Programming Smooth Muscle Plasticity With
Chromatin Dynamics
Oliver G. McDonald, Gary K. Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Abstract—Smooth muscle cells (SMCs) possess remarkable phenotypic plasticity that allows rapid adaptation to
fluctuating environmental cues. For example, vascular SMCs undergo profound changes in their phenotype during
neointimal formation in response to vessel injury or within atherosclerotic plaques. Recent studies have shown that
interaction of serum response factor (SRF) and its numerous accessory cofactors with CArG box DNA sequences within
promoter chromatin of SMC genes is a nexus for integrating signals that influence SMC differentiation in development
and disease. During development, SMC-restricted sets of posttranslational histone modifications are acquired within the
CArG box chromatin of SMC genes. These modifications in turn control the chromatin-binding properties of SRF. The
histone modifications appear to encode a SMC-specific epigenetic program that is used by extracellular cues to influence
SMC differentiation, by regulating binding of SRF and its partners to the chromatin template. Thus, SMC differentiation
is dynamically regulated by the interplay between SRF accessory cofactors, the SRF–CArG interaction, and the
underlying histone modification program. As such, the inherent plasticity of the SMC lineage offers unique glimpses
into how cellular differentiation is dynamically controlled at the level of chromatin within the context of changing
microenvironments. Further elucidation of how chromatin regulates SMC differentiation will undoubtedly yield
valuable insights into both normal developmental processes and the pathogenesis of several vascular diseases that
display detrimental SMC phenotypic behavior. (Circ Res. 2007;100:1428-1441.)
Key Words: smooth muscle cells (SMCs) 䡲 serum response factor (SRF) 䡲 myocardin
䡲 chromatin 䡲 histone modifications
D
rioocclusive disease, in which SMC pathophysiology plays a
prominent role. Many other parenchymal cell lineages (including endothelial cells, various epithelial cell types, fibroblasts, hepatocytes, chondrocytes, and glial cells) display
similar reactive changes in their phenotype under pathophysiological conditions, suggesting that some if not many molecular principles underlying SMC physiology might apply to
these cell lineages as well.
ifferentiation of vascular smooth muscle cells (SMCs)
from embryonic stem cells (ESCs) during development
is characterized by the appearance of proteins (eg, smooth
muscle [SM] ␣-actin, SM–myosin heavy chain [MHC])
whose expression is restricted to the SMC lineage and
required for SMC contraction and regulation of blood pressure under adult physiological conditions. However, unlike
skeletal and cardiac myocytes, which are terminally differentiated, SMCs within adult animals readily switch phenotypes
in response to changes in local environmental cues.1 For
example, vascular SMCs express high levels of SMC-specific
contractile proteins and do not generally proliferate, migrate,
or secrete significant amounts of extracellular matrix. However, in response to extracellular cues released at sites of
vascular injury or within atherosclerotic lesions, SMCs exhibit decreased expression of SMC-specific contractile proteins and increased migration, proliferation, and production
of extracellular matrix components as well as matrix metalloproteases. These changes presumably evolved as an important survival mechanism to repair vascular damage, and the
process appears to be fully reversible if the pathological
stimuli dissipate.2 Arteries are especially predisposed to this
process, often with fatal consequences resulting from arte-
Transcriptional Control of SMC
Differentiation by Serum Response Factor
Transcriptional control of SMC gene expression plays a
central role in the maintenance of SMC differentiation and
control of SMC phenotypic plasticity. For example, studies in
transgenic mice have demonstrated that SMC-specific contractile gene promoters are transcriptionally active within
vascular SMCs under physiological conditions, whereas they
are transcriptionally repressed in response to vascular injury3
and atherosclerotic plaque formation.4 After resolution of the
injury, SMCs resume transcription of these genes and regain
their fully differentiated phenotype.3,5 Thus, there is great
interest in identifying molecular mechanisms that regulate
SMC gene transcription, so that vascular pathologies might
Original received November 11, 2006; resubmission received February 5, 2007; accepted March 20, 2007.
From the Department of Molecular Physiology and Biological Physics, University of Virginia Health Sciences Center, Charlottesville.
Correspondence to Gary K. Owens, Department of Molecular Physiology and Biological Physics, University of Virginia Health Sciences Center, PO
Box 800736, Charlottesville, VA 22903. E-mail [email protected]
© 2007 American Heart Association, Inc.
Circulation Research is available at http://circres.ahajournals.org
DOI: 10.1161/01.RES.0000266448.30370.a0
1428
McDonald and Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
be better understood and potentially controlled. Strikingly,
nearly all SMC-restricted contractile protein genes examined
to date contain evolutionarily conserved CArG box DNA
sequences within their promoters, and these sequences are
required for SMC gene transcription in vivo.6 Paradoxically,
many other genes important for induction of SMC phenotypic
switching within vascular lesions, including genes important
for migration, proliferation, and extracellular matrix production, also contain CArG boxes within their promoters.7 Thus,
it appears that CArG box DNA sequences may have evolved
within the context of SMCs for coordinate regulation of the
multiple transcriptional programs that underlie differentiation
and plasticity of the SMC phenotype.
CArG boxes serve as binding sites for serum response
factor (SRF). The original biological function of SRF appears
to be that of a generic DNA-binding transcriptional activator
important for expansion of cell populations, as the yeast SRF
homolog MCM1 regulates DNA synthesis via binding DNA
sequences near replication origins and by transcriptional
activation of several cell-cycle control genes important for
proliferative growth of this organism.8 Indeed, this function
has been evolutionarily conserved as SRF was so named
because it was first found to regulate transcription of immediate early genes important for initiation of mitogenesis in
response to the addition of serum into the media of various
cultured mammalian cell lines.9 It has been appreciated for
quite some time that SRF homodimers perform this function
via binding to CArG box DNA sequences within the promoters of immediate early genes (eg, c-fos) and forming a ternary
complex with various serum-activated Ets-domain proteins
(ie, the ternary factors [TCFs] Elk-1, Sap-1, Fli-1) that assist
SRF in stimulating transcription of these genes.10 However,
during evolution from unicellular to complex multicellular
organisms, SRF acquired multiple other functions in addition
to induction of cellular proliferation, including regulation of
myogenesis,1 cell migration,11–13 cell– cell adhesion,12 cytoskeletal assembly and organization,11,12,14 and extracellular
matrix production.15 This is consistent with the coevolution of
CArG boxes in the promoters of numerous genes important
for these diverse cellular operations.15 For a comprehensive
consideration of these and related topics, see several seminal
reviews by Miano and colleagues.6,7 In addition, a plethora of
environmental cues have been found to regulate SRF transcriptional activity in mammals, perhaps because the SRF–
CArG interaction was selected by nature to control such a
large and varied repertoire of cellular behaviors. It is no
surprise then that SRF activity has emerged as a paradigm for
how diverse cellular processes are coordinated at the level of
transcription for adaptation to changing microenvironments,
especially within the context of SMCs.
Interest in how environmental cues regulate SRF activity in
SMCs was intensified by the recognition that there are 3
major paradoxes regarding the transcriptional activity of
SRF.6 First, as noted above, SRF has the ability to simultaneously activate transcription of genes involved in opposing
cellular processes, such as differentiation and proliferation.
Second, SRF itself is a weak transcriptional activator, probably because its transcriptional activation domain (TAD)
does not effectively recruit the basal transcription machinery
Smooth Muscle Chromatin
1429
relative to other transcription factors with strong TADs.
Third, SRF is expressed in all cell lineages, yet only activates
transcription of SMC-restricted contractile genes in SMC,
raising questions as to how this ubiquitously expressed
transcription factor can selectively activate SMC-specific
gene expression in SMCs and not in non-SMCs. To explain
these issues, it was thought that SRF must bind to SMC gene
promoters and subsequently recruit other muscle-specific
promyogenic accessory factors with strong TADs (ie, similar
to the ternary complex factors at growth genes) and that this
interaction was amenable to regulation by environmental cues
that influence the SMC phenotype.
This idea was validated in dramatic fashion by the discovery of the SRF coactivator myocardin,16 perhaps the most
potent transcriptional coactivator yet identified in nature.17
Myocardin is exclusively expressed in SMCs and cardiomyocytes, possesses a powerful C-terminal TAD with the capability to selectively activate transcription of cardiac and
SMC-specific contractile genes to levels never imagined (eg,
several thousand fold by in vitro transient transfection assays), and physically associates with SRF to form a ternary
complex on CArG box DNA.18 Several extracellular cues
have since been demonstrated to positively and negatively
modulate the expression and transcriptional activity of myocardin in SMCs, and numerous other SRF accessory factors
have subsequently been shown to transmit these extracellular
signals to myocardin/SRF.18 Thus, available data suggest a
model whereby SRF functions as a protein platform that
generically binds CArG box DNA to recruit other downstream accessory factors that either stimulate or repress
transcription of antagonistic gene subsets in response to
various environmental cues. Although this model will likely
prove essentially correct at a simplified level, there is abundant evidence that the story is much more complicated than
this model portends, as outlined below.
SRF Activity Within the Context of
SMC Chromatin
It has become widely appreciated over the past decade that
transcription takes place on promoter DNA within the context
of chromatin (ie, not a naked linear strand of DNA) and that
the structure of this chromatin plays an active and fundamental role in transcriptional control.20 Ironically, although much
of the original work demonstrating that chromatin was an
important regulator of transcription was performed by experiments examining the SRF-dependent c-fos promoter nearly 2
decades ago,21 very little is known regarding how SRF
activates transcription within the context of intact chromatin.
Because SRF and its accessory factors must operate within
this chromatin environment, future revisions to the current
model explaining SRF activity will be incomplete without an
accurate description of this process.
At its most fundamental level, chromatin is organized into
repeating units of nucleosomes, the basic building blocks of
chromatin. The nucleosome is composed of 146 base pairs of
genomic DNA wrapped around an octamer of histone proteins (2 copies each of histone H2A, H2B, H3, and H4).
Nucleosomes are in turn connected to one another by variable
lengths of so called linker DNA, forming the characteristic
1430
Circulation Research
May 25, 2007
A
DNA
+
=
Nucleosomes
Histones
B
No n - H i s t o n e
Me
Nucleosomes
=
+
Me
TFs
Ac
TFs
TFs
N o n - Hi s t o n e
TFs
TFs
TFs
Ac
N o n - H is t o n e
Chromatin
Figure 1. Fundamental mechanisms regulating
chromatin structure. A, Free DNA is wrapped
around histone proteins (octamers of 2 copies
each of histone H2A, H2B, H3, and H4) into a
“beads on a string” polymer of nucleosomes. DNA
is wrapped in approximately 1.67 turns around
each octamer. B, Nucleosomes are in turn complexed with other DNA (eg, transcription factors
[TFs]) and histone-binding proteins (nonhistone)
and modified with posttranslational modifications
to the histones and DNA. All of these components
are complexed together to form native chromatin.
Gray cylinders represent histone octamers; black
lines wrapped around and connecting them represent DNA molecules. Ac indicates acetylation; Me,
methylation.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Chromatin DNA- binding Other Chromatin
Associated
Modifications
Proteins
Proteins
“beads on a string” morphology that is visible by electron
microscopy.22 Furthermore, there are numerous other nonhistone proteins complexed with nucleosomes and linker DNA
within the chromatin polymer. The basics of chromatin
structure are outlined in Figure 1.
Although folding of linear chromatin fibers into the higher
order 3D configurations that form chromosomes is not well
understood, it is known that chromatin exists primarily in
either of 2 rudimentary conformations along a chromosome23:
euchromatin and heterochromatin. These 2 forms of chromatin display properties that are opposite from one another.
Euchromatin tends to exist in a conformation that is uncondensed and “open” to the nuclear environment in that it does
not form clumped intranuclear aggregates microscopically.
Indeed, nucleosomes within euchromatin are not tightly
packed; they are variably spaced and highly mobile, providing plentiful free DNA to proteins, such as transcription
factors that require physical contact with DNA. It is therefore
not surprising that euchromatic DNA sequences are generich. Hence there is an abundance of active transcription and
mRNA production from euchromatic areas of the genome. In
contrast, heterochromatin exists in a condensed conformation
that manifests as densely staining clumped material microscopically. Here, nucleosomes are packed closely together
and display restricted mobility. There is little free DNA and
access to proteins that bind DNA is severely inhibited.
Transcription is a very inefficient and rare event. Much of the
gene-poor repetitive DNA sequences originating from past
viral integration events are stored here, and deposition of
these sequences into condensed heterochromatin is thought to
guard against widespread genomic instability. However, recent studies have revealed that foci of heterochromatin are
also formed within euchromatic regions of higher eukaryotic
genomes; here heterochromatin functions to inhibit transcription of cell-specific genes in cells where they must be kept
silent.20,24 –26 The opposing properties of heterochromatin and
euchromatin are depicted in Figure 2. If chromatin holds
influence over silencing of SMC genes in non-SMCs, heterochromatin is the form that one would expect SMC genes to
take within non-SMC lineages. Indeed, one could easily
envision a scenario whereby CArG box DNA sequences are
wrapped into euchromatin or heterochromatin-like structures,
thereby dictating the ability of SRF to bind its cognate cis
element and activate transcription in SMCs versus nonSMCs. Because of this, investigations into whether chromatin
might influence the DNA-binding properties of SRF to SMC
gene promoters were initiated.
Chromatin Regulates SRF Binding to CArG
Box DNA
The first piece of evidence that suggested chromatin has the
potential to regulate SRF binding was implicitly provided by
publications describing of the x-ray crystal structure of SRF
bound to CArG box DNA27 and the crystal structure of the
nucleosome core particle itself.28 If one compares these
structures, it is apparent that SRF makes several critical
contacts with DNA minor groove phosphates that would be
obscured if this DNA were nucleosomal. This is because
acidic minor groove phosphates required for interaction with
the SRF MADS box are buried within basic histone amino
acids as a functional anchor for DNA as it winds around
histone octamers in solution. From thermodynamic considerations alone, the DNA-binding activity of SRF would be
greatly inhibited if not impossible if SRF molecules were to
encounter CArG box DNA sequences tightly wrapped around
histone octamers. The extreme bending of the DNA template
as it tracks around the histones would be expected to exert a
high degree of steric hindrance as well. The next piece of
evidence that chromatin might regulate the ability of SRF to
bind CArG box DNA was provided by comparing the ability
of SRF to bind SMC gene promoters by electrophoretic
mobility assays (EMSAs) and chromatin immunoprecipitation (ChIP) assays.29 –31 The EMSAs, which used radiolabeled, naked DNA templates not assembled into chromatin as
probes to monitor SRF binding in cell lysates in vitro, showed
that SRF promiscuously binds to SMC gene promoter DNA
probes with equal affinity in lysates from cultured SMCs and
non-SMCs alike. This is consistent with the notion that SRF
McDonald and Owens
Smooth Muscle Chromatin
1431
Figure 2. Heterochromatin and euchromatin possess opposite properties
Me
“Closed” Heterochromatin
regarding nucleosome spacing, histone
Versus “Open” Euchromatin
modifications, transcription factor bindMe
TF
ing, and transcriptional activity. MethylMe
ated histones in heterochromatin (red
TF
pentagons) recruit repressive methylMe
binding proteins (MBPs) (eg, HP1, PRC)
Me
that function to tightly compact chromatin and block binding of transcription factors (TFs) to DNA. In euchromatin, H4
acetylation (Ac) (orange flags) functions
to inhibit tight nucleosome packing,
thereby providing abundant free DNA for
transcription factor binding. Methylation
E n vi r o nm en t a l
(Me) of certain histone residues, such as
Cues
Transcriptionally Repressed
H3 Lys4 (green pentagons), further creTranscriptionally Active
Euchromatin
Euchromatin
ates a transcriptionally permissive environment by providing docking sites for
protein–protein contacts with MBPs (eg,
TF
TF
CHD1) that recruit transcription factors.
Me MBP
When transcription is transiently
c
A
Me
TF
repressed within euchromatin in
TF
Pol II
TF
response to various environmental cues,
HDACs
TBP
HDACs transiently remove acetyl groups,
Me
which results in chromatin compaction
HATs
and inhibition of transcription factor
binding. The presence of methylated histones within repressed chromatin might preserve crucial information regarding memory of
which SMC genes require reactivation of transcription on termination of repressive signals. Not depicted are histone demethylation
events, which may also contribute to reversible changes in transcription.
Me
M
T r a ns c r i p t i o n a l l y S i l en t
Heterochromatin
BP
M
BP
M
BP
M
Ac
BP
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
is bound to SMC gene promoters in non-SMCs but does not
activate transcription because of the absence of cofactors
such as myocardin. In stark contrast, results obtained by
several independent studies using ChIP assays revealed that
SRF binds to endogenous SMC gene promoters much more
effectively in SMCs than in non-SMCs within the context of
native chromatin.30,31–33 Furthermore, micrococcal nuclease
digestion experiments demonstrated that the CArG box chromatin of SMC gene promoters exists in a form that is much
more accessible to digestion in SMCs than in non-SMCs,
suggesting that SMC genes are euchromatin-like in SMCs
and heterochromatin-like in non-SMCs.32 These results revealed that although SRF is expressed in all cell types and has
the inherent ability to bind SMC gene CArG box DNA
equally in SMCs and non-SMCs, SRF preferentially binds to
accessible chromatin of endogenous SMC genes in SMCs as
opposed to non-SMCs. The final piece of evidence that
implicitly suggested that chromatin might be an important
regulator of the ability of SRF to activate SMC gene
expression (in the context of the above findings) was the
discovery of myocardin-related transcription factors.34 These
factors behave in ways that are similar to myocardin in terms
of strong activation of SMC gene transcription, but these
proteins are expressed in numerous different embryonic and
adult cell types (rather than muscle-restricted, like myocardin). Because of this, these proteins are theoretically predicted to ectopically activate SMC gene expression in nonSMCs if SMC gene promoters were constitutively occupied
with SRF because of the absence of a heterochromatin
conformation, which is not the case under native conditions.18
Collectively, these data argue that the reason SRF does not
activate SMC gene expression in non-SMCs is not simply
attributable to the absence of myogenic SRF coactivators in
these cells. Rather, it appears that there might be some
property of the promoter chromatin that inhibits SRF binding
to CArG boxes of SMC genes in non-SMCs, which in turn
precludes recruitment of SRF coactivators under native conditions. Thus, the process that dictates whether SMC genes
exist in a euchromatin or heterochromatin conformation in
SMCs versus non-SMCs might represent a fundamental
mechanism that underlies SMC identity, via regulation of
SMC-restricted SRF binding to SMC genes. Furthermore, if
chromatin-based control of SRF binding is important for
specifying the SMC lineage, this mechanism might also be
used by environmental factors that influence the phenotypic
plasticity of adult SMCs, in conjunction with the wellrecognized role of SRF cofactors.
Histone Modifications Regulate Chromatin
Structure and Function
Recent studies have attempted to characterize components of
CArG box chromatin that regulate the SMC-restricted binding activity of SRF. These efforts have thus far primarily
focused on identification and functional characterization of
posttranslational modifications to histone octamers within the
chromatin of SMC genes. This is because a wealth of data
obtained from classically studied cell systems (eg, yeast,
Drosophila, HeLa cells) have unequivocally demonstrated
that chromatin structure and function is profoundly influenced by posttranslational modifications to the histone proteins within the nucleosomes, such as methylation, acetylation, phosphorylation, and ubiquitination of histones by the
action of nuclear protein complexes with catalytic activity
toward specific histone amino acid residues.20,35 Over the past
decade, numerous studies have documented that many of
these modifications are found solely within areas of heterochromatin, whereas others are found exclusively in euchromatin. Because studies in SMCs have thus far focused on
1432
Circulation Research
May 25, 2007
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
acetylation and methylation of histone lysine residues, we
have limited our discussion of the vast subject of histone
modifications to these. These modifications are thought to
serve 2 purposes. First, histone methylation patterns are
thought to represent an epigenetic program that preserves
information regarding cellular identity when cells are induced
to divide or exposed to changes in the native environment.35
Second, histone acetylation is thought to represent a means by
which chromatin structure can be rapidly and reversibly
adjusted to dynamically regulate transcription factor binding
in response to fluctuating environmental cues.36 These concepts are detailed below.
There is a plethora of data supporting the notion that
different histone methylation patterns across eukaryotic genomes function as specialized binding surfaces within the
chromatin that function to attract protein complexes that
contain chromatin remodeling and transcriptional activation/
silencing activity. It is also thought that in some circumstances, methylation of histone lysine residues in chromatin
represents a means by which information storage regarding
basal transcriptional competence is preserved during periods
when cells are induced to proliferate or change patterns of
gene expression away from baseline.37 Four observations
support these concepts. First, bulk turnover of methylated
histones is extremely low, indicating that histone methylation
is a stable modification that is not often removed once it is
deposited into chromatin.20 This implies that once histone
methylation patterns are programmed across the genome,
they are present for an extended or even indefinite time over
a cells lifetime. Second, methylated histones serve as docking
sites for protein modules that are members of multisubunit
complexes, thereby directly tethering these complexes to the
chromatin template.35 Remarkably, some methylated lysine
residues recruit complexes that promote assembly of heterochromatin whereas others recruit protein complexes that
activate transcription, and these antagonistic modifications
are found almost exclusively in areas of heterochromatin and
euchromatin, respectively (eg, see the Table). Third, there are
data indicating that under some circumstances histone methylation patterns at specific loci remain relatively constant
amid fluctuations in transcriptional activity from baseline. In
particular, a “methyl/phos” binary switch has been proposed38
whereby reversible phosphorylation of serine/threonine residues located adjacent to methylated lysines can function to
transiently displace methyl-docking proteins, thereby potentially antagonizing the action of these protein complexes.
However, once the phosphate groups are removed by phosphatases, methyl lysine– binding proteins can relocate to the
locus and resume their activities.39 – 41 Therefore, histone
methylation has potential to retain information regarding
baseline transcriptional competence of a gene, via directing
reactivation or resilencing after termination of cellular events
that trigger histone phosphorylation or other mechanisms that
displace methyl-binding modules off the chromatin template.42 Fourth, histone methylation patterns and their functional consequences are faithfully inherited from parent to
daughter cell, along with the associated genomic DNA during
both mitotic and meiotic cell divisions.24,43 This strongly
suggests that histone methylation evolved in part to provide
SMC-Specific Histone Modifications Distinguish SMCs from
Non-SMCs
SMC Gene Modifications
Cell Type/Chromatin Type
H3 Lys4 methylation
SMCs/Euchromatin
H3 Lys9 acetylation
SMCs/Euchromatin
H3 Lys14 acetylation
SMCs/Euchromatin
H3 Lys79 methylation
SMCs/Euchromatin
H4 acetylation
SMCs/Euchromatin
H3 Lys9 methylation
Non-SMCs/Heterochromatin
H3 Lys27 methylation
Non-SMCs/Heterochromatin
H4 Lys20 methylation
Non-SMCs/Heterochromatin
stable units of epigenetic inheritance for information transmission across generations.
In light of these amazing observations, histone methylation
has emerged in the literature as an attractive candidate for a
bona fide “epigenetic memory” of long-term transcriptional
competence and therefore a primary determinant of cellular
identity. Although there is a wealth of data supporting this
notion, it must be noted that this idea has been challenged by
the recent discovery of several histone demethylase enzymes,
although it is still far from clear as to how and to what degree
these enzymes operate in vivo.37 Therefore it is currently very
controversial as to whether histone lysine methylation is just
another dynamic entity similar phosphorylation/acetylation or
if it really does represent a genomic indexing system that
resists erasure. The answer probably lies somewhere in
between, in that there may be some developmental contexts,
genomic loci, or histone residues that are resistant to demethylation, whereas others are amenable to it.37 To further
complicate matters, histone lysines can be mono-, di-, or
trimethylated (abbreviated hereafter as H#K#Me#; eg, H3
Lys4 dimethylation is H3K4Me2, H3 Lys27 trimethylation is
H3K27Me3, and so on), and different histone methyltransferases and demethylases appear to possess varying levels of
enzymatic activity toward these different states in addition to
their different specificities toward histone residues. In light of
these findings, other components of the chromatin fiber have
been proposed to carry out the functions classically ascribed
to histone methylation,44 although it is still believed that
histone methylation in some form holds these properties even
amid the existence of demethylases.37
Whereas methylated histones appear to function primarily
as targeting platforms within chromatin, histone acetylation
has been shown to directly relieve structural chromatin
compaction, possibly through disruption of interactions between adjacent nucleosomes and through loosening of contacts between histones and DNA.28,45– 47 The presence of
acetylated histones within promoter chromatin is thought to
promote local unfolding of the otherwise condensed chromatin fiber coupled with partial or complete unwrapping of
DNA from histones, and the degree to which these adjustments to chromatin structure occur are a function of the
quantitative levels of acetylation.36 There is also evidence that
acetylation events may also facilitate the action of ATPdependent chromatin remodeling proteins that mobilize DNA
from histone octamers48 as well as directly recruit members
McDonald and Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
of the basal transcription apparatus.49 All of these effects are
thought to synergize to expose the DNA template to transcription factors that require physical contact with DNA to
activate transcription (eg, SRF). Acetyl groups added to
histones by histone acetyltransferases (HATs) are readily and
often removed by the action of other enzymes with histone
deacetylase (HDAC) activity in vivo. Therefore, histone
acetylation by HATs actively facilitates chromatin unfolding
and presentation of DNA to transcription factors, whereas the
removal of acetyl groups from histones by HDACs reverses
these effects. The result is 2-fold. First, euchromatin is
enriched with acetylated histones, whereas heterochromatin is
relatively devoid of acetylated histones. Acetylation therefore
provides 1 explanation as to why these 2 forms of chromatin
take on opposite properties in terms of chromatin condensation and transcription factor accessibility. In effect, the
presence of acetylated histones primes euchromatin for transcriptional activation. Second, quantitative regulation of acetylation levels within euchromatin provides a means whereby
chromatin compaction and transcription factor binding can be
rapidly and reversibly adjusted in a graded fashion in response to fluctuating extracellular cues that signal to HATs
and HDACs.36 This provides an efficient mechanism for fine
titration of transcription of genes within euchromatin, so that
cells may effectively adapt to fluxes in the extracellular
microenvironment.
A Histone Modification Program Specifies the
SMC Lineage
Although there are numerous molecular features that underlie
cell lineage specification, there are certain features about
histone modifications that make them particularly strong
candidates to participate in this function. All histone modifications are in intimate association with DNA, and histone
methylation patterns are inherited through cell division, along
with the genetic information contained in the DNA sequence.
It is therefore conceivable that different histone modification
patterns that code for the assembly of euchromatin or heterochromatin are present at the same genetic loci in different cell
lineages, thereby giving rise to the cell-specific gene expression potential characteristic of the multitude of diverse cell
types found in higher eukaryotes.20 Indeed, several recent
observations support this concept, including studies examining histone modification patterns in SMCs. SMC-specific sets
of histone modifications are acquired at the promoter chromatin of SMC genes during development from multipotential
stem cells to SMCs. Several studies from our laboratory and
others have shown that acetylation to histone H3 and H4
residues are enriched at SMC gene CArG boxes in cultured
SMCs, whereas stem cell lines such as cultured ECSs, P19
cells, and A404 SMC progenitors possess low levels of
acetylation at these promoters.32,33,50,51 In addition, these
genes display high levels of H3 Lys4 and Lys79 dimethylation in SMCs relative to non-SMC lineages, including stem
cells and somatic cell lineages.32 In contrast to SMCs, SMC
gene promoters are enriched for H3 Lys27 trimethylated
histones in ESCs, a well-studied modification that promotes
assembly of heterochromatin via recruitment of polycomb
group proteins. Similar to ESCs, SMC gene chromatin in
Smooth Muscle Chromatin
1433
non-SMC somatic cells (eg, endothelial cells, skeletal muscle
myotubes, leukocytes) also contains high levels of H3 Lys27
trimethylation, in addition to H4 Lys20 dimethylation (elsewhere32 and O.G.M. and G.K.O., unpublished observations,
2005). These data closely fit the micrococcal nuclease and
SMC-restricted SRF-binding observations. Collectively,
these findings suggest that SMC gene promoter DNA is
initially wrapped into chromatin harboring modified histones
that direct heterochromatin assembly in ESCs (H3K27Me3).
At some point during the developmental transition from
undifferentiated stem cells to differentiated SMCs,
H3K27Me3 is replaced with euchromatic modifications,
which are expected to encode chromatin that is permissive
to SRF binding. Other non-SMC lineages retain
H3K27Me3 (and/or gain H4K20Me2) during their developmental histories, thereby effectively maintaining SMC
gene chromatin in a heterochromatic state refractory to
SRF binding. Figure 3A and 3B and the Table summarize
this data for histones H3 and H4. Clearly, there is an
abundance of exciting work that lies ahead to further
characterize how histone modification patterns are programmed into SMC gene chromatin during development
and how these contribute to SMC fate determination.
At the very least, these distinct histone modification
patterns can be used to distinguish SMCs from non-SMCs
and vice versa. Even more fundamental, SMC-specific
patterns of histone modifications might represent an epigenetic program that defines the SMC lineage through
selective control of SRF binding to SMC gene chromatin.
All of the results described thus far reflect ChIP data around
the 5⬘-CArG box regions of these genes. Interestingly, we
mapped the distribution of these modifications across the
SMC ␣-actin locus for ⬇3 kb on either side of the CArG box
in cultured SMCs, which captures regions outside the promoter on the 5⬘ end and extends to regions inside the coding
region on the 3⬘ end (Figure 3C). We found that all of the
euchromatic modifications examined were essentially absent
from chromatin 5⬘ of the promoter, whereas the heterochromatic modifications were highly enriched. At a point approximately 1 kb 5⬘ of the CArG boxes, these distributions
abruptly reversed (elsewhere32 and O.G.M. and G.K.O.,
unpublished observations). These results suggest that these
modification patterns are specifically programmed into this
locus around the CArG box promoter region (by yet unknown
factors), presumably to serve some functional role in activating transcription. Drawing from our earlier discussions regarding histone acetylation and methylation, it can be hypothesized that some of the histone modifications present around
the CArG boxes in SMCs function to decompact the chromatin and expose CArG box DNA to SRF (eg, acetylation),
whereas others directly recruit and/or stabilize SRF binding
to the chromatin via a tethering function (eg, methylation). In
contrast, the absence of these modifications combined with
the presence of heterochromatic modifications observed in
the non-SMC lineages may direct the assembly of heterochromatin at SMC loci, thereby precluding efficient SRF binding
(Figures 2 and 3). In this way, SMC-specific gene expression
is accomplished. Much work is needed to test these possibilities and demonstrate causality, both in SMCs and in non-
1434
Circulation Research
May 25, 2007
Figure 3. A and B, Histone modification
patterns found on histones H3 and H4
SMC-specific Histone
within the promoter chromatin of SMC
H4: NH2—S G R G K G G K G L G K G G A K R H R K V--H4 Modifications
genes in non-SMCs (A) and SMCs (B).
5
8
12
16
20
The string of amino acids represent the
Me 3
Me 3
N-terminal tail domains of each histone,
and those residues known to undergo
H3: NH2--- A R T K Q T A R K S T G G K A P R K Q L A T K A A R K S---H3
modification in SMCs are numbered by
9
14
27
4
their position from the N terminus. The
Me
Me
Me
Me
methylation state (di- or tri-) for each
Heterochromatic Histone Modifications
methylated amino acid is depicted next
SRF SRF
Low SRF Binding to CArG Chromatin
to the modification as a 2 or 3. Levels of
SRF binding to the SMC gene chromatin
in these different cell types are illusB S MCs
Ac
Ac
Ac
Ac
trated underneath the histone modification profiles. It should be noted that we
have thus far observed only H3K9Me3
H4: NH2—S G R G K G G K G L G K G G A K R H R K V--H4
(K79)
5
16
20
8
12
at SM ␣-actin in ESCs, whereas we
Ac
Me 2
Ac
Me 2
have generally observed the others
within CArG box chromatin of multiple
SMC-specific contractile genes in sevH3: NH2--- A R T K Q T A R K S T G G K A P R K Q L A T K A A R K S---H3
eral SMC and non-SMC lines (and from
4
9
14
27
rat tissues in vivo). One exception is that
Me
c
A
H4K20Me2 is not found at SMC genes
SRF
E
u
c
h
r
o
m
a
t
i
c
H
i
s
t
o
n
e
M
o
d
i
f
i
c
a
t
i
o
n
s
SRF
in ESCs, although it is found in nonHigh SRF Binding to CArG Chromatin
SMC somatic cells. C, Distribution of
Me
heterochromatic (red) and euchromatic
(green) histone modification enrichments
across the SM ␣-actin locus. SRF,
C SM a-actin locus (locus-specificity in SMCs)
TATA-binding protein/TFIID, and RNA
polymerase II distributions are also
included, as are sites with the highest
Pol
II
Pol
II
Pol
II
Pol
II
TBP
SRF
probability of nucleosome occupancy
(gray cylinders) as predicted by micrococcal nuclease digestion experiments.
+2400
+400
+800
+1200
+1600
+2000
+1
-2400
-2000
-1600
-1200
-800
-400
-2800
Data in this figure represent previous
ChIP-mapping experiments32 and
unpublished observations by O.G.M.
and G.K.O., 2005. Note: modifications to different histone residues are color coded in all figures according to Figure 3. Ac indicates
acetylation; Me, methylation.
Non-SMCs
Me
2
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Ac
A
SMCs. In this regard, experiments in undifferentiated myoblasts demonstrated that the absence of acetylated histones
and the presence of H3K27Me3 within CArG box chromatin
of skeletal muscle genes effectively precluded SRF binding to
these promoters, thereby blocking myocyte differentiation.26
Similar functional studies addressing silencing of SMC genes
non-SMCs are needed.
SMC Histone Modifications Regulate the
Chromatin-Binding Properties of SRF
Recent evidence has demonstrated a role for histone modifications in control of chromatin-binding activity of SRF, as
would be expected if these modifications played a role in
SMC specification. From these studies a sequence of stepwise molecular events within chromatin that direct activation
of SMC transcription can be derived as depicted in Figure 4
and the text below. In ESCs, SMC gene chromatin contains
H3 histones methylated at Lys27, which maintains these
genes in a heterochromatin conformation that is refractory to
SRF binding (Figure 4A).32 ESCs that are destined for
non-SMC lineages retain H3K27Me3 (and/or acquire
H4K20Me2). However, in ESC subsets that commit to the
SMC lineage, H3K27Me3 is replaced with the permissive
modifications characteristic of differentiated SMCs. A key
challenge to understanding this process is deciphering the
temporal sequence in which these different modifications are
programmed into the SMC chromatin during development, as
well as what (if any) roles they play in promoting SRF
binding and/or other events during transcriptional activation.
Assuming that activation of transcription is a linear step-wise
event, 52 any histone modifications that regulate the
chromatin-binding properties of SRF must be added to the
chromatin before SRF can bind. That is, these modifications
are predicted to lie upstream of SRF binding during transcriptional activation within chromatin. Under this premise, we
identified H3 Lys4 methylation and H4 acetylation 2 two
modifications that are programmed into SMC gene chromatin
upstream of SRF binding (Figure 4B and 4C).32 Indeed, much
of the data that have uncovered a role for histone modifications in upstream control of SRF binding are based on
experiments aimed at dissecting the roles of H3 Lys4 dimethylation (H3K4Me2) and H4 acetylation (H4Ac) within
SMC gene promoter chromatin. These studies have suggested
a model whereby H4Ac “relaxes” or “opens” the chromatin to
provide CArG box DNA accessible to SRF binding (Figure
4C through 4E), whereas H3K4Me2 provides a docking site
to tether and/or stabilize myocardin/SRF complexes to the
chromatin (Figure 4D and 4E). Myocardin/SRF complexes
subsequently recruit other cofactors such as CBP/p300 that
further modify other histone residues (eg, H3 Lys9/14 acetylation) and recruit members of the basal transcription machinery (Figure 4F),33,51 which fully activates the SMC gene
McDonald and Owens
Smooth Muscle Chromatin
1435
Figure 4. Step-wise assembly of SMCspecific chromatin at SMC gene promotStep-wise Assembly
ers. In this model, SMC-specific
ESCs
Of SMC-specific
?HMT
H3K4Me2 and H4 acetylation (Ac) comr
n
C
h
o
m
a
t
i
n
D
u
r
i
g
Me
Me
Me
Me
bine to encode a chromatin structure perMe
SMC Development
?HMT
missive for binding of myocardin/SRF.
Myocardin/SRF then recruit other tranH3 K 4 M e2
scription factors to activate transcription.
Ac
ESCs and other non-SMC lineages contain heterochromatic histone modifica?MBP
T
n
A
o
tions and compacted chromatin, thereby
ati
?H
?HAT
Myo
tyl
e
inhibiting SRF binding (A). Methylation
D
SRF
Ac
C
S
R
4
SRF
F SR
H
(Me) of H3 Lys4 by unknown histone
F
P
Me
Me
?MB
methyltransferases is programmed into
c
c
Myo
A
A
SMCs during development of SMCs from
ESCs (B) to provide a docking site for
Myocardin
Deg
D
e
g
either myocardin or some myocardinCAr enerate
CAr enerate
GB
GB
oxe
g
oxe
associated factor that binds methylated
n
s
i
s
ind
histones (MBP). H4 acetylation by as of
dB
e
yet unknown HAT(s) is also required to
iliz
b
F
a
D i f fe r e n t i a t e d S M C s
E
“open” up the chromatin and expose
St
?MBP
?MBP
degenerate CArG box DNA accessible to
Myo
Myo
Me
Me
SRF (C). However, SRF binding to these
c
R ec r u i t m e n t o f
A
Ac
SRF
SRF
degenerate CArG boxes is weak and
Basal TFs
Pol II
SRF
SRF p300
transient in the absence of myocardin.
Deg
P
B
T
e
CAr nerate
The appearance of myocardin and/or
GB
Me
oxe
putative MBPs that interact with myocars
din therefore stabilizes binding of SRF to
the degenerate CArG boxes via their
interaction with nearby histones harboring H3K4Me2 (D and E). In this way, H4 acetylation and H3K4Me2 synergize to assemble chromatin that is permissive for binding myocardin/SRF complexes. Myocardin/SRF then recruits other transcription factors (eg, CBP/p300)
via the TAD of myocardin that subsequently add other histone modifications (eg, H3 acetylation, H3K79Me2) and recruit the basal transcription machinery (TATA-binding protein/TFIID [TBP], RNA polymerase II [Pol II], etc) downstream of SRF to robustly activate transcription (F).
SRF SRF
S MC L in e a g e
Com mitment
B
SRF S RF
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Ac
Me
A
expression program.51 The upstream contributions of
H3K4Me2 and H4 acetylation in relation to SRF binding are
highlighted below.
Remarkably, histone H3 molecules dimethylated at Lys4
appear to serve as docking sites for myocardin/SRF ternary
complexes, as these complexes specifically associate with
histones and SMC gene CArG box chromatin in a myocardinand H3K4Me2-dependent fashion.32 This suggests that addition of H3K4Me2 to SMC gene CArG box chromatin during
development might provide an upstream docking site that is
used by either myocardin itself or some myocardin-associated
protein that in turn assists myocardin/SRF complexes in
locating and physically associating with SMC gene chromatin
(Figure 4D and 4E). These data are consistent with current
models that describe histone methylation as information
storage units for cell identity, via tethering specialized protein
complexes to cell-specific genomic loci (eg, SMC contractile
genes). Indeed, data from classic cell systems strongly suggest that modules in various protein complexes physically
associate with H3K4Me2 to gain extra thermodynamic stability beyond that provided through binding DNA in the
chromatin. It is thought that this extra stability assists to
anchor these protein complexes within the chromatin template.53,54 In this regard, multiple SMC contractile genes
harbor substitutions in their CArG box sequences that display
weakened binding affinity for SRF (so-called “degenerate”
CArG boxes1,6), such that other thermodynamic contacts are
probably needed to stably hold SRF within chromatin that
carries these sequences. Remarkably, SMC genes remain
transcriptionally active even in the absence of myocardin (eg,
after vascular injury) if these degenerate CArG boxes are
replaced with high-affinity consensus CArG boxes.55 Does
this occur because the increased SRF affinity conferred by
consensus CArG boxes compensates for loss of stability
provided by interaction of myocardin with H3 Lys4 methylated histones? Degenerate CArG boxes at SMC gene promoters have been conserved over hundreds of millions of years of
evolution. Could it be that these sequences were selected to
provide a mechanism whereby SRF binding could be disrupted by signaling pathways that target myocardin or
H3K4Me2, via disabling the extra stability conferred by
myocardin–H3K4Me2 needed to hold SRF within chromatin
containing these low-affinity degenerate CArG boxes? This
hypothesis predicts that the weak multivalent interactions of
myocardin with H3K4Me2 and SRF with degenerate CArG
boxes synergize to stabilize myocardin/SRF complexes
within SMC gene chromatin and that these complexes subsequently act as a scaffold to recruit other transcriptional
coactivators (eg, p300) via the strong TAD of myocardin
(Figure 4D through 4F). Although much work is needed to
define whether this concept is an accurate explanation of the
limited available data and numerous other interpretations
exist, a testable prediction is that any process that results in
loss of myocardin from myocardin/SRF complexes, myocardin mutations that disrupt its association with methylated
histones or methyl-binding proteins (MBPs), loss of
H3K4Me2 from chromatin containing degenerate CArG box
DNA, or replacement of degenerate with consensus CArG
sequences would all influence the ability of SRF to bind
chromatin. Investigation into these questions will likely
provide rewarding insights into the nature of both SMC
physiology and chromatin biology.
1436
Circulation Research
May 25, 2007
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Numerous investigations into the functions of H4 acetylation in traditional cell systems have demonstrated a role for
this modification in direct regulation of chromatin compaction. Indeed, one of the first studies to demonstrate H4
acetylation is a prerequisite for activation of transcription in
response to physiological stimuli did so in the context of
SRF-dependent activation of c-fos.56 This concept appears
applicable to SMCs as well. Two proteins, homeodomainonly protein 1 (HOP) and Kruppel-like factor 4 (KLF4) have
been found to physically associate with HDAC2 and recruit
this protein to SMC gene promoters.32,57 HDAC2 possesses
deacetylase activity specific to histone H4 residues,58 and
expression of both HOP and KLF4 in cultured cells results in
loss of H4 acetylation from SMC gene chromatin. This in turn
is accompanied by chromatin compaction, loss of SRF
binding, and transcriptional repression consistent with an
upstream role for H4 acetylation in control of SRF binding to
SMC gene promoter chromatin. Indeed, these effects are
blocked by the HDAC inhibitor trichostatin A. Remarkably,
both KLF4 and HOP potently antagonize the transcriptional
activity that myocardin conveys to SRF, and this action is
also dependent on the ability of these proteins to recruit
HDAC activity to SMC genes and block SRF association
with CArG box chromatin. Furthermore, treatment of SMCs
with trichostatin A under native conditions results in H4
hyperacetylation and increased binding of SRF within chromatin of SMC genes.33 These studies reveal that the SMCspecific presence of H4 acetylation likely functions to maintain SMC gene chromatin in a conformation in which CArG
box DNA sequences are accessible to SRF (Figure 4C and
4E). The absence of this modification, as in non-SMCs or in
SMCs expressing proteins such as HOP or KLF4, results in
compaction of this chromatin into a conformation that blocks
binding of SRF, most likely by tight packaging of CArG box
DNA with histone octamers, which is predicted to strongly
inhibit SRF binding based on crystal structure data (Figure
4A and 4B). It is also likely that the quantitative levels H4
acetylation within SMC gene euchromatin are adjusted in
response to various environmental cues for reversible, graded
regulation of SRF binding to SMC genes. This would allow
fine tuning of SMC gene expression levels in adult SMCs for
adaptation to fluctuating extracellular signals.
Collectively, available evidence implies that H3K4Me2
serves as a docking site for myocardin/SRF complexes within
chromatin, whereas H4 acetylation functions to “open up” the
chromatin and provide CArG box DNA suitable for binding
SRF. These 2 modifications operate in combination to establish a chromatin structure that is permissive to SRF binding
(Figure 4), consistent with the histone code hypothesis.20,59
Several pieces of data support this concept, although a
definitive demonstration of causality has yet to be realized.
This is attributable to a number of confounding experimental
obstacles that are commonly encountered in chromatin biology, most notably the inherent difficulties in minimizing
indirect effects caused by loss-of-function strategies targeted
toward chromatin-based processes, as these events play
universal roles in a variety of cell behaviors besides transcription of gene subsets. Nevertheless, progress on these fronts
will require development of effective strategies to test cau-
sality, such as in vitro reconstitution of SMC gene promoters
wrapped into nucleosomes harboring various histone modifications or identification of SMC-restricted proteins that
recruit HATs and histone methyltransferases to SMC gene
promoters.
Other Factors Regulating SRF Activity
Within SMC Chromatin
There are unequivocal data that the SMC-specific histone
modification program is just 1 component of a fascinating
multilayered transcriptional regulatory process that has
evolved around SRF within SMC chromatin. Indeed, a truly
integrated model is emerging in the literature that suggests
control of SMC differentiation involves complex interplay
among (1) SRF cofactors, (2) the SRF–CArG interaction
itself, and (3) histone modifications within CArG box chromatin. Although much work regarding interconnections between these 3 processes is needed, we favor the view that all
3 of these regulatory levels (and others yet discovered)
operate simultaneously in synergy and/or provide redundancy
with one another to provide tight regulatory control over the
SMC phenotype. As such, below, we very briefly review
nonhistone based regulatory control of SRF activity within
chromatin to provide a comprehensive glimpse into what is
known about the chromatin dynamics underlying SMC
transcription.
First, numerous SRF accessory proteins have been discovered that either displace promyogenic cofactors off of SRF to
repress transcription or facilitate interactions SRF with promyogenic factors to activate transcription. These events occur
while SRF is bound to CArG box DNA, consistent with the
model that SRF is a platform for recruitment of other
accessory modules. For example, Olson and colleagues60
completed very elegant studies showing that activation of the
extracellular signal-regulated kinase (ERK) signaling pathway in SMCs results in phosphorylation of the TCF Elk-1, a
protein that has the ability to form a ternary complex with
SRF on CArG box DNA within promoters of both SMCspecific genes and growth-responsive immediate early genes.
Phosphorylation of Elk-1 directs relocalization of this protein
from cytoplasm to the nucleus, where Elk-1 proceeds to
compete with myocardin for a common docking surface on
SRF. Displacement of myocardin off of SRF by Elk-1 has the
resultant effect of blocking the strong transcriptional activation conveyed to SRF by myocardin.60 Whereas SRFdependent growth-responsive genes (which do not use myocardin) are transcriptionally stimulated by the presence Elk/
SRF ternary complexes, SMC-specific contractile genes are
repressed after replacement of myocardin/SRF with Elk/
SRF,60 albeit to different degrees.61 Other SRF-binding partners have been found to operate similar to Elk-1 in response
to other signaling pathways.62,63 In contrast, other SRF
accessory factors such as cysteine-rich proteins (CRP1/2),64
myocardin-related transcription factors,34 Prx1,19 and GATA
factors64 have been found to stimulate the transcriptional activity
of myocardin/SRF ternary complexes via their association with
SRF on SMC gene promoters. This leads to increased levels of
SMC gene transcription. Putting this together, tipping the balance between the activities of positive and negative SRF
McDonald and Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
cofactors by different signaling pathways offers a dynamic
means by which SMC gene expression might be controlled.
Second, the ability of SRF proteins to directly bind CArG
box DNA molecules is regulated through multiple mechanisms. Yin-yang 1 (YY1) is a transcriptional repressor that
possesses CArG box DNA-binding activity. YY1 in turn has
the ability to compete with SRF for binding to CArG box
DNA by EMSAs. Exchange of YY1 for SRF on SMC gene
promoters thus has the effect of repressing SMC gene
transcription.65 HERP1 also has the ability to interfere with
SRF binding to DNA, but via physical association with the
SRF MADS box rather than with CArG DNA.66 The MADSbox of SRF can also be phosphorylated at sites that inhibit
SRF binding to DNA specifically at SMC gene promoters but
not at promoters of growth-responsive genes.67 Finally, SRF
can translocate from nucleus to cytoplasm in response to
various signals, which may also reduce levels of SRF binding
to CArG box DNA over time via reductions in the numbers of
SRF molecules available to bind SMC gene promoters.68,69 In
contrast, other proteins, such as CRP2,64 PIAS1,70 and various homeodomain proteins (eg, Phox1)19,71 enhance binding
of SRF to CArG box DNA to stimulate transcription. Importantly, the mechanisms described in this passage were demonstrated to be operational not only via EMSAs (ie, raw SRF
DNA-binding activity) but also within the context of CArG
box DNA wrapped into euchromatin by ChIP assays.
The Chromatin Dynamics of SMC Plasticity
To truly integrate chromatin into the current model describing
the myogenic activity of SRF, investigations into the interplay between SRF cofactors, the SRF–CArG interaction,
histone modifications, and environmental cues are required.
Currently, there are only a handful studies that have attempted to make connections between these regulatory layers.
As such, future work in the field should attempt to address
modular, synergistic, and/or redundant functions for these
different processes relative to each other. Treatment of
cultured SMCs with platelet-derived growth factor BB
(PDGF-BB) has shed some light on what might be revealed
by studies such as these. PDGF-BB is a molecule released at
sites of vascular injury and atherosclerosis that potently
induces phenotypic switching of cultured SMCs, including
transcriptional repression of SMC-specific contractile genes,
proliferation, and migration.1 These effects are completely
reversible after removal of PDGF-BB from the media. Treatment of cultured SMCs with PDGF-BB results in rapid (⬇30
minutes) activation of the ERK pathway and phosphorylation/nuclear translocation of Elk-1, thereby putting competition between Elk and myocardin for SRF into motion.60
Expression of KLF4 is also upregulated in response to
PDGF-BB,32,72 and 24 hours after treatment, there are reversible reductions in H4 acetylation and SRF binding from the
chromatin of SMC genes.32 In contrast, growth-responsive
genes such as c-fos retain high levels of acetylation,32 and
Elk-1/SRF complexes stimulate transcription within the chromatin of these genes.60 Finally, there is evidence that SRF
partially translocates out of the nucleus into the cytoplasm
several hours after treatment of SMCs with PDGF-BB.68
These processes may operate to infuse redundancy into the
Smooth Muscle Chromatin
1437
actions of PDGF-BB, such that if 1 pathway is inhibited,
genes remain partially or fully repressed because of compensation by the other pathways.
Alternatively, the diverse events that are initiated in response to PDGF-BB signaling might operate synergistically
within a linked pathway. This is expected given that several
studies indicate that inhibition of multiple different PDGFBB–triggered events can antagonize PDGF-BB–mediated
repression of SMC gene expression, although it is difficult to
know how efficiently these diverse processes repress SMC
transcription relative to each other without comparative
studies. This concept is depicted in Figure 5, where physiological SMC chromatin (assembled during development as in
Figure 4) is reorganized in response to PDGF-BB into a
transcriptionally repressive configuration. This action is accomplished synergistically by the numerous events that are
triggered by PDGF-BB signaling (eg, KLF4, Elk-1, Herp,
phosphorylation, etc), thereby creating a pathological SMC
chromatin environment that inhibits binding of myocardin/
SRF complexes. These events are transient and fully reversible after removal of PDGF-BB from the culture medium,
which allows reconstruction of physiological SMC chromatin
that is permissive to myocardin/SRF. Reconstruction would
proceed similar to Figure 4 via reacetylation of histone H4
coupled with reintroduction of myocardin to SRF complexes.
This is predicted to make SMC chromatin transcriptionally
permissive by both “reopening” chromatin (via acetylation)
to SRF coupled with stabilization of SRF binding to CarG
box DNA (via interaction of myocardin or a myocardinassociated protein with H3K4Me2). To illustrate, data from
our laboratory indicate that the ability of KLF4 to repress
SMC gene expression is dependent on activation of the ERK
pathway, and we have found a strong physical interaction
between Elk-1 and KLF4 in extracts isolated from SMCs
where the ERK pathway has been activated (R. Deaton,
O.G.M., and G.K.O., unpublished observations, 2006). Although speculative, it is possible that displacement of myocardin off SRF by Elk-1 might be the sole mediator of the
PDGF-BB response early (eg, 30 minutes), when myocardin
levels are high and KLF4 levels are low. Once KLF4 is
upregulated, occupancy of Elk-1 within SMC gene chromatin
might also attract KLF4/HDAC2 complexes to SMC promoters leading to deacetylation of histones, chromatin compaction, and ejection of SRF from the chromatin template,
further enhancing the repressive effects of PDGF-BB. Any
other potential transcriptional activation by residual myocardin/SRF binding attributable to stochastic/transient opening
of compacted chromatin would continue to be antagonized by
the competitive function of Elk-1. Translocation of SRF out
of the nucleus might further reduce any residual SRF binding.
Theoretically, the synergistic actions of these processes
would generate tight repression of SMC gene expression.
Again, removal of PDGF-BB from the culture media terminates these repressive molecular forces and allows reconstruction of physiological SMC chromatin, thereby restoring
SMC differentiation. Many of these cell culture observations
have been confirmed by experimental observations in rats and
mice, including demonstration of reversible loss of SRF binding
1438
Circulation Research
Synergistic
A ct i va t i o n
?MBP
Me
Myo
SRF
SRF
p300
Ac
Ac
May 25, 2007
Pol II
TBP
Me
Physiological SMC Chromatin
The Chromatin
Dynamics Underlying
SMC Plasticity
N eo
in
Sig timal
n a ls
Add
iti
PDG on of
F-BB
SRF
P
Cytosol
Elk-1
Myo
Re-differentiation
B
?M
P
KL
F4
SRF SRF
HD
AC HOP
HE R P
2
P
?P M e
Ac
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
B
?M
?HAT
Re-localization
Via H3K4Me2
P
Myo
SRF SRF
2
1
Ac
Ac
Synergistic
R epres sion
Me
on
luti
o
s
Re
l of
ova B
m
e
R
F -B
PDG
Pathological SMC Chromatin
Reconstruction
Figure 5. Histone methylation (Me) provides epigenetic memory of SMC identity during synergistic repression of SMC gene transcription. In this model, various pathways that have been shown to repress SMC gene expression are linked together in synergy. These
events are dependent on each other to achieve the full function of inhibiting SRF-dependent transcription via positive feedback. In
addition, some or all of these processes may be physically linked in a step-wise manner over time (eg, Elk-1 binds SRF, then recruits
KLF4/HDAC2, which subsequently deacetylates H4). During these events, histone H3K4Me2 is preserved within pathological chromatin,
thereby providing a means by which the original physiological chromatin configuration can be restored by directing relocalization of
myocardin/SRF and/or H4 HATs to SMC gene promoters after termination of repressive environmental cues (ie, the reconstruction
phase). The black arrow adjacent to myocardin represents decreased myocardin expression, as the event has been documented in
some SMC lines in response to PDGF-BB and during vascular injury/atherosclerotic plaque formation in vivo. This event would be predicted to decrease SRF binding to SMC genes via disruption of the myocardin–H3K4Me2 interaction. Finally, the hypothetical situation
of a “methyl/phos” switch to temporarily displace myocardin/SRF from SMC gene promoters during transcriptional repression is also
included, which might also antagonize myocardin–H3K4Me2. Ac indicates acetylation; P, phosphorylation.
and H4 acetylation from the CArG box chromatin of SMCspecific contractile genes in response to vascular injury in vivo.32
SMC Plasticity: A Model System to Study
Chromatin Dynamics
The profound phenotypic plasticity of the SMC lineage has
historically represented a confounding obstacle for study of
SMC behavior. However, we feel that this property represents
a unique advantage for the study of how cell-specific gene
expression/transcription is dynamically controlled at the level
of chromatin. Differentiated SMCs can be induced to reversibly switch their phenotype virtually on command, representing a potential model system whereby chromatin is reversibly
remodeled into pathophysiological configurations and yet
still retains information critical for restoration back to its
original physiological form. Given the gross rearrangements
in chromatin structure and replacement of myocardin/SRF
with other repressive factors at SMC genes in response to
PDGF-BB (Figure 5), how is the original chromatin/transcription factor configuration restored within the chromatin
of these genes when PDGF-BB is removed from the media
and SMCs are induced to redifferentiate? Consistent with the
observations regarding the behavior of histone methylation
documented earlier, H3K4Me2 is (thus far) the only aspect of
the original transcriptionally active SMC gene chromatin that
is not erased or reconfigured during PDGF-BB treatment.
Does the persistence of this modification within transcriptionally repressed SMC gene chromatin offer a means by which
these genes can be reactivated on termination of PDGF-BB
signaling and initiation of SMC redifferentiation? Given that
myocardin or some myocardin-associated factor can specifically associate with this modification and control binding of
myocardin/SRF complexes to chromatin, H3K4Me2 may
function as a permanent signal within SMC gene chromatin
(at least during the PDGF-BB response) that has the ability to
direct relocalization of myocardin/SRF complexes back to
SMC genes once Elk-1/KLF4 activity is extinguished. It is
also possible that H3K4Me2 is involved in attracting HATs
back to SMC genes for restoration of H4 acetylation and
re-opening of the CArG box chromatin to SRF (Figure 5). In
McDonald and Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
our opinion, investigation into this phenomenon is very
important, as H3K4Me2 (or some other modification not yet
tested) patterns across SMC genomes could represent a
molecular memory of which genes must retain the ability to
become reactivated on resolution of conditions (eg, vascular
injury) in which SMCs undergo reversible phenotypic switching (ie, cell lineage memory). This could also provide a
practical means by which to follow neointimal SMCs in
vessels exposed to various stimuli such as atherosclerotic
plaque formation, an exercise that has historically been
confounded because of loss of markers such as SMC-specific
contractile proteins that are typically used to identify the
SMC lineage in vivo.
A simple first step to explore this hypothesis would be to
transfect SMCs with histone peptides methylated at H3 Lys4
and determine whether these “squelch” myocardin and/or
SRF from SMC promoters. Immunohistochemistry of sections from arteries exposed to vascular injury or atherosclerotic plaque with H3K4Me2 antisera conjugated to fluorescent DNA probes homologous to SMC gene sequences could
confirm whether these genes remain methylated under these
conditions. Although certainly technically challenging, these
experiments are envisioned to proceed in a fashion similar to
standard in situ cytogenetic protocols, except that conditions
would be such that hybridization would depend on binding of
the antisera in close proximity to DNA probe targets, in
addition to binding of the DNA probe to the target itself (in
this case, the targets/probes are SMC gene promoters).
Subsets of H3 Lys4 methylated histones have been found to
be phosphorylated at Thr3 in vivo, and the presence of this
modification inhibits binding of proteins that dock to H3
Lys4 methylated histones via the “methyl/phos” binaryswitch process described earlier (Figure 5).41 It is also
possible that myocardin or a myocardin-associated protein is
modified so as to inhibit interaction with H3K4Me2. Antibodies that recognize histones with dual methyl and phosphoryl groups are becoming commercially available (Upstate,
Abcam), providing a starting point for investigation into
whether a methyl/phos switch might be operational at SMC
gene promoters. Eventually, purification of all proteins present in myocardin/SRF complexes, along with identification
and functional characterization of any modules that bind
methylated histones, will provide the most definitive experiments addressing this issue. Clearly, opportunities are plentiful for investigation into chromatin dynamics underlying
SMC phenotypic plasticity.
Closing Remarks
The inherent phenotypic plasticity of the SMC lineage has
presented remarkable insights and unique challenges regarding our understanding of transcriptional control of cellular
differentiation. Indeed, the unfolding story documenting the
molecular behavior of SMCs under physiological conditions
versus times of pathophysiological stress is emerging as a
general paradigm for how fluctuating environmental cues
dynamically regulate differentiation and plasticity of most if
not all cell types in higher organisms. The model that is
emerging from studies of SMCs (and other cell lineages)
suggests that cellular differentiation is tightly regulated at the
Smooth Muscle Chromatin
1439
level of chromatin through a complex, synergistic combination of DNA-binding transcription factors (eg, SRF), accessory cofactors for the DNA-binding proteins (eg, myocardin/
Elk-1), the direct interaction of DNA and transcription factor
complexes (eg, the SRF–CArG interaction), and histone
modifications present within promoter chromatin (eg, SMCspecific H3K4Me2 and H4 acetylation at CArG boxes). In
SMCs, this multilayered orchestra evolved around SRF to
provide multiple avenues by which environmental cues and
signaling pathways may signal to chromatin to dynamically
control gene expression.
Further investigation into the issues addressed in this
review will undoubtedly yield important insights into the
SMC component of vascular development and the multitude
of disease processes in which SMC pathology is a prominent
component (eg, atherosclerosis). In addition to the multiple
experimental directions proposed earlier, several clever cell
systems that capture the transit from undifferentiated ESCs to
differentiated SMCs have recently been developed.50,73,74
These systems still largely await investigation into how
histone modifications are programmed into SMC gene chromatin during normal SMC development. These latter experiments are critical as they will provide a wealth of information necessary to engineer healthy SMCs in vitro that may be
used for therapeutic purposes in vivo. Future studies regarding this issue should focus on identification of trans factors
and cis elements important for programming histone modifications into SMC gene promoters during development. Continued efforts in combining mutation of promoter cis elements with ESC knockout of trans factors identified by yeast
2-hybrid assays or bioinformatics methodology are the most
powerful initial approaches to address these issues. In addition, expansion of our knowledge regarding epigenetic regulation of SMC chromatin is a must. Investigations must
include the following: how other histone modifications regulate SMC differentiation, identification of new mechanisms
whereby histone modifications operate in SMCs, examination
of other epigenetic processes such as ATP-dependent chromatin remodeling, micro-RNAs, and DNA methylation in
SMCs, nuclear packaging/localization of SMC genes, and
assembly of heterochromatin at SMC gene loci in non-SMCs.
Over the past decade, efforts aimed at elucidation of the
molecular processes that underlie SMC differentiation have
intensified at an exponential rate. These studies have begun
painting what promises to be an elegant picture depicting an
incredibly complex and finely tuned molecular process that
bestows SMCs with their remarkable cellular properties. We
anticipate that future studies have the potential to offer
particularly instructive insights into how cell differentiation is
controlled at the level of chromatin, both within the context of
normal SMC development and SMC phenotypic plasticity
within dynamic tissue microenvironments.
Sources of Funding
This work was supported by NIH grants P01 HL19242, R37
HL57353, and R01 HL38854 (to G.K.O., principle investigator).
O.G.M. was partially supported by NIH Medical Scientist Training
Program grant T32 GM07267-27 (to G.K.O., principle investigator)
during the completion of a PhD.
1440
Circulation Research
May 25, 2007
Disclosures
None.
References
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
1. Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular
smooth muscle cell differentiation in development and disease. Physiol
Rev. 2004;84:767– 801.
2. Owens GK. Regulation of differentiation of vascular smooth muscle cells.
Physiol Rev. 1995;75:487–517.
3. Regan CP, Adam PJ, Madsen CS, Owens GK. Molecular mechanisms of
decreased smooth muscle differentiation marker expression after vascular
injury. J Clin Invest. 2000;106:1139 –1147.
4. Wamhoff BR, Hoofnagle MH, Burns A, Sinha S, McDonald OG, Owens
GK. A G/C element mediates repression of the SM22a promoter within
phenotypically modulated smooth muscle cells in experimental atherosclerosis. Circ Res. 2004;95:981–988.
5. Aikawa M, Sakomura Y, Ueda M, Kimura K, Manabe I, Ishiwata S,
Komiyama N, Yamaguchi H, Yazaki Y, Nagai R. Redifferentiation of
smooth muscle cells after coronary angioplasty determined via myosin
heavy chain expression. Circulation. 1997;96:82–90.
6. Miano JM. Serum response factor: toggling between disparate programs
of gene expression. Mol Cell Cardiology. 2003;35:577–593.
7. Sun Q, Chen G, Streb JW, Long X, Yang Y, Stoeckert CJ Jr, Miano JM.
Defining the mammalian CArGome. Genome Res. 2006;16:197–207.
8. Chang VK, Donato JJ, Chan CS, Tye BK. Mcm1 promotes replication
initiation by binding specific elements at replication origins. Mol Cell
Biol. 2004;24:6514 – 6524.
9. Treisman R. The serum response element. Trends Biochem Sci. 1992;17:
423– 426.
10. Hill CS, Treisman R. Transcriptional regulation by extracellular signals:
mechanisms and specificity. Cell. 1995;199 –211.
11. Somogyi K, Rorth P. Evidence for tension-based regulation of Drosophila
MAL and SRF during invasive cell migration. Dev Cell. 2004;7:85–93.
12. Schratt G, Philippar U, Berger J, Schwarz H, Heidenreich O, Nordheim A.
Serum response factor is crucial for actin cytoskeletal organization and focal
adhesion assembly in embryonic stem cells. J Cell Biol. 2002;156:737–750.
13. Tarnawski AS. Cellular and molecular mechanisms of gastrointestinal
ulcer healing. Dig Dis Sci. 2005;50(suppl 1):S24 –S33.
14. Miralles F, Posern G, Zaromytidou AI, Treisman R. Actin dynamics
control SRF activity by regulation of its coactivator MAL. Cell. 2003;
113:329 –342.
15. Zhe X, Yang Y, Schuger L. Imbalanced plasminogen system in lymphangioleiomyomatosis: potential role of serum response factor. Am J
Respir Cell Mol Biol. 2005;32:28 –34.
16. Wang D, Chang PS, Wang Z, Sutherland L, Richardson JA, Small E, Krieg
PA, Olson EN. Activation of cardiac gene expression by myocardin, a
transcriptional cofactor for serum response factor. Cell. 2001;105:851–862.
17. Spiegelman BM, Heinrich R. Biological control through regulated transcriptional coactivators. Cell. 2004;119:157–167.
18. Wang DZ, Olson EN. Control of smooth muscle development by the
myocardin family of transcriptional coactivators. Curr Opin Genet Dev.
2004;14:558 –566.
19. Yoshida T, Hoofnagle MH, Owens GK. Myocardin and Prx1 contribute
to angiotensin II-induced expression of smooth muscle alpha-actin. Circ
Res. 2004;94:1075–1082.
20. Jenuwein T, Allis CD. Translating the histone code. Science. 2001;293:
1074 –1080.
21. Chen TA, Allfrey VG. Rapid and reversible changes in nucleosome
structure accompany the activation, repression, and superinduction of
murine fibroblast protooncogenes c-fos and c-myc. Proc Natl Acad Sci
U S A. 1987;84:5252–5256.
22. Kornberg RD, Lorch Y. Twenty-five years of the nucleosome, fundamental particle of the eukaryote chromosome. Cell. 1999;98:285–294.
23. Khorasanizadeh S. The nucleosome: from genomic organization to
genomic regulation. Cell. 2004;116:259 –272.
24. Hall IM, Shankaranarayana GD, Noma K, Ayoub N, Cohen A, Grewal SI.
Establishment and maintenance of a heterochromatin domain. Science.
2002;297:2232–2237.
25. Lunyak VV, Burgess R, Prefontaine GG, Nelson C, Sze SH, Chenoweth
J, Schwartz P, Pevzner PA, Glass C, Mandel G, Rosenfeld MG.
Corepressor-dependent silencing of chromosomal regions encoding
neuronal genes. Science. 2002;298:1747–1752.
26. Caretti G, Di Padova M, Micales B, Lyons GE, Sartorelli V. The
Polycomb Ezh2 methyltransferase regulates muscle gene expression and
skeletal muscle differentiation. Genes Dev. 2004;18:2627–2638.
27. Pellegrini L, Tan S, Richmond TJ. Structure of serum response factor core
bound to DNA. Nature. 1995;376:490 – 498.
28. Luger K, Mader AW, Richmond RK, Sargent DF, Richmond TJ. Crystal
structure of the nucleosome core particle at 2.8 A resolution. Nature.
1997;389:251–260.
29. Mack CP, Thompson MM, Lawrenz-Smith S, Owens GK. Smooth
muscle alpha-actin CArG elements coordinate formation of a smooth
muscle cell-selective, serum response factor-containing activation
complex. Circ Res. 2000;86:221–232.
30. Manabe I, Owens GK. The smooth muscle myosin heavy chain gene
exhibits smooth muscle subtype selective modular regulation in vivo.
J Biol Chem. 2001;276:39076 –39087.
31. Manabe I, Owens GK. CArG elements control smooth muscle subtypespecific expression of smooth muscle myosin in vivo. J Clin Invest.
2001;107:823– 834.
32. McDonald OG, Wamhoff BR, Hoofnagle MH, Owens GK. Control of
SRF binding to CARG-box chromatin regulates smooth muscle gene
expression in vivo. J Clin Invest. 2006;116:36 – 48.
33. Qiu P, Li L. Histone acetylation and recruitment of serum responsive
factor and CREB-binding protein onto SM22 promoter during SM22 gene
expression. Circ Res. 2002;90:858 – 865.
34. Wang DZ, Li S, Hockemeyer D, Sutherland L, Wang Z, Schratt G,
Richardson JA, Nordheim A, Olson EN. Potentiation of serum response
factor activity by a family of myocardin-related transcription factors.
Proc Natl Acad Sci U S A. 2002;99:14860.
35. Margueron R, Trojer P, Reinberg D. The key to development: interpreting
the histone code? Curr Opin Genet Dev. 2005;15:163–176.
36. Clayton AL, Hazzalin CA, Mahadevan LC. Enhanced histone acetylation
and transcription: a dynamic perspective. Mol Cell. 2006;23:289 –296.
37. Trojer P, Reinberg D. Histone lysine demethylases and their impact on
epigenetics. Cell. 2006;125:213–217.
38. Fischle W, Wang Y, Allis CD. Binary switches and modification cassettes
in histone biology and beyond. Nature. 2003;425:475– 479.
39. Fischle W, Tseng BS, Dormann HL, Ueberheide BM, Garcia BA,
Shabanowitz J, Hunt DF, Funabiki H, Allis CD. Regulation of HP1chromatin binding by histone H3 methylation and phosphorylation.
Nature. 2005;438:1116 –1122.
40. Flanagan JF, Mi LZ, Chruszcz M, Cymborowski M, Clines KL, Kim Y,
Minor W, Rastinejad F, Khorasanizadeh S. Double chromodomains
cooperate to recognize the methylated histone H3 tail. Nature. 2005;438:
1181–1185.
41. Hirota T, Lipp JJ, Toh BH, Peters JM. Histone H3 serine 10 phosphorylation by Aurora B causes HP1 dissociation from heterochromatin.
Nature. 2005;438:1176 –1180.
42. Eissenberg JC, Elgin SC. Molecular biology: antagonizing the
neighbours. Nature. 2005;438:1090 –1091.
43. Bantignies F, Cavalli G. Cellular memory and dynamic regulation of
polycomb group proteins. Curr Opin Cell Biol. 2006;18:275–283.
44. Hake SB, Allis CD. Histone H3 variants and their potential role in
indexing mammalian genomes: the “H3 barcode hypothesis”. Proc Natl
Acad Sci U S A. 2006;103:6428 – 6435.
45. Dorigo B, Schalch T, Bystricky K, Richmond TJ. Chromatin fiber
folding: requirement for the histone H4 N-terminal tail. J Mol Biol.
2003;327:85–96.
46. Fan JY, Rangasamy D, Luger K, Tremethick DJ. H2A.Z alters the
nucleosome surface to promote HP1alpha-mediated chromatin fiber
folding. Mol Cell. 2004;16:655– 661.
47. Narlikar GJ, Fan HY, Kingston RE. Cooperation between complexes that
regulate chromatin structure and transcription. Cell. 2002;108:475– 487.
48. Syntichaki P, Topalidou I, Thireos G. The Gcn5 bromodomain
co-ordinates nucleosome remodelling. Nature. 2000;404:414 – 417.
49. Jacobson RH, Ladurner AG, King DS, Tjian R. Structure and function of
a human TAFII250 double bromodomain module. Science. 2000;288:
1422–1425.
50. Manabe I, Owens GK. Recruitment of serum response factor and hyperacetylation of histones at smooth muscle-specific regulatory regions
during differentiation of a novel P19-derived in vitro smooth muscle
differentiation system. Circ Res. 2001;88:1127–1134.
51. Cao D, Wang Z, Zhang CL, Oh J, Xing W, Li S, Richardson JA, Wang
DZ, Olson EN. Modulation of smooth muscle gene expression by association of histone acetyltransferases and deacetylases with myocardin.
Mol Cell Biol. 2005;25:364 –376.
McDonald and Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
52. Metivier R, Penot G, Hubner MR, Reid G, Brand H, Kos M, Gannon F.
Estrogen receptor-alpha directs ordered, cyclical, and combinatorial
recruitment of cofactors on a natural target promoter. Cell. 2003;115:
751–763.
53. Pray-Grant MG, Daniel JA, Schieltz D, Yates JR III, Grant PA. Chd1
chromodomain links histone H3 methylation with. Nature. 2005;433:
434 – 438.
54. Wysocka J, Swigut T, Xiao H, Milne TA, Kwon SY, Landry J, Kauer M,
Tackett AJ, Chait BT, Badenhorst P, Wu C, Allis CD. A PHD finger of
NURF couples histone H3 lysine 4 trimethylation with chromatin remodelling. Nature. 2006;442:86 –90.
55. Hendrix J, Wamhoff BR, McDonald T, Sinha S, Yoshida T, Owens GK.
5⬘ CArG degeneracy in smooth muscle alpha-actin is required for injuryinduced gene suppression in vivo. J Clin Invest. 2005;115:418 – 427.
56. Alberts AS, Geneste O, Treisman R. Activation of SRF-regulated chromosomal templates by Rho-family GTPases requires a signal that also
induces H4 hyperacetylation. Cell. 1998;92:475– 487.
57. Kook H, Lepore JJ, Gitler AD, Lu MM, Wing-Man YW, Mackay J, Zhou
R, Ferrari V, Gruber P, Epstein JA. Cardiac hypertrophy and histone
deacetylase-dependent transcriptional repression mediated by the atypical
homeodomain protein Hop. J Clin Invest. 2003;112:863– 871.
58. Yang SH, Sharrocks AD. SUMO promotes HDAC-mediated transcriptional repression. Mol Cell. 2004;13:611– 617.
59. Strahl BD, Allis CD. The language of covalent histone modifications.
Nature. 2000;403:41– 45.
60. Wang Z, Wang DZ, Hockemeyer D, McAnally J, Nordheim A, Olson EN.
Myocardin and ternary complex factors compete for SRF to control
smooth muscle gene expression. Nature. 2004;428:185–189.
61. Zhou J, Hu G, Herring BP. Smooth muscle-specific genes are differentially sensitive to inhibition by Elk-1. Mol Cell Biol. 2005;25:9874 –9885.
62. Liu ZP, Wang Z, Yanagisawa H, Olson EN. Phenotypic modulation of
smooth muscle cells through interaction of Foxo4 and myocardin. Dev
Cell. 2005;9:261–270.
63. Philippar U, Schratt G, Dieterich C, Muller JM, Galgoczy P, Engel FB,
Keating MT, Gertler F, Schule R, Vingron M, Nordheim A. The SRF
target gene Fhl2 antagonizes RhoA/MAL-dependent activation of SRF.
Mol Cell. 2004;16:867– 880.
64. Chang DF, Belaguli NS, Iyer D, Roberts WB, Wu SP, Dong XR, Marx
JG, Moore MS, Beckerle MC, Majesky MW, Schwartz RJ. Cysteine-rich
LIM-only proteins CRP1 and CRP2 are potent smooth muscle differentiation cofactors. Dev Cell. 2003;4:107–118.
Smooth Muscle Chromatin
1441
65. Itoh S, Katoh Y, Konishi H, Takaya N, Kimura T, Periasamy M, Yamaguchi
H. Nitric oxide regulates smooth-muscle-specific myosin heavy chain gene
expression at the transcriptional level—possible role of SRF and YY1
through CArG element. J Mol Cell Cardiol. 2001;33:95–107.
66. Doi H, Iso T, Yamazaki M, Akiyama H, Kanai H, Sato H, Kawai-Kowase
K, Tanaka T, Maeno T, Okamoto E, Arai M, Kedes L, Kurabayashi M.
HERP1 inhibits myocardin-induced vascular smooth muscle cell differentiation by interfering with SRF binding to CArG box. Arterioscler
Thromb Vasc Biol. 2005;25:2328 –2334.
67. Iyer D, Chang D, Marx J, Wei L, Olson EN, Parmacek MS, Balasubramanyam A, Schwartz RJ. Serum response factor MADS box serine-162
phosphorylation switches proliferation and myogenic gene programs.
Proc Natl Acad Sci U S A. 2006;103:4516 – 4521.
68. Kaplan-Albuquerque N, Garat C, Desseva C, Jones PL, Nemenoff RA.
Platelet-derived growth factor-BB-mediated activation of Akt suppresses
smooth muscle-specific gene expression through inhibition of mitogenactivated protein kinase and redistribution of serum response factor.
J Biol Chem. 2003;278:39830 –39838.
69. Camoretti-Mercado B, Liu HW, Halayko AJ, Forsythe SM, Kyle JW, Li
B, Fu Y, McConville J, Kogut P, Vieira JE, Patel NM, Hershenson MB,
Fuchs E, Sinha S, Miano JM, Parmacek MS, Burkhardt JK, Solway
J. Physiological control of smooth muscle-specific gene expression
through regulated nuclear translocation of serum response factor. J Biol
Chem. 2000;275:30387–30393.
70. Kawai-Kowase K, Kumar MS, Hoofnagle MH, Yoshida T, Owens GK.
PIAS1 activates the expression of smooth muscle cell differentiation
marker genes by interacting with serum response factor and class I basic
helix-loop-helix proteins. Mol Cell Biol. 2005;25:8009 – 8023.
71. Hautmann M, Thompson MM, Swartz EA, Olson EN, Owens GK. Angiotensin II-induced stimulation of smooth muscle alpha-actin expression
by serum response factor and the homeodomain transcription factor
MHox. Circ Res. 1997;81:600 – 610.
72. Liu Y, Sinha S, McDonald OG, Shang Y, Hoofnagle MH, Owens GK.
Kruppel-like factor 4 abrogates myocardin-induced activation of smooth
muscle gene expression. J Biol Chem. 2005;280:9719 –9727.
73. Sinha S, Wamhoff BR, Hoofnagle MH, Thomas J, Neppl RL, Deering T,
Helmke BP, Bowles DK, Somlyo AV, Owens GK. Assessment of contractility of purified smooth muscle cells derived from embryonic stem
cells. Stem Cells. 2006;24:1678 –1688.
74. Rodriguez LV, Alfonso Z, Zhang R, Leung J, Wu B, Ignarro LJ. Clonogenic multipotent stem cells in human adipose tissue differentiate into
functional smooth muscle cells. Proc Natl Acad Sci U S A. 2006;103:
12167–12172.
Programming Smooth Muscle Plasticity With Chromatin Dynamics
Oliver G. McDonald and Gary K. Owens
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Circ Res. 2007;100:1428-1441
doi: 10.1161/01.RES.0000266448.30370.a0
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2007 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/100/10/1428
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/