Short CourSe I

yalechurlishAI and Robotics

Nov 7, 2013 (3 years and 5 months ago)




Short CourSe I
Epigenetic Control and Neuronal Function
organized by Paolo Sassone-Corsi, PhD
Short Course I
Epigenetic Control and Neuronal Function
Organized by Paolo Sassone-Corsi, PhD



Please cite articles using the model:
[CHAPTER TITLE] In: Epigenetic Control and Neuronal Function.
(Sassone-Corsi P, ed) pp. [xx-xx]. Washington, DC: Society for Neuroscience.
All articles and their graphics are under the copyright of their respective authors.
Cover graphics and design © 2009 Society for Neuroscience.
Table of Contents
Epigenetic Inheritance and Replicative Cellular Aging
Shelley L. Berger, PhD ...........................................................1
Linking Metabolism to Epigenetics: Chromatin Remodeling and Circadian Control
Paolo Sassone-Corsi, PhD .........................................................7
Functional Packaging of Chromatin into Loops
Terumi Kohwi-Shigematsu, PhD, Shutao Cai, PhD, Yoshinori Kohwi, PhD ................17
The Epigenetic Impact of Early Life Adversity
Moshe Szyf, PhD ......................................................27
Regulation of Histone Deacetylases: the Role of Nitric Oxide
Alexi Nott, PhD, Antonella Riccio, MD, PhD ........................................35
The Role of Chromatin-Modifying Enzymes in Long-term Memory Processes
Marcelo A. Wood, PhD .........................................................45
Small-Molecule Probes of Chromatin-Mediated Neuroplasticity:
Histone Deacetylase Inhibitors as Memory Enhancers
Stephen J. Haggarty, PhD ........................................................55
Neurons are cells submitted to an exceptional variety of stimuli that are able to convert them into
high-order functions, such as storing memories, controlling behavior, and governing consciousness.
These unique properties are based on highly plastic processes, which are intimately dependent on the
complex molecular machinery that controls gene expression. Evidence is accumulating that neuronal
functions have more than a solely genetic basis. Epigenetic control, which largely involves events of
chromatin remodeling, appears to govern some of the more distinctive features of neuronal responses,
guiding, for example, dynamic plasticity and long-lasting cellular memory.
Understanding the molecular pathways of chromatin transitions in neurons is therefore critical
because it will provide fundamental insights into how plasticity is achieved. What are the epigenetic
pathways leading to specific responses in neurons? The presence of an “epigenetic indexing code” has
been postulated, but how this may operate still needs to be elucidated.
The purpose of this course is to provide neuroscientists with the conceptual basis of epigenetics and
chromatin remodeling and to discuss how central findings accumulating at an exponential rate in the
field are changing our perspective of brain function.
Course Organizer: Paolo Sassone-Corsi, PhD, Department of Pharmacology, School of Medicine,
University of California, Irvine. Faculty: Shelley L. Berger, PhD, Department of Cell and Developmental
Biology, University of Pennsylvania; Stephen J. Haggarty, PhD, Department of Neurology, Harvard
Medical School, Center for Human Genetic Research, Massachusetts General Hospital, and Stanley
Center for Psychiatric Research, Broad Institute of Harvard and Massachusetts Institute of Technology;
Terumi Kohwi-Shigematsu, PhD, Life Sciences Division, Lawrence Berkeley National Laboratory,
University of California, Berkeley; Antonella Riccio, MD, PhD, MRC Laboratory for Molecular Cell
Biology, Department of Neuroscience, Physiology and Pharmacology, University College London;
Moshe Szyf, PhD, Department of Pharmacology and Therapeutics, McGill University; Marcelo A.
Wood, PhD, Center for the Neurobiology of Learning and Memory, Department of Neurobiology and
Behavior, University of California, Irvine.
© 2009 Berger
Epigenetic Inheritance and
Replicative Cellular Aging
Shelley L. Berger, PhD
Department of Cell and Developmental Biology
University of Pennsylvania
Philadelphia, Pennsylvania
Epigenetic Inheritance and Replicative Cellular Aging
© 2009 Berger
An Introduction to
Epigenetic States
Understanding the mechanisms involved in the
initiation, maintenance, and inheritance of epigenetic
states is a key focus of current biomedical research.
Indeed, it is important to distinguish chromatin-
based regulatory mechanisms (as described below)
from true epigenetic memory of genomic states.
Epigenetics can be defined as phenotypic traits that
are stably inherited as a consequence of changes in
a chromosome without alterations in the underlying
DNA sequence. Thus, there is often a component of
heritability of a phenotype, passed on either through
mitosis or meiosis. However, epigenetics can also
refer to long-term chromatin changes that establish
a particular cell type, and this view of epigenetics
is relevant to end-stage differentiated cells such
as neurons.
Epigenetic inheritance comprises three distinct but
interrelated molecular pathways, which involve the
following steps, resulting in the establishment of a
stably heritable epigenetic state (Fig. 1) (Berger et
al., 2009). First, there is a signal, the “Epigenator,”
which derives from the cell-external environment
and leads to an intracellular pathway. Second, there
is an “Epigenetic Initiator,” which responds to the
Epigenator and leads to the localization within the
genomic chromatin. Third, there is an “Epigenetic
Maintainer,” which maintains the chromatin
change through multiple cell divisions, or in stable
differentiated cells.
Here, a specific example is described of epigenetic-
based changes in histone lysine acetylation during
replicative aging in the model eukaryote, the budding
yeast Saccharomyces cerevisiae (Dang et al., 2009). We
discovered that telomeric changes in chromatin are
centrally involved in replicative aging and are caused
by increased histone H4 acetylation. The result of
this altered level of acetylation is decompaction of
the telomeric chromatin in old cells. This change
taking place in the old cells is not transmitted to the
young daughter cells.
Chromatin Regulation and Histone
The long strands of nuclear DNA are associated with
packaging proteins, called histones, into a structure
known as chromatin, akin to the way thread is
organized around a spool. Changes occur in this
chromatin structure via chemical modification of the
histone proteins. These targeted structural changes
conceptually resemble the unraveling of the thread
to reach specific, buried sections. The changes in
chromatin caused by these histone modifications
that persist through cell division from one cell into
two daughter cells, i.e., the epigenetic changes, are of
particular interest because they are key to normal and
abnormal growth. They occur during development
and differentiation into multicellular tissues and
organs, and are typically disrupted during abnormal
reversal of tissue specialization and growth control,
as in cancer, as well as during the aging of cells and
entire organisms.
Histone modifications regulate genomic functions,
including transcription of genes, DNA replication
during cell division, repair of DNA mutations as a
result of DNA damage, and other processes (Berger,
2007). There are a plethora of histone post-
translational modifications (hPTMs), including
acetylation, phosphorylation, methylation, ubiqui-
tylation, and sumoylation. Many modification sites
are found within the core histones and variant
histones; most localize within the amino-terminal
and carboxyl-terminal tails, and a few localize within
the histone globular domains.
Figure 1. A pathway for epigenetic inheritance. Three
categories of epigenetic signals are shown. The “Epigenator”
integrates environmental changes, such as differentiation/
metabolic signals and temperature variations. These signals
are transduced through the cytoplasm into the nucleus, where
they are manifested into chromatin via two mechanisms. The
“Initiator” provides location information via sequence-specific
association of DNA binding factors and noncoding RNAs.
The “Maintainer” sustains the signal via histone and DNA
modifications and incorporation of histone variants.
© 2009 Berger
Lysine is a key substrate residue in histone biochemistry
because multiple, exclusive modifications occur there,
including acetylation, methylation, ubiquitylation,
and sumoylation. To help us understand their role,
there are several ways to classify these modifications.
Acetylation and methylation involve small chemical
groups, whereas ubiquitylation and sumoylation
add large moieties, occupying two-thirds the size
of the histone proteins themselves; these proteins,
owing to their bulk, may lead to more profound
changes in chromatin structure. (Note that histones
are mono-ubiquitylated, leading to signaling,
rather than poly-ubiquitylated, for the purpose of
mediating proteolysis.)
A second general theme of histone biochemistry
is whether or not a particular modification has a
strictly consistent effect on transcription. Abundant
evidence shows that acetylation is activating, whereas
sumoylation appears to be repressing, and these
processes may mutually interfere. In clear contrast,
methylation and ubiquitylation have variable effects,
depending on the precise residues and contexts.
Each of the histone hPTMs is removable. Histone
deacetylase (HDAC) enzymes remove acetyl groups,
ubiquitin proteases remove mono-ubiquitin on
H2B, and two classes of lysine demethylase remove
methyl groups.
Two general, nonexclusive proposals have been put
forth for mechanisms underlying these modifications.
One idea is that chromatin compaction is directly
altered (via a change either in electrostatic charge
or in internucleosomal contacts) to open/close the
polymer to control access of DNA binding proteins.
For example, the histone H4 tail is structurally critical
to internucleosomal contacts. Indeed, acetylation of
H4 K16 within a completely naïve chromatin array
assembled from bacterial recombinant histones
results in relaxation of the array. A second concept
is that the chemical moieties create an altered
nucleosome surface to promote the association of
chromatin-binding proteins, i.e., effector proteins.
Initial evidence emerged for the role of acetyl
lysine, which has been shown to associate with
bromodomains. For example, acetylated histone H3
stabilizes binding of the histone acetyltransferase
Gcn5 through its bromodomain. In addition, lysine
methylation provides an important switch for binding
of representatives of the royal family of domains,
including chromodomains and tudordomains. For
example, methyl histone H3K9 provides a surface for
association of the chromodomain of HP1 to promote
binding to heterochromatin.
Chromatin Changes During
Replicative Aging
Cells undergoing developmental processes such as
differentiation, apoptosis, and gametogenesis are
characterized by persistent but nongenetic alterations
in chromatin, termed “epigenetic changes.” These
changes are represented by distinct patterns of DNA
methylation and hPTMs. Sirtuins are a group of
conserved NAD
-dependent deacetylases or ADP-
ribosylases that promote longevity in yeast, worms,
and flies; however, their molecular mechanisms for
regulating aging remain poorly understood. The yeast
S. cerevisiae Sir2 (silencing information regulator 2),
the founding member of the family, establishes and
maintains silencing within yeast heterochromatic-
like regions at several sites—telomeres, recombinant
DNA (rDNA), and silenced mating type loci
(HM)—by removing H4 lysine 16 acetylation
(H4K16ac) and bringing in other silencing proteins.
An age-associated decrease takes place in Sir2
protein abundance, accompanied by an increase in
H4K16ac and loss of histones at specific subtelomeric
regions in replicatively old cells; these changes result
in compromised transcriptional silencing at these
loci. Deletion of Sas2, the histone acetyltransferase
(HAT) targeting H4K16 at subtelomeric regions,
stabilizes Sir2 levels in old cells and extends life
span. Finally, mutations to H4K16 negatively affect
life span, downstream of Sir2, in a pathway distinct
from the accumulation of extrachromosomal circles
in the nucleolus, a previously established mechanism
for yeast replicative aging.
These findings establish H4K16ac as the critical
target for Sir2 in regulating life span and indicate
that Sir2 opposes replicative aging in yeast through
an additional pathway by maintaining low H4K16ac
and silenced chromatin at telomere elements
(summarized in Fig. 2). This pathway may represent
an evolutionarily conserved function of sirtuins, i.e.,
the regulation of replicative aging by maintaining
intact telomeric chromatin.
Conclusion: Significance of
the Epigenetic Pathway in
Replicative Aging
Returning to the general pathway for establishing
epigenetic states in chromatin, we can propose a
specific pathway for the epigenetic state in daughter
yeast cells (Fig. 3), distinct from aging mother cells.
Here, the environmental signal is the partitioning
of Sir2 into daughter cells. Rap1 is a DNA binding
protein that binds to telomeres, hence providing
Epigenetic Inheritance and Replicative Cellular Aging
© 2009 Berger
sequence-specific information to initiate the
localization of Sir2. Sir2 associates with Sir3 and
Sir4 to form a complex, which is recruited by Rap1
to telomeres to maintain low acetylation and stable
telomeric chromatin. Each epigenetic pathway in
eukaryotes can be characterized using these three
categories: Epigenator, Initiator, and Maintainer.
Berger SL (2007) The complex language of
chromatin regulation during transcription. Nature
Berger SL, Kouzarides T, Shiekhattar R, Shilatifard
A (2009) An operational definition of epigenetics.
Genes Dev 23:781-783.
Dang W, Steffen KK, Perry R, Dorsey JA, Johnson
FB, Shilatifard A, Kaeberlein M, Kennedy BK,
Berger SL (2009) Histone H4 lysine 16 acetylation
regulates cellular lifespan. Nature 459:802-807.
Figure 2. Chromatin decompaction at the ends of
chromosomes is centrally involved in aging. A schematic
picture is shown for the chromatin status at telomeres in young
compared with old yeast cells. The HDAC Sir2 is localized to
the telomeres in young cells, and the histone acetylase Sas2 is
present at the boundaries between telomeric heterochromain
and euchromatin. This maintains low acetylation at telomeres
and higher acetylation at boundaries. In old cells, Sir2 levels
drop and Sas2 leaks into the telomeres, causing the acetylation
level to increase, leading to chromatin destabilization.
Figure 3. An epigenetic inheritance pathway maintaining
compacted young yeast cells. Three categories of epigenetic
signals can be specified. The Epigenator signal is the
partitioning of Sir2 into daughter cells. The Sir2 complex (with
Sir3 and Sir4 proteins) serves as the Maintainer to sustain
low acetylation levels. Sir 2 is recruited by the Inititator Rap1,
which is a DNA binding factor and thus provides sequence-
specific information.
© 2009 Sassone-Corsi
Linking Metabolism to Epigenetics:
Chromatin Remodeling and Circadian Control
Paolo Sassone-Corsi, PhD
Department of Pharmacology
School of Medicine
University of California, Irvine
Irvine, California
Linking Metabolism to Epigenetics: Chromatin Remodeling and Circadian Control
© 2009 Sassone-Corsi
Introduction to Circadian Rhythms
Neurons are cells that are submitted to an exceptional
variety of stimuli and are able to convert them into
high-order functions, such as storing memories,
controlling behavior, and governing consciousness.
These unique properties are based on highly plastic
processes, which intimately depend on the complex
molecular machinery that controls gene expression.
Evidence is accumulating that neuronal functions
have more than a solely genetic basis. Epigenetic
control, which largely involves events of chromatin
remodeling, appears to govern some of the more
distinctive features of neuronal responses (Borrelli et
al., 2008).
Circadian rhythms of 24 h govern a number of
fundamental physiological functions in almost all
organisms, from prokaryotes to humans (Dunlap,
1999; Cermakian and Sassone-Corsi 2000; King and
Takahashi, 2000; Young and Kay, 2001; Reppert and
Weaver, 2002). The circadian clocks are intrinsic
time-tracking systems with which organisms can
anticipate environmental changes and adapt
themselves to the appropriate time of day. In mammals,
circadian rhythms are generated in pacemaker
neurons within the suprachiasmatic nuclei
(SCN) of the hypothalamus and are entrained by
environmental cues, principally light. Disruption
of these rhythms can have a profound influence on
human health and has been linked to insomnia,
depression, coronary heart diseases, various
neurodegenerative disorders, and cancer (Fu and
Lee, 2003; Hastings et al., 2003) (Fig. 1).
The molecular mechanism of the circadian clock is
based on interlocked transcriptional–translational
feedback loops, as revealed by molecular and genetic
analyses in Drosophila and mammals (Dunlap, 1999;
Cermakian and Sassone-Corsi 2000; King and
Takahashi, 2000; Young and Kay, 2001; Reppert and
Weaver, 2002). To date, various core circadian-clock
genes have been identified in mammals: Clock, Bmal1,
casein kinase I epsilon (CKIe), cryptochromes 1 and 2
(Cry1, Cry2), Period 1, 2 and 3 (Per1, Per2, Per3), and
Rev-erb-a. Interaction of clock proteins occurs via the
PAS domains (named after the proteins PER-ARNT-
SIM), which provide heterodimerization surfaces.
The Clock and Bmal1 genes encode basic-helix-loop-
helix (bHLH)–PAS transcription activators, which
heterodimerize and induce the expression of Per and
Cry genes by binding to E-box elements (CACGTG)
present in their promoters. Once the PER and CRY
Figure 1. Schematic model of the circadian clock (pacemaker) and its input signals, the most prominent being light. The light
signaling directly influences neurons in the central clock in the hypothalamic SCN, thereby modulating the self-sustained clock
circadian regulation. The outputs of the circadian system include a large array of physiological, metabolic, and neuronal functions.
Disruption of clock function may cause dramatic pathophysiological effects, including neurodegeneration and cancer.
© 2009 Sassone-Corsi
proteins are synthesized, they form heterodimeric
complexes, which in turn translocate to the nucleus
and inhibit CLOCK-BMAL1–mediated transcription
through direct protein-protein interactions (Reppert
and Weaver, 2002).
The result of these complex regulatory pathways is
that the messenger RNAs (mRNAs) and protein
levels of most circadian genes (except Clock and
CKIe) oscillate with a 24 h period. Importantly,
the CLOCK-BMAL1 heterodimer regulates the
transcription of many clock-controlled genes
(CCGs), which in turn influence a wide array of
physiological functions external to the oscillatory
mechanism. This genetic influence mediates the
output function of the clock, thereby controlling
food intake, hormonal synthesis and release, body
temperature, metabolism, and other functions.
Remarkably, 10-15% of all mammalian transcripts
undergo circadian fluctuations in their expression
levels (Akhtar et al., 2002; Duffield et al., 2002;
Panda et al., 2002), indicating that a global circadian
chromatin remodeling system must operate.
Plasticity in Circadian Regulation
by Chromatin Remodeling
How is the oscillatory expression of clock-controlled
genes regulated so that transcription-permissive
chromatin states are dynamically established in a
circadian, time-specific manner? The first evidence
that chromatin remodeling may be intimately
implicated in circadian control was obtained almost 10
years ago, by studying histone modifications induced
by a light pulse in neurons of the suprachiasmatic
nucleus (Crosio et al., 2000). Subsequently, the
activation of CCGs by CLOCK:BMAL1 was
correlated to circadian changes in histone acetylation
at their promoters (Etchegaray et al., 2003; Curtis et
al., 2004; Naruse et al., 2004; Nakahata et al., 2007).
A finding that paved the way toward understanding
how circadian chromatin remodeling could occur was
that CLOCK induces histone acetylation (Doi et al.,
2006). The carboxy-terminal glutamine-rich region
of CLOCK, a region implicated in transactivation
function (Allada et al., 1998; Gekakis et al., 1998),
displays a significant sequence homology with the
carboxy-terminal domain of ACTR, a domain
previously shown to have histone acetyltransferase
(HAT) activity (Chen et al. 1997). Our analyses
indicated that CLOCK could have intrinsic
enzymatic HAT activity, and we did establish this
premise using an in-gel HAT assay. This biochemical
assay is ideal for unequivocally determining that a
protein exhibits HAT function.
In our experiment, we immunoprecipitated a Myc-
mCLOCK fusion protein expressed in cultured cells
and used it in in-gel HAT assays with a mixture of
purified histones as substrate (Brownell et al., 1999).
Proteins resolved by SDS-PAGE were subjected to
in-gel enzymatic reaction. Detection of histones
C]-acetylated in situ demonstrated that acetylation
took place specifically in a position corresponding to
where the Myc-CLOCK protein had migrated. As
control, it is helpful to use truncated and/or mutated
forms of the protein suspected to have HAT activity.
In this experiment, also to rule out the possibility
that a contaminant HAT comigrating with Myc-
CLOCK would be responsible for the acetylation,
we used an N-terminally truncated mCLOCK (Myc-
mCLOCK∆N) protein in the in-gel HAT assay. This
truncated CLOCK protein lacks the N-terminal
residues 1–242 but has an intact C-terminal region
and still displays efficient HAT activity in the gel.
In a series of experiments using cultured cells that
recapitulate circadian clock regulation, it was shown
that the HAT function of CLOCK is essential for
the circadian control of CCGs (Doi et al., 2006).
This system is based on mouse embryonic fibroblast
(MEF) cells derived from homozygous Clock mutant
(c/c) mice (Pando et al., 2002). Because Clock is
essential for circadian rhythm, as expected, MEF
c/c cells show no cyclic expression of clock genes
(Pando et al., 2002), whereas ectopic expression of
CLOCK is able to rescue the circadian phenotype.
On the contrary, ectopic expression of a HAT-
deficient CLOCK failed to restore the circadian
gene expression, demonstrating the essential role
of CLOCK’s HAT activity (Doi et al., 2006). These
findings underscore the importance of chromatin
remodeling in circadian regulation and reveal the
molecular pathways by which such essential control
is achieved.
Acetylation of nonhistone
Acetylation of nonhistone proteins is achieved
by various HATs (Glozak et al., 2005; Zhang and
Dent, 2005) and is demonstrated to have profound
physiological significance. In a survey aimed at
identifying proteins that might be acetylated
rhythmically in vivo, we analyzed various clock
proteins, such as BMAL1, CLOCK, and PER1, in
the mouse liver at different circadian times. While,
as expected, these proteins oscillate in terms of
abundance and phosphorylation levels (Lee et al.,
2001; Matsuo et al., 2003), acetylation of BMAL1
displays a robust acetylation with a circadian peak at
ZT 15 (Hirayama et al., 2007). Significantly, the other
Linking Metabolism to Epigenetics: Chromatin Remodeling and Circadian Control
© 2009 Sassone-Corsi
clock proteins are not acetylated. Ongoing studies
in our laboratory on a variety of nuclear proteins
and transcription factors indicate that BMAL1 is
one of the few substrates for CLOCK, underscoring
the specificity of the assay. Importantly, CLOCK
is directly responsible for BMAL1 acetylation in
cultured mammalian cells (Grimaldi et al., 2007;
Hirayama et al., 2007).
To identify the site of CLOCK-mediated acetylation,
we generated several mutant proteins with Lys>Arg
substitutions in the putative acetylated sites. All
mutants displayed acetylated levels comparable with
wild-type BMAL1, with the exception of K537,
a highly conserved residue among all vertebrate
BMAL1s. By using MEFs, with an approach
analogous to the one described above, we could also
demonstrate that acetylation at K537 is critical for
circadian function (Grimaldi et al., 2007; Hirayama
et al., 2007).
This finding suggests that CLOCK may have several
putative targets and that their identification is likely
to provide significant clues about the neuronal
pathways influenced by the circadian clock. In this
respect, another protein may play a relevant regulatory
function: NPAS2. This is an alternative partner of
BMAL1, whose structure is loosely similar to CLOCK
(Reick et al., 2001). Interestingly, NPAS2 displays a
neuronal-specific distribution, being abundant in the
forebrain areas, including the cortex, hippocampus,
striatum, amygdala, and thalamus (Garcia et al.,
2000). Although it is yet unclear whether NPAS2
may have acetyltransferase activity, its function as
a substitute for CLOCK in the dimerization with
BMAL1 confers on it a potential role in indirectly
regulating CLOCK’s HAT activity.
The Flip Side of Acetylation: SIRT1,
a Circadian Histone Deacetylase
A hallmark of chromatin remodeling factors is that
they may have both positive and negative enzymatic
activities, functions that we have previously
described as “writers” and “erasers” of specific histone
modifications (Borrelli et al., 2008). Thus, as soon
as CLOCK was found to be a HAT, the quest for
its counterbalancing histone deacetylase (HDAC)
had begun.
CLOCK expression is not rhythmic (Gekakis et al.,
1998), whereas its induced acetylation is, indicating
that its chromatin remodeling activity is critical for
circadian physiology (Doi et al., 2006). There are four
classes of HDACs, and the subdivision is based on
their protein structure (Yang and Seto, 2008). SIRT1
belongs to the family of sirtuins that constitutes the
so-called class III HDACs. These are the only HDACs
whose enzymatic activity is NAD
-dependent and
that have been intimately linked to the control of
metabolism and aging (Bordone and Guarente, 2005).
SIRT1 directly associates with CLOCK and functions
as a rheostat in modulating the acetylation state of
histone H3 and BMAL1 (Nakahata et al., 2008).
These observations are relevant to establishing a
direct molecular coupling between circadian control
and energy metabolism (Fig. 2).
The CLOCK–SIRT1 complex is, in fact, regulated
by the NAD
/nicotinamide balance in the cell
(Nakahata et al., 2008). This finding provides a novel
perspective on the links that have been historically
reported between circadian rhythms, metabolism,
and cellular reduction-oxidation pathways (Wijnen
and Young, 2006; Collis and Boulton, 2007). The
finding that SIRT1 acts as a rheostat in the context
of circadian acetylation is of interest because it may
be linked to other, recently described functions
of this regulator in aging and neurodegeneration
(Gan and Mucke, 2008). For example, inhibitors
of SIRT1 rescue

-synuclein–mediated toxicity in
animal models of Parkinson’s disease (Outeiro et al.,
2007). As the role of dopamine in neurotoxicity and
neuroprotection is established (Bozzi and Borrelli,
2006), the link between dopamine-mediated signaling
and the circadian machinery (Yujnovsky et al., 2006)
will take on new significance. Furthermore, SIRT1
has been found to contribute to the redox-dependent
fate of neural progenitors (Prozorovski et al., 2008).
The Central Role of Metabolites in
Chromatin Remodeling
Accumulating evidence shows that the enzymatic
machinery that elicits histone modifications
operates under the control of a variety of neuronal
stimuli. These stimuli link physiological variations
to modulated chromatin remodeling, and thereby
to controlled gene expression. One important
consideration relates to the intracellular pathways
involved in marking these posttranslational
modifications (Borrelli et al., 2008). Interestingly, all
of the modifications use physiological metabolites.
This indicates that the dynamic process of chromatin
remodeling may “sense” cellular metabolism and
changes in energy levels (Table 1), which are
highly controlled and essential to functioning
neuronal responses.
The example of SIRT1 is paradigmatic because its
enzymatic activity is under the control of metabolic
cofactors and inhibitors. While NAD
is SIRT1’s
© 2009 Sassone-Corsi
Figure 2. Schematic model of CLOCK-mediated histone acetylation and its regulation by the NAD
-dependent histone
deacetylase SIRT1.The HAT function of CLOCK regulates promoters of CCGs and clock genes (such as Per1) by inducing locally open or-
ganization of the chromatin. CCGs include a large number of genes involved in cellular metabolism, proliferation, and cell cycle (exam-
ples are indicated), underscoring the critical function of the CLOCK–SIRT1 complex in neurodegeneration and physiological control.
Strikingly, the expression of 15% of all mammalian transcripts oscillates in a circadian manner. CLOCK acetylates H3 and
BMAL1, its natural heterodimerization partner, in order to regulate promoters of CCGs. Acetylation by CLOCK is thought to elicit
chromatin remodeling by inducing a transcription-permissive state. The acetyltransferase enzymatic activity of CLOCK has a dual
function: It regulates the circadian machinery by targeting both histone and nonhistone proteins. We envisage a scenario in which
circadian control of chromatin remodeling by CLOCK may be influenced by the dynamic assembly of a multiprotein regulatory
complex. SIRT1 associates with CLOCK and, in response to the metabolic changes in intracellular NAD levels, modulates CCGs
by virtue of its HDAC enzymatic activity. Thus, metabolic, nutritional, and environmental cues modulate the circadian machinery
via chromatin remodeling.
table 1. The importance of posttranslational modifications (PTMs) of histones in chromatin remodeling is well established. One
critical facet of these modifications is that they are elicited by specific enzymatic activities that depend on the intracellular levels
of essential metabolites; these metabolites sense cellular metabolism, nutrients, and energy levels in the cell. PTMs target specific
sites on the histone tails, indicating that the transient states of chromatin remodeling are under the dynamic regulation of cellular
physiology (Borrelli et al., 2008).
© 2009 Sassone-Corsi
Linking Metabolism to Epigenetics: Chromatin Remodeling and Circadian Control
natural cosubstrate, the oxidated form NADH and
the by-product of NAD
consumption, nicotinamide
(NAM), repress the activity of SIRT1 (Bordone and
Guarente, 2005), generating an enzymatic feedback
loop on the HDAC function of this enzyme. Indeed,
it has been established that fluctuations in NAD
changes in the NAD
/NADH ratio and nicotinamide
concentrations directly influence SIRT1 function.
Two main systems determine NAD+ levels in the
cell: the de novo biosynthesis from tryptophan and
the NAD
salvage pathway.
A critical step of this latter pathway is controlled by
the enzyme nicotinamide phosphoribosyltransferase
(NAMPT), also known as visfatin or PBEF
(Rongvaux et al., 2003). NAMPT catalyzes the first
step in the biosynthesis of NAD
from nicotinamide.
In mammalian cells, NAMPT slows down senescence
and promotes survival during genotoxic stress
(Anderson et al., 2002; Revollo et al., 2004; van der
Veer et al., 2007; Yang et al., 2007). Importantly,
Nampt gene expression is dynamic because it is
inducible by various agents and responds to specific
cellular stresses (e.g., DNA damage) and nutrients.
This resilience indicates that its control is central to
governing the intracellular NAD
:NAM balance.
Interestingly, SIRT1 and CLOCK protein levels
do not seem to oscillate (Nakahata et al., 2008),
whereas the acetylation of the targets of CLOCK-
mediated HAT function (e.g., K14 of histone H3
and K537 of BMAL1) is circadian (Hirayama et al.,
2007; Nakahata et al., 2008). Thus, one possibility
is that circadian oscillation would be determined
by oscillating levels of intracellular NAD
. Indeed,
intracellular NAD
levels cycle with a 24 h rhythm,
and this oscillation is driven by the circadian clock.
CLOCK:BMAL1 regulates the circadian expression
of the Nampt gene in concert with SIRT1, which
thereby contributes to the circadian synthesis of its
own coenzyme (Nakahata et al., 2009). Using the
specific inhibitor of NAMPT enzyme activity, FK866,
researchers have also demonstrated that NAMPT is
required to modulate circadian gene expression as
well as BMAL1 circadian acetylation (Nakahata et
al., 2009).
These results uncover two self-regulating, interlocking
systems, in which a classical transcription circadian
loop is coupled to an enzymatic feedback loop
(Fig. 2). In addition, at least one other enzyme may
be critically influenced by oscillatory NAD
the poly(ADP-ribose) polymerase-1 (PARP-1),
which utilizes NAD
as a coenzyme to exert its role in
recovery from DNA damage. A functional interplay
between SIRT1 and PARP-1 has been reported
(Kolthur-Seetharam et al., 2006).
The implications of these findings are myriad. First,
they reveal an intriguing link between cellular
metabolism and circadian physiology (Eckel-Mahan
and Sassone-Corsi, 2009). In addition, they
highlight possible molecular pathways that could
bridge the circadian machinery to various diseases.
Finally, they may open up novel avenues for pharma-
cological intervention.
Akhtar RA, Reddy AB, Maywood ES, Clayton JD,
King VM, Smith AG, Gant TW,
Allada R, White NE, So WV, Hall JC, Rosbash
M (1998) A mutant Drosophila homolog of
mammalian Clock disrupts circadian rhythms
and transcription of period and timeless. Cell
Anderson RM, Bitterman KJ, Wood JG, Medvedik
O, Cohen H, Lin SS, Manchester JK, Gordon JI,
Sinclair DA. (2002) Manipulation of a nuclear
NAD+ salvage pathway delays aging without
altering steady-state NAD+ levels. J Biol Chem
Bordone, L, Guarente L (2005) Calorie restriction,
SIRT1 and metabolism: understanding longevity.
Nat Rev Mol Cell Biol 6:298-305.
Borrelli E, Nestler EJ, Allis CD, Sassone-Corsi
P (2008) Decoding the epigenetic language of
neuronal plasticity. Neuron 60:961-974.
Bozzi Y, Borrelli E (2006) Dopamine in neurotoxicity
and neuroprotection: What do D2 receptors have
to do with it? Trends Neurosci 29:167-174.
Brownell JE, Mizzen CA, Allis CD (1999) An SDS-
PAGE–based enzyme activity assay for the detection
and identification of histone acetyltransferases.
Methods Mol Biol 119:343-353.
Cermakian N, Sassone-Corsi P (2000) Multilevel
regulation of the circadian clock. Nat Rev Mol
Cell Biol 1:59-67.
Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash
A, Nagy L, Privalsky ML, Nakatani Y, Evans
RM (1997) Nuclear receptor coactivator ACTR
is a novel histone acetyltransferase and forms a
multimeric activation complex with P/CAF and
CBP/p300. Cell 90:569-580.
Collis SJ, Boulton SJ (2007) Emerging links between
the biological clock and the DNA damage response.
Chromosoma 116:331-339.
Crosio C, Cermakian N, Allis CD, Sassone-Corsi
P (2000) Light induces chromatin modification
in cells of the mammalian circadian clock. Nat
Neurosci 3:1241-1247.
© 2009 Sassone-Corsi
Curtis AM, Seo SB, Westgate EJ, Rudic RD,
Smyth EM, Chakravarti D, FitzGerald GA,
McNamara P (2004) Histone acetyltransferase–
dependent chromatin remodeling and the vascular
clock. J Biol Chem 279:7091-7097.
Doi M, Hirayama J, Sassone-Corsi P (2006) Circadian
regulator CLOCK is a histone acetyltransferase.
Cell 125:497-508.
Duffield GE, Best JD, Meurers BH, Bittner A,
Loros JJ, Dunlap JC (2002) Circadian programs of
transcriptional activation, signaling, and protein
turnover revealed by microarray analysis of
mammalian cells. Curr Biol 12:551-557.
Dunlap JC (1999) Molecular bases for circadian
clocks. Cell 96:271-290.
Eckel-Mahan K, Sassone-Corsi P (2009) Metabolism
control by the circadian clock and vice versa. Nat
Struct Mol Biol 16:462-467.
Etchegaray JP, Lee C, Wade PA, Reppert SM
(2003) Rhythmic histone acetylation underlies
transcription in the mammalian circadian clock.
Nature 421:177-182.
Fu L, Lee CC (2003) The circadian clock: pacemaker
and tumour suppressor. Nat Rev Cancer
Gan L, Mucke L (2008) Paths of convergence:
Sirtuins in aging and neurodegeneration. Neuron
Garcia JA, Zhang D, Estill SJ, Michnoff C, Rutter J,
Reick M, Scott K, Diaz-Arrastia R, McKnight SL
(2000) Impaired cued and contextual memory in
NPAS2-deficient mice. Science 288:2226-2230.
Gekakis N, Staknis D, Nguyen HB, Davis FC,
Wilsbacher LD, King DP, Takahashi JS, Weitz
CJ (1998) Role of the CLOCK protein in the
mammalian circadian mechanism. Science
Glozak MA, Sengupta N, Zhang X, Seto E (2005)
Acetylation and deacetylation of non-histone
proteins. Gene 363:15-23.
Grimaldi B, Nakahata Y, Sahar S, Kaluzova M,
Gauthier D, Pham K, Patel N, Hirayama J,
Sassone-Corsi P (2007). Chromatin remodeling
and circadian control: master regulator CLOCK is
an enzyme. Cold Spring Harbor Symp Quant Biol
Hastings MH, Reddy AB, Maywood ES (2003) A
clockwork web: circadian timing in brain and
periphery, in health and disease. Nat Rev Neurosci
Hirayama J, Sahar S, Grimaldi B, Tamaru T,
Takamatsu K, Nakahata Y, Sassone-Corsi P (2007).
CLOCK-mediated acetylation of BMAL1 controls
circadian function. Nature 450:1086-1090.
King DP, Takahashi JS (2000) Molecular genetics
of circadian rhythms in mammals. Annu Rev
Neurosci 23:713-742.
Kolthur-Seetharam U, Dantzer F, McBurney MW, de
Murcia G, Sassone-Corsi P (2006) Control of AIF-
mediated cell death by the functional interplay of
SIRT1 and PARP-1 in response to DNA damage.
Cell Cycle 5:873-877.
Lee C, Etchegaray JP, Cagampang FR, Loudon AS,
Reppert SM (2001) Posttranslational mechanisms
regulate the mammalian circadian clock. Cell
Matsuo T, Yamaguchi S, Mitsui S, Emi A, Shimoda F,
Okamura H (2003) Control mechanism of the
circadian clock for timing of cell division in vivo.
Science 302:255-259.
Nakahata Y, Grimaldi B, Sahar S, Hirayama J,
Sassone-Corsi P (2007) Signaling to the circadian
clock: plasticity by chromatin remodeling. Curr
Opin Cell Biol 19:230-237.
Nakahata Y, Kaluzova M, Grimaldi B, Sahar S,
Hirayama J, Chen D, Guarente L, Sassone-Corsi P
(2008) The NAD+ dependent deacetylase
SIRT1 modulates CLOCK-mediated chromatin
remodeling and circadian control. Cell
Nakahata Y, Sahar S, Astarita G, Kaluzova M,
Sassone-Corsi P (2009) Circadian control of
the NAD+ salvage pathway by CLOCK-SIRT1.
Science 324:654-657.
Naruse Y, Oh-hashi K, Iijima N, Naruse M, Yoshioka
H, Tanaka M (2004) Circadian and light-induced
transcription of clock gene Per1 depends on
histone acetylation and deacetylation. Mol Cell
Biol 24:6278-6287.
Outeiro TF, Kontopoulos E, Altmann SM, Kufareva
I, Strathearn KE, Amore AM, Volk CB, Maxwell
M, Rochet JC, McLean PJ, Young AB, Abagyan
R, Feany MB, Hyman BT, Kazantsev AG (2007)
Sirtuin 2 inhibitors rescue alpha-synuclein–
mediated toxicity in models of Parkinson’s disease.
Science 317:516-519.
Panda S, Antoch MP, Miller BH, Su AI, Schook AB,
Straume M, Schultz PG, Kay SA, Takahashi JS,
Hogenesch JB (2002) Coordinated transcription
of key pathways in the mouse by the circadian
clock. Cell 109:307-320.
© 2009 Sassone-Corsi
Linking Metabolism to Epigenetics: Chromatin Remodeling and Circadian Control
Pando MP, Morse D, Cermakian N, Sassone-Corsi P
(2002) Phenotypic rescue of a peripheral clock
genetic defect via SCN hierarchical dominance.
Cell 110:107-117.
Prozorovski T, Schulze-Topphoff U, Glumm R,
Baumgart J, Schröter F, Ninnemann O, Siegert E,
Bendix I, Brüstle O, Nitsch R et al. (2008) Sirt1
contributes critically to the redox-dependent fate
of neural progenitors. Nat Cell Biol 10:385-394.
Reick M, Garcia J, Dudley C, McKnight SL (2001)
NPAS2: an analog of clock operative in the
mammalian forebrain. Science 293:506-509.
Reppert SM, Weaver DR (2002) Coordination
of circadian timing in mammals. Nature
Revollo JR, Grimm AA, Imai S (2004) The NAD
biosynthesis pathway mediated by nicotinamide
phosphoribosyltransferase regulates Sir2 activity in
mammalian cells. J Biol Chem 279:50754-50763.
Rongvaux A, Andris F, Van Gool F, Leo O (2003)
Reconstructing eukaryotic NAD metabolism.
Bioessays 25:683-690.
van der Veer E, Ho C, O’Neil C, Barbosa N,
Scott R, Cregan SP, Pickering JG (2007) Extension
of human cell lifespan by nicotinamide phosphori-
bosyltransferase. J Biol Chem 282:10841-10845.
Wijnen H, Young MW (2006) Interplay of circadian
clocks and metabolic rhythms. Annu Rev Genet
Yang XJ, Seto E (2008) The Rpd3/Hda1 family of
lysine deacetylases: from bacteria to yeast to mice
and men. Nat Rev Mol Cell Biol 9:206-218.
Yang H, Yang T, Baur JA, Perez E, Matsui T, Carmona
JJ, Lamming DW, Souza-Pinto NC, Bohr VA,
Rosenzweig A, de Cabo R, Sauve AA, Sinclair DA
(2007) Nutrient-sensitive mitochondrial NAD+
levels dictate cell survival. Cell 130:1095-1107.
Young MW, Kay SA (2001) Time zones: a
comparative genetics of circadian clocks. Nat Rev
Genet 2:702-715.
Yujnovsky I, Hirayama J, Doi M, Borrelli E,
Sassone-Corsi P (2006) Signaling mediated by
the dopamine D2 receptor potentiates circadian
regulation by CLOCK:BMAL1. Proc Natl Acad
Sci USA 103:6386-6391.
Zhang K, Dent SY (2005) Histone modifying enzymes
and cancer: going beyond histones. J Cell Biochem
© 2009 Kohwi-Shigematsu
Functional Packaging of
Chromatin into Loops
terumi Kohwi-Shigematsu, PhD, Shutao Cai, PhD,
and Yoshinori Kohwi, PhD
Life Sciences Division
Lawrence Berkeley National Laboratory, University of California
Berkeley, California
© 2009 Kohwi-Shigematsu
Functional Packaging of Chromatin into Loops
All the cells in our body contain the same genetic
information at the level of primary DNA sequence,
yet we are composed of a great diversity of cell types.
This diversity is particularly evident in the brain. It
is now well known that the activation or repression
of specific sets of genes by different combinations of
transcription factors plays a crucial role in this diver-
sity, influencing processes from cell-fate specification
or differentiation to synaptic plasticity. Epigenetic
modifications such as DNA methylation and post-
translational histone modifications are another
essential component in the control of gene expres-
sion. Also, the three-dimensional organization
of chromatin itself (intra- and interchromosomal
interactions, or higher-order chromatin organiza-
tion) represents an important component of the epig-
enome and is now widely accepted as an important
mechanism underlying gene regulation (Yasui et al.,
2002; Cai et al., 2003; Kosak and Groudine, 2004;
Spilianakis and Flavell, 2004; Horike et al., 2005;
Spilianakis et al., 2005; Cai et al., 2006; Fraser, 2006;
Wallace and Felsenfeld, 2007; de Laat et al., 2008).
This is because chromatin organization influences
the accessibility of DNA to gene regulatory compo-
nents and the way genetic information is utilized but
does not affect primary DNA sequences.
Chromatin architecture in eukaryotic nuclei is highly
dynamic. In contrast to the classic view that genes
are static where highly mobile, sequence-specific
DNA-binding proteins travel to each gene locus to
regulate its expression individually, genes from dif-
ferent chromosomes can move within the nucleus to
be brought together, and in some cases, assembled
with transcription or other chromatin factors for
transcriptional coregulation. To understand genetic
activity from the viewpoint of chromatin architec-
ture, it is useful to understand not only recent con-
ceptual developments but also the technologies and
experimental controls used in chromatin structure
analysis. When studying chromatin association with
specific transcription factors and chromatin looping
by nuclear proteins in vivo, the quality and reproduc-
ibility of the data depend heavily on the way cells
are crosslinked, the abundance of factors in the cells
used, how they bind chromatin, and the quality of
antibodies used. On account of these many variables,
experimentalists need to carefully optimize methods
for their biological system of interest. This chapter
introduces basic methods of chromatin structure
analysis and offers important tips to facilitate studies
of chromatin structure.
We will mainly use the genome organizer SATB1 as
an example (Dickinson et al., 1992; Alvarez et al.,
2000) to describe methods for several procedures:
chromatin crosslinking, purification, chromatin
immunoprecipitation (ChIP), cloning of individual
genomic sequences bound to the protein, Chromo-
some Conformation Capture (3C) mediated by
SATB1 (ChIP-3C), and the controls necessary for
these experiments. Although not described in this
chapter, we will cover in the lecture the methods for
identifying interaction of one genomic site of choice
with all other genomic sequences genomewide (ChIP
one-to-all, or Chromosome Conformation Capture-
on-Chip [4C]).
Crosslinking of Chromatin
and Purification
Formaldehyde is typically used for crosslinking chro-
matin at a final concentration of 1%. However, it is
important to use formaldehyde that does not contain
methanol (e.g., Pierce #28906 [Thermo Scientific,
Rockford, IL]). Methanol is added as a stabilizer
because it slows down the formation of paraformal-
dehyde (formaldehyde polymer), though its ability
to precipitate proteins is to be avoided in this
application. To study in vivo binding of histones
with a specific epigenomic mark to genomic DNA,
one could find published conditions that work with-
out the need for further optimization (Milne et al.,
2009). For proteins such as histones, owing to their
high abundance and tight association with DNA, a
wide range of conditions would work unless the cells
are overly crosslinked. On the other hand, in order
to yield the best results for transcription factors that
are expressed at much lower levels than histones, it
is important to test a series of crosslinking conditions
using the cells that express the factor, by varying
temperature and time of incubation of the cells with
formaldehyde. It is best to employ the mildest condi-
tion that enables the detection of specific signals for
ChIP, followed by PCR amplification. This is because
the solubility of crosslinked chromatin decreases
dramatically as cells are subjected to more extensive
crosslinking, resulting in a low yield of final solubi-
lized crosslinked chromatin to work with. However,
if formaldehyde-crosslinking is insufficient, there
will be no protein crosslinked to the genomic DNA,
leading to no positive signals after PCR amplification
of ChIP samples.
The optimal crosslinking condition depends on
specific factors of interest and their level of expres-
sion; it also appears to depend on cell type to some
extent. In the case of SATB1, we use 1% formalde-
hyde crosslinking condition in either a 37°C water
bath for 5 min in a 50 ml tube or we place cells in a
37°C dry incubator for 10 min with rocking/agitation.
© 2009 Kohwi-Shigematsu
An additional incubation at 4°C is optional in the
latter case. One could vary conditions, referring to
these examples for other proteins of interest.
In addition, starting with tissue, such as brain, pre-
paring the cell for crosslinking is an important step.
To isolate cells from brain tissue before crosslinking,
the sample is first cut into small pieces, ground, and
filtered through a 100 μm cell strainer, then suspend-
ed in PBS at room temperature, and followed by a
second filtering through a 70 μm cell strainer to make
a single-cell suspension.
Purification of crosslinked chromatin
We have been using a urea-gradient ultracentrifuga-
tion method to purify crosslinked chromatin (Kohwi-
Shigematsu et al., 1998). Starting with uncrosslinked
cells in 4% SDS layered on top of 5–8 M urea
gradient, prepared in 10 mM Tris, pH 8.0, 1 mM
EDTA, followed by centrifugation at 30,000 rpm for
16 h in a Beckman SW41 rotor (Beckman Coulter,
Fullerton, CA), this method removes all proteins
that are bound to chromatin, leaving a pellet of pure
unsheared genomic DNA on the bottom of the tube.
However, starting with crosslinked cells, only pro-
teins that were crosslinked to genomic DNA will be
found in the DNA pellet. Therefore, this method
cleanly separates crosslinked chromatin from free
proteins, RNA, and other small cellular debris. ChIP
experiments using urea-gradient-purified, crosslinked
chromatin as the template have much reduced back-
ground signals. Also, with this method it is possible to
generate high-quality data with much reduced back-
ground for 3C, as described below.
The crosslinked chromatin purified by urea-
gradient ultracentrifugation can be subjected to
water-bath sonication (e.g., Bioruptor [Diagenode,
Liège, Belgium]) or to digestion with restriction
enzymes. The pellet formed by urea-gradient centrif-
ugation is pure crosslinked chromatin and resembles
a contact lens. To facilitate solubility of chromatin,
this “contact lens” is carefully removed from the
bottom of the centrifuge tube and crushed with a
Pellet Pestle (Thomas Scientific [Swedesboro, NJ]
7495211590) or disrupted by passing it through an
insulin syringe and dissolved in a buffer used for the
restriction enzyme to be used. Alternatively, if one
chooses sonication to cleave chromatin, the “contact
lens” is crushed or disrupted by syringe in sonication
buffer (1% SDS, 10mM EDTA, 50 mM Tris, pH 7.5)
before sonication. DNA that is either isolated from
restriction enzyme-digested, crosslinked chromatin
or cleaved by sonication must be analyzed by running
a de-crosslinked aliquot on an agarose gel to confirm
the extent of digestion or shearing. Only samples
lacking any large-molecular-weight DNA are consid-
ered ready to be processed for further experiments.
Chromatin immunoprecipitation
Before using any antibody for a ChIP experiment,
it is essential to confirm that the antibody can
immunoprecipitate its antigen efficiently. This prop-
erty can be tested by immunoprecipitation of the
antigen from cell or nuclear extracts using different
concentrations of the antibody, followed by Western
blot analysis (on immunoprecipitated sample and the
soluble fraction). This experiment will determine
the optimal level of antibody to use for ChIP. Two
important points are that sufficient amounts of initial
chromatin samples should be used for immunoprecip-
itation, and that control experiments should succeed
(Kohwi-Shigematsu et al., 1998). For a transcription
factor, it is best to use crosslinked chromatin,
comparable with a minimum of 20–40 micrograms
of DNA, for immunoprecipitation. For a rabbit
antiserum, it is best to use the preimmune serum as a
control from the same rabbit before immunization. If
this is not available, one has to obtain nonimmune
serum that does not generate nonspecific signals.
Before performing chromatin immunoprecipitation,
crosslinked chromatin samples will be precleared
twice with either preimmune or nonimmune
serum by incubation with protein A-Sepharose 4B
beads (Pharmacia, Uppsala, Sweden) or Protein A
Dynabeads (Invitrogen, Carlsbad, CA). Precleared
samples will be equally divided, to be immunopre-
cipitated by either antibody or control serum. For an
experiment to be considered successful, 1/100 of the
DNA immunoprecipitated from the negative con-
trol samples should not generate a signal after 30–45
cycles of PCR, whereas the experimental immuno-
precipitation should generate a strong signal before
the twenty-eighth PCR cycle (when analyzed on an
agarose gel). An example of using this approach for
determining in vivo SATB1-binding sites in G10.
G4.1 (a T-helper-2 [T
2] cell line) within a 200 kb
2 cytokine cluster region is shown in Figure 1 (Cai
et al., 2006).
With ChIP followed by quantitative PCR
(ChIP-qPCR), it is possible to determine relative
fold difference in quantity of target sequence in
immunoprecipitated (IP) DNA prepared from
resting and activated cells, using input DNA as
a common reference for each primer set. For this
purpose, IP DNA and input DNA, after reverse
crosslinking, are quantified using the Quant-iT
PicoGreen dsDNA Assay Kit (Invitrogen,
© 2009 Kohwi-Shigematsu
Functional Packaging of Chromatin into Loops
Carlsbad, CA) or Qubit Quantitation
Platform (Invitrogen), followed by real-
time PCR on IP DNA. An absolute quan-
tification method is employed that deter-
mines the amount of the DNA fragments
of interest in the target genomic locus in
IP DNA using the standard curve gener-
ated by real-time PCR for each primer pair
from input DNA (Carter et al., 2002). We
determined the ratio R, where R =
(moles of target sequence in IP fraction/
moles of total IP DNA) / (moles of tar-
get sequence in input DNA/moles of
total input DNA). The denominator is 1
if the target sequence is unique. The data
we obtained using this formula are shown
in Figure 1c (Cai et al., 2006).
Chromatin Conformation
Capture (3C) and ChIP-3C
Formaldehyde crosslinking of cells can
capture chromatin loops that form in
vivo. If remote DNA sequences are
physically brought into close proximity
by chromatin looping in vivo, they can be
found in the same crosslinked chromatin
fragment, and the DNA sequences at the
stem of chromatin loops can be identified.
As shown in Figure 2a, after digesting
crosslinked chromatin with a restriction
enzyme that removes the loop portion of
chromatin, the two sequences initially
located at the stem of a loop can then be
ligated intramolecularly, after diluting the
chromatin sample to avoid intermolecular
ligation events. After reverse crosslinking
(to remove proteins), purified genomic
DNA that had been intramolecularly
ligated (excluding the loop portion of DNA) can be
amplified by PCR, using primer pairs designed from
the two distal sequences of interest. The ligation
products detected by PCR indicate that physical
interactions took place between DNA fragments
held together by higher-order chromatin structure.
The 3C method (Dekker et al., 2002) enables
us to estimate the frequency with which two
remote genomic sequences interact in space in a
given nucleus.
The more recently devised ChIP-loop assay (or
ChIP-3C) (Horike et al., 2005) is a modified 3C
analysis. ChIP-3C studies the chromatin fragments
that were immunoprecipitated with an antibody
specific for a chromatin-associated protein of interest.
The difference between 3C and ChIP-3C is that,
whereas whole crosslinked chromatin fragments are
used as templates for the 3C assay, the ChIP-3C assay
uses immunoprecipitated crosslinked-chromatin
fragments as templates. Although both assays are
used to study long-range DNA interactions that
might occur between genomic sites within the same
chromosome or between different chromosomes, the
ChIP-3C assay allows us to examine a specific group
of chromatin loops that are fastened at their base
with a specific protein. It also allows for the study
of chromatin loops that have histone modifications
associated with either transcriptionally active or silent
chromatin, depending on antibodies used against
specific modifications of histone that are correlated
with the transcriptional status of chromatin (Horike
et al., 2005).
Figure 1. ChIP analysis for SATB1 binding sites in the T
2 cytokine locus.
This analysis revealed nine SATB1-binding sequences within the 200 kb
2 locus (SBS-C1-C9). In addition, two other regulatory sequences (CNS-1
and CNS-2) were bound to SATB1. a, Schematic representation of the T
cytokine locus; b, Semiquantitative PCR; and c, Real-time PCR analyses of
crosslinked chromatin from resting and activated D10.G4.1 cells purified
by urea-gradient centrifugation, digested with Sau3A, and precipitated
with anti-SATB1.
© 2009 Kohwi-Shigematsu
As Figure 2a illustrates, for 3C, experiments include
the following steps:
(1) In vivo crosslinking;
(2) Urea-gradient ultracentrifugation;
(3) Digestion of purified crosslinked chromatin with
a restriction enzyme;
(4) Ligation of digested crosslinked chromatin over-
night at 16°C using T4 DNA ligase;
(5) Reversal of crosslinking; and
(6) PCR amplification: Multiple primers are designed
along the regions of interest.
For ChIP-3C, experiments include the following five
steps, as for typical ChIP experiments:
(1) In vivo crosslinking;
(2) Urea-gradient ultracentrifugation;
(3) Digestion of purified crosslinked chromatin with
a restriction enzyme;
(4) Preclearing of chromatin sample with control
nonimmune serum; and
(5) Immunoprecipitation.
The following three steps are additional and required
for ChIP-3C experiments:
(6) Ligation: ChIP-DNA on the beads are suspended
in ligation buffer, and DNA is allowed to be ligated
overnight at 16°C using T4 DNA ligase (Note:
Alternatively, the ligation step can be done
immediately after step 3 without affecting the
final results);
(7) Reversal of crosslinking; and
(8) PCR amplification: Multiple primers are designed
along regions of interest (Cai et al., 2006).
We will use the 200 kb murine T
2 cytokine gene
cluster locus (T
2 locus), shown in Figure 2a, as an
example. In this case, we used Sau3A restriction
enzyme-digested fragments (1–20) as well as BamH1/
BglII restriction enzyme-digested fragments (A–S),
which were used for the verification purpose. Two
primers (forward and backward) in each of DNA frag-
ments 2–19 were designed, as well as a forward primer
in DNA fragment 1 and a backward primer in DNA
fragment 2. The primers were designed to test the
looping events through SATB1-binding sequences
(SBSs) and control nonbinding sequences. The
genomic DNA prepared for either ChIP-3C or 3C
assaying, as described above, was subjected to PCR
amplification using various combinations of forward
and backward primer pairs derived from the DNA
fragments (1–20 or A–S). These PCR products were
derived from the 200 kb T
2 cytokine locus using
genomic DNA and are designated T
2. The purifica-
tion of crosslinked chromatin through urea-gradient
centrifugation allows more quantitative digestion
with restriction enzymes. Therefore, the pattern of
ligation products remains unchanged even after an
additional 10–20 cycles of PCR amplification. The
sequences need to be confirmed for all the ligation
products by cloning the PCR products, followed by
DNA sequencing of the insert.
When two fragments of interest are very closely
located in a linear configuration, some PCR
products are inevitably generated; typically, two
sequences have to be <4 kb apart to assess indepen-
dent looping events.
Controls and data analysis
The crosslinking frequencies of any two DNA frag-
ments need to be determined by the intensity of the
PCR signals of a ligation product. Their frequency
depends on the relative proximity of the two DNA
fragments to each other at a given time point. Rela-
tive ligation crosslinking frequencies of any two
DNA fragments were calculated as described below
in order to normalize various parameters, such as
PCR amplification efficiencies, ligation, crosslinking
efficiencies, and the amount of the template initially
used (Fig. 2b).
For analysis on the T
2 locus, the

-actin locus
(using primers S2 and S3 or B2 and B3, for Sau3A or
BamH1/BglII digested/re-ligated, crosslinked chroma-
tin templates, respectively) was used as an internal
control to normalize any difference in the amounts
of genomic DNA, and crosslinking and ligation
efficiencies. Upon either Sau3A or BamH1/BglII
digestion and ligation, a PCR product of 149bp or
131bp is generated, respectively. The PCR products
using S2 and S3 or B2 and B3 primers are labeled as
Actin. To correct for the amount of genomic DNA
used for the ChIP-loop assay, we added the plasmid
DNA containing the actin locus (see below) in the
original immunoprecipitated DNA samples to correct
for the amount of the starting DNA template used.
To correct for ligation and amplification efficiency
of different primers, we used a mixture of plasmid
DNA containing the

-actin locus between S1
and S4 primers or B1 and B4 primers, and a control
set prepared from two BAC (bacteria artificial
chromosome) clones covering the 200 kb T
cytokine locus. The latter was prepared by digesting
BAC clones with a restriction enzyme, either
Sau3A or BamH1/BglII, and re-ligated so that all
possible ligation products are present in the sample.
Functional Packaging of Chromatin into Loops
© 2009 Kohwi-Shigematsu
These control templates (from

-actin locus and
digested/re-ligated BAC clones) are subjected to
PCR amplification with the same series of primer
pair combination from different DNA fragments.
Relative crosslinking frequency is calculated using
the formula shown. The PCR products derived from
3C or ChIP-3C templates prepared from D10.G4.1
cells (abbreviated as D10) are indicated as T
2 or
Actin. The PCR products derived from the BAC
clones and Actin plasmid DNA are indicated by
2* and Actin*, respectively (Fig. 2b).
The relative crosslinking frequency corrects for
several factors: any differences in PCR amplification
efficiencies, crosslinking and ligation efficiencies,
the amounts of the template initially used, and the
size of the PCR products. The PCR products were
labeled by including 0.1 μl [

P] dATP (10 mCi/
ml) and 0.1 μl [

P] dCTP (10 mCi/ml) in each
reaction, resolved with 6% PAGE, and quantified
using a Storm Phosphorimager and ImageQuant
software (GE Healthcare, Uppsala, Sweden). We
confirmed that premixing primers had no effects on
the final PCR products (Fig. 2c). For the ChIP-3C
assay, relative crosslinking frequency was calculated
in a similar manner as for the 3C assay, and we used
20 ng of DNA for the 3C assay and ~1 ng of IP DNA
for the ChIP-3C assay. Actual data and the relative
crosslinking frequencies are shown for two SATB1-
binding sequences (SBS-C1 and SBS-1C) in Figure 3
(Cai et al., 2006).
By performing the ChIP and ChIP-3C assays, we
determined that, upon T
2 cell activation, SATB1
is rapidly induced and forms a unique transcription-
ally active chromatin conformation at the 200 kb
2 cytokine locus on mouse chromosome 11 (Cai
et al., 2006). This structure consists of chromatin
folded into many small loops, all anchored to SATB1
at their base. Further, transcription factors (GATA3,
STAT6, c-Maf), chromatin-remodeling enzyme
Brg1, and RNA polymerase I are all bound across the
200 kb region. By knocking down SATB1 using RNA
interference, the dense chromatin looping at the T
locus was inhibited, and Il-4, Il-5, I1-13, and c-Maf
Figure 2. 3C and ChIP-3C assays to determine higher-order chromatin structure. These assays were applied to determine intrac-
hromosomal interactions at the T
2 cytokine locus upon activation. a, The experimental strategies of ChIP-3C and 3C assays. b,
Analysis of 3C and ChIP-3C data. c, Verification that premixing primers from the T
2 cytokine gene locus and the actin locus does
not affect the final results of the 3C assay.
© 2009 Kohwi-Shigematsu
Figure 3. Chromatin looping involving SBS-C1 and SBS-C2 before and after activation of T
2 cells. a, Schematic representation
of T
2 cytokine locus and positions of 20 Sau3A DNA fragments used in the ChIP assay are shown. Black lines indicate chromatin
looping detected in resting cells. Violet lines indicate DNA fragments showing enhanced interactions upon cell activation. Red
lines indicate positions brought in close proximity after activation. Relative crosslinking frequencies (blue bars in the histograms)
are shown between fragment 2 (SBS-C1) as a fixed reference point (red bar in the histogram) or b, fragment 20 (SBS-C9) as a
fixed reference point and c, the other fragments of the locus. The experiments are as described in Figure 2a, b, and c.
Figure 4. Three-dimensional, transcriptionally active complex containing dense looping. a, Summary of 3C and ChIP-3C assays of
the cytokine regions in resting and activated D10.G4.1 cells. Black lines connecting two positions indicate juxtaposition of these
sites in resting cells. Pink lines represent two positions that show much increased crosslinking frequencies after activation. Red
lines represent two positions that are newly brought into close proximity upon activation. Black vertical arrowheads show direct
SATB1-binding sites. The crosslinking frequency for each ligation product generated between any two positions is represented
by the peak of the parabola connecting the positions. b, A schematic diagram based on the looping events shown in a, assum-
ing that all looping events can occur in a single cell. In this model, all small loops converge onto a common core base bound to
SATB1 (blue spheres). As a consequence, the total physical volume of the active transcriptional complex is reduced, enhancing
the accessibility of factors to genomic sites.
Functional Packaging of Chromatin into Loops
© 2009 Kohwi-Shigematsu
were not induced upon activation. These findings
indicate that SATB1 is required for T
2 activation
through the establishment of a specific transcription-
ally competent chromatin architecture at this locus
(Fig. 4).
Using the ChIP-3C assay, we previously reported
that methyl CpG-binding protein 2 (MeCP2)
regulates higher-order chromatin organization
and is responsible for chromatin looping of the
transcriptionally silent chromatin at the locus
of one of its target genes, Dlx5/6 (Fig. 5). This
experiment (then called ChIP-loop) used antibody
against a histone modification mark associated with
transcriptionally active or repressed chromatin
for chromatin immunoprecipitation (Horike et
al., 2005).
In contrast to the ChIP-3C assay, in which one
examines the interaction of any two genomic
sequences of choice mediated by a protein of choice,
ChIP-one-to-all identifies interacting partners
genomewide for a single, specific DNA sequence
of choice (as a bait). Several methods for capturing
interacting partners from a single bait locus have
been published (Simonis et al., 2006; Zhao et al.,
2006; Gondor et al., 2008). ChIP-one-to-all involves
high-throughput sequencing, and the protocol needs
to be well devised in order to reduce associated back-
ground problems. We will introduce our approach in
the lecture.
Alvarez JD, Yasui DH, Niida H, Joh T, Loh DY,
Kohwi-Shigematsu T (2000) The MAR-binding
protein SATB1 orchestrates temporal and
spatial expression of multiple genes during T-cell
development. Genes Dev 14:521-535.
Cai S, Han HJ, Kohwi-Shigematsu T (2003) Tissue-
specific nuclear architecture and gene expression
regulated by SATB1. Nat Genet 34:42-51.
Cai S, Lee CC, Kohwi-Shigematsu T (2006) SATB1
packages densely looped, transcriptionally active
chromatin for coordinated expression of cytokine
genes. Nat Genet 38:1278-1288.
Carter D, Chakalova L, Osborne CS, Dai YF,
Fraser P (2002) Long-range chromatin regulatory
interactions in vivo. Nat Genet 32:623-626.
de Laat W, Klous P, Kooren J, Noordermeer D,
Palstra RJ, Simonis M, Splinter E, Grosveld F
(2008) Three-dimensional organization of gene
expression in erythroid cells. Curr Top Dev Biol
Dekker J, Rippe K, Dekker M, Kleckner N (2002)
Capturing chromosome conformation. Science
Dickinson LA, Joh T, Kohwi Y, Kohwi-Shigematsu T
(1992) A tissue-specific MAR/SAR DNA-binding
protein with unusual binding site recognition. Cell
Fraser P (2006) Transcriptional control thrown for a
loop. Curr Opin Genet Dev 16:490-495.
Gondor A, Rougier C, Ohlsson R (2008) High-
resolution circular chromosome conformation
capture assay. Nat Protoc 3:303-313.
Horike S, Cai S, Miyano M, Cheng JF, Kohwi-
Shigematsu T (2005) Loss of silent-chromatin
looping and impaired imprinting of DLX5 in Rett
syndrome. Nat Genet 37:31-40.
Figure 5. MeCP2 mediates chromatin looping of
transcriptionally repressed chromatin at the Dlx5/6 locus. By
ChIP assay performed on 1-d-old, wild-type mouse brain,
MeCP2 binding shows peaks at the Dlx5/Dlx6 locus. ChIP-3C
assay using either anti-MeCP2 or antihistone H3 dimethylated
at K9 (a histone mark for transcriptionally repressed chromatin)
detected a looping event connecting positions F9 and R2 in
the 58 kb genomic region containing the Dlx5/Dlx6 locus.
Using antihistone H3 acetylated at K9/14 (a histone mark for
transcriptionally active chromatin) detected looping between
F1 and R5, as well as between F1 and R2. With MeCP2-null
mouse brain, the looping between F9 and R2 was undetected
with ChIP-3C using antihistone H3 dimethylated at K9.
© 2009 Kohwi-Shigematsu
Kohwi-Shigematsu T, deBelle I, Dickinson LA,
Galande S, Kohwi Y (1998) Identification of
base-unpairing region-binding proteins and
characterization of their in vivo binding sequences.
Methods Cell Biol 53:323-354.
Kosak ST, Groudine M (2004) Form follows
function: The genomic organization of cellular
differentiation. Genes Dev 18:1371-1384.
Milne TA, Zhao K, Hess JL (2009) Chromatin
immunoprecipitation (ChIP) for analysis of
histone modifications and chromatin-associated
proteins. Methods Mol Biol 538:409-423.
Simonis M, Klous P, Splinter E, Moshkin Y,
Willemsen R, de Wit E, van Steensel B, de Laat W
(2006) Nuclear organization of active and inactive
chromatin domains uncovered by chromosome
conformation capture-on-chip (4C). Nat Genet
Spilianakis CG, Flavell RA (2004) Long-range
intrachromosomal interactions in the T helper
type 2 cytokine locus. Nat Immunol 5:1017-1027.
Spilianakis CG, Lalioti MD, Town T, Lee GR,
Flavell RA (2005) Interchromosomal associations
between alternatively expressed loci. Nature
Wallace JA, Felsenfeld G (2007) We gather
together: insulators and genome organization. Curr
Opin Genet Dev 17:400-407.
Yasui D, Miyano M, Cai S, Varga-Weisz P, Kohwi-
Shigematsu T (2002) SATB1 targets chromatin
remodelling to regulate genes over long distances.
Nature 419:641-645.
Zhao Z, Tavoosidana G, Sjolinder M, Gondor A,
Mariano P, Wang S, Kanduri C, Lezcano M,
Sandhu KS, Singh U, Pant V, Tiwari V,
Kurukuti S, Ohlsson R (2006) Circular chro-
mosome conformation capture (4C) uncovers
extensive networks of epigenetically regulated
intra- and interchromosomal interactions. Nat
Genet 38:1341-1347.
© 2009 Szyf
The Epigenetic Impact
of Early Life Adversity
Moshe Szyf, PhD
Department of Pharmacology and Therapeutics
McGill University
Montreal, Quebec, Canada
© 2009 Szyf
the Epigenetic Impact of Early Life Adversity
Both animal and human studies have demonstrated
that exposure to social adversity early in life has a
long-lasting impact on behavior. A critical question
is, What are the mechanisms that preserve memory
of early life adversity? Recent studies suggest that
early life adversity results in a change in chromatin
modification and DNA methylation. These changes
alter gene expression programming in the brain in a
way that lasts into adulthood. It has been proposed
that these changes in DNA methylation and
chromatin structure result in long-lasting phenotypic
changes. Although the mechanisms mediating the
impact of early life experience on the epigenome are
unknown, signaling pathways elicited by exposure
to certain stimuli likely target sequence-specific
transcription factors to specific regulatory positions
in the genome. These, in turn, deliver chromatin and
DNA methylation enzymes to these positions.
The Epigenetic Programming of
Gene Expression
The genome is programmed to express an appropriate
set of genes in specific cells at specific time points in
life. The programming of the genome is accomplished
by the epigenome. Two elements of epigenomic
control are as follows: the different chemical
modifications of histones; and DNA methylation,
a covalent coating of the DNA with methyl groups
at discrete cytosines in the genome. The patterns
of chromatin modifications and DNA methylation
are tissue-specific. It has been well established that,
during gestation, a highly organized process creates
the patterns of DNA methylation and chromatin
modification. Recent data suggest that the DNA
methylation pattern is dynamic even in postmitotic
neurons (Levenson et al., 2006; Miller and Sweatt,
2007) and that different environmental exposures
during both gestation and early childhood can
modulate DNA methylation patterns that could lead
to long-term changes in gene expression.
Gene function, and thus phenotype, can be influenced
not only by variation in the gene sequence but also
by the epigenetic programming of gene expression.
Thus, aberrant human behavior could be caused
by either genetic polymorphisms or epigenetic
changes. There is a fundamental difference between
genetic variation, which involves random chance
mutations, and epigenetic changes, which appear
to be a coordinated response by the epigenome to
specific environmental exposures. Thus, it has been
hypothesized that epigenetic changes leading to
behavioral disorders are more prevalent and more
targeted than genetic sequence variations. It has
also been proposed that epigenetic reprogramming
early in life is an adaptive response that trains the
genome to be consistent with the social, biospheric,
and physical environments anticipated to exist
during that individual’s life course. This adaptive
response becomes maladaptive, however, when the
real environment differs from the environment
anticipated early in life. Epigenetic changes are
potentially reversible by pharmacological (Szyf, 2001)
and perhaps even cognitive interventions. Indeed,
several epigenetic drugs are now at different stages
of clinical trials for cancer (Weidle and Grossmann,
2000; Kramer et al., 2001) and psychiatric disease
(Simonini et al., 2006).
The Epigenome
The epigenome consists of the machinery for
programming long-term gene expression profiles and
thus defines gene function. Its known components
are histone modification and remodeling; a covalent
modification by methylation of cytosine rings found at
the dinucleotide sequence CG in DNA (Razin, 1998);
and small noncoding RNAs termed microRNAs,
which were recently discovered (Bergmann and
Lane, 2003). microRNAs regulate gene expression at
different levels: silencing chromatin, degradation of
mRNA, and blocking translation. microRNAs were
found to play an important role in cancer (Zhang
et al., 2007) and could potentially play a key role
in behavioral pathologies as well (Vo et al., 2005).
Recently, it has become clear that larger noncoding
RNA might be playing important roles in genome
regulation as well (Guttman et al., 2009). Histone
tails are modified by a panel of enzymatically catalyzed
chemical modifications such as methylation,
acetylation, ubiquitination, phosphorylation, and
sumoylation (Wade et al., 1997; Jenuwein, 2001;
Shilatifard, 2006).
DNA methylation in critical regulatory regions
can suppress gene expression. It does so either by
recruiting methylated DNA binding proteins such as
MeCP2, which in turn attract chromatin-modifying
enzymes (Nan et al., 1997), or by interfering with the
binding of transcription factors to genes (Comb and
Goodman, 1990; Inamdar et al., 1991). Although
the correlation between DNA methylation and gene
silencing might not be true for all promoters (Weber
et al., 2007), it has been well established for a large
subset of promoters (Rauch et al., 2009). However,
most methylated cytosines are not found in regulatory
regions of genes, and therefore, their role in genome
function is still unclear.
© 2009 Szyf
Links with the early life
social environment
Several principles of epigenome function are impor-
tant for understanding the link between the early life
social environment, DNA methylation variations,
and the resulting behavioral phenotypes.
Levels of epigenomic regulation
are interactive
All three levels of epigenomic regulation interact.
For example, an inverse relationship exists between
histone acetylation and DNA methylation:
Methylation of promoters triggers hypoacetylation
of histones through recruitment of methylated
DNA binding proteins and histone deacetylases
(Nan et al., 1997), and histone hyperacetylation
triggers DNA demethylation (Cervoni and Szyf,
2001). Because histone acetylation could trigger
DNA demethylation, there is a conduit through
which signaling pathways triggered by neuronal
activation can trigger DNA demethylation. For
example, activation of cAMP triggers recruitment
of CREB-binding protein (CBP), a histone
acetyltransferase (HAT) to specific genes, causing
gene-specific hyperacetylation (Parekh and Maniatis,
1999; Weaver et al., 2007). This bilateral relation
also has pharmacological implications: DNA
methylation could be altered by drugs; in turn,
interventions that alter histone acetylation and
drugs that alter DNA methylation would change
chromatin modification.
Specificity of DNA
methylation response
Most of the enzymes that catalyze chromatin and
DNA methylation are not gene-specific and need to
be delivered to specific genes by sequence transcrip-
tion factors (Weaver et al., 2007). This specificity of
DNA methylation response to an environmental cue
would be defined by sequence-specific factors that lie
downstream to the signaling pathway activated by
the extracellular signal.
Chromatin modification and DNA
methylation: implications for therapy
Chromatin modification and DNA methylation
states are a dynamic balance of modifying and
demodifying enzymes. Histone acetylation involves
an equilibrium between HAT enzymes and
histone deacetylase (HDAC) activities; likewise,
histone methylation is a balance between histone
methyltransferases and histone demethylases. DNA
methylation also balances DNA methyltransferases
(DNMTs) and demethylases (Szyf, 2009).
The concept that epigenetic changes are reversible
has clear implications for therapy. First, the
epigenetic equilibrium could be tilted in either
direction using inhibitors of the modifying or
the demodifying enzymes. For example, HDAC
inhibitors were recently shown to ameliorate the
severity of symptoms in preclinical models relevant
to Huntington’s disease, Parkinson’s disease, anxiety
and mood disorders, Rubinstein-Taybi syndrome, and
Rett syndrome (Abel and Zukin, 2008). Moreover,
since it is becoming clear that the epigenetic
enzymes lie downstream to signaling pathways that
are triggered by neuronal activation, it might be
possible to modulate epigenetic states by behavioral
rather than therapeutic interventions. To this end,
it has been shown in a mouse model of neuronal
degeneration that environmental enrichment could
improve learning behavior after significant brain
atrophy and neuronal loss had already occurred; this
improvement was associated with increased histone
acetylation (Fischer et al., 2007). Although DNA
methylation has not been directly implicated in
these studies, it stands to reason that this process
would respond as well to behavioral therapy.
Epigenetic signatures
So-called epigenetic signatures of early life exposures
appear to be programmed responses to a set of altered
environmental or physiological cues rather than
errors that occurred at random. Unraveling the
gene circuitries involved in the epigenomic response
would provide insight into the links between
environmental cues, the adaptive readjustment of
genomic programs, and the resulting physiological
and pathological consequences.
Genomic Adaptation to
Changing Environments
Genomes adapt to changing environments at several
different time scales: an evolutionary time scale that
spans numerous generations, a transgenerational time
scale that spans two or several generations, a single-
generation time scale, and a physiological time scale
ranging from minutes to years. Random mutations in
the gene sequence that provide selective advantage
in changing environments could serve as a mecha-
nism of adaptation on an evolutionary time scale.
However, random genetic drift cannot be at work in
a single generation or even a few generations.
The twentieth century has seen a dramatic increase
in several late-onset diseases and mental health
issues that cannot be explained by genetic variation.
A reasonable hypothesis is that dramatic changes in
the social biosphere and physical environment in the
© 2009 Szyf
the Epigenetic Impact of Early Life Adversity
past century have resulted in an adaptive epigenomic
response that involves several systems and gene
circuitries, and that this adaptive response can
become maladaptive under changed circumstances
later in life. Different levels of social adversity early
in life have been proposed as cues for this adjustment
of epigenomic programming. Although the evidence
for this hypothesis is scant, some data from rodent
and human studies are consistent with it.
Evidence from animal models
The long-term effects of maternal behavior in
the rat (as well as other mammals) on the stress
responsiveness and behavior of the offspring during
adulthood are well documented. In the rat, the adult
offspring of mothers that exhibit increased levels of
pup licking/grooming (High LG mothers) during
the first week of life show several positive changes:
increased hippocampal glucocorticoid receptor
(GR) expression, enhanced glucocorticoid feedback
sensitivity, decreased hypothalamic corticotrophin-
releasing factor (CRF) expression, and more modest
HPA (hypothalamic-pituitary-adrenal) stress
responses, compared with animals reared by Low
LG mothers (Liu et al., 1997; Francis et al., 1999).
Cross-fostering studies suggest that maternal care has
direct effects on both gene expression and stress and
therefore support an epigenetic mechanism (Liu et
al., 1997; Francis et al., 1999). In accordance with
this hypothesis, differences in DNA methylation and
histone acetylation in the regulatory regions of the
glucocorticoid receptor (GR) exon 1
promoter gene
were observed in the hippocampus of the offspring
of High LG and Low LG mothers. These differences
in epigenetic programming emerge early in life in
response to differences in maternal LG and remain
stable into adulthood (Weaver et al., 2004).
This animal model allows for testing plausible
mechanisms that link social cues and physical
changes, such as covalent modification of DNA. Our
fundamental hypothesis is that the brain responds
to social cues by eliciting signaling pathways in
neurons, and that these in turn signal changes
in DNA methylation. For example, maternal
behavior triggers a signaling pathway that involves
the serotonin receptor, an increase in cAMP, and
recruitment of the transcription factor NGFI-A
(nerve growth factor–inducible protein A). This
transcription factor in turn recruits the histone
acetyltransferase CBP and the methylated DNA
binding protein and candidate DNA demethylase
MBD2 (methylcytosine binding domain 2) (Weaver
et al., 2007) to the GR promoter. Our hypothesis is
that the increased histone acetylation triggered by
CBP facilitates the demethylation of the gene by
MBD2 or other DNA demethylases (our unpublished
data). Pharmacologically induced histone acetylation
was previously shown to trigger DNA demethylation
(Cervoni and Szyf, 2001; Cervoni et al., 2002). Thus,
these data chart a course for how maternal behavior
results in epigenetic modification of a specific gene
in the brain. A similar paradigm has been proposed
to explain how epigenetic response mediates other
social and physical environments and at other points
in life.
As discussed above, the epigenomic readjustment to
early life social environment is not limited to one gene
or one exon; rather, it is consistent with a systemic
response. Recent high-density epigenome mapping of
chromosome 18 in the adult rat offspring of High LG
and Low LG mothers revealed broad differences in
DNA methylation and histone acetylation covering
wide regions of chromosome 18. High maternal care
resulted in hypomethylation of some regions and
hypermethylation of others, and an inverse picture
was observed with histone acetylation (P. McGowan,
M. Meaney, M. Suderman, M. Hallett, and M. Szyf,
unpublished data). These findings explain why the
adult offspring of High LG and Low LG mothers
exhibit widespread differences in gene expression
(Weaver et al., 2006).
The rat model also allowed us to test the hypothesis
that epigenetic programming is reversible using
pharmacological epigenetic modulators. Injecting
the HDAC inhibitor trichostatin A (TSA) into the