chemical process, loss of methylation (DNA demeth- genetic status in the daughter cells (Zhu and Reinberg 2011).
ylation) can occur either passively or actively. Passive DNA
demethylation consists of gradual loss of methylation pat-
terns during DNA replication. During the S phase of the Noncoding RNA
cell cycle, the newly synthesized DNA strand is methylated
according to the existing pattern from the parent strand, as MicroRNAs are noncoding RNAs up to 25 nucleotides in
described previously. Failure of maintenance mechanisms of length that regulate gene expression through RNA interfer-
DNA methylation would result in loss of methylation over ence (Sato et al. 2011). MicroRNAs modulate the gene

Figure 1 Methylation events during mouse development

Volume 53, Number 3/4 2012 271

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 expression of several key proteins involved in the epigenetic DNA methylation or chromatin modifications), different re-
machinery responsible for DNA and histone methylation and quirements of how the tissues should be harvested and pre-
acetylation (a few examples are Dnmt3a/b, HDAC1/4, and served require the availability of more than one animal. In
MeCP2) (Sato et al. 2011). Moreover, some microRNAs can the context of nutritional epigenetics and perinatal develop-
also be epigenetically regulated, and, therefore, their gene ment, it is important to have multiple pups that were exposed
expression is highly dependent upon the DNA methylation to the same maternal nutrition cues (from one dam). One
status of their promoters (reviewed in Sato et al. 2011). The example is the experiment design in which tissue from the
main effect of microRNAs on gene expression is gene same organ is needed for multiple assays, which requires dif-
silencing by means of posttranscriptional repression (small ferent chemical treatments specific to each assay (e.g., the
interfering RNA). However, it has been indicated that exog- liver from one pup is required to be harvested for DNA,
enous small interfering RNA targeted to the promoter region whereas another pup has to be infused with a fixation agent
can induce DNA hypermethylation and increased dimethyl- for successful immunohistological testing). The usually
ation of H3K9 (H3K9me2), suggesting that microRNAs equal distribution of male and female pups also allows for
can directly alter the epigenetic status of promoters (Fabbri sex-specific determination of epigenetic alterations under
2011). the same nutritional cues.
Genetic background plays an important role when decid-
ing which mouse strain is best to use, in accordance with the
Types of Mouse Studies in aims and hypothesized mechanisms. A very recent study us-
Nutritional Epigenetics ing seven mouse strains exposed to 1,3-butadiene identified
major epigenetic differences in response to exposure to this
Mouse models have been extensively used in nutrition studies contaminant. When males from seven strains (CAST/EiJ,
in which epigenetic mechanisms were described in details. In NOD/LtJ, A/J, WSB/EiJ, PWK/PhJ, 129S1/SvImJ, and
fact, most of the information about the role of epigenetic C57BL/6J) were exposed to 1,3-butadiene for 2 weeks, global
mechanisms in intrauterine development and in aging is the DNA methylation increased in only three strains (PWK/PhJ,
result of mouse and, to some extent, rat studies (Faulk and 129S1/SvImJ, and C57BL/6J), and LINE1 methylation
Dolinoy 2011; Wiedmeier et al. 2011), with some notable ad- (considered as a proxy measurement of global DNA meth-
ditions on the ontogeny of other species (such as honeybee ylation) increased only in 129S1/SvImJ and C57BL/6J
and Caenorhabditis elegans) (Kaminen-Ahola et al. 2011; males (Koturbash et al. 2011). Interestingly, histone modifi-
Niculescu and Lupu 2011). However, human epidemiologic cations appeared to be only partially correlated to DNA
studies, even when lacking mechanistic approaches (because methylation changes. Trimethylated histones H3K9me3,
of obvious ethical constraints), have recently significantly in- H3K27me3, and H4K20me3 increased only in CAST/EiJ
creased their contribution to understanding the role that epi- males and decreased in C57BL/6J males. In addition, in-
genetic mechanisms have in nutrition-related outcomes creased H3K27me3 was reported only in 129S1/SvImJ
(Feinberg et al. 2010; Heijmans et al. 2008; Pons et al. 2011; males (Koturbash et al. 2011). Although 1,3-butadiene is not
Tobi et al. 2011; Zhu, Hou, et al. 2010). a nutrient, this study demonstrated the importance of the ge-
netic background on environmentally induced epigenetic
alterations. It is plausible that exposure to various nutritional
Advantages of the Mouse Model cues that are already known to induce epigenetic alterations
(see below) would similarly induce different epigenetic out-
The versatility of nutritional epigenetic studies in mice is re-
comes, with potentially different functional consequences
lated to reduced costs and time necessary for the develop-
on gene expression. Other examples point toward the impor-
ment of experiments as well as to biological and genetic
tance of localized mutations that are prone to epigenetic
features (Table 1).
alterations induced by nutrition (the Axin(Fu) and Avy geno-
Short lifespan (up to 3 years) allows for the rapid testing,
types) (Dolinoy 2008; Waterland et al. 2006).
establishment, or development of nutritional experimental
models and paradigms. Specific to nutritional epigenetics
are the short times necessary for studies involving intrauterine Types of Nutrition Studies Involving
development, lactation, or the role of nutrition in epigenetics Epigenetics
and aging, including cognition. In addition, the short life-
span allows investigation of the epigenetic role of nutrition over Studies about the role of specific nutrients or diets in
multiple generations and makes possible a better understand- the epigenetic milieu range from prenatal to postnatal to
ing of the role of nutrition in transgenerational epigenetic aging studies, and some are aimed at identifying transgen-
inheritance, as exemplified by studies involving high-fat diets erational outcomes. In this light, careful consideration of
(Dunn and Bale 2011) experiment design has to include several factors: the homo-
Litter size is an important feature when tissue availabil- geneity of the epigenetic status across mice from same strain,
ity is a limiting factor. Because sound experimental design the age of exposure to nutrients, gender specificity, the poten-
does not stop at only addressing epigenetic changes (either tial influence of postnatal maternal care upon the epigenetic

272 ILAR Journal

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 Table 1 Versatility of mouse models for nutritional epigenetics
Features Components Advantages

Lifespan (1.3–3 y)a Short gestation Allows for rapid generation of samples. Fetal
(18.5–20.5 days)b development can be studied in its dynamics over a
Short lactation short period. Rapid establishment of proof of concept,
(approximately 20 days)a followed by development and refinement of an
Allows studies on aging experimental model. Allows the study of
transgenerational epigenetic inheritance
Litter size Sample size Provides necessary replicates for different types
(up to 10 pups)a of assays
Stochasticityc Allows for the study of stochastic distribution of
epigenetic patterns under the influence of identical
maternal nutrition cues
Genetics Availability of a wide range Pathway-focused epigenetic studies
of genetically engineered
mouse lines
Inbred mice Relatively homogenous individual genomes
(however, genetic drift could influence the
epigenetic component)
Gene ontology highly similar The possibility to generate inference upon the role
to humans of epigenetic mechanisms, with significance to
humans. Building plausibility for further human studies
Genomes unraveled for Essential for epigenetic studies (gene-specific DNA
many strains methylation, chromatin immunoprecipitation,
development of genome-wide platforms)
Controlled genetic backgrounds Interaction between epigenetic status and allelic
diversity (under development)d
Costs Affordable colony maintenance
Diets Well-defined and varied diets Relatively cheap; possible to completely customize
diets at low cost
Assay methods All genome-wide platforms Validated methods; costs are decreasing and new
are available (genome, methods are available
epigenome, transcriptome,
high-throughput sequencing)
Antibodies available for both Possible to validate relevance of gene expression
immunohistochemistry and at protein levels and further study posttranslational
chromatin immunoprecipitation modifications

aMGI Mouse Facts. Available at http://www.informatics.jax.org/mgihome/other/mouse_facts1.shtml. Accessed November 1, 2012.

bMurray et al. 2010.
cFaulk and Dolinoy 2011.
dPhilip et al. 2011.

status in the offspring, and genetic background. For studies were linked with epigenetic outcomes. Methyl-donors
involving transgenerational inheritance of acquired epigen- (betaine, folate, choline), methionine, and related cofac-
etic traits caused by nutrient exposures, the type of imprint- tors (vitamins B12 and B6) were described to induce epi-
ing (paternal or maternal) also has to be considered for genetic modifications at different exposure times and
imprinted genes. ages. Mice fed a folate-deficient diet for 32 weeks exhibited
Studies have applied various nutrients or diets to the global DNA hypomethylation (Linhart et al. 2009). Cho-
mouse models to determine their impact upon the epigenetic line and betaine were reported to alter both global and gene-
status. Table 2 presents several such studies, in which dietary specific DNA methylation (Cdkn3, Calb1), with possible
cues were applied either before and during gestation or post- roles in fetal brain development (Mehedint et al. 2010;
natally, respectively. Niculescu et al. 2006). The epigenetic role of methyl donors
In regards to nutritional epigenetic studies in mouse has also been studied in mouse cancer models (Min mice) as
models, it should be pointed out that the list of nutrients reported by Sibani and colleagues (2002). Specifically, folate
or diets used is very diverse and various mechanisms supplementation has been directly linked with Nat2 promoter

Volume 53, Number 3/4 2012 273

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 Table 2 Examples of nutrition studies with epigenetic outcomes
Exposure
Nutrient/diet period Mouse strain Tissues Outcomes Epigenetic outcomes

Protein Late gestation FVB/NJ Brain in offspring IUGR Ace1 promoter

restriction hypomethylation
(maternal)a
Protein Entire C57BL/6J Fetal liver Function (gene Lxra promoter
restriction gestation (gestation day expression and hypermethylation
(maternal)b 19.5) epigenetics)
Folate Before and C57BL/6J Fetal gut DNA methylation Slc394a hypomethylation
deficiencyc during (gestation day
gestation 17.5)
Folate Young adult C57BL/6J Colon DNA methylation, Trend for decreased
deficiency gene expression genomic DNA
and ethanold methylation. Folate is
potential modulator
for ethanol-induced
outcomes.
Choline Late gestation C57BL/6J Fetal brain Brain development Cdkn3 hypomethylation,
deficiencye,f (gestation Calb1 promoter
day 17), neuron hypermethylation;
precursors H3K9me1, H3K9me2
hypomethylation
Genisteing,h Gestation 129/SvJ:C57BL/6J Adult and fetal Hematopoietic Hypomethylation of
blood, fetal liver, alterations repetitive elements in
bone marrow bone marrow
Avy Endodermal, Obesity prevention Avy IAP hypermethylation
mesodermal,
and ectodermal
tissues
High-fat dieti,j Prebreeding 129/SvJ:C57BL/6J Liver Body weight, glucose Persistence of
(F0 generation) (F3 generation) metabolism, increased body
imprinted genes weight in F3 females
Early gestation DBA/2:C57BL/6J Gestation day Imprinted genes (by means of paternal
15.5 placenta, lineage)
various
fetal tissues Hypo- and
hypermethylation of
some imprinted genes,
sexual dimorphism

IAP, intracisternal A particle; IUGR, intrauterine growth restriction; F0, parent generation; F3, third generation of progeny.
aGoyal et al. 2010.
bvan Straten et al. 2010.
cMcKay et al. 2011.
dSauer et al. 2010.
eNiculescu et al. 2006.
fMehedint et al. 2010.
gVanhees et al. 2011.
hDolinoy et al. 2006.
iDunn and Bale 2011.
jGallou-Kabani et al. 2010.

hypermethylation and subsequent decrease in gene expres- The epigenetic role of retinoids (reviewed in Vieira
sion (Wakefield et al. 2010). Other epigenetic roles of methyl 2011) was recently validated in a mouse model (Apc mice),
donors were linked to the reversal of DNA methylation in which Rxr␣ promoter was hypomethylated in mice fed
changes induced by bisphenol A (Dolinoy et al. 2007). green tea (Volate et al. 2009).

274 ILAR Journal

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
by guest
on 09 March 2018
Table 3 Recent studies on the relationship between epigenetics and obesity and metabolic syndrome in mouse models

Volume 53, Number 3/4

Nutrient/diet Exposure period Mouse strain Tissues Outcomes Epigenetic outcomes

2012
High-fat dieta Postweaning C57BL/6J:DBA/2J Reward-related brain regions Obesity µ-opioid receptor (MOR)
promoter hypermethylation
and increased MeCP2
binding; increased H3K9
methylation and decreased
H3 acetylation
High-fat dietb Early gestation C57BL/6J:DBA/2J Gestation day 15.5 placenta, Imprinted genes Hypo- and hypermethylation
various fetal tissues of some imprinted genes,
sexual dimorphism
High-fat dietc Before gestation C57BL/6J:DBA/2J Reward-related Obesity DAT, MOR, and
through lactation brain regions PENK promoter DNA

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123

hypomethylation
High-fat dietd Long-term exposure BFMI and B6 Postnatal brain Obesity Mc4r exon hypomethylation
Cyclophosphamide-induced 8 wk old NOD Lymphocytes Induced diabetes; 5-Aza-2’-deoxycytydine–
diabetes in nonobese micee 5-Aza-2’-deoxycytydine induced Foxp3
modulation hypomethylation in the
first intron.
High-fat dietf Not specified C57BL/6J ob/ob Adipose Dnmt3a related Dnmt3a overexpression; Sfrp1
hypomethylation in a
Dnmt3a transgenic mouse.

aVucetic et al. 2011.

bGallou-Kabani et al. 2010.
cVucetic et al. 2010.
d Widiker et al. 2010.
eZheng et al. 2009.
f Kamei et al. 2009.

275
 Isoflavones (flavonoids) are a class of plant compounds division (Cerny and Quesenberry 2004). However, in the
with estrogenic activity. With the exception of studies on human late blastocyst stage, different types of cells are dif-
genistein, only few mouse studies are available. Maybe the ferentially methylated: whereas the inner cell mass is less
most relevant research in this area is the role of genistein in the methylated, the trophectodermal (TE) cells are heavily meth-
reversal of epigenetic changes induced by bisphenol A in the ylated to levels similar to that for the two- to four-cell stage
agouti model (Dolinoy et al. 2006, 2007) (Table 2). Another (Fulka et al. 2004). These patterns are different in the mouse
study indicated that soy isoflavones, administered perina- embryo (inverse relationship between trophectoderal cells and
tally, induced Acta1 promoter hypomethylation in C3H inner cell mass) (Fulka et al. 2004), suggesting that the use of
mice (Guerrero-Bosagna et al. 2008). mouse models may have important limitations (Figure 2).
Isothiocyanates, found in cruciferous vegetables, have Other differences exist in the epigenetic regulation of
become of interest lately for nutritionally induced epigenetic various molecular processes. One example is the difference
modifications, specifically upon histone modifications. in the expression of telomerase (Tert), required for telomere
Sulforaphane (found in such vegetables as broccoli and length maintenance. Whereas telomere shortening in hu-
broccoli sprouts) inhibited histone deacetylase activity in mans serves as a mitotic clock and directs the senescence
mouse prostate and peripheral blood mononuclear cells and process, this mechanism does not appear to be critical in the
inhibited spontaneous intestinal polyps in Apcmin mice mouse (Forsyth et al. 2005). The chromatin environment
(Myzak, Dashwood, et al. 2006; Myzak, Hardin, et al. 2006). also differs between human (hTERT) and mouse (mTert),
(-)-Epigallocatechin-3-gallate (present in green tea) was with the human gene located within a more condensed chro-
protective against ultraviolet B–induced global DNA hypo- matin than the mouse one (Zhu, Zhao, et al. 2010). These dif-
methylation and inhibited the malignant transformation of ferences point toward obvious limitations when considering
ultraviolet B–induced papillomas to carcinomas in the SKH-1 the relevance of mouse studies in the epigenetic regulation
hairless mouse model (Mittal et al. 2003). of telomeric length and cellular senescence.
Epigenetic involvement in obesity and metabolic syn- Other challenges (discussed below), which are not lim-
drome has become of great interest because of the hypothe- ited only to the mouse model, are the result of the extraordi-
sized (and largely confirmed) role of early life nutrition in nary progress in understanding the epigenetic mechanisms
the outset of chronic disease later in life and in aging (devel- brought by the most recent discoveries, like hydroxymeth-
opmental origins of health and disease theory) (Gluckman ylation and its relationship with active demethylation (see
et al. 2011; Niculescu and Lupu 2011). The most-used diets “DNA Methylation”).
were the high-fat diets administered either before and during Because epigenetic phenomena are dynamic during the
gestation or postnatally. Table 3 presents the latest scientific life cycle of mammals, the selection of the proper exposure
developments in this field, limited to mouse models only. period to nutritional cues is essential for the successful
accomplishment of epigenetic studies. Such an example is
the use of the same mouse model in two different studies
Challenges and Limitations of the (see Table 3) in which opposite epigenetic outcomes were
Mouse Model
Although mouse models are very versatile for studying
the relationship between nutrition and epigenetics, several
limitations, as well as newly discovered challenges, which
sometimes limit the relevance of this model to human epi-
genetics, exist. Nonetheless, mouse and rats remain the most
used models in nutrition studies. The mouse model, as opposed
to the rat model, benefits from better support in the availability
of assay platforms (especially arrays) and from numerous
transgenic mouse lines.
One limitation consists in the relatively different pattern of
DNA methylation in the late mouse embryo compared with the
human embryo. Following implantation, in both humans and
mice, embryonic stem cells are subjected to de novo methyla-
tion catalyzed by Dnmt3a and Dnmt3b genes (de novo methyl-
ases), with the exception of tissue-specific genes that remain
Figure 2 Methylation patterns in human and mouse early embryo-
unmethylated (Cerny and Quesenberry 2004; Paulsen and
genesis. Global DNA methylation decreases from the two-cell stage
Ferguson-Smith 2001). Once the new pattern is established (it to morula in both human and mouse embryos. During blastocyst
can, however, be further modulated during fetal development), development, overall methylation levels increase to levels similar
cell proliferation is paralleled by the maintenance of methyla- to those for the two-cell stage, but the methylation pattern in the
tion. This is catalyzed by Dnmt1, which restores the methyla- human embryo differs from that in the mouse embryo. Bl, blasto-
tion pattern of the newly synthesized DNA in the S phase of cell cyst; ICM, inner cell mass; TE, trophectodermal cells.

276 ILAR Journal

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 reported for the same gene (MOR) according to the exposure Cerny J, Quesenberry PJ. 2004. Chromatin remodeling and stem cell theory
period. When a high-fat diet was administered prior to gesta- of relativity. J Cell Physiol 201:1-16.
Chen ZX, Riggs AD. 2011. DNA methylation and demethylation in mam-
tion, during gestation, and continuing in lactation, the µ-opioid mals. J Biol Chem 286:18347-18353.
receptor promoter was hypomethylated in specific brain re- Comings DE. 1972. Methylation of euchromatic and heterochromatic
gions associated with reward (Vucetic et al. 2010). However, DNA. Exp Cell Res 74:383-390.
when the high-fat diet was administered to the pups only Cucu NV. 2011. DNA methylation. In: Niculescu MD, Haggarty P, eds.
after weaning, hypermethylation of the µ-opioid receptor Nutrition in Epigenetics. 1st ed. Ames IA: John Wiley & Sons.
Dolinoy DC. 2008. The agouti mouse model: An epigenetic biosensor for
promoter was reported in the same areas (Vucetic et al. nutritional and environmental alterations on the fetal epigenome. Nutr
2011). These opposite results point toward the importance Rev 66:S7-S11.
of different development periods for the establishment of Dolinoy DC, Huang D, Jirtle RL. 2007. Maternal nutrient supplementation
nutrient-driven epigenetic patterns. counteracts bisphenol A-induced DNA hypomethylation in early devel-
Another challenge is the hypothesized connection between opment. PNAS 104:13056-13061.
Dolinoy DC, Weidman JR, Waterland RA, Jirtle RL. 2006. Maternal genis-
genetics and epigenetics, with transgenerational consequences. tein alters coat color and protects Avy mouse offspring from obesity by
An interesting example is the study by Nelson and colleagues modifying the fetal epigenome. Environ Health Perspect 114:567-72.
(2010) in which genetic differences in the noninherited, pater- Dunn GA, Bale TL. 2011. Maternal high-fat diet effects on third-generation
nal chromosome Y led to phenotypic alterations in the female female body size via the paternal lineage. Endocrinology 152:2228-2236.
progeny (which did not inherit the chromosome Y). The au- Fabbri M. 2011. Roles of RNAi and other micro-RNAs in the regulation of
epigenetic processes. In: Niculescu MD, Haggarty P, eds. Nutrition in
thors suggested that paternal epigenetic alterations, which Epigenetics. 1st ed. Ames IA: John Wiley & Sons.
were acquired as a result of the aforementioned genetic defect Faulk C, Dolinoy DC. 2011. Timing is everything: The when and how of
in the parent, could be inherited by the female offspring even in environmentally induced changes in the epigenome of animals. Epi-
the absence of the female progeny of the causal genetic defect, genetics 6:791-797.
but there is not enough evidence yet to support this hypothesis Feinberg AP, Irizarry RA, Fradin D, AryeeMJ, Murakami P, Aspelund T,
Eiriksdottir G, Harris TB, Launer L, Gudnason V, Fallin MD. 2010.
at the present, nor is their evidence that nutrition could play a Personalized epigenomic signatures that are stable over time and covary
role in this type of hypothesized epigenetic alteration. with body mass index. Sci Transl Med 2:49ra67.
Given these limitations and challenges, the following Ferguson-Smith AC. 2011. Genomic imprinting: The emergence of an epi-
questions should be first addressed when designing a nutri- genetic paradigm. Nat Rev Genet 12:565-575.
tion experiment with an epigenetic component: Forsyth NR, Elder FF, Shay JW, Wright WE. 2005. Lagomorphs (rabbits,
pikas and hares) do not use telomere-directed replicative aging in vitro.
Mech Ageing Dev 126, 685-691.
1. How relevant is a mouse model in comparison with Fulka H, Mrazek M, Tepla O, Fulka J Jr. 2004. DNA methylation pattern in
humans for that particular metabolic, molecular, or human zygotes and developing embryos. Reproduction 128:703-708.
developmental process? Gallou-Kabani C, Gabory A, Tost J, Karimi M, Mayeur S, Lesage J,
2. Which tissue(s) is (are) the most relevant in regard to Boudadi E, Gross M-S, Taurelle J, Vigé A, Breton C, Reusens B,
hypothesized outcome? Remacle C, Vieau D, Ekström TJ, Jais J-P, Junien C. 2010. Sex- and
diet-specific changes of imprinted gene expression and DNA methyla-
3. Which mouse strain or transgenic model (if available) tion in mouse placenta under a high-fat diet. PLoS ONE 5:e14398.
would be best suited? Gluckman PD, Hanson MA, Low FM. 2011. The role of developmental
4. What should be the period of exposure? plasticity and epigenetics in human health. Birth Defects Res C Embryo
5. If relevant, which parental lineage (maternal vs. paternal) Today 93:12-18.
should be exposed to the nutritional cue? Goyal R, Goyal D, Leitzke A, Gheorghe CP, Longo LD. 2010. Brain renin-
angiotensin system: Fetal epigenetic programming by maternal protein
restriction during pregnancy. Reprod Sci 17:227-238.
Gravina S, Vijg J. 2010. Epigenetic factors in aging and longevity. Pflugers
Conclusion Arch 459:247-258.
Guerrero-Bosagna C, Sabat P, Valdovinos F, Valladares L, Clark S. 2008.
Nutritional epigenetics is a complex field with multiple chal- Epigenetic and phenotypic changes result from a continuous pre and
lenges. The mouse model is a convenient alternative to human post natal dietary exposure to phytoestrogens in an experimental popu-
models because of its low cost, well-defined molecular and lation of mice. BMC Physiology 8:17.
Hake SB. 2011. Chromatin modifications. In: Niculescu MD, Haggarty P,
physiologic mechanisms, short lifespan, and tissue accessibil-
eds. Nutrition in Epigenetics. 1st ed. Ames IA: John Wiley & Sons.
ity. Although the epigenetic mechanisms in the mouse have Heijmans BT, Tobi EW, Stein AD, Putter H, Blauw GJ, Susser ES,
contributed greatly to the understanding of human epigenetics, Slagboom PE, Lumey LH. 2008. Persistent epigenetic differences as-
several challenges and limitations exist. Accordingly, the use sociated with prenatal exposure to famine in humans. PNAS 105:17046-
of mouse models in nutrition studies with epigenetic outcomes 17049.
Hotchkiss RD. 1948. The quantitative separation of purines, pyrimidines,
should be carefully considered at the early stages of experi-
and nucleosides by paper chromatography. J Biol Chem 175:315-32.
ment design and hypothesis testing. Johnson TB, Coghill RD. 1925. Researches on pyrimidines. C111. The dis-
covery of 5-methyl-cytosine in tuberculinic acid, the nucleic acid of the
tubercle bacillus 1. J Am Chem Soc 47:2838-2844.
References Kamei Y, Suganami T, Ehara T, Kanai S, Hayashi K, Yamamoto Y, Miura S,
Ezaki O, Okano M, Ogawa Y. 2009. Increased expression of DNA meth-
Bonifer C, Cockerill PN. 2011. Chromatin mechanisms regulating gene yltransferase 3a in obese adipose tissue: Studies with transgenic mice.
expression in health and disease. Adv Exp Med Biol 711:12-25. Obesity 18:314-321.

Volume 53, Number 3/4 2012 277

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018
 Kaminen-Ahola NJ, Ahola AI, Whitelaw E. 2011. Epigenetic inheritance: tion and folate are determinants of genomic DNA methylation, p16 pro-
Both mitotic and meiotic. In: Niculescu MD, Haggarty P, eds. Nutrition moter methylation and the expression of p16 in the mouse colon. Br J
in Epigenetics. 1st ed. Ames IA: John Wiley & Sons. Nutr 104:24-30.
Koturbash I, Scherhag A, Sorrentino J, Sexton K, Bodnar W, Swenberg JA, Sibani S, Melnyk S, Pogribny IP, Wang W, Hiou-Time F, Deng L, Trasler J,
Beland FA, Pardo-Manuel Devillena F, Rusyn I, Pogribny IP. 2011. Epi- James SJ, Rozen R. 2002. Studies of methionine cycle intermediates
genetic mechanisms of mouse inter-strain variability in genotoxicity of (SAM, SAH), DNA methylation and the impact of folate deficiency on
the environmental toxicant 1,3-butadiene. Tox Sci 122:448-456. tumor numbers in Min mice. Carcinogenesis 23:61-65.
Linhart HG, Troen A, Bell GW, Cantu E, Chao WH, Moran E, Steine E, He Skinner MK. 2011. Role of epigenetics in developmental biology and trans-
T, Jaenisch R. 2009. Folate deficiency induces genomic uracil misincor- generational inheritance. Birth Defects Res C Embryo Today 93:51-55.
poration and hypomethylation but does not increase DNA point muta- Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y,
tions. Gastroenterology 136:227-235 e3. Agarwal S, Iyer LM, Liu DR, Aravind L, Rao A. 2009. Conversion of
McKay JA, Wong YK, Relton CL, Ford D, Mathers JC. 2011. Maternal fo- 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by
late supply and sex influence gene-specific DNA methylation in the fetal MLL partner TET1. Science 324:930-935.
gut. Mol Nutr Food Res 55:1717-1723. Tobi EW, Heijmans BT, Kremer D, Putter H, Delemarre-van de Waal HA,
Mehedint MG, Niculescu MD, Craciunescu CN, Zeisel SH. 2010. Choline Finken MJ, Wit JM, Slagboom PE. 2011. DNA methylation of IGF2,
deficiency alters global histone methylation and epigenetic marking at GNASAS, INSIGF and LEP and being born small for gestational age.
the Re1 site of the calbindin 1 gene. FASEB J 24:184-195. Epigenetics 6:171-176.
Mittal A, Piyathilake C, Hara Y, Katiyar SK. 2003. Exceptionally high van Straten EME, Bloks VW, Huijkman NCA, Baller JFW, Meer HV,
protection of photocarcinogenesis by topical application of (–)- Lutjohann D, Kuipers F, Plosch T. 2010. The liver X-receptor gene pro-
epigallocatechin-3-gallate in hydrophilic cream in SKH-1 hairless mouse moter is hypermethylated in a mouse model of prenatal protein restric-
model: Relationship to inhibition of UVB-induced global DNA hypo- tion. Am J Physiol Regul Integr Comp Physiol 298:R275-282.
methylation. Neoplasia 5:555-565. Vanhees K, Coort S, Ruitjers EJB, Godschalk RWL, van Schooten FJ, van
Mohr F, Dohner K, Buske C, Rawat VP. 2011. TET genes: New players in Doorn-Khosrovani SBVW. 2011. Epigenetics: Prenatal exposure to ge-
DNA demethylation and important determinants for stemness. Exp nistein leaves a permanent signature on the hematopoietic lineage.
Hematol 39:272-281. FASEB J 25:797-807.
Murray SA, Morgan JL, Kane C, Sharma Y, Heffner CS, Lake JU, Donahue Vieira A. 2011. Epigenetic regulation by retinoids. In: Niculescu MD, Haggarty
LR. 2010. Mouse gestation length is genetically determined. PLoS One P, eds. Nutrition in Epigenetics. 1st ed. Ames IA: John Wiley & Sons.
5:e12418. Volate SR, Muga SJ, Issa AY, Nitcheva D, Smith T, Wargovich MJ. 2009.
Myzak MC, Dashwood WM, Orner GA, Ho E, Dashwood RH. 2006. Sul- Epigenetic modulation of the retinoid X receptor ␣ by green tea in the
foraphane inhibits histone deacetylase in vivo and suppresses tumori- azoxymethane-ApcMin/+ mouse model of intestinal cancer. Mol Carcinog
genesis in Apcmin mice. FASEB J 20(3):506-508. 48:920-933.
Myzak MC, Hardin K, Wang R, Dashwood RH, Ho E. 2006. Sulforaphane Vucetic Z, Kimmel J, Reyes TM. 2011. Chronic high-fat diet drives postna-
inhibits histone deacetylase activity in BPH-1, LnCaP and PC-3 pros- tal epigenetic regulation of [mu]-opioid receptor in the brain. Neuropsy-
tate epithelial cells. Carcinogenesis 27:811-819. chopharmacology 36:1199-1206.
Nelson VR, Spiezio SH, Nadeau JH. 2010. Transgenerational genetic ef- Vucetic Z, Kimmel J, Totoki K, Hollenbeck E, Reyes TM. 2010. Maternal
fects of the paternal Y chromosome on daughters’ phenotypes. Epig- high-fat diet alters methylation and gene expression of dopamine and
enomics 2:513-521. opioid-related genes. Endocrinology 151:4756-4764.
Niculescu MD. 2011. Epigenetic transgenerational inheritance: Should Wakefield L, Boukouvala S, Sim E. 2010. Characterisation of CpG meth-
obesity-prevention policies be reconsidered? Synesis 2:G18-G26. ylation in the upstream control region of mouse Nat2: Evidence for a
Niculescu MD, Craciunescu CN, Zeisel SH. 2006. Dietary choline defi- gene-environment interaction in a polymorphic gene implicated in
ciency alters global and gene-specific DNA methylation in the develop- folate metabolism. Gene 452:16-21.
ing hippocampus of mouse fetal brains. FASEB J 20:43-49. Waterland RA, Dolinoy DC, Lin JR, Smith CA, Shi X, Tahiliani KG. 2006.
Niculescu MD, Lupus DS. 2011. Nutritional influence on epigenetics and Maternal methyl supplements increase offspring DNA methylation at
effects on longevity. Curr Opin Clin Nutr Metab Care 14:35-40. Axin fused. Genesis 44:401-406.
Niculescu MD, Zeisel SH. 2002. Diet, methyl donors and DNA methyla- Widiker S, Kärst S, Wagener A, Brockmann G. 2010. High-fat diet leads to
tion: Interactions between dietary folate, methionine and choline. J Nutr a decreased methylation of the Mc4r gene in the obese BFMI and the
132:2333S-2335S. lean B6 mouse lines. J Appl Genet 51:193-197.
Paulsen M, Ferguson-Smith AC. 2001. DNA methylation in genomic im- Wiedmeier JE, Joss-Moore LA, Lane RH, Neu J. 2011. Early postnatal nu-
printing, development, and disease. J Pathol 195:97-110. trition and programming of the preterm neonate. Nutr Rev 69:76-82.
Philip VM, Sokoloff G, Ackert-Bicknell CL, Striz M, Branstetter L, Williams K, Christensen J, Pedersen MT, Johansen JV, Cloos PA,
Beckmann MA, Spence JS, Jackson BL, Galloway LD, Barker P, Rappsilber J, Helin K. 2011. TET1 and hydroxymethylcytosine in tran-
Wymore AM, Hunsicker PR, Durtschi DC, Shaw GS, Shinpock, Manly KF, scription and DNA methylation fidelity. Nature 473:343-348.
Miller DR, Donohue KD, Culiat CT, Churchill GA, Lariviere WR, Wyatt GR. 1950. Occurrence of 5-methylcytosine in nucleic acids. Nature
Palmer AA, O’Hara BF, Voy BH, Chesler EJ. 2011. Genetic analysis in the 166:237-238.
collaborative cross breeding population. Genome Res 21:1223-1238. Zheng Q, Xu Y, Liu Y, Zhang B, Li X, Guo F, Zhao Y. Induction of Foxp3
Pons D, Trompet S, De Craen AJ, Thijssen PE, Quax PH, De Vries MR, Wierda demethylation increases regulatory CD4 T cells and prevents the occur-
RJ, van den Elsen PJ, Monraats PS, Ewing MM, Heijmans BT, Slagboom rence of diabetes in mice. J Mol Med 87:1191-1205.
PE, Zwinderman AH, Doevendans PA, Tio RA, De Winter RJ, De Maat Zhu B, Reinberg D. 2011. Epigenetic inheritance: Uncontested? Cell Res
MP, Iakoubova OA, Sattar N, Shepherd J, Westendorp RG, Jukema JW. 21:435-441.
2011. Genetic variation in PCAF, a key mediator in epigenetics, is associ- Zhu J, Zhao Y, Wang S. 2010. Chromatin and epigenetic regulation of the
ated with reduced vascular morbidity and mortality: Evidence for a new telomerase reverse transcriptase gene. Protein Cell 1:22-32.
concept from three independent prospective studies. Heart 97:143-150. Zhu Z-Z, Hou L, Bollati V, Tarantini L, Marinelli B, Cantone L, Yang AS,
Sato F, Tsuchiya S, Meltzer SJ, Shimizu K. MicroRNAs and epigenetics. Vokonas P, Lissowska J, Fustinoni S, Pesatori AC, Bonzini M, Apostoli
FEBS J 278:1598-1609. P, Costa G, Bertazzi PA, Chow W-H, Schwartz J, Baccarelli A. 2010.
Sauer J, Jang H, Zimmerly EM, Kim KC, Liu Z, Chanson A, Smith DE, Predictors of global methylation levels in blood DNA of healthy sub-
Mason JB, Friso S, Choi SW. 2010. Ageing, chronic alcohol consump- jects: A combined analysis. Int J Epidemiol 41:126-139.

278 ILAR Journal

Downloaded from https://academic.oup.com/ilarjournal/article-abstract/53/3-4/270/654123
by guest
on 09 March 2018