IN VIVO ANALYSIS OF THE MOLECULAR MECHANISMS OF LONG- AND SHORT-RANGE
TRANSCRIPTIONAL REPRESSION

By
Li Li

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Microbiology and Molecular Genetics

2010

ABSTRACT
IN VIVO ANALYSIS OF THE MOLECULAR MECHANISMS OF LONG- AND SHORT-RANGE
TRANSCRIPTIONAL REPRESSION
By
Li Li

Transcription repression is essential for establishing precise patterns of gene expression
during development. Studies of Drosophila transcriptional regulation reveal that transcriptional
repressors fall into two classes: those acting locally as “short-range” repressors, and those
acting dominantly as “long-range” repressors. The types of transcription factors and cofactors
involved are highly conserved in all metazoans. One area in which our knowledge has lagged is
in understanding how these different classes of repressors function at a mechanistic level.
Understanding of repression mechanisms in the developmental circuits will shed light on key
biological processes including stem cell reprogramming, cancer, and development.
In this work, I describe studies that reveal molecular mechanisms of long- and shortrange repression during early Drosophila development. My chromatin studies of the prototypic
long-range repressor Hairy and the short-range repressor Knirps suggest that these two modes
of repression induce distinct chromatin states on the repressed genes. These results indicate
that transcriptoinal corepressors can be deployed in a context-dependent manner to effect
transcription. To discern the importance of individual corepressors, my studies have also
examined the role of the conserved Groucho corepressor in Hairy-mediated long-range
repression.

ACKNOWLEDGEMENTS

I would like to acknowledge everyone who contributed to my thesis work. First of all, I would
like to express my deepest gratitude to my mentor, Dr. David N. Arnosti, for his enthusiasm,
patience, support, and excellent guidance during my graduate study. David invested an
incredible amount of time in my research projects, and helped me tremendously in becoming a
better scientist. I would also like to thank all my lab members for their insight and discussions,
with special thanks to Dr. Sandhya Payankaulam for being a supportive colleague and a great
friend. My research project was substantially enriched by the thoughtful contribution of my
committee members: Dr. Andy Amalfitano, Dr. Lee Kroos, Dr. Richard Schwartz, and Dr. Michael
Thomashow. My deepest gratitude goes to my parents in China, whose supports and
encouragements inspired me during the course of this work, and made it all worth it.

iii

TABLE OF CONTENTS
LIST OF TABLES..............................................................................................................................vi
LIST OF FIGURES...........................................................................................................................vii
KEY TO ABBREVIATION.................................................................................................................ix
CHAPTER I
Introduction...................................................................................................................................1
Section I: Transcriptional repression: conserved and evolved features........................................2
Section II: Long- and short-range transcriptional repressors.......................................................16
Section III: Long- and short-range repression: two sides of the same coin..................................25
Conclusion....................................................................................................................................30
References....................................................................................................................................31
Chapter II
Long- and short-range transcriptional repressors induce distinct chromatin states on repressed
genes
Abstract.......................................................................................................................................41
Introduction................................................................................................................................42
Results.........................................................................................................................................44
Discussion....................................................................................................................................67
Materials and Methods...............................................................................................................71
Appendices :................................................................................................................................74
References....................................................................................................................................96
Chapter III
Analysis of Hairy-mediated long-range transcriptional repression in a reporter system
Abstract......................................................................................................................................102
Introduction................................................................................................................................103
Materials and Methods..............................................................................................................109
Results and Discussion................................................................................................................111
Appendices: Cofactor contributions to Hairy repression activities............................................125
References..................................................................................................................................142
Chapter IV
Conclusions and Future Directions
Conclusions.................................................................................................................................147
Future Directions........................................................................................................................149
References..................................................................................................................................154

iv

LIST OF TABLES

Supplemental Table 2-1:

Numbers of embryos within each category after heat shock
titration of Hairy protein are summarized. ................................ 76

Supplemental Table 2-2-1: p-values for Student’s t-test comparing before repression and
after repression chromatin immunoprecipitation signals of
each data point in Figure 2-2,2-3 and 2-4 (n ≥ 3)....................... 77
Supplemental Table 2-2-2: p-values for Student’s t-test comparing before repression and
after repression Micrococcal Nuclease Digested/Undigested
values of each data point in Figure 2C (n ≥ 3).............................. 78
Supplemental Table 2-2-3: p-values for Student’s t-test comparing before repression and
after repression Micrococcal Nuclease Digested/Undigested
values of each data point in Figure 2D (n ≥ 3)............................ 79
Supplemental Table 2-3:

List of primer pairs used in chromatin immunoprecipitation
and micrococcal nuclease sensitivity analysis............................ 80

vi

LIST OF FIGURES
Figure 1-1:

Distinct repression mechanisms used by prokaryotic repressors...................

Figure 1-2:

Distinct repression mechanisms used by eukaryotic repressors..................... 9

Figure 1-3:

Modulation of corepressor activity through posttranslational modification
and small molecules........................................................................................ 15

Figure 1-4:

Long-and short-range repression....................................................................

Figure 1-5:

Both long-range repressor Hairy and short-range repressor Knirps recruit
multiple corepressors...................................................................................... 20

Figure 1-6:

Possible mechanisms of short-range repression.............................................

Figure 1-7:

Mechanisms of long-range transcriptional repression.................................... 29

Figure 2-1:

Repression of ftz by Hairy and eve by Knirps................................................... 46

Figure 2-2:

Hairy and Knirps-mediated transcriptional repression result in differential
changes in histone H3 occupancy and micrococcal nuclease sensitivity......... 50

Figure 2-3:

Hairy induces global, while Knirps induces local changes in histone
acetylation levels. For both repressors, H3K27 tri-methylation are little
changed overall............................................................................................... 55

Figure 2-4:

Activator occupancy of ftz and eve before and after repression....................

Figure 2-5:

Pol II and TATA binding protein occupancy at ftz and eve promoter and
transcribed regions.......................................................................................... 64

Figure A-1:

Chromatin sensitivity to MN ase digestion in the Drosophila embryo,
before and after induction of Knirps repressor............................................... 94

Figure 3-1:

Hairy-regulated reporter system.....................................................................

Figure 3-2:

Promoter occupancy by Gal4 activator and LexA-Hairy repressor was
measured by chromatin immunoprecipitation followed by Real-time PCR
analysis............................................................................................................ 112

Figure3-3:

Repression of lacZ reporter expression by LexA-Hairy results in inhibition of
RNA
polymerase
II
recruitment
measured
by
chromatin
immunoprecipiation using antibodies again (A) unphosphorylated, (B) ser-5
vii

4

18

26

61

104

phosphorylated, (C) ser-5 and ser-2 phosphorylated forms of RNA
polymerase II................................................................................................... 114
Figure 3-4:

Histone H3 lysine 4 and lysine 36 methylation marks track with
transcription activity of the lacZ reporter measured by chromatin
immunoprecipitation using antibodies specific to (B) Histone H3 Lysine 4
and (A) Lysine 36 methylation......................................................................... 116

Figure 3-5:

Histone variant H2A.Z tracks with histone H3 during activation and
repression of the lacZ reporter measured by immunoprecipitation using
antibody against H2A.Z.................................................................................... 118

Figure3-6:

Repression of twist 2 X PE enhancer by LexA-Hairy or endogenous Hairy
protein from a distance................................................................................... 120

Figure3-7:

Closed chromatin environment inhibits LexA-Hairy recruitment.................... 123

Figure B-1:

Mutant Hairy accumulates at a lower level than the wild-type Hairy
protein, despite that the mRNA levels are comparable................................. 136

Figure B-2:

Abundance of ftz mRNA in embryos with overexpressed wild-type Hairy or
mutant Hairy lacking the Groucho-binding WRPW motif (Hairy w/a) relative
to embryos lacking the Hairy transgne (yw).................................................... 137

Figure B-3:

Expression patterns of early Drosophila genes before and after 20 minute
induction of Hairy protein analyzed by in situ hybridization........................... 138

Figure B-4:

Abundance of paird and achaete mRNA in embryos with overexpressed
wild-type Hairy or mutant Hairy lacking the Groucho-binding WRPW motif
(Hairy w/a) relative to embryos lacking the Hairy transgne (yw).................... 139

Figure B-5:

Mutation in the Orange domain rendered Hairy less efficient in repression
of ftz................................................................................................................ 140

Figure B-6:

Groucho migrates in molecular masses from around 158 KDa to above 670
KDa.................................................................................................................. 141

Figure 4-1

Mechanisms of Hairy and Knirps-mediated transcriptional repression..........

Figure 4-2

Possible mechanisms underlying Groucho-mediated context-dependent
repression........................................................................................................ 152

viii

150

KEY TO ABBREVIATIONS
aa: amino acid
bp: base pair (s)
cDNA: complementary deoxyribonucleic acid
ChIP: chromatin immunoprecipitation
ChIP-chip: chromatin immunoprecipitation followed by microarray analysis
ChIP-seq: chromatin immunoprecipitation followed by sequencing analysis
CtBP: C-terminal binding protein
DBD: DNA binding domain
DNA: deoxyribonucleic acid
EDTA: ethylenedinitrilo tetraacetic acid
eve: even-skipped
ftz: fushi tarazu
HAT: histone acetyltransferase
HDAC: histone deacetylase
HEPES: N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid
HMT: histone methyltransferase
hsp: heat shock promoter
PcG: polycomb group
PCR: polymerase chain reaction
Pol II: polymerase II
Rb: Retinoblastoma tumor suppressor protein
RNA: ribonucleic acid
TBP: TATA box binding protein
ix

Chapter I
Introduction

Transcriptional repression is essential for establishing precise patterns of gene
expression during development. One of the first major findings stemming from analysis of early
Drosophila transcriptional regulation was the realization that transcriptional repressors appear
to fall into two classes: those acting locally as “short-range” repressors, and those acting
globally as “long-range” repressors. The types of transcription factors and cofactors involved
point strongly to conservation of these pathways in all metazoans. One area in which our
knowledge has lagged is in understanding how these different classes of repressors function at
a mechanistic level. The interest in this topic is huge, because the interaction of transcriptional
corepressors with developmental circuits is central to key areas of biology including stem cell
reprogramming, cancer, and development.
In this work, I describe studies that shed light on the mechanisms of long- and shortrange repression during early Drosophila development. My chromatin studies of the prototypic
long-range repressor Hairy and the short-range repressor Knirps provide mechanistic
information about short- and long-range repression at the chromatin level. My work indicates
that transcriptional corepressors can be differentially deployed to mediate chromatin changes
in a context-dependent manner. To discern the importance of individual corepressors, my
studies have also examined the role of the conserved Groucho corepressor in Hairy-mediated
long-range repression.
In section I of this chapter, I will first review the literature regarding known general
transcriptional repression mechanisms from an evolutionary perspective. Then in section II, I

1

will introduce the distinctions between long- and short-range transcriptional repressors, with
focus on the prototypic long-range repressor Hairy and short-range repressor Knirps, as well as
their corepressors. Finally in section III, I will discuss the previous knowledge on the
mechanisms of long- and short-range repression, which serve as a background to my own
discoveries.

Section I: Transcriptional repression: conserved and evolved features

1

Transcriptional repression embodies the original mechanism discovered for gene
regulation, dating to the pioneering work of Jacob and Monod. Subsequent discoveries filled
out a picture showing that bacterial genes are regulated by sophisticated layers of activation
and repression, as well as complex posttranscriptional mechanisms. From basic biophysical
considerations, simple bacterial cis regulatory transcriptional elements should be capable of
generating complex outputs of the kind seen with eukaryotic genes [1]. Nature is not so
parsimonious, however; the evolution of more sophisticated gene expression programs in
eukaryotes has been accompanied by the appearance of more complex machinery and
mechanisms of transcriptional control, including repression. A multitude of cellular processes
control transcriptional activity of genes, and include negative regulatory pathways that limit
production and activity of activator proteins. Dedicated DNA-binding transcription repressors
that actively block transcription play equally crucial roles, however. Focusing on this latter
aspect, here we survey our current understanding of their repression mechanisms, and suggest
1

Section I has been published as part of a review article for which I was a coauther: Payankaulam, S., Li, L.M. and
Arnosti, D.N. (2010) Transcriptional repression: conserved and evolved features. Current Biology 20, R764-R771.

2

how evolving complexity is matched by adoption of multitiered repressor/corepressor systems
that themselves are subject to considerable elaboration and modification.
Transcriptional repression in bacteria: complex responses from streamlined systems
Bacterial RNA polymerase action has been shown to involve binding of the enzyme to
the promoter to form a closed complex, melting of the DNA to form the open complex,
promoter escape after formation of the first few phosphodiester bonds, and elongation and
termination [2, 3]. All of these steps are possible points for control (Figure 1-1). With the
exception of antitermination mechanisms that involve proteins complexed to the nascent
transcript, bacterial repression pathways invariably involve the direct action of DNA-binding
repressor proteins on the transcription machinery to block the activity of RNA polymerase at
various steps, including interfering with recruitment, initiation, or elongation. A common
mechanism involves an either/or occupancy of promoter regions by inhibitors and the
polymerase. The well-studied LacI repressor, for example, utilizes a primary binding site close to
the transcription initiation site to block RNA polymerase access to the promoter; this
mechanism is greatly enhanced by loop formation involving distal auxiliary sites [4].
Interestingly, placing binding sites just 3′ of the initiation site can switch the mode of repression
to blockage of promoter escape, suggesting a flexibility in possible effects mediated by these
simple components [5]. Other bacterial repressors, including phage lambda cI repressor and the
LexA repressor also target promoter proximal sites to block RNA polymerase binding [6, 7]. In
cases where the repressor binding site does not overlap with the promoter, repressor binding
can nucleate the binding of additional repressor molecules to occlude RNA polymerase, as seen

3

A

B

C

D

Figure 1-1. Distinct repression mechanisms used by prokaryotic repressors. A bacterial
repressor (R) directly targets the core transcription machinery in the following stages: (A)
repressor binds to the promoter proximal sites to block the binding of RNA polymerase (Pol);
(B) and (C) repressor binds simultaneously to the promoter with RNA polymerase, inhibiting its
transition from closed to open complex, or preventing promoter escape; (D) inhibition of RNA
polymerase promoter escape can also occur when repressor binds 3′ of the initiation site. For
interpretation of the references to color in this and all other figures, the reader is referred to
the electronic version of this dissertation.

4

with DnaA binding to the Escherichia coli dnaA gene [8]. Binding of this protein within the gene
can also induce premature termination [9].
A second common mechanism involves simultaneous occupancy of RNA polymerase and
repressor, with inhibitory consequences. The phage φ29 protein p4 binds to sites 5′ of the viral
A2b promoter and interacts with the α subunit of the RNA polymerase to prevent promoter
escape, although abortive initiation can occur [10]. The MerR repressor acts at an earlier step;
this repressor and the polymerase bind simultaneously to the merT promoter, resulting in a
complex that is blocked in the transition from closed to open complex in the absence of
mercury [11, 12]. Upon binding of the metal, allosteric alterations in MerR change DNA bending
to permit open complex formation [13]. Similar allosteric controls of RNA polymerase closed to
open complex formation have been reported for the E. coli GalR repressor and the plasmidborne KorB repressor. The latter protein occupies promoter proximal sites to exert a moderate
level of repression, but strongly synergizes with two other plasmid encoded repressors, KorA
and TrbA, that can functionally interact with KorB at a distance, suggesting that distally-acting
multiprotein complexes can also be utilized to repress bacterial genes [14].
These examples show that a major theme of transcriptional regulation in bacteria is the
direct physical interactions of repressors (and activators) with the basal transcription
machinery, programmed by the structure of cis regulatory regions. More complex responses
are effected by utilization of additional auxiliary sites, as with the
LacI repressor, but there are relatively few instances known where two distinct DNA binding
proteins are jointly employed to effect repression (or activation), and in no cases do the
repressors need to mediate their control of polymerase through intermediate protein
5

complexes. This direct interaction contrasts strongly with the situation in eukaryotes, as we
discuss below.
Archaeal transcription — complex core machinery, simple repression mechanisms
Recent studies of the third domain of life, the archaea, indicate that the transcriptional
machinery of these organisms exhibits similarities with the machineries of both bacteria and
eukaryotes. The archaeal RNA polymerase of Methanococcus jannaschi has twelve subunits
homologous to those of eukaryotic RNA polymerase II; a core complex containing just the
homologs to bacterial ββ′α is sufficient for in vitro initiation [15]. The holoenzyme is positioned
by basal factors TFB and TBP that bind to the basal promoter, similar to the case in eukaryotes
[16].
Some Euryarchaea contain H3 and H4-like histones that serve to compact DNA, but
these histones lack the tail regions frequently modified for gene regulatory purposes in
eukaryotes, and they bind less tightly to DNA than eukaryotic histones [17]. Other archaeal
genomes lack these eukaryotic-like histone genes, but appear to have basic DNA-binding
proteins that serve a similar purpose [18, 19]. It is not clear if archaeal histones are as
inherently inhibitory to transcription as those present in eukaryotes, and whether all types of
histone analogs in archea have similar repressive potential.
Archaeal genomes contain numerous homologs to bacterial transcription factors, and
insight into transcriptional repression in archaea was first obtained through characterization of
one of these, the metal-dependent regulator Mdr1 of Archaeoglobus fulgidus. In a similar
manner bacterial repressors, the Mdr1 protein regulates transcription of its own gene by
blocking recruitment of the RNA polymerase to the promoter [20, 21]. Another transcriptional

6

repressor, the Phr protein of Pyrococcus furiosus, acts through promoter-proximal binding sites
to similarly prevent RNA polymerase from accessing the promoter [22]. The basal factors bound
at the transcription start site can also be targeted; Lrs14, an archaeal homolog of the bacterial
leucine responsive regulatory protein family, inhibits the promoter binding of archaeal general
transcription factors TBP and TFB, thus interfering with early steps of transcription initiation
[23, 24]. In vitro experiments have revealed that NpR, a regulator of nitrogen metabolism in
Methanococcus maripaludis can also inhibit binding of TFB and TBP.
Thus, even though the archaeal core transcription machinery bears a strong
resemblance to that of eukaryotes, transcriptional repression appears to follow the simple
scenario seen in bacteria, whereby DNA-binding proteins directly target basal machinery at the
promoter. These mechanistic studies, however, have relied largely on in vitro assays in the
absence of archaeal nucleosome-like proteins, where possible effects of covalent modification
of nucleosome-like proteins are neglected [25]. A more complete picture is likely to emerge as
biochemical studies progress. One potentially unifying notion is that the small bacterial-sized
genomes found in archaea may demand no more highly sophisticated repression mechanisms
than those present in bacteria. However, it is striking that the bacterial-sized genome of the
budding yeast Saccharomyces cerevisiae does feature a full panoply of chromatin-mediated
repression mechanisms. It is possible that other features of the eukaryotic lifestyle dictate
elaboration of more complex mechanisms for transcriptional control.
Repression features layers of regulation in eukaryotes
To consider how eukaryotic repression mechanisms compare with these bacterial and
archaeal mechanisms, we should note key general features of eukaryotic transcription control.

7

The distinctly different set of proteins and mechanisms involved in eukaryotic repression
correlates with enhanced structural and functional demands in eukaryotes (Figure 1-2).
Especially in multicellular eukaryotes, achieving correct temporal- and tissue- specific gene
expression poses enormous functional challenges. In addition, the large size of many eukaryotic
genomes dilutes regulatory sequences in a sea of non-coding elements, making establishment
of the correct interactions between regulatory proteins and target genes problematic.
Presumably as an adaptation to these demands, the chromatin structure of eukaryotes is
profoundly different from that of bacteria and archaea. Eukaryotic histone proteins, assembled
on DNA as nucleosomes, pose a formidable intrinsic barrier to access of transcription
machinery, reducing background transcriptional noise that fortuitous interactions may generate
[26]. Eukaryotic histones, with flexible ‘tail’ regions, also provide a platform for elaboration of
various modification marks of regulatory significance.
An additional, recently appreciated feature of eukaryotic transcriptional regulation is
the presence of RNA polymerase at many promoters even when the genes are in a quiescent
state, suggesting that promoter escape may be the rate-limiting step [27, 28]. In bacteria, such
stalled complexes reflect either a deficient closed-to-open complex transition, a function of the
DNA template that is compensated for by activators, or the active intervention by repressors
that prevent an otherwise favorable reaction. In eukaryotes, it is not clear whether stalled
polymerases are influenced by DNA binding repressors, or if such stalling is solely a
consequence of basal promoter sequences and chromatin composition [29].
Eukaryotic transcriptional systems may have developed a number of specific features to
provide additional complexity necessary for gene regulatory demands in these organisms. First,
8

Figure 1-2. Distinct repression mechanisms used by eukaryotic repressors. Because of the
increased genome complexity and presence of chromatin, eukaryotic repressors (R) rely largely
on recruitment of corepressors and chromatin modifying enzymes, including HMTs (histone
methyltransferases), HDACs (histone deacetylases), KDMs (lysine demethylases) and chromatin
remodeling factors with ATPase activity. In some cases these corepressors interact directly with
activators (A) or the basal machinery.

9

increased complexity in a transcriptional system may be generated by increasing the number
of transcription factors, a trend also noted in bacteria, where those species with complex
developmental life cycles tend to possess a large number of RNA polymerase binding sigma
factors. Genomic studies have indeed revealed lineage-specific expansions of specific
eukaryotic transcription families, suggesting that there are advantages in allocating functions
among more actors [30]. When viewed at a global level, however, the percentage of genes
annotated as encoding DNA-binding transcription factors is similar in eukaryotes and bacteria,
ranging from 7% in E. coli to 12% in Streptomyces coelicolor, compared with 5.9% in the plant
Arabidopsis thaliana and 10% in humans [31, 32]. Some eukaryotes, such as the nematode
Caenorhabditis elegans, have a smaller percentage of genes devoted to DNA-binding
transcription factors (5%) than do many bacteria [33]. Alternative splicing and posttranslational
modifications can of course increase the number of isoforms present in a cell, but this is not the
sole source of complexity.
A second layer of complexity is the nature of eukaryotic cis regulatory elements; many
bacterial elements consist of only a few binding sites, whereas eukaryotic promoters and
enhancers commonly have dozens of motifs responsible for assembling core and regulatory
machinery. Such combinatorial interactions provide an effective way to differentiate real from
spurious cis elements in large genomes, facilitate cooperative mechanisms to overcome the
barrier of histones, and offer more complex switch responses for fine-scale control. Yet with
respect to transcriptional repressors, these proteins are generally capable of mediating
effective transcriptional responses from rather simple cis elements; often from one or a few

10

binding sites for transcription factors. Hwa’s predictions that cis element structures need not be
elaborate to generate proper readouts appear to be accurate in this context [34].
A third major area of difference between bacterial and eukaryotic systems is the
elaboration of core transcriptional machinery, which is defined as the set of proteins required
for assembly and initiation at a basal promoter. In addition to the twelve polypeptides
associated with the RNA polymerase II responsible for transcription of protein coding genes, at
least 60–80 other polypeptides are considered critical for the essential activity of
transcriptional initiation [35]. Clearly, this complexity should offer additional levels of
regulation. In fact, recent discoveries of tissue-specific forms of this basal transcriptional
machinery suggest that developmental gene regulation does involve alterations to this core set
of factors [36]. It is less clear that with a larger complex of proteins assembling at promoters, a
proportionally greater number of control points or available protein surfaces are directly
contacted by different transcription factors. Many activator effects funnel through a limited
number of targets in the basal machinery, and as we note below, repressors generally appear
to predominantly interact with the chromatin.
The nature of the interaction between transcription factor and transcription
machinery/DNA template is where bacterial and eukaryotic regulation shows the greatest
divergence. Eukaryotic transcriptional activators appear to function through two general levels,
namely, recruiting coactivator complexes that can modify chromatin, and interacting with core
transcriptional machinery [37]. In the latter category, some transcriptional activators directly
target components of the basal machinery, such as TAF proteins of the TFIID complex, to recruit
or allosterically affect these proteins, while other transcriptional factors interact with core
11

machinery by means of coactivator complexes. DNA-binding eukaryotic transcriptional
repressors, on the other hand, have only in rare instances been documented to directly contact
and modify the activity of core machinery [38]. Some in vitro systems have documented
competitive interactions between repressors and the basal transcriptional machinery, but the in
vivo relevance of such studies remains uncertain [39]. Physical competition may provide a
mechanism, as DNA binding repressors have also been shown to compete for overlapping
activator binding sites, but the generality of this model is uncertain, for most repressor binding
sites do not directly overlap activator sites [40]. Most DNA-binding transcriptional repressors
appear to rely instead on indirect interactions, mediated by corepressors.
Corepressors — definition, function and conservation
The term corepressor refers to either a single protein or a scaffolding protein mediating
the assembly of a multi-subunit complex that is recruited to a gene by a transcription factor.
Repressors have discreet peptide motifs often contained within transcriptional repression
domains which permit interaction with specific corepressors. It has been assumed that the
transcription factor–corepressor complex is modular, because direct tethering of corepressors
to DNA via a heterologous DNA-binding domain elicits transcriptional repression [40-42]. This
assay neglects important context effects, however, as we discuss below. Corepressor
complexes can exhibit a range of activities; they frequently contain histone deacetylase (HDAC)
activity essential for stripping activation marks from histone tails. Corepressor complexes can
contain other enzymatic activities as well, including histone methyl transferase or demethylase
activities that can reset chromatin marks (Figure1-2).

12

Another conserved group of cofactors are SWI/SNF nucleosome remodeling complexes
which generally serve an activating role, but in some cases are recruited by transcription factors
to induce gene repression and thus can be considered a special type of corepressor. The Mi2–
NuRD corepressor complex combines nucleosome remodeling activity with histone deacetylase
activity, and does appear to play a dedicated repressive role [43-45]. In addition to nucleosome
modification, corepressors have been shown to directly interfere with coactivators such as the
CBP/p300 histone acetyltransferase or the mediator complex [46, 47].
These latter examples appear to be exceptions involving nonhistone targets, but in
general, histone modifiying enzymes are capable of modifying transcription factors with the
same posttranscriptional modifications as seen on histone tails. Nonetheless, our current
understanding is that corepressors predominantly target chromatin structures. With hundreds
of transcription factors in the cell, but only a dozen or so corepressor complexes, it may appear
that implementing transcriptional repression through these cofactors produces a bottleneck of
limited, generic activities. However, corepressors are not simply interchangeable levers to
lower the nucleosome barrier, but sensitive control points that provide additional regulatory
modulations through structural diversification, by sensing of signaling cascades, and contextdependent responses. These elaborations tune cell/tissue specific gene responses that are keys
for development.
Corepressors link signal pathways with transcription activity
Signaling pathways link environmental and physiological stimuli to transcription activity
through posttranslational modifications of transcription factors, for example phosphorylation,
acetylation, hydroxylation, and sumoylation. Corepressor complexes are similarly targets of
13

signaling pathways and may provide unique channels of communication with the cellular
environment. Posttranslational modifications of corepressors can either change their
subcellular localization or regulate the binding to transcription factors. In addition, some
corepressors can bind to small molecule compounds in the environment, which can modulate
repression potencies (Figure 1-3).
A number of posttranslational modifications affect subcellular localization of
corepressors. Mammalian CtBP1 is SUMOylated, a modification that is essential for its
accumulation in the nucleus and repression of the E-cadherin promoter [48]. Acetylation of
CtBP2 by p300 can play a similar role; this modification is required for nuclear retention and
repression activity [49]. The SMRT corepressor is phosphorylated by kinases of the MAP kinase
pathway, leading to subcellular redistribution [50].
Posttranslational modifications of corepressors also affect the recruitment of
corepressor complexes by transcription factors. Phosphorylation is one of the most prevalent
post-translational modifications that modulate corepressor activities. During G1–S phase
transition, the Rb corepressor is phosphorylated at multiple residues by the combinatorial
effects of several cyclin-dependent kinases (CDKs), which leads to the disruption of the Rb–E2F
complex [51]. Groucho activity is also modulated by phosphorylation, in response to signaling
pathways. MAPK pathway phosphorylation attenuates Groucho’s binding to the Eyeless
transcription factor, while in contrast phosphorylation of the Groucho homolog TLE1 by casein
kinase II enhances its association with the Hairy homolog Hes1 and with chromatin [52-54]. This
mechanism is suggested to allow Groucho to play an integrating role between the EGF receptor
and Notch signaling pathways [55]. A distinct modification, polyADP ribosylation by PARP1, is
14

Figure 1-3. Modulation of corepressor activity through posttranslational modification and
small molecules. (A) Subcellular localization of corepressors is affected by posttranslational
modifications, such as phosphorylation, SUMOylation or acetylation. (B) Posttranslational
modifications of corepressors alter a protein-protein interaction surface, either inhibiting or
enhancing the recruitment of corepressors. (C) Binding to cellular compounds may impact the
corepressor activity: CtBP is suggested to be a more potent corepressor when bound to NADH
compared to NAD+, thus allowing it to function as a cellular redox state sensor.

15

suggested to mediate the dismissal of the TLE1 corepressor complex from HES1-regulated
promoter during neural stem cell differentiation [56].
In addition to these covalent modifications, binding of ligands has been suggested to
provide a way for corepressors to respond to the changes in the cellular environment. CtBP
exhibits remarkable sequence similarity to dehydrogenases and has a Rossman fold for binding
to NAD+/NADH [57]. Binding of the cofactor is essential for CtBP’s interaction with cellular and
viral transcriptional repressors, and it was reported that NADH has a much higher affinity than
NAD+ for CtBP [58]. CtBP might serve as a redox sensor to regulate transcription according to
changes in nuclear NAD+/NADH ratios. These modifications provide a sampling of some of the
ways that corepressor activity can be modulated by posttranslational effects that link
corepressor function to cellular physiology.

Section II: Long- and short-range transcriptional repressors

Analysis of transcriptional regulation in Drosophila indicates a further level of
complexity regarding the transcription repression mechanisms in eukaryotes. Studies revealed
that transcriptional repressors can be classified into two classes according to their range of
action [59]. Long-range repressors act over a range of more than 1 kb to dominantly silence the
whole targeted locus. In Drosophila, Hairy and Dorsal repressors have been found to repress
enhancers from a distance in transgenic reporter assays in Drosophila embryos [60, 61]. Related
examples of long-range repression in mammals include: a silencer element contained within
16

the mouse CD4 gene, a neuron-restrictive silencer element regulating the mouse neuronal
growth associated protein scg-10, a retinoic acid response element within the mouse Hoxb-1
locus, and the mammalian E2F-Rb complex binding element [62-66]. Although repressors
binding to most of these elements are not very well characterized, the elements have been
demonstrated to function in a direction- and orientation- independent manner to mediate
dominant repression. On the other hand, short-range repressors work over a distance of less
than 100 – 150 bp to inhibit activators in a local fashion [67]. Prototype short-range repressors
in Drosophila include Knirps, Kruppel and Snail [59-61, 68, 69] (Figure1-4).
Long- and short-range repressors identified in Drosophila have homologs throughout
metazoa, including highly conserved functional motifs, such as cofactor binding surfaces and
DNA binding domains. I will introduce the prototypic long-range repressor Hairy and shortrange repressor Knirps in more detail, as they were used as models to study mechanisms of
long- and short-range repression in the following chapters.
The long-range repressor Hairy
1. The Hairy and Enhancer of Split related proteins
Hairy is the founding member of the Hairy and Enhancer of Split (HES) related protein
family. Other proteins within this family include Deadpan, HES, Hey and Stra 13. Most of the
family members function as DNA binding transcriptional repressors, which contain a tandem
arrangement of the basic helix-loop-helix domain and an adjacent sequence known as the
Orange domain. [70]. Except for the Stra 13 proteins, proteins in this family also have a
conserved C-terminal tetrapeptide motif, either WRPW or YXXW that mediates interactions
with the Groucho corepressor [70].
17

Figure 1-4. Long-and short-range repression. (A) Long-range repressors like Hairy can
dominantly repress all enhancers within the targeted locus. Shown in the figure is the
schematic representation of Hairy regulation of its canonical target fushi tarazu (ftz). (B) Shortrange repressors like Knirps can only locally regulate the enhancers they are located in. Shown
in the figure is the schematic representation of Knirps regulation of its canonical target even
skipped (eve). The boxes represent stripe-specific enhancers, and the numbers underneath the
boxes represent the targeted stripes.

18

HES proteins regulate multiple biological processes in Drosophila as well as vertebrates,
including mesoderm segmentation, vertebrate neurogenesis, neural cell fate determination,
blood vessel formation, myogenesis, and inhibition of oncogenesis [70]. Understanding Hairy
repression will not only shed light on the basic long-range repression mechanisms, but also
provide insights into many important biological processes.
2. Hairy –mediated transcriptional repression in Drosophila
In Drosophila, Hairy directly binds to DNA and functions as a homodimer [71, 72]. In vivo
transgenic reporter studies revealed that Hairy mediates long-range dominant repression from
isolated binding sites, although later studies suggested that this repression can be potentiated
by adjacent activators [60, 73]. Like many repressors in eukaryotes, Hairy repression is mainly
mediated by corepressors, and the Groucho corepressor has been suggested to account for
most of Hairy repression activity [74, 75].
Other than Groucho, Hairy has also been shown to interact with corepressor dSir2
through a motif RRAR in the basic region of Hairy and with corepressor C-terminal binding
protein (CtBP) through the PLSLV motif located in close proximity to the Groucho binding motif,
only nine amino acids away from the WRPW motif [76, 77] (Figure 1-5, A). The roles of these
different corepressors in Hairy-mediated repression are still unclear. One possibility is that
Hairy utilizes these corepressors in a context-dependent manner; in vivo binding analysis
carried out using DamID suggested that the Groucho and CtBP corepressors do not entirely
overlap with Hairy binding [78]. It is also possible that these corepressors interact with each
other to fine-tune the output of Hairy. dCtBP and Groucho are seen to have antagonistic effects
in Hairy-mediated repression [79].
19

A

B

Figure 1-5. Both long-range repressor Hairy and short-range repressor Knirps recruit multiple
corepressors. Schematic representations of functional domains of Hairy and Knirps are shown.
(A) Hairy binds to DNA through bHLH domain, which also recruits dSir2 corepressor. Orange
domain is conserved among Hairy and Enhancer of split family members, and is important for
Hairy function. The C-terminus of Hairy protein contains the Groucho recruitment motif
‘WRPW’ and the CtBP recruitment motif ‘PLSLV’. These two motifs are nine amino acids apart.
(B) Knirps binds DNA through Zinc Finger at its N-teminus. Groucho is recruited to Knirps
through the collaborative action of two motifs located at 85-91aa and 169-189aa. CtBP is
brought to the complex through an atypical ‘PMDLSMK’ motif.

20

Despite the importance of Hairy and its related family members in Drosophila, only a
handful of Hairy targets have been identified so far. Hairy was first identified as a
transcriptional repressor through its ability to repress fushi tarazu expression upon ectopic
expression using a hsp70 promoter [80]. During development, Hairy is expressed in an anteriorposterior seven stripe segmental pattern that serves to stabilize the striped expression pattern
of ftz. The initial ftz pattern is established by regional cues provided by gap genes [81]. Hairy
stabilizes this pattern by repressing this gene in the interstripe regions. Hairy also inhibits
sensory organ formation by direct repression of achaete, through binding sites located 300 bp
upstream of this gene’s basal promoter [72]. Recent studies have suggested that Hairy also
regulates the expression of TCA cycle enzymes, functioning as a metabolic switch to mediate
hypoxic tolerance [82].
The short-range repressor Knirps
Knirps is a member of the nuclear receptor superfamily, possessing this family’s
characteristic zinc finger DNA binding domain, which is coupled to a unique C-terminal motif
‘PMDLSMK’ that interacts with CtBP [83, 84]. Knirps is an important regulator for striped
patterns of pair-rule gene expression in the Drosophila embryo [85]. even skipped is a canonical
Knirps target, with extensive information about the role of this repressor on the modular eve
stripe 3+7 and 4+6 enhancers [86]. During later stages of development, Knirps and its related
proteins determine the position of the second longitudinal wing vein (L2) [87]. They also
control cell fates and branch morphogenesis during tracheal development [88].
Knirps has been shown to mediate both local quenching and direct repression in
Drosophila embryos in both a CtBP-dependent and -independent manner [59, 89] (Figure1-5,
21

B). The Rpd3 deacetylase has also been shown to exist in the Knirps corepressor complex
through its interaction with CtBP [90]. Earlier studies suggested that similar to the distinction
between transcription repressors, distinct corepressors are required for long- and short-range
repression: Groucho is the mediator for long-range repression, and CtBP is required for shortrange repression [84, 91]. Although functioning as a dedicated short-range repressor, Knirps
also recruits and functionally interacts with the Groucho corepressor, which had previously
been suggested to be a dedicated long-range corepressor [92].
Long- and short-range transcriptional corepressors
The “Long-range” corepressor Groucho
1. Groucho and the Transducin-like Enhancer of split (TLE) proteins
The Drosophila groucho (gro) gene was identified through a viable mutation that results
in clumps of extra bristles above the adult eyes, reminiscent of the bushy eyebrows of Groucho
Marx [93] . Groucho is the prototype of the transducin-like Enhancer of split (TLE) protein
family, whose members are found in most metazoans. Members within this family contain 40
amino acid tandem repeats with characteristically located tryptophan and aspartate residues,
so called the WD-40 motif [94]. They function mainly during segmentation, central and
peripheral neurogenesis, as well as epithelial differentiation in Drosophila and mammals [74,
95].
2. Mechanisms of Groucho/TLE-mediated repression
Groucho’s role as a transcriptional corepressor was initially revealed through its
interaction with Hairy. Later studies revealed that Groucho interacts with many other
sequence-specific repressors including Dorsal, Runt, Brinker and Engrailed [74, 96-99].
22

Courey and colleagues proposed that Groucho functions by spreading along the
chromatin, based on the evidence that Groucho can oligomerize in vitro by means of its Nterminal glutamine-rich-domain, which is important for its function. Biochemical studies
indicated that Groucho interacts with the Rpd3 histone deacetylase and with deacetylated
histones [41, 91, 100, 101]. Our laboratory has demonstrated that Groucho spreads along the
transcribed region of a Hairy-repressed reporter in vivo [102].
The fact that Groucho interacts with Hairy and Dorsal, two well-characterized longrange repressors, as well as the molecular characteristics of Groucho, led to the prevalent view
that Groucho is a long-range corepressor [91]. However, accumulating evidence suggests that
Groucho also is involved in short-range repression mediated by Slp1 and Knirps [92, 103]. Thus,
rather than functioning as a dedicated long-range repressor, Groucho is more likely to function
in a context-dependent manner to mediate repression.
The short-range corepressor CtBP
1. CtBP family protein
CtBP was discovered during a search for cellular proteins that complex with the Cterminal region of E1A implicated in negative modulation of oncogenic transformation. CtBP
interacts with a short sequence motif (PLDLS) conserved among the E1A proteins of all human
and primate adenoviruses [104]. The Drosophila CtBP has been shown to be essential for wildtype function of short-range repressors, and it binds to the short-range repressors Knirps,
Kruppel and Snail [84]. Although CtBP has also been shown to interact with the long-range
repressor Hairy, this interaction appears to antagonize with Hairy repression, through
competition with Groucho corepressor [79].
23

2. Mechanisms of CtBP repression
Proteomic analysis of the mammalian CtBP1 complex revealed that CtBP1 recruits
multiple enzymatic components, including

the histone deacetylase Rpd3, histone

methyltransferases G9a and GLP which methylate histone H3 at lysine 9, and the histone
demethylase LSD1, which demethylates di-methylated histone H3 lysine 4 [105, 106]. It is
possible that CtBP-mediated repression is a combinatorial effect of decreased activating
chromatin marks and elevated repressive chromatin marks.
Context-dependent activity of corepressors

2

Assuming a cofactor is in the correct subcellular location and in an active state, rather
than simply switching a gene off, corepressors can be deployed in a context-dependent manner
to mediate distinct types of repression. Recent studies of the Groucho corepressor have
provided a tantilizing insight into yet another level of functional complexity affecting repression
systems. Although considered as a long-range corepressor for a long time, recent evidence
indicates that Groucho is also an essential corepressor for the Slp1 and Knirps short-range
repressors in the Drosophila embryo [78,79]. The same component of the cellular corepression
machinery can thus evince distinct functions —to what may we attribute this flexibility? Several
molecular features of the protein and its complexes provide clues.
In some cases, transcription factors interact with multiple corepressors to elicit contextdependent effects. Different corepressor complexes are recruited to generate developmental
stage-specific, or gene- or enhancer- specific effects. The Drosophila Runt repressor requires
different corepressors for repression of engrailed at different developmental stages: initially
2

This section has been published as part of a review article for which I was a coauther: Payankaulam, S., Li, L.M. and
Arnosti, D.N. (2010) Transcriptional repression: conserved and evolved features. Current Biology 20, R764-R771.

24

repression is dependent on Tramtrack, and later Groucho, CtBP and Rpd3 are required for the
maintenance of repression after blastoderm stage [80]. The Drosophila repressor Brinker
protein displays gene specific effects, relying on either CtBP or Groucho complexes to repress
different target genes [81].
Even distinct enhancers from the same gene can show cofactor specific effects. The
even skipped stripe 3/7 enhancer is repressed by Knirps even in the absence of CtBP, while this
same repressor requires CtBP to inactivate the stripe 4/6 enhancer during the same period of
development. Knirps relies on Groucho for repression of the stripe 3/7 enhancer [79,82]. Such
combinatorial usage of corepressor complexes is not restricted to transcription factors in
Drosophila but is also observed in higher eukaryotes. For example, the Xenopus Tcf3 (XtCF3)
interacts with both the Groucho and CtBP corepressors to regulate target gene expression [83].
Section III. Long- and short-range repression: two sides of the same coin?
The fact that both long- and short-range repressors can recruit CtBP and Groucho raises
several interesting questions. Is there indeed a distinction between long- and short-range
repression? What are the underlying differences between long- and short-range repression? In
order to better dissect the two modes of repression during early Drosophila development, it is
important to review previous knowledge of long- and short-range repression mechanisms.

Mechanisms of short-range repression
Transgenic reporter assays performed in Drosophila embryos have revealed several
possible mechanisms of short-range repression. Short-range repression can involve direct
inhibition of activator binding as seen for example by Snail repressor binding sites that overlap
with
25

A

B

C

Figure 1-6. Possible mechanisms of short-range repression. Analysis of short-range repression
in Drosophila embryos suggests that there might be three different mechanisms. (A) Shortrange repressors (R) compete for the same binding sites with activators (A). (B) Short-range
repressors (R) disrupt the crosstalk between the activators (A) and the basal transcription
machinery (Pol II). (C) Short-range repressors (R) can directly interact and interfere with the
basal transcription machinery (Pol II).

26

Twist activator binding sites in the dorsal-ventral enhancers regulating genes expressed in the
neuroectodem of the embryo (Figure 1-6). Short-range repression can also involve a
“quenching” mechanism, which need not involve binding sites that overlap with activator sites.
Alternatively, when suitably positioned proximal to the transcription start site, short-range
repressors can directly interact with the basal transcriptional machinery [59, 68].
These conclusions are largely based on experimental approaches that involve
manipulation of repressor binding sites and measurement of the associated transcriptional
readouts. Many mechanistic questions remain to be answered. Do activators and components
of the basal transcriptional machinery stay bound upon repression? How might the repressors
inhibit the crosstalk between activators and the basal transcriptional machinery? How do the
repressors regulate the basal transcriptional machinery and affect chromatin structure?
Genome-wide ChIP-chip studies reveal that the Snail short-range repressor co-occupied
repressed targets with Twist activator and RNA polymerase II, providing direct evidence for the
quenching model in which the crosstalk between activators and the basal transcriptional
machinery after recruitment was interrupted by the short-range repressor [28, 107].
Mechanisms of long-range repression
The two most popular models to explain long-range repression are mainly through
based on studies of Polycomb group proteins and Sir family proteins. The first model is the
‘spreading’ mechanism, in which binding of a repressor at one site facilitates the adjacent
binding of additional repressors or corepressors until the resulting protein complex extends
throughout the targeted locus, somehow interfering with gene expression (Figure 1-7, A). The
best example for this mechanism by the Polycomb group proteins-mediated repression,
27

occupancy of a locus by the Polycomb group complex is associated with transcription silencing,
despite continued occupancy of the basal transcriptional machinery, as well as wide-spread
H3K27, H3K9 and H4K20 methylation. Though the exact roles of these epigenetic marks are still
unclear [108, 109]. Another well-studied example for a spreading mechanism is exemplified by
SIR-dependent repression in budding yeast, in which the Sir2/Sir3/Sir4 proteins assemble a
silent chromatin region of around 3 kb at telomeres and the silent mating type loci [110].
Similarly, spreading of Drosophila Groucho protein has been observed on a reporter gene
regulated by Hairy long-range repression, which is thought to deliver the histone deactylase
Rpd3 along the chromatin to produce an inaccessible chromatin environment [102].
An alternative model for long-range repression involves a looping mechanism, in which
the intervening DNA longs out to allow contact between repressors and their targeted
enhancers and promoters (Figure 1-7, B). One alternative model for the role of H3K27
methylation in PRC1-mediated repression is that H3K27 facilitates intralocus looping to bring
PRE (polycomb response element)-bound complexes into contact with the body of the silenced
gene [111]. This model is supported by the studies showing that one insulator between the PRE
and promoter blocks silencing, but two insulators in tandem allow PRC repression to bypass this
blockage [112]. The observation suggests the continuous chromatin context is not necessary for
Polycomb repression, and the PRC can directly repress the target promoter through looping out
of the DNA in-between (Figure 1-7, B).

28

Figure 1-7. Mechanisms of long-range transcriptional repression. There are two mechanisms
by which long-range repressors (R) can reach distal promoters. (A) Spreading Mechanism:
Binding of the long-range repressors might facilitate the binding of additional repressors or
corepressors to adjacent sites, or modifications of chromatin nearby, until the resulting protein
complex or modified chromatin extend throughout the targeted locus. (B) Looping Mechanism:
The repressors (R) might interact with proteins binding to the basal promoters, allowing direct
contacts between the enhancers and promoters, and looping out the chromatin in-between.

29

Conclusions

Examination of the mechanisms used by DNA-binding transcriptional repressors
indicates that a relatively limited set of direct interactions is sufficient to regulate the activity of
bacterial and archaeal promoters, which feature compact cis regulatory elements. Eukaryotic
genes, whether because of their more complex cis regulatory structure, chromatin packaging,
or regulatory demands tend to require the action of corepressors, proteins that directly or
indirectly mediate histone modifications. Other than simply dictating an ‘OFF’ state of the
targeted genes, transcriptional repressors can be classified into long-and short-range repressors
based on their ranges of action. To understand the molecular distinctions between long-and
short-range transcriptional repressors, it is necessary to determine how they regulate the
chromatin environment, activators and the basal transcriptional machinery. A single
corepressor can be recruited by multiple transcriptional repressors. On the other hand, a single
transcription repressor is usually capable of recruiting multiple corepressors to differentially
repress their target genes. It is important to understand the roles of different corepressors
within a single repressor complex, in order to better predict the repression output mediated by
the particular repressor.

30

References

31

References
1.

Bintu, L., et al., Transcriptional regulation by the numbers: applications. Curr Opin Genet
Dev, 2005. 15(2): p. 125-35.

2.

Chamberlin, M.J., The selectivity of transcription. Annu Rev Biochem, 1974. 43(0): p.
721-75.

3.

Krummel, B. and M.J. Chamberlin, RNA chain initiation by Escherichia coli RNA
polymerase. Structural transitions of the enzyme in early ternary complexes.
Biochemistry, 1989. 28(19): p. 7829-42.

4.

Lewis, M., Response: DNA Looping and Lac Repressor--CAP Interaction. Science, 1996.
274(5294): p. 1931-2.

5.

Lopez, P.J., et al., On the mechanism of inhibition of phage T7 RNA polymerase by lac
repressor. J Mol Biol, 1998. 276(5): p. 861-75.

6.

Hawley, D.K., A.D. Johnson, and W.R. McClure, Functional and physical characterization
of transcription initiation complexes in the bacteriophage lambda OR region. J Biol
Chem, 1985. 260(14): p. 8618-26.

7.

Bertrand-Burggraf, E., et al., Promoter properties and negative regulation of the uvrA
gene by the LexA repressor and its amino-terminal DNA binding domain. J Mol Biol,
1987. 193(2): p. 293-302.

8.

Lee, Y.S. and D.S. Hwang, Occlusion of RNA polymerase by oligomerization of DnaA
protein over the dnaA promoter of Escherichia coli. J Biol Chem, 1997. 272(1): p. 83-8.

9.

Tesfa-Selase, F. and W.T. Drabble, Specific binding of DnaA protein to a DnaA box in the
guaB gene of Escherichia coli K12. Eur J Biochem, 1996. 241(2): p. 411-6.

10.

Monsalve, M., et al., Activation and repression of transcription at two different phage
phi29 promoters are mediated by interaction of the same residues of regulatory protein
p4 with RNA polymerase. EMBO J, 1996. 15(2): p. 383-91.

11.

Heltzel, A., et al., Activator-dependent preinduction binding of sigma-70 RNA
polymerase at the metal-regulated mer promoter. Biochemistry, 1990. 29(41): p. 957284.

12.

Summers, A.O., Untwist and shout: a heavy metal-responsive transcriptional regulator. J
Bacteriol, 1992. 174(10): p. 3097-101.

32

13.
14.

Ansari, A.Z., J.E. Bradner, and T.V. O'Halloran, DNA-bend modulation in a repressor-toactivator switching mechanism. Nature, 1995. 374(6520): p. 371-5.
Williams, D.R., M. Motallebi-Veshareh, and C.M. Thomas, Multifunctional repressor
KorB can block transcription by preventing isomerization of RNA polymerase-promoter
complexes. Nucleic Acids Res, 1993. 21(5): p. 1141-8.

15.

Korkhin, Y., et al., Evolution of Complex RNA Polymerases: The Complete Archaeal RNA
Polymerase Structure. PLoS Biol, 2009. 7(5): p. e102.

16.

Zhang, G., et al., Crystal structure of Thermus aquaticus core RNA polymerase at 3.3 A
resolution. Cell, 1999. 98(6): p. 811-24.

17.

Wilkinson, S.P., M. Ouhammouch, and E.P. Geiduschek, Transcriptional activation in the
context of repression mediated by archaeal histones. Proc Natl Acad Sci U S A. 107(15):
p. 6777-81.

18.

Li, W.T., et al., MJ1647, an open reading frame in the genome of the hyperthermophile
Methanococcus jannaschii, encodes a very thermostable archaeal histone with a Cterminal extension. Extremophiles, 2000. 4(1): p. 43-51.

19.

Fahrner, R.L., et al., An ancestral nuclear protein assembly: crystal structure of the
Methanopyrus kandleri histone. Protein Sci, 2001. 10(10): p. 2002-7.

20.

Bell, S.D. and S.P. Jackson, Mechanism of autoregulation by an archaeal transcriptional
repressor. J Biol Chem, 2000. 275(41): p. 31624-9.

21.

Napoli, A., et al., An Lrp-like protein of the hyperthermophilic archaeon Sulfolobus
solfataricus which binds to its own promoter. J Bacteriol, 1999. 181(5): p. 1474-80.

22.

Vierke, G., et al., A novel archaeal transcriptional regulator of heat shock response. J Biol
Chem, 2003. 278(1): p. 18-26.

23.

Lie, T.J., G.E. Wood, and J.A. Leigh, Regulation of nif expression in Methanococcus
maripaludis: roles of the euryarchaeal repressor NrpR, 2-oxoglutarate, and two
operators. J Biol Chem, 2005. 280(7): p. 5236-41.

24.

Lie, T.J. and J.A. Leigh, A novel repressor of nif and glnA expression in the methanogenic
archaeon Methanococcus maripaludis. Mol Microbiol, 2003. 47(1): p. 235-46.

25.

Bell, S.D., et al., The interaction of Alba, a conserved archaeal chromatin protein, with
Sir2 and its regulation by acetylation. Science, 2002. 296(5565): p. 148-51.

26.

Struhl, K., Fundamentally different logic of gene regulation in eukaryotes and
prokaryotes. Cell, 1999. 98(1): p. 1-4.
33

27.

Fuda, N.J., M.B. Ardehali, and J.T. Lis, Defining mechanisms that regulate RNA
polymerase II transcription in vivo. Nature, 2009. 461(7261): p. 186-92.

28.

Zeitlinger, J., et al., RNA polymerase stalling at developmental control genes in the
Drosophila melanogaster embryo. Nat Genet, 2007. 39(12): p. 1512-6.

29.

Hendrix, D.A., et al., Promoter elements associated with RNA Pol II stalling in the
Drosophila embryo. Proc Natl Acad Sci U S A, 2008. 105(22): p. 7762-7.

30.

Madan Babu, M. and S.A. Teichmann, Evolution of transcription factors and the gene
regulatory network in Escherichia coli. Nucleic Acids Res, 2003. 31(4): p. 1234-44.

31.

Bentley, S.D., et al., Complete genome sequence of the model actinomycete
Streptomyces coelicolor A3(2). Nature, 2002. 417(6885): p. 141-7.

32.

Vaquerizas, J.M., et al., A census of human transcription factors: function, expression
and evolution. Nat Rev Genet, 2009. 10(4): p. 252-63.

33.

Riechmann, J.L., et al., Arabidopsis transcription factors: genome-wide comparative
analysis among eukaryotes. Science, 2000. 290(5499): p. 2105-10.

34.

Buchler, N.E., U. Gerland, and T. Hwa, On schemes of combinatorial transcription logic.
Proc Natl Acad Sci U S A, 2003. 100(9): p. 5136-41.

35.

Sikorski, T.W. and S. Buratowski, The basal initiation machinery: beyond the general
transcription factors. Curr Opin Cell Biol, 2009. 21(3): p. 344-51.

36.

D'Alessio, J.A., K.J. Wright, and R. Tjian, Shifting players and paradigms in cell-specific
transcription. Mol Cell, 2009. 36(6): p. 924-31.

37.

Weake, V.M. and J.L. Workman, Inducible gene expression: diverse regulatory
mechanisms. Nat Rev Genet. 11(6): p. 426-37.

38.

Kim, J., G. de Haan, and D.J. Shapiro, DNA bending between upstream activator
sequences increases transcriptional synergy. Biochem Biophys Res Commun, 1996.
226(3): p. 638-44.

39.

Austin, R.J. and M.D. Biggin, A domain of the even-skipped protein represses
transcription by preventing TFIID binding to a promoter: repression by cooperative
blocking. Mol Cell Biol, 1995. 15(9): p. 4683-93.

40.

Wang, H. and D.J. Stillman, Transcriptional repression in Saccharomyces cerevisiae by a
SIN3-LexA fusion protein. Mol Cell Biol, 1993. 13(3): p. 1805-14.
34

41.

Chen, G., P.H. Nguyen, and A.J. Courey, A role for Groucho tetramerization in
transcriptional repression. Mol Cell Biol, 1998. 18(12): p. 7259-68.

42.

Sutrias-Grau, M. and D.N. Arnosti, CtBP contributes quantitatively to Knirps repression
activity in an NAD binding-dependent manner. Mol Cell Biol, 2004. 24(13): p. 5953-66.

43.

Zamir, I., J. Zhang, and M.A. Lazar, Stoichiometric and steric principles governing
repression by nuclear hormone receptors. Genes Dev, 1997. 11(7): p. 835-46.

44.

Burke, L.J. and A. Baniahmad, Co-repressors 2000. FASEB J, 2000. 14(13): p. 1876-88.

45.

Perissi, V., et al., Deconstructing repression: evolving models of co-repressor action. Nat
Rev Genet. 11(2): p. 109-23.

46.

Kim, J.H., et al., CtBP represses p300-mediated transcriptional activation by direct
association with its bromodomain. Nat Struct Mol Biol, 2005. 12(5): p. 423-8.

47.

Gromoller, A. and N. Lehming, Srb7p is a physical and physiological target of Tup1p.
EMBO J, 2000. 19(24): p. 6845-52.

48.

Lin, X., et al., Opposed regulation of corepressor CtBP by SUMOylation and PDZ binding.
Mol Cell, 2003. 11(5): p. 1389-96.

49.

Zhao, L.J., et al., Acetylation by p300 regulates nuclear localization and function of the
transcriptional corepressor CtBP2. J Biol Chem, 2006. 281(7): p. 4183-9.

50.

Hong, S.H. and M.L. Privalsky, The SMRT corepressor is regulated by a MEK-1 kinase
pathway: inhibition of corepressor function is associated with SMRT phosphorylation
and nuclear export. Mol Cell Biol, 2000. 20(17): p. 6612-25.

51.

Poznic, M., Retinoblastoma protein: a central processing unit. J Biosci, 2009. 34(2): p.
305-12.

52.

Choi, C.Y., et al., Phosphorylation by the DHIPK2 protein kinase modulates the
corepressor activity of Groucho. J Biol Chem, 2005. 280(22): p. 21427-36.

53.

Nuthall, H.N., et al., Role for Hes1-induced phosphorylation in Groucho-mediated
transcriptional repression. Mol Cell Biol, 2002. 22(2): p. 389-99.

54.

Hasson, P., et al., EGFR signaling attenuates Groucho-dependent repression to
antagonize Notch transcriptional output. Nat Genet, 2005. 37(1): p. 101-5.

35

55.

Hasson, P. and Z. Paroush, Crosstalk between the EGFR and other signalling pathways at
the level of the global transcriptional corepressor Groucho/TLE. Br J Cancer, 2006. 94(6):
p. 771-5.

56.

Ju, B.G., et al., Activating the PARP-1 sensor component of the groucho/ TLE1
corepressor complex mediates a CaMKinase IIdelta-dependent neurogenic gene
activation pathway. Cell, 2004. 119(6): p. 815-29.

57.

Schaeper, U., et al., Molecular cloning and characterization of a cellular phosphoprotein
that interacts with a conserved C-terminal domain of adenovirus E1A involved in
negative modulation of oncogenic transformation. Proc Natl Acad Sci U S A, 1995.
92(23): p. 10467-71.

58.

Zhang, Q., D.W. Piston, and R.H. Goodman, Regulation of corepressor function by
nuclear NADH. Science, 2002. 295(5561): p. 1895-7.

59.

Arnosti, D.N., et al., The gap protein knirps mediates both quenching and direct
repression in the Drosophila embryo. EMBO J, 1996. 15(14): p. 3659-66.

60.

Barolo, S. and M. Levine, hairy mediates dominant repression in the Drosophila embryo.
EMBO J, 1997. 16(10): p. 2883-91.

61.

Ip, Y.T., et al., The dorsal morphogen is a sequence-specific DNA-binding protein that
interacts with a long-range repression element in Drosophila. Cell, 1991. 64(2): p. 43946.

62.

Sawada, S., et al., A lineage-specific transcriptional silencer regulates CD4 gene
expression during T lymphocyte development. Cell, 1994. 77(6): p. 917-29.

63.

Siu, G., et al., A transcriptional silencer controls the developmental expression of the
CD4 gene. EMBO J, 1994. 13(15): p. 3570-9.

64.

Schoenherr, C.J. and D.J. Anderson, The neuron-restrictive silencer factor (NRSF): a
coordinate repressor of multiple neuron-specific genes. Science, 1995. 267(5202): p.
1360-3.

65.

Studer, M., et al., Role of a conserved retinoic acid response element in rhombomere
restriction of Hoxb-1. Science, 1994. 265(5179): p. 1728-32.

66.

Weintraub, S.J., et al., Mechanism of active transcriptional repression by the
retinoblastoma protein. Nature, 1995. 375(6534): p. 812-5.

67.

Gray, S. and M. Levine, Transcriptional repression in development. Curr Opin Cell Biol,
1996. 8(3): p. 358-64.
36

68.

Gray, S., P. Szymanski, and M. Levine, Short-range repression permits multiple
enhancers to function autonomously within a complex promoter. Genes Dev, 1994.
8(15): p. 1829-38.

69.

Gray, S. and M. Levine, Short-range transcriptional repressors mediate both quenching
and direct repression within complex loci in Drosophila. Genes Dev, 1996. 10(6): p. 70010.

70.

Davis, R.L. and D.L. Turner, Vertebrate hairy and Enhancer of split related proteins:
transcriptional repressors regulating cellular differentiation and embryonic patterning.
Oncogene, 2001. 20(58): p. 8342-57.

71.

Van Doren, M., et al., Negative regulation of proneural gene activity: hairy is a direct
transcriptional repressor of achaete. Genes Dev, 1994. 8(22): p. 2729-42.

72.

Ohsako, S., et al., Hairy function as a DNA-binding helix-loop-helix repressor of
Drosophila sensory organ formation. Genes Dev, 1994. 8(22): p. 2743-55.

73.

Nibu, Y., H. Zhang, and M. Levine, Local action of long-range repressors in the
Drosophila embryo. EMBO J, 2001. 20(9): p. 2246-53.

74.

Paroush, Z., et al., Groucho is required for Drosophila neurogenesis, segmentation, and
sex determination and interacts directly with hairy-related bHLH proteins. Cell, 1994.
79(5): p. 805-15.

75.

Fisher, A.L., S. Ohsako, and M. Caudy, The WRPW motif of the hairy-related basic helixloop-helix repressor proteins acts as a 4-amino-acid transcription repression and
protein-protein interaction domain. Mol Cell Biol, 1996. 16(6): p. 2670-7.

76.

Rosenberg, M.I. and S.M. Parkhurst, Drosophila Sir2 is required for heterochromatic
silencing and by euchromatic Hairy/E(Spl) bHLH repressors in segmentation and sex
determination. Cell, 2002. 109(4): p. 447-58.

77.

Poortinga, G., M. Watanabe, and S.M. Parkhurst, Drosophila CtBP: a Hairy-interacting
protein required for embryonic segmentation and hairy-mediated transcriptional
repression. EMBO J, 1998. 17(7): p. 2067-78.

78.

Bianchi-Frias, D., et al., Hairy transcriptional repression targets and cofactor recruitment
in Drosophila. PLoS Biol, 2004. 2(7): p. E178.

79.

Zhang, H. and M. Levine, Groucho and dCtBP mediate separate pathways of
transcriptional repression in the Drosophila embryo. Proc Natl Acad Sci U S A, 1999.
96(2): p. 535-40.
37

80.

Ish-Horowicz, D. and S.M. Pinchin, Pattern abnormalities induced by ectopic expression
of the Drosophila gene hairy are associated with repression of ftz transcription. Cell,
1987. 51(3): p. 405-15.

81.

Yu, Y. and L. Pick, Non-periodic cues generate seven ftz stripes in the Drosophila
embryo. Mech Dev, 1995. 50(2-3): p. 163-75.

82.

Zhou, D., et al., Mechanisms underlying hypoxia tolerance in Drosophila melanogaster:
hairy as a metabolic switch. PLoS Genet, 2008. 4(10): p. e1000221.

83.

Oro, A.E., et al., The Drosophila gene knirps-related is a member of the steroid-receptor
gene superfamily. Nature, 1988. 336(6198): p. 493-6.

84.

Nibu, Y., H. Zhang, and M. Levine, Interaction of short-range repressors with Drosophila
CtBP in the embryo. Science, 1998. 280(5360): p. 101-4.

85.

Kosman, D. and S. Small, Concentration-dependent patterning by an ectopic expression
domain of the Drosophila gap gene knirps. Development, 1997. 124(7): p. 1343-54.

86.

Clyde, D.E., et al., A self-organizing system of repressor gradients establishes segmental
complexity in Drosophila. Nature, 2003. 426(6968): p. 849-53.

87.

Lunde, K., et al., The knirps and knirps-related genes organize development of the
second wing vein in Drosophila. Development, 1998. 125(21): p. 4145-54.

88.

Chen, C.K., et al., The transcription factors KNIRPS and KNIRPS RELATED control cell
migration and branch morphogenesis during Drosophila tracheal development.
Development, 1998. 125(24): p. 4959-68.

89.

Struffi, P., et al., Quantitative contributions of CtBP-dependent and -independent
repression activities of Knirps. Development, 2004. 131(10): p. 2419-29.

90.

Struffi, P. and D.N. Arnosti, Functional interaction between the Drosophila knirps short
range transcriptional repressor and RPD3 histone deacetylase. J Biol Chem, 2005.
280(49): p. 40757-65.

91.

Courey, A.J. and S. Jia, Transcriptional repression: the long and the short of it. Genes
Dev, 2001. 15(21): p. 2786-96.

92.

Payankaulam, S. and D.N. Arnosti, Groucho corepressor functions as a cofactor for the
Knirps short-range transcriptional repressor. Proc Natl Acad Sci U S A, 2009. 106(41): p.
17314-9.

38

93.

Chen, G. and A.J. Courey, Groucho/TLE family proteins and transcriptional repression.
Gene, 2000. 249(1-2): p. 1-16.

94.

Hartley, D.A., A. Preiss, and S. Artavanis-Tsakonas, A deduced gene product from the
Drosophila neurogenic locus, enhancer of split, shows homology to mammalian Gprotein beta subunit. Cell, 1988. 55(5): p. 785-95.

95.

Dehni, G., et al., TLE expression correlates with mouse embryonic segmentation,
neurogenesis, and epithelial determination. Mech Dev, 1995. 53(3): p. 369-81.

96.

Flores-Saaib, R.D., S. Jia, and A.J. Courey, Activation and repression by the C-terminal
domain of Dorsal. Development, 2001. 128(10): p. 1869-79.

97.

Jimenez, G., Z. Paroush, and D. Ish-Horowicz, Groucho acts as a corepressor for a subset
of negative regulators, including Hairy and Engrailed. Genes Dev, 1997. 11(22): p. 307282.

98.

Aronson, B.D., et al., Groucho-dependent and -independent repression activities of Runt
domain proteins. Mol Cell Biol, 1997. 17(9): p. 5581-7.

99.

Hasson, P., et al., Brinker requires two corepressors for maximal and versatile repression
in Dpp signalling. EMBO J, 2001. 20(20): p. 5725-36.

100.

Chen, G., et al., A functional interaction between the histone deacetylase Rpd3 and the
corepressor groucho in Drosophila development. Genes Dev, 1999. 13(17): p. 2218-30.

101.

Flores-Saaib, R.D. and A.J. Courey, Analysis of Groucho-histone interactions suggests
mechanistic similarities between Groucho- and Tup1-mediated repression. Nucleic Acids
Res, 2000. 28(21): p. 4189-96.

102.

Martinez, C.A. and D.N. Arnosti, Spreading of a corepressor linked to action of longrange repressor hairy. Mol Cell Biol, 2008. 28(8): p. 2792-802.

103.

Andrioli, L.P., et al., Groucho-dependent repression by sloppy-paired 1 differentially
positions anterior pair-rule stripes in the Drosophila embryo. Dev Biol, 2004. 276(2): p.
541-51.

104.

Schaeper, U., et al., Interaction between a cellular protein that binds to the C-terminal
region of adenovirus E1A (CtBP) and a novel cellular protein is disrupted by E1A through
a conserved PLDLS motif. J Biol Chem, 1998. 273(15): p. 8549-52.

105.

Shi, Y., et al., Coordinated histone modifications mediated by a CtBP co-repressor
complex. Nature, 2003. 422(6933): p. 735-8.
39

106.

Shi, Y., et al., Histone demethylation mediated by the nuclear amine oxidase homolog
LSD1. Cell, 2004. 119(7): p. 941-53.

107.

Zeitlinger, J., et al., Whole-genome ChIP-chip analysis of Dorsal, Twist, and Snail suggests
integration of diverse patterning processes in the Drosophila embryo. Genes Dev, 2007.
21(4): p. 385-90.

108.

Schwartz, Y.B., et al., Genome-wide analysis of Polycomb targets in Drosophila
melanogaster. Nat Genet, 2006. 38(6): p. 700-5.

109.

Papp, B. and J. Muller, Histone trimethylation and the maintenance of transcriptional
ON and OFF states by trxG and PcG proteins. Genes Dev, 2006. 20(15): p. 2041-54.

110.

Strahl-Bolsinger, S., et al., SIR2 and SIR4 interactions differ in core and extended
telomeric heterochromatin in yeast. Genes Dev, 1997. 11(1): p. 83-93.

111.

Schwartz, Y.B. and V. Pirrotta, Polycomb silencing mechanisms and the management of
genomic programmes. Nat Rev Genet, 2007. 8(1): p. 9-22.

112.

Comet, I., et al., PRE-mediated bypass of two Su(Hw) insulators targets PcG proteins to a
downstream promoter. Dev Cell, 2006. 11(1): p. 117-24.

40

Chapter II
Long- and Short-Range Transcriptional Repressors induce Distinct Chromatin States
on Repressed Genes3
Abstract
Transcriptional repression is essential for establishing precise patterns of gene
expression during development. Repressors governing early Drosophila segmentation can be
classified into short- or long-range factors based on their ranges of action, acting either locally
to quench adjacent activators, or broadly to silence an entire locus. Paradoxically, these
repressors recruit common corepressors, Groucho and dCtBP, despite their different ranges of
repression. To reveal the mechanisms underlying these two distinct modes of repression, we
performed chromatin analysis, using the prototypical long-range repressor Hairy and the shortrange repressor Knirps. Chromatin immunoprecipitation and micrococcal nuclease mapping
studies reveal that Knirps causes local changes of histone density and acetylation, and the
inhibition of activator recruitment, without affecting the recruitment of basal transcriptional
machinery. In contrast, Hairy induces wide-spread histone deacetylation and inhibits the
recruitment of basal machinery without inducing chromatin compaction. Our study provides
unprecedented mechanistic information into short- and long-range repression at the chromatin
level, and suggests that the transcriptional corepressors can be differentially deployed to
mediate chromatin changes in a context-dependent manner.
Key words
chromatin/ Hairy/Knirps/repression/transcription
3

Chapter II is presented in the form of a manuscript to be submitted for publication as: Li, L.M., and Arnosti, D.N
Long- and short-range transcription repressors induce distinct chromatin states on repressed genes.

41

Introduction
Eukaryotic development involves the establishment of precise temporal and spatial
patterns of gene expression. The segmentation process in Drosophila is a paradigm for
developmental regulation, and is governed by a cascade of transcription factors, including
maternal, gap, pair-rule, and segment-polarity genes [1]. Among these factors, numerous
transcriptional repressors play central roles, setting the boundaries for gene expression [2].
A unifying principle emerging from molecular studies of these disparate proteins relates
to their ranges of action [3]. Long-range repressors such as Hairy and Dorsal cause dominant
inactivation of activators over a range of more than 1000 bp, whereas short-range repressors
such as Knirps, Krϋppel, and Snail interfere with locally bound activators, leaving untouched the
function of more distal enhancers [3, 4]. Short-range repression underlies enhancer autonomy,
in which different enhancers work independently of one another within a complex cisregulatory region [5]. It is not known if the special attributes of long-range repressors are
uniquely harnessed to specific gene structures, however.
The different activities of long- and short-range repressors have been proposed to
reflect the recruitment of distinct corepressors, namely Groucho and the C-terminal binding
protein (CtBP). Groucho, a conserved WD-40 motif protein related to the human TLE and the
mouse Grg protein, was initially identified as a corepressor of Hairy and later other repressors
such as Deadpan and Engrailed [6, 7]. Groucho is capable of oligomerization, and interacts with
histones and the histone deacetylase Rpd3 [8-10]. Based on these observations, it has been
proposed that long-range repression involves spreading of the Groucho corepressor along the
chromatin [11]. Indeed, Groucho spreading has been observed on a Hairy target reporter gene
42

[12]. CtBP, originally identified as an E1a-interacting protein in vertebrate cells, was found to
bind the Knirps, Krϋppel and Snail short-range repressors in Drosophila and mediate their
activity [13]. CtBP can interact with histone deacetylases and demethylases, suggesting that it
acts via chromatin modifications [14, 15].
While extensive data points to Groucho’s and CtBP’s involvement in numerous
repression systems, recent studies have shown that the distinction in repression range is not a
simple function of recruiting different cofactors. For example, the short-range repressor Knirps
utilizes both CtBP and Groucho to repress even skipped [16]. This suggests that different types
of repressors can recruit the same cofactors to effect short- or long-range repression.
Mechanistic insights into how these processes differ have been lacking: long-range repression
has been associated with Groucho cofactor spreading, but almost no mechanistic information is
available for short-range repressors, which have been alternately suggested to act via
competition for the binding sites with the activators, interfering with the basal transcriptional
machinery, or quenching the interaction between activators and the basal transcriptional
machinery [4, 17].
In this study, we investigated chromatin mechanisms underlying the functional
differences between long- and short-range repression. Using the long-range repressor Hairy
and short-range repressor Knirps as models, we provide evidence that the two classes of
repressors have distinct effects on chromatin structure, as well as activator and basal
transcriptional machinery recruitment, despite the common corepressors they recruit. These
results suggest that these transcriptional corepressors cause context-specific (in this case,

43

repressor-specific) chromatin changes, and also provide insights into how developmental
transcription factors regulate RNA polymerase II.

Results
Local and global repression
Hairy directly represses fushi tarazu (ftz), a secondary pair-rule gene expressed in the
blastoderm embryo in a seven-stripe pattern [18]. ftz is regulated by both regionally acting gap
genes, as well as the hairy pair-rule gene [19]. Two dense clusters of peaks for Caudal,
Hunchback, Knirps, Giant, Huckebein, Kruppel and Tailless transcription factors mapping on the
5’ regulatory region of the ftz locus represent presumptive regulatory elements, in addition to
the promoter-proximal Zebra, and the stripe 1+5 enhancer located 3’ of the gene *20-22](Figure
2-1A). Hairy is found to bind in vivo to several of these regions. In the blastoderm embryo, the
ftz gene is active in some nuclei and repressed in others by Hairy. In order to obtain a
population of entirely repressed nuclei for chromatin studies, we overexpressed Hairy protein
in embryos using a heat shock driver, which results in complete repression of ftz (Figure 2-1A).
Hairy repression of ftz relies on the Groucho corepressor, as a mutant version of Hairy that does
not bind to Groucho failed to repress ftz (Figure 2-1E, F).
Interestingly, a titration of heat shock induction times resulted in a non-uniform,
progressive loss of specific ftz stripes, with stripe 4 being the most sensitive and stripe 1+5 the
least (Figure 2-1C; differential repression quantified in Supplemental Table 2-1). This result
points to the intriguing possibility that Hairy can act locally, at least very transiently, although
the end result of Hairy repression is complete silencing of all enhancer elements. This
44

asynchronous repression of the ftz locus also suggests that Hairy-mediated long-range
repression does not act solely by direct targeting the basal promoter, a previous model for this
class of repressor, because this mechanism would cause uniform inhibition of stripe elements
[17].
Similar to ftz, the pair-rule gene even-skipped (eve) is also expressed in a seven-stripe
pattern and is regulated by multiple modular enhancers. eve is a well-characterized target of
the short-range repressor Knirps, which sets posterior boundaries of eve stripe 3 and 4 and
anterior borders of eve stripe 6 and 7 [23, 24]. After substantial overexpression of Knirps (20
minute heat shock induction), the repressor is able to repress all of the eve stripe enhancers
except for stripe 5 enhancer (Figure 2-1B). When the induction is titrated, Knirps represses
individual enhancers in a step-wise manner, with the most sensitive enhancers down-regulated
earliest, at a low dose of Knirps [25]. Together, these experiments indicate that Hairy can
initially act locally, but ultimately in a globally dominant fashion, while Knirps acts in a restricted
manner (Figure 2-1D).

45

Figure 2-1. Repression of ftz by Hairy and eve by Knirps. (A) Overexpression of Hairy longrange repressor by 20 minute heat shock represses ftz expression. Hairy protein (top panel) and
ftz mRNA (lower panel) expression pattern in wild-type and hs-hairy transgenic embryos after
20 minute heat shock are shown. (B) Substantial heat shock induction (20 minute) of Knirps
represses all eve enhancers except the stripe 5 enhancer. knirps (top panel) and eve (lower
panel) mRNA expression pattern in wild-type and hs-knirps transgenic embryos after 20 minute
heat shock are shown.

46

Figure 2-1 continued

Figure 2-1. Repression of ftz by Hairy and eve by Knirps. (C) Overexpression of the long-range
Hairy repressor by titrated heat shock results in progressive repression of ftz enhancers shown
by in situ hybridization. The order of repression is stripe 4, stripe 2+7, stripe 3+6, stripe 1+5
(from top to bottom). Details of the heat shock titration experiment are listed in Supplemental
Table 2-1. (D) Heat shock titration of Knirps represses eve in a step-wise manner, with stripe
3+7 enhancer the most sensitive and stripe 1 enhancer the least sensitive (from top to bottom).
Details of the heat shock titration experiments have been published and can be found in Struffi
et al. 2004.

47

Figure 2-1 continued

F

Figure 2-1. Repression of ftz by Hairy and eve by Knirps. (E) Groucho binding deficient Hairy
fails to repress ftz. Hairy protein (top panel) and ftz mRNA (lower panel) expression pattern in
wild-type , hs-hairy or hs-hairy (wrpw-) (Groucho binding deficiency version of Hairy)
transgenic embryos after 20 minute heat shock are shown. (F) ftz mRNA levels in wild-type, hshairy or hs-hairy (wrpw-) embryos are quantified by real-time PCR. ftz mRNA level is lower only
in embryos containing the hs-hairy transgene. Heat shock alone, in the absence of the hs-hairy
or hs-knirps transgenes, had no effect on ftz and eve.
48

Hairy and Knirps differentially affect chromatin structure
To compare the effects of repression by Hairy and Knirps, we studied chromatin changes
associated with repression of ftz and eve using chromatin immunoprecipitation. We observed
little or no significant change of histone H3 occupancy at most regions sampled throughout the
ftz locus after Hairy overexpression with one possible exception at -5 kbp (Figure 2-2A;
Supplemental Table 2-2). In contrast, Knirps repression of eve resulted in increased histone H3
density in two of the three regions corresponding to the Knirps-sensitive enhancers, namely
stripe 4+6 and stripe 2 (Figure 2-2B; Supplemental Table 2-2). Little change was noted in the
promoter region, transcribed region, or the stripe 1 and 5 enhancers, which are not readily
repressed by Knirps. An apparent increase in histone H3 density on the repressed stripe 3+7
enhancer, although of low statistical significance, correlates with other alterations common to
repressed enhancers, noted below (Figure 2-2B; Supplemental Table 2-2).
To provide a more detailed picture of chromatin structure, we adapted a micrococcal
nuclease (MNase) mapping protocol used in yeast and cultured cells for Drosophila embryos
(Sekinger et al., 2005; Petesch and Lis, 2008; Li and Arnosti, 2010). MNase mapping showed
that Hairy repression had little effect on chromatin accessibility throughout the ftz locus (Figure
2-2C; Supplemental Table 2-2), whereas Knirps induced a significant increase in MNase
insensitivity specifically at eve stripe 3+7, 2, 4+6 enhancers, and a minor increase in stripe 1
protection (Figure 2-2D; Supplemental Table 2-2). The promoter and the eve stripe 5 enhancer
were not much changed, mirroring the patterns noted for overall histone H3 occupancy. The
changes noted for the eve locus appear to be specific; as Knirps did not induce any change of a
non-targeted site on the third chromosome. Hairy also had no effect at this locus (Figure 2-2C).

49

Figure 2-2. Hairy and Knirps-mediated transcriptional repression result in differential changes
in histone H3 occupancy and micrococcal nuclease sensitivity. (A, B) Histone H3 occupancy
was measured by chromatin immunoprecipitation in regions of ftz and eve before (solid line)
and after repression (dashed line). Relative immunoprecipitation signal was obtained by
normalization to the actin5C promoter region, which was not affected by Hairy or Knirps
repression.

50

Figure 2-2 continued

Figure 2-2. Hairy and Knirps-mediated transcriptional repression result in differential changes
in histone H3 occupancy and micrococcal nuclease sensitivity. (A, B) Histone H3 occupancy
was measured by chromatin immunoprecipitation in regions of ftz and eve before (solid line)
and after repression (dashed line). Relative immunoprecipitation signal was obtained by
normalization to the actin5C promoter region, which was not affected by Hairy or Knirps
repression.

51

Figure 2-2 continued

Figure 2-2. Hairy and Knirps-mediated transcriptional repression result in differential changes
in histone H3 occupancy and micrococcal nuclease sensitivity. (C, D) MNase sensitivity of the
ftz and eve loci. Hairy induced repression does not significantly change the overall MNase
protection pattern in any of the regions tested in the ftz locus (solid line before, dashed line
after repression). Knirps repression is associated with increased resistance to MNase digestion
at the eve 3+7, 2 and 4+6 enhancers. Little or no change is observed at the promoter, stripe 1
enhancer and stripe 5 enhancer. The specificity of MNase digestion was also shown by
digestion pattern of a 450 bp intergenic region on the 3 rd chromosome as shown in (C), for
embryos with no repressor overexpression (solid line), Hairy overexpression (long dashed line),
and Knirps overexpression (short dashed line). All results in A-D and in following figures
represent at least three biological replicates; error bars show standard errors. Areas under the
lower plots are shadowed for the clarity of presentation. For this and later figures, the
statistical significance of the differences between each pair of points is shown in Supplemental
Table 1; for histone H3 occupancy, the p values are < 0.05 for eve stripes 2 and 4+6; for ftz, no
points reached this level of significance. Nonrepressed embryos (lacking the heat shockinducible repressor gene) were heat shocked to control for any effect of the heat induction
itself. In this and subsequent figures, the Y-axis shows the amount of immunoprecipitate as
percentage of the input.

52

Figure 2-2 continued

Figure 2-2. Hairy and Knirps-mediated transcriptional repression result in differential changes
in histone H3 occupancy and micrococcal nuclease sensitivity. (C, D) MNase sensitivity of the
ftz and eve loci. Hairy induced repression does not significantly change the overall MNase
protection pattern in any of the regions tested in the ftz locus (solid line before, dashed line
after repression). Knirps repression is associated with increased resistance to MNase digestion
at the eve 3+7, 2 and 4+6 enhancers. Little or no change is observed at the promoter, stripe 1
enhancer and stripe 5 enhancer. The specificity of MNase digestion was also shown by
digestion pattern of a 450 bp intergenic region on the 3rd chromosome as shown in (C), for
embryos with no repressor overexpression (solid line), Hairy overexpression (long dashed line),
and Knirps overexpression (short dashed line). All results in A-D and in following figures
represent at least three biological replicates; error bars show standard errors. Areas under the
lower plots are shadowed for the clarity of presentation. For this and later figures, the
statistical significance of the differences between each pair of points is shown in Supplemental
Table 1; for histone H3 occupancy, the p values are < 0.05 for eve stripes 2 and 4+6; for ftz, no
points reached this level of significance. Nonrepressed embryos (lacking the heat shockinducible repressor gene) were heat shocked to control for any effect of the heat induction
itself. In this and subsequent figures, the Y-axis shows the amount of immunoprecipitate as
percentage of the input.

53

The consistent results from overall histone H3 density and MNase mapping suggest that
Hairy-mediated long-range repression does not involve compaction of chromatin on the ftz
locus. Previous studies showed that repression of slp1 locus by Runt, another Grouchodependent repressor, does not involve changes in H3 density, thus it is possible that Grouchodependent long-range repression does not involve chromatin compaction in general [26]. In
contrast, these findings indicate that repression by Knirps is associated with an increase in the
histone density of targeted enhancer regions, which may result either from Knirps recruitment
of factors that mediate chromatin condensation, or the blocking of proteins responsible for
loosening of chromatin.

Hairy mediates wide-spread histone deacetylation, whereas Knirps only causes local histone
deacetylation
Histone acetylation is dynamically regulated on actively transcribed genes in eukaryotes,
with histone acetylation generally correlated with active loci [27]. The histone deacetylase Rpd3
is a component of both Hairy and Knirps corepressor complexes [9, 28]. We assayed histone
acetylation levels across the eve and ftz genes before and after repression. Hairy repression
resulted in wide-spread histone H4 deacetylation throughout the ftz locus (Figure 2-3A;
Supplemental Table 2-2). Using anti H3-acetylation antibodies, similar widespread H3
deacetylation was also noted (data not shown). This distributed effect on the ftz locus
correlates with the prior observations that Hairy-mediated long-range repression might involve
a Groucho-mediated “spreading” mechanism [12]. By this means, Rpd3 may be delivered to
extensive areas of a gene.
54

Figure 2-3. Hairy induces global, while Knirps induces local changes in histone acetylation
levels. For both repressors, H3K27 tri-methylation are little changed overall. (A, C) H4K5,8,12,16 acetylation was assayed by chromatin immunoprecipitation on the ftz and eve loci
before (solid line) and after (dashed line) repression by Hairy and Knirps respectively. Significant
reduction in H4 acetylation was observed at all loci tested in ftz, while reductions in H4
acetylation on eve was limited to repressed enhancers. (B) H4 deacetylation is localized around
Hairy binding sites after brief 5 minute heatshock induction.
55

Figure 2-3 continued

Figure 2-3. Hairy induces global, while Knirps induces local changes in histone acetylation
levels. For both repressors, H3K27 tri-methylation are little changed overall. (A, C) H4K5,8,12,16 acetylation was assayed by chromatin immunoprecipitation on the ftz and eve loci
before (solid line) and after (dashed line) repression by Hairy and Knirps respectively. Significant
reduction in H4 acetylation was observed at all loci tested in ftz, while reductions in H4
acetylation on eve was limited to repressed enhancers. (B) H4 deacetylation is localized around
Hairy binding sites after brief 5 minute heatshock induction.

56

Figure 2-3 continued

D

Figure 2-3. Hairy induces global, while Knirps induces local changes in histone acetylation
levels. For both repressors, H3K27 tri-methylation are little changed overall. (D, E) Trimethylated H3 lysine 27 levels are around 5-10 times higher on eve and ftz than tubulin,
actin5C and hsp70. In a repressed state, the levels are only modestly affected in portions of the
ftz gene, and no significant reduction was observed for the eve gene.

57

Figure 2-3 continued

Figure 2-3. Hairy induces global, while Knirps induces local changes in histone acetylation
levels. For both repressors, H3K27 tri-methylation are little changed overall. (D, E) Trimethylated H3 lysine 27 levels are around 5-10 times higher on eve and ftz than tubulin,
actin5C and hsp70. In a repressed state, the levels are only modestly affected in portions of the
ftz gene, and no significant reduction was observed for the eve gene.

58

To test whether a spreading of histone deacetylation might correlate with the
successive inhibition of ftz enhancers we noted in Figure 2-1, we investigated histone
acetylation levels across ftz after a short 5 minute heat shock followed by immediate fixing. In
this setting, deacetylation was mostly concentrated around Hairy binding sites, consistent with
a spreading action of this repressor (Figure 2-3B).
A different picture emerged from studies of Knirps acting on eve. Here, repression led to
selective decreases in H3 and H4 acetylation levels, concentrated over the eve stripe 4+6 and
stripe 2 enhancers, with lesser decreases noted at stripe 3+7 and stripe 1 enhancers (Figure 23C; Supplemental Table 2-2). A local change in acetylation was also noted near the
transcriptional initiation site, but not immediately 5’ and 3’ of this area. The reductions in
histone acetylation levels seen on both eve and ftz are consistent with Hairy and Knirps
recruiting deacetylases to their target genes. However, it is striking that the broad
deacetylation mediated by Hairy on ftz is not associated with changes in histone density or
resistance to nuclease accessibility, whereas increased histone density is associated with Knirps
repression on eve. It is possible that in addition to inducing deacetylation, Knirps triggers
additional histone modifications or interacts with nucleosome remodeling complexes to
increase the density of histones at the enhancers.

Repression does not involve changes in H3 lysine 27 trimethylation
The H3 lysine 27 trimethylation mark is established by the Enhancer of Zeste
methyltransferase, and is an essential repression mark for establishing and maintaining
polycomb-mediated long-term developmental regulation of Hox genes [29, 30]. Therefore, we
59

investigated whether repression mediated by Hairy and Knirps in the blastoderm embryo
involves modulation of H3K27 trimethylation level. Before induction of the repressors, we
detected robust levels of H3K27 trimethylation on the ftz and eve loci, at least 5-10 times
higher than levels on housekeeping genes such as tubulin and actin5C (Figure 2-3E). However,
only modest changes were observed on the ftz locus in response to Hairy. Similarly, the levels
of H3 lysine 27 tri-methylation on the eve locus were not significantly affected by Knirps (Figure
2-3D, 2-3E; Supplemental Table 2-2). Although, a change in H3 lysine 27 tri-methylation is not
required for repression by these early developmental repressors, the enriched levels of H3K27
trimethylation observed on the eve and ftz genes do suggest the possible involvement of
Polycomb-group proteins in the regulation of these pair-rule genes; this chromatin modification
might prime these loci for later, permanent inactivation.

Differential effects of Hairy and Knirps on activator recruitment
Our previous studies indicated that Hairy can effectively repress a reporter gene without
displacing the activators [12]. We sought to test whether this was the case on an endogenous
gene, ftz, by examining occupancy by Caudal, a transcription factor that also activates eve.
Caudal activates the posterior stripes of both ftz and eve, and we found that Caudal binds the
ftz 5’ regulatory region and the promoter-proximal Zebra element, consistent with a recent
global study [31-33]. Repression of the locus by Hairy did not affect Caudal binding pattern
(Figure 2-4A; Supplemental Table 2-2), similar to the results obtained with a Hairy regulated
reporter gene [12]. In contrast, Knirps repression decreased Caudal occupancy specifically at
eve 3+7 and 4+6
60

Figure 2-4. Activator occupancy of ftz and eve before (solid line) and after (dashed line)
repression. (A) Caudal protein occupancy of ftz measured by chromatin immunoprecipitation.
Caudal activator levels do not decrease on the ftz locus in reponse to repression by Hairy.

61

Figure 2-4 continued

Figure 2-4. Activator occupancy of ftz and eve before (solid line) and after (dashed line)
repression. (B) Caudal occupancy decreases at eve stripe 3+7 and 4+6 enhancers after
repression by Knirps. (C) Bicoid occupancy of eve is unchanged after Knirps repression. No
Bicoid binding was detected on the ftz locus.
62

in occupancy down to near-baseline levels. This decrease is not an effect of global decrease of
Caudal occupancy, as the Caudal binding peak at the eve promoter is not affected. A similar
decrease in Caudal occupancy was also observed on a hunchback enhancer after repression by
Knirps (data not shown). Interestingly, Bicoid occupancy of the eve stripe 2 and stripe 1
enhancers was not altered by Knirps, although these enhancers were repressed (Figure 2-4C). It
is possible that Knirps utilizes multiple mechanisms for repression: on the eve stripe 3+7
enhancer, the high density of Knirps binding sites may permit competitive binding with
activators such as Caudal [24, 34]. Interestingly, this enhancer exhibited a relatively modest
decrease in histone acetylation upon repression, thus acetylation levels may not directly reflect
the abundance of recruited repressors. In the case of the stripe 2 enhancer, which exhibited
more significant changes in H4 acetylation, Knirps might mainly repress by altering chromatin
structure via deacetylation, without reducing Bicoid activator binding.
Distinct effects on RNA polymerase II by long- and short-range repressors
New insights have suggested many developmental genes, including those regulated by
short-range repressors such as Snail, feature RNA polymerase paused in the promoter region
even in their inactive state, suggesting post-recruitment levels of regulation [35]. We analyzed
components of the core machinery before and after repression by Hairy and Knirps. Upon Hairy
repression, a marked decrease of RNA polymerase II occupancy was observed at the ftz locus.
The same trend is observed for the pre-initiating, initiating, and elongation forms of RNA
polymerase II (Fig 2-5A-C). These results suggest that Hairy directly or

63

Figure 2-5. Pol II and TATA binding protein occupancy at ftz and eve promoter and transcribed
regions. (A-C) ChIPs were performed using antibodies against unphosphorylated, initiating (Ser5P)
and elongating (Ser2P) Pol II on ftz before (black bars) and after repression (gray bars). Strong
decreases in all forms of Pol II were noted. (Asterisk denotes where p<0.05).

64

Figure 2-5 continued

Figure 2-5. Pol II and TATA binding protein occupancy at ftz and eve promoter and
transcribed regions. (D-F) A similar analysis of eve did not show significant changes in Pol II
occupancy after Knirps repression.

65

Figure 2-5 continued

Figure 2-5. Pol II and TATA binding protein occupancy at ftz and eve promoter and transcribed
regions. (G, H) TBP occupancy of the ftz promoter decreased after repression by Hairy, but no
significant changes were noted for TBP occupancy of the eve gene after Knirps repression.

66

indirectly blocks recruitment of RNA polymerase II. Similar decreases were noted with levels of
TATA-box binding protein (TBP) at the promoter (Fig. 2-5G).
In contrast, induction of Knirps does not change Pol II occupancy at the eve transcription
unit, even under condition where most enhancers were repressed (Fig. 2-5D-F). (Under
conditions tested here, over three-quarters of the embryos have shut down expression of all
but stripe 1 and/or 5). Similarly, TBP occupancy remains at the comparable level before and
after Knirps repression (Fig. 2-5H). The constant level of RNA polymerase on the eve
transcription unit was a surprise, in light of the sharp reduction in mRNA production, as
measured by in situ hybridization. However, there is precedence for this effect: Runt repression
of Slp1 appears to act through elongation control, which causes no change of the concentration
of RNA polymerase II on slp1 [26]. Knirps may produce a similar effect by inducing a slower
transit rate of RNA polymerase II on the repressed eve locus. Similar observations have been
made at the hsp70 gene upon depletion of elongation factors such as Spt6 or Paf1 [36, 37].
Discussion
Our study provides molecular insights into general mechanisms mediated by two major
classes of repressors vital for early Drosophila development. Long-range repression mediated
by Hairy is dependent upon Groucho, whose extensive interaction with large tracts of
chromatin have led to a “spreading” model of repression *11+. Widespread histone
deacetylation is consistent with a picture of a globally repressed locus, but interestingly, rapid
kinetic “snapshots” of partially repressed states indicate that chromatin modification is
nucleated around Hairy binding sites, with initial inhibition of separate cis-elements. The

67

progressive repression of ftz by Hairy suggests the Hairy-mediated repression is not simply
directly targeting the basal promoter, because in this case all ftz enhancers would be repressed
at the same time. Pol II and TBP occupancy is inhibited when the whole locus is repressed,
which may either reflect a complete loss of cis-regulatory signaling with the silencing of all the
enhancers or Groucho-mediated chromatin modification may in the end also target the basal
promoter blocking access by the preinitiation complex.
The global mechanism of Hairy-mediated repression resembles the deacetylated
chromatin domain mediated by Sir2 in both the telomeres and mating type loci in yeast,
although the latter system farms a stable state that is propagated for generations [38]. Hairy in
contrast appears to generate a relatively labile state that is reset as soon as the nucleating
transcription factor is turned over (Martinez & Arnosti, 2008 and data not shown). This
transient long-range repression might fit into the role of Hairy during early development, which
allows for effective yet dynamic regulation of its targets. Polycomb protein-mediated longrange repression also involves similar wide-spread chromatin modifications, including H3K27,
K9 and K20 methylation chromatin marks [39]. We did note an enrichment of H3K27
methylation of ftz (as well as eve), but Hairy-mediated repression did not involve dynamic
modulation of this chromatin mark.
There is extensive information about cis-regulatory requirements for short-range
repressors such as Knirps, but we have had limited insight into the molecular machanisms of
these transcriptional regulators, other than indications that chromatin deacetylation should
play a role, above and beyond the simple competitive binding models first proposed for those
repressors [4, 28]. Our results strongly point to important local changes occurring on cis68

regulatory elements that coincide with silencing of gene function, including increases in histone
density, local resistance of chromatin to nuclease attack, deacetylation and loss of activator
binding. This picture strongly supports a local ‘quenching’ model of short-range repression, and
is at odds with the alternative ‘hitchhiking’ model that posited long-range, albeit transient,
interactions with the basal transcription machinery [4, 17]. At the same time, the lack of change
in Pol II occupancy indicates that these local modifications can influence polymerase elongation,
a feature that needs further investigation. Other transcriptional repressors have also been
linked to repression of Pol II elongation [26, 37].
The local increase in histone density may indicate an inhibition of nucleosome
disassembly or recruitment of cofactors that deposit nucleosomes at the enhancer regions by
Knirps. These mechanistic insights may provide key discoveries to facilitate mathematical
modeling of cis-regulatory elements controlled by these short-range repressors. Modeling can
in turn provide a basis for understanding the evolution of the developmental enhancers [40].
With the recent demonstration that transcriptional factors considered to be short- and
long-range repressors utilize shared cofactors, namely CtBP and Groucho, there has been a
question of whether long-range repression is actually functionally distinct from short-range
repression [16]. Our study provides evidence that the chromatin states associated with longand short-range repressors are distinct in several ways. We do not yet know whether the
effects seen on ftz are observed for all Hairy targets, although the similarity of changes
observed with a lacZ reporter subject to Hairy repression suggests they are conserved [12].
Similarly, the reproducibility of Knirps-induced changes at different eve enhancers indicates
that this protein can effect related chromatin changes a cis regulatory modules bound by
69

different activators. Snail, another short-range repressor, also appears to mediate localized
deacetylation and activator displacement, thus this mechanism may be a common feature of
this entire class of proteins (Yutaka Nibu, personal communication). The highly divergent
activities of Knirps and Hairy demonstrated in this study not only underscore the fact that these
proteins do mediate biochemically divergent events, but also raise interesting questions about
how similar cofactors can mediate such distinct effects in a context-dependent manner. It is
possible that the corepressors adopt distinct conformations when recruited by different
repressors, or the corepressor may form distinct complexes with unique activities [41]. In
addition to determining how cis- and trans-acting factors affect repression pathways, these
mechanistic insights will provide important contextual information for interpretation of
genome-wide transcription factor binding and chromatin modifications.

70

Materials and Methods
Plasmid construction
Transgenic flies carrying inducible Hairy genes were generated using pCasper-hs
transformation vector [25].

The genes were created by joining a EcoRI/XbaI fragment

containing a Kozak sequence, initiator ATG, and coding sequence for either wildtype or the
mutant (WRPW/AAAA) Hairy protein (primer sequences are available upon request) amplified
from a pGEX-2T vector containing hairy cDNA [6].
In situ hybridization and antibody staining of Drosophila embryos
Embryos were fixed for in situ hybridization and stained using anti-digoxigenin-UTP-label RNA
probe to ftz or eve as previously described [25].
Embryo Collection
Embryos used for chromatin immunoprecipitation and micrococcal nuclease protection
experiments were 2-3 hour old, exposed to 20 minute heat shock to induce maximal repression,
and allowed no recovery period after heat shock treatment. To control for possible nonspecific
effects of heat shock, wild-type embryos without hs-trangenes were similarly treated to
generate the chromatin profiles of ftz and eve in the unrepressed state (heat shock alone has
no effect on the expression patterns of eve or ftz; Struffi et al., 2004 and data not shown).
Embryos containing either the hs-hairy or hs-knirps transgene were used to generate the “after
repression”chromatin.
Chromatin immunoprecipitation
Heat shocks and chromatin immunoprecipitations were performed as previously
described [12], with the exceptions that embryos were sonicated for 20s (60% duty cycle) and

71

cooled on ice for 30s 15 times; beads were washed twice with the low-salt buffer, twice with
the high-salt buffer, and twice with Tris-EDTA. We used the following antibodies: mouse
monoclonal IgG (10 ul; Upstate), rabbit anti-H3 (1ul; Abcam), rabbit anti-acetyl H4 (1ul;
Upstate); rabbit anti-trimethyl H3K27 (2ul; Abcam); 8WG16 (10ul; Covance; antiunphosphorylated Pol II CTD); H5 (10ul; Covance; anti-Ser2-phosphorylated CTD); H14 (10ul;
Covance; anti-Ser5-phosphorylated CTD); TBP (2ul; Abcam); rabbit anti-mouse IgM (10ul;
Abcam); rabbit anti-Bicoid serum (10 ul; gift from M. Biggin); rabbit anti-Caudal serum (10ul;
gift from M. Biggin).
Micrococcal nuclease mapping in Drosophila embryos
Micrococcal nuclease mapping in Drosophila embryos was performed as previously
described [42].
Quantitative PCR analysis
The samples from ChIP and MNase mapping were analyzed using real-time PCR (Applied
Biosystem 7500). Primer pairs had a Tm in the range of 58-60⁰C, and amplicons ranged from 50
to 150 bp. Primer sequences are listed in Supplemental Table 3. For ChIP samples, a standard
curve was generated by serially diluting input samples to quantify IP samples. For MNase
digests, a ratio was calculated between MNase digested and undigested samples. All values
used were collected from the linear range of amplification.
Acknowledgments
We thank Xiaoyong Li and Mark Biggin for the Bicoid and Caudal antisera; Zachary
Burton and Mark Schroeder for critical reading of the manuscript; Sandhya Payankaulam for

72

valuable advice on Knirps-related experiments and the manuscript; and other Arnosti lab
members for useful discussions. This work is supported by NIH grant GM56976 to D.N.A.

73

APPENDICES

74

Appendix A-1

75

Supplementary Tables for Chapter II

Supplemental Table 2.1 Numbers of embryos within each category after heat shock titration of
Hairy protein are summarized. Staining patterns for embryos in each category can be referred to Figure
1C. For each experiment, total n=100 .

Induction(min) Recovery (min) i
5
0
10
0
20
0
5
30

ii

iii

71
52
6
2

29
21
13
35

76

iv
0
11
4
26

v
0
1
3
15

0
15
74
24

Supplemental Table 2.1 p-values for Student’s t-test comparing before repression and after
repression chromatin immunoprecipitation signals of each data point in Figure 2-2,2-3 and 24 (n ≥ 3).
Hairy repression on ftz

Position
Histone H3 (Figure 22A)

Zebra

+1 1kb 1+5

-7593 -5350 -4813 -4171 3352 -2697 1570 -562

0.68 0.21 0.14 0.52 0.43

0 1000 4121 6624

0.72 0.31 0.81 0.33 0.15 0.84 0.19

Histone H4 acetylation
(Figure 2-3A)
0.034 0.058 0.17 0.064 0.007 0.033 0.06 0.013 0.069 0.031 0.057 0.058
Histone H4 acetylation
(Figure2- 3B)
0.59 0.41 0.009 0.091 0.13 0.0002 0.11 0.019 0.14 0.87 0.007 0.10
Histone H3 K27 me
(Figure 2-3D)
Caudal (Figure 2-4A)

Knirps repression on
eve
Position

0.19 0.36 0.049 0.85 0.15 0.056 0.23 0.11 0.38 0.94 0.18 0.21
0.80

3+7

0.12

2

+1

-3522 -2330 -1281 -583

1kb

0.97 0.67

4+6

0.11

1

5

83 1194 2856 4934 5876 6882 7821

Histone H3 (Figure 22B)

0.20 0.17 0.009 0.70 0.53 0.65 0.70 0.031 0.99 0.74 0.57

Histone H4 acetylation
(Figure 2-3C)

0.39 0.045

Histone H3 K27 me
(Figure 2-3D)

0.63

0.86

0.07 0.34 0.74 0.54

0.60 0.44

Caudal (Figure 2-4B)

0.015

0.38

0.96 0.90 0.96 0.11

0.90 0.95

Bicoid (Figure 2-4C)

0.29

0.96

0.99 0.94 0.97 0.42

0.76 0.56

0.038 0.66 0.16 0.90 0.15 0.15 0.28 0.29 0.16

77

Supplemental Table 2.2: p-values for Student’s t-test comparing before repression and after
repression Micrococcal Nuclease Digested/Undigested values of each data point in Figure 2C
(n ≥ 3).
ftz promoter
Position
p-value

-503
0.29

-404
0.50

-334
0.66

-259
0.86

-113
0.37

-10
0.88

5' regulatory_2
Position
-6592 -6535 -6411 -6279 -6201 -6147
p-value
0.79
0.54
0.35
0.22
0.22
0.27
5' regulatory_1
Position
-7822 -7794 -7593 -7519 -7414 -7228
p-value
0.50
0.49
0.82
0.17
0.52 0.076
5'_regulatory_3
Position
-3997 -3783 -3697 -3490 -3403
p-value
0.27
0.33
0.15
0.59
0.77
ftz_1+5
Position
p-value

3722
0.36

3819
0.18

4007
0.13

4122
0.23

78

4231 4349
0.31 0.082

72
0.61

219
0.89

304
0.91

422
0.048

Supplemental Table 2.3: p-values for Student’s t-test comparing before repression and after
repression Micrococcal Nuclease Digested/Undigested values of each data point in Figure 2D
(n ≥ 3).
eve 2
Position
p-value

-1456
0.005

-1372
0.041

-1334
0.097

-1297
0.074

-1190
0.016

-1119
0.04

eve4+6
Position
p-value

4723
0.06

4768
0.099

4895
5014
0.018 0.00002

5061
0.015

5161
0.025

eve3+7
Position
p-value

-3655
0.015

-3593
0.34

-3458
0.031

-3376
-3298
0.002 0.00005

eve 1
Position
p-value

6654
0.20

6780
0.41

6877
0.50

6966
0.086

7112
0.10

7206
0.83

eve5
Position
p-value

7340
0.21

7464
0.40

7545
0.29

7669
0.65

7794
0.13

7873
0.38

eve p
Position
p-value

-423
0.26

-358
0.78

-309
0.75

-216
0.72

-21
0.70

78
0.90

79

182
0.97

310
0.99

436
0.22

Supplemental Table 2-3. List of primer pairs used
micrococcal nuclease sensitivity analysis
Primer pairs for chromatin immunoprecipitation
ftz locus
Location FWR
Sequence
DA
CGAGTTTTTTAAGCGACAATAT
-7593 2156
GTCT
DA
AAGGAAAGTCTGTTTTGGGTAA
-5350 1956
ACA
DA
-4813 2158
TCAGCGGAATCGCATTGAC
DA
CAGGAGCAATTACAGCCTTATC
-4171 1958
CT
DA
-3352 2160
GAAAACCGCCGGAGGAGAT
DA
TCCTAATGATCTTGGTCACTTTG
-2697 1960
G
DA
-1570 2162
ACATCGGAGAGATGCGAACTG
DA
-562 1962
CCAGGGATCGGACGTAATGT
DA
TCTCCGTGCCTGCAAGGACATT
0 1882
T
DA
1000 2013
TCCAAAACCGACGCATGAA
DA
GATCGTAAAGTTATCCTTTCGC
4121 2164
AAA
DA
GGGTTCTGAGGCCATTATGATT
6624 1964
T
eve
locus
-3523 DA2205
-2330 DA2782
-1282 DA2207
-583 DA2784

TTTGTTCCGGGCTCAGAAAT
TGTAAAAAACTGAACTGGCACT
CTTC
AGTTGCAGCGTTTCGCTTTC
CAGCAGTTCCCAAATGGTTATG

84 DA2209
1194 DA2211

GCAGAGCGCAGCGGTATAA
AGCGGGAGCAGGAACTGAT

2857 DA2786
4935 DA2213

CGCGATCTTAGCCGGAATC
CCAGGCAATTGTCCTTTTTTG
80

in chromatin immunoprecipitation and

Rev
DA
2157
DA
1957
DA
2159
DA
1959
DA
2161
DA
1961
DA
2163
DA
1963
DA
1883
DA
2014
DA
2165
DA
1965

Sequence
TCCGTCTGTCCATATGAACGAT
CCTTGATTCGTCCGTGGAAA
TCGTGGGACATCAACATCTAAT
AAG
CGAGGTCCTGTGCGTCCTA
CTCGGGATTTGTTTGGTTTTCT
TTCTGGCTGCTGTTGAAGTCA
CCTTACCCCATCCGCATACTC
AAGACTGGCTGCCAGTTGTTC
TAGCTGTAGTGGCTCTGGCTGT
TT
GTGTAGCCGGCACCACAGT
TGCGCGGACAAAGGTGTT
TCGATTAGGTGCATGTCTTATGT
AGA

DA2206 CTCTCGCGGCAGGACACT
GTTGGAATGGGTTGAGTAAGTG
DA2783 ATT
DA2208 AGTATGGCACAGCGTGTTACCA
DA2785 CCATGGCTGCCATGACTTT
AGGATATTAACGAAGGCAGTTA
DA2210 GTTGT
DA2212 TCACGCCAAGGTGTTCGA
AGCGCTTAGTGTTTTATGGAGG
DA2787 AT
DA2214 CGATCCCAGGCGAATATCAT
Table continued next page

Supplemental Table 2-3. List of primer pairs used in chromatin immunoprecipitation and
micrococcal nuclease sensitivity analysis
5877 DA2215
6883 DA2501
7822 DA2503
actin5c

CGTGGCGCAGGTTGATAAT
CCCAGAGAATGGGCAACAAG
ATAACCCTACGAGCGGCTCTT

p
DA2095 CCACTTTCAGTCGGTTTATTCCA
1kb
DA2217 AGTCCGGCCCCTCCATT
Mnase footprinting primers
ftz
promoter
-503 DA2266
-404 DA2268
-334 DA2270
-259 DA2272
-113 DA2274
-10 DA2276
72 DA2278
219 DA2280
304 DA2282
422 DA2284
ftz 3+6 enhancer
-7822 DA2905
-7794 DA2907
-7593 DA2909
-7519 DA2911
-7414 DA2913
-7228 DA2915
ftz 2+7 enhancer
-6592 DA2770

CGGGTTCCGGGAACAAC
CCTCTGCCCCGAAGGAA
AATTAGAGGGAGATCGGCTGA
GA
CGTCTCAAGGTCGCCGAGTA
TGCCTGCAAGGACATTTCG
CTCTGATTTTGCTATATATGCAG
GATCT
ACATCGCAGAGTTAGAGAAGA
AATCTAG
GCAGCAATGCCTACTATCAGAA
CA
GGAGAGCTGCTACTACTACAAC
AATCAG
AGGAGCGACCCAGCACACT

DA2216 CAGCTAGTCGCATAATTTTGCAA
DA2502 GTCCCGCTTCGGATGAACT
DA2504 GCAGTTACGCGGAAGTCGAA
AAATTCAAGGCGTGAAAACTAC
DA2096 TGT
DA2218 CTGATCCTCTTGCCCAGACAA

DA2267
DA2269
DA2271
DA2273
DA2275

GGTTCCCTAAAAATTTCTCTGGA
TT
CCGATCTCCCTCTAATTGCAAT
TGAGACGGTGCTGCAGTAATTA
C
TACCCATCGCGCCCATTAT
ATAAATATCCCTGCGCATGACA

DA2277 CACGCAACGCTGGTGAGTT
DA2279 TGTTGTCGGCGTAGCTGTAGTG
TTGTAGTAGTAGCAGCTCTCCG
DA2281 AGTAAC
DA2283 GCGATGATGGAAGCAGCAT
GCTCGACGGTTGTGTAGAAATA
DA2285 GTC

GGCACATTTTTTGTGTGCTTTAT
T
ACGATTTTTAGCGGTATCATAC
GAT
CGAGTTTTTTAAGCGACAATAT
GTCT
TCCCGAGATCACATCGTTCA
CACTTTCATCCTTTGATACATAC
TTTTCAT
CTCCCATCGGGTTTTGACTGT

GAGTGTATTCTTTGTAAAATTGT
DA2906 CAATGG
TGTAGGAGTGTATTCTTTGTAAA
DA2908 ATTGTCA

GCCTCCTTTGGTAGCCTCAA

DA2771 TCGCAGGATACGCGCTTAT

81

DA2910 TCCGTCTGTCCATATGAACGAT
DA2912 CGGGATTACTGGCCAATTCA
GCAATAACTGGGAGTGGTGAAT
DA2914 AA
DA2916 GCCTCAAGCGGCGTAGATAA

Table continued next page

Supplemental Table 2-3. List of primer pairs used in chromatin immunoprecipitation and
micrococcal nuclease sensitivity analysis

-6535 DA2772

GCGCGTATCCTGCGAATTAA

-6411 DA2774

TTGTCTTGATGGCATTGTTTCTG DA2775

-6279 DA2776

AACAATTCCGGCGTTCTCAA

DA2777

-6201 DA2778

TGCATTGTTTTGGAATGTTTCC

DA2779

-6147 DA2780
ftz inter enhancer
-3997 DA2875

CGCCACACACTTTTCGTTACAA

DA2781

TCGGCATCCTTGACTTTGATT

-3783 DA2879

TTTCCCCAGCGCAGACTTAA
GAATAATTCTCGAGCATCGCTA
CTAA
TAATTTTTCTCGGCCCTTTACTA
TTT
CACCCCTTCCCCAATTATGTG

DA2876 GGTCGTCTCAATGGGTGGATA
TGCGTTATGGGATTTGTTTGAG
DA2880 T

-3697 DA2881
-3490 DA2883
-3403 DA2885

DA2773

AAACCCATACGTTGCATAGACTT
G
GAAGGTGACGTTTTATTGACTG
TCTTC
TTAAATTAGAGCAAACGGCGAC
TA
TGGGTCATAAAAACGAAAGACT
TG
GCAACATCTATTCAAAAAATCGT
TCA

DA2882 GTTCGGATCGCAGGCAAT
CAATCGAGTATATCACACTCATT
DA2884 ATCACAGT
DA2886 TCCTTTGAAGTTGCTGCTGATC

ftz 1+5 enhancer
3722 DA2887
3819 DA2889
4007 DA2891
4122 DA2893
4231 DA2895
4349 DA2897
eve
promoter
-423 DA2290
-358 DA2292

CGTTAACTCTTCCTGTGTCCTTC
TG
GAGCCATAAACCCCGAGATTC
CGAATAGGCGTGCCATTGT
GATCGTAAAGTTATCCTTTCGC
AAA
AGTTGGCGAATTATCCGTTCA
GGAAGGAACACGACAGGCTAG
A

-309 DA2294
-216 DA2296

GCTGGCAGGCGATCGATA
ACCCGACTGAGCGGCATA
GATGAAGCCGATAAAATCCCAT
TAT
CGCTCATCCGCTATGAATACC

-21 DA2298

ACTCTCAGCACCGCACGATTAG
82

DA2888 CCGACTCCGAGGACCTCTAAT
DA2890 GACTGAAACCGGGTTTGCA
GCTGGTGGAAGGGATAGAGAT
DA2892 G
DA2894 TGCGCGGACAAAGGTGTT
DA2896 GCCCATCGCGAATTTCCTA
DA2898 CCGCTCCGTGTTTATCTTTAACA

DA2291 TGGCATTATGCCGCTCAGT
DA2293 GGCGGGCCCCCTTAA
DA2295 CGGATGAGCGAATCTTTTGTCT
DA2297 TCTGCTCGGCAGGACCAT
AGGCAGTTAGTTGTTGACTGTG
DA2299 CG
Table continued next page

Supplemental Table 2-3. List of primer pairs used in chromatin immunoprecipitation and
micrococcal nuclease sensitivity analysis

78 DA2300
182 DA2302
310 DA2304

TGCAGAGCGCAGCGGTATAAA
TTGAATCACAAGACGCATACCA
A
CGGCAAGCCCCAGACA

436 DA2306

ACACCCAATCTTTCCTCTGTCCA

eve 3+7
-3656 DA2694
-3594 DA2696
-3458 DA2698
-3377 DA2700
-3298 DA2702
eve 2
-1457 DA2654
-1373 DA2656
-1335 DA2658
-1298 DA2660
-1191 DA2662
-1120
eve 4+6
4724
4768
4896
5014

DA2664
DA2666
DA2668
DA2670
DA2672

5061 DA2674
5161 DA2676
eve 1
6655 DA2728

TGTTTTGGCCGACCGATTT
GGCCGCGTTCCCATTT
TGTCGCCCGTATTAGGAAAGTA
G
CCAGCGAACTGCTCTAATTTTTT
AA
CGCGGACAAGGTTATAACGCTC
TACTT

CGAAGGCAGTTAGTTGTTGACT
DA2301 GTGC
DA2303 CATGGTGGCTCTCCATGTTG
DA2305 CGTTTGCTCGGCTTTATCTTTAA
TGTAGAACTCCTTCTCCAAGCGA
DA2307 C

DA2695 TGGGAACGCGGCCATA
DA2697 CGCCGGGACCCATAAAA
DA2699 AAAAACTAGCGCAGCGAAAAA
DA2701 ACCTTGTCCGCGTTTTTCC
GTGAATGCCATCGCGGACAAAC
DA2703 AA

TAACTGGCAGGAGCGAGGTAT
C
DA2655 GAGAAACCAGGCCAGCTTTTC
CCGGTACTGCATAACAATGGAA DA2657 CTGATGGCAAACGGATTAACAC
TTGTTCAAGATGCTGCAATAAA
CGTTTGCCATCAGCGAGATT
DA2659 GTC
GCAGCGTTTCGCTTTCGT
DA2661 TGAAAGTATGGCACAGCGTGTT
GGACCCTGGACTATAATCGCAC
GCAGAAGATGGCGATGGCTAG
AA
DA2663 AT
CATCTTCTGCGGGCGTTTGTTT
GGGCTAGGGATCGGGATTGGA
GT
DA2665 TT
TCCGCCGACCCTCCATA
TCTCGTTTTTTTCGCGTTATTTT
GCCTCGAGCAGGACTCTTTG
GGCGTTTTATGGCGGAGAT
GGCATAAAAAAACTGCATTGG
AA
CAGAATGCCAGAATGGTCAGA
A

CCGTACCTGAGCCCACTGAT
83

DA2667
DA2669
DA2671
DA2673

CGGTCATTAAAGCCCCTAAAAA
AATCAGTGCGCGAAATGTGA
CGCTGGACGAAAAAACTGAGA
CCGCCTAATTTTTATGGCCTAA

DA2675 CCGCGAAATGCGAATCTAG
DA2677 ACCGCAGCGAGTCAATTTTTT
AGATTAGAGCCAAAGAGTTGAC
DA2729 AATATTT
Table continued next page

Supplemental Table 2-3. List of primer pairs used in chromatin immunoprecipitation and
micrococcal nuclease sensitivity analysis
6780
6877
6966
7112

DA2730
DA2732
DA2734
DA2736

7206 DA2738
eve 5
7341
7465
7545
76670

DA2740
DA2742
DA2744
DA2746

7794 DA2748
7873 DA2750
Intergenic control
0 DA2806
87 DA2808
194 DA2810
300.5 DA2812
447.5 DA2814

TGGCGGAGAGGGAGATGAG
GCAACAAGTAGCGGCGAATT
CCGCGCATCTCAGTTCATC
AGAAACCGCCGAGGACACT
ACTGGGTCCGCCCCTAATC

GCCAATCTCGACTGCCACTT
TGCGCGGGCAAGAGTT
CGCCTCGAAATTAATGCAAAC
CAGTTGCAAATGCGAAAGGA
GGTAATCGCAGATAATCCCATT
DA2739 ACC

AGCGAAGTCAACTAAATCCAAT
CC
CTGTCCTGCACAAATGCTGAA
CGGCAGCAAATTTCCCTTT
GCGTGCAATTATGGACATCCT

DA2741
DA2743
DA2745
DA2747

CAGCAGCTTGCGTAAAATTTCA
TCCTTTTTACCTCTTGCCAATCC
CTGCCATGCGCTTTTGTCT
GGCCGTATTACAAACGAGTATG
C
AAGTCGCACTCGGTAGCTCTCT
GGCCGGCTGCAACATG
TTTGTGTAGGCGACATTGTTGT
AA

84

DA2731
DA2733
DA2735
DA2737

GGGCAACTTAATTGCCAGAAAT
GCGGCCTTGGGATATCTGA
CCCAAACAATGCAGCGAGTT
GGGTTCTGTAGGCGGATGAA
AAAAAGGATAATGCCAGGGATT
DA2749 C
DA2751 CTGGAAATCGCCGTCCTTT
DA2807 GGCGGAGTAGTTGGATTTTACG
GCGGAAGTGCTTTATTTAAGCT
DA2809 AGA
DA2811 AAGGTCCTGGCACGATCCT
DA2813 CCCCTTCACCATCAACATCCT
DA2815 CTGTATGGGCCTCGGTGTTT

Appendix A-2

85

Fine mapping of chromatin structure in Drosophila melanogaster embryos using
micrococcal nuclease4
Abstract
The structure of chromatin in eukaryotes exerts significant influences on many DNA related
processes, including transcription, replication, recombination and repair. A useful tool for
mapping chromatin structure is micrococcal nuclease (MNase), which induces double-strand
breaks within nucleosome linker regions, and with more extensive digestion, single-strand nicks
within the nucleosome itself. Many studies, carried out largely with microbes and cell cultures,
have used MNase to determine the positions of nucleosomes within a region of DNA to identify
dynamic changes induced during gene regulation. To measure similar processes in a
developmental context, we turned to a tractable model system, the Drosophila embryo. Here
we describe a protocol that enables MNase mapping of the enhancer chromatin structure in
the embryo, and show how it can be used to identify structural changes on a cis-regulatory
element targeted by the Knirps repressor.
Key words
Micrococcal nuclease mapping, Drosophila embryo, enhancer

4

This work was published as the following manuscript: Li, L.M. and Arnosti, D.N. (2010) Fine mapping of
chromatin structure in Drosophila melanogaster embryos using micrococcal nuclease. Fly (Austin). 2010 Jul
29;4(3).

86

Introduction
In eukaryotes, DNA is packaged into chromatin, which imposes a natural barrier for various
DNA-related processes including transcription and replication [43]. Eukaryotic cells deploy
multiple mechanisms to alter the positions of nucleosomes in order to enhance the accessibility
of DNA [44]. In the case of transcription, regulatory factors recruit chromatin remodeling
complexes such as SWI/SNF and ISWI to remove, transfer, or slide a nucleosome along the DNA
template [45]. Alternatively, histone modifying enzymes can also be recruited by transcription
factors to acetylate, methylate, phosphorylate, monoubiquitinate, sumoylate, or ADP-ribosylate
histones, or carry out reverse reactions [44]. Such modifications can directly affect inter- and
intranucleosomal interactions, or recruit downstream effectors to modulate the structure of
nucleosomes [46, 47]. Thus, measurement of chromatin structure and changes associated with
transcriptional activation and repression is essential for understanding of gene regulatory
mechanisms.
Chromatin immunoprecipitation (ChIP) is used to analyze histone occupancy and modifications,
and can pinpoint whether a specific histone protein or modification is associated with a
particular piece of DNA; however the method does not provide direct information on overall
chromatin structure [48]. Nuclease digestion does provide such structural information; DNase I
has been used to map hypersensitive sites created by the loss or remodeling of nucleosomes,
providing an indication of an active regulatory function [49]. Alternatively, fine structure
mapping can be obtained by the use of micrococcal nuclease (MNase). This nuclease can induce
double-stranded breaks within the nucleosome linker region, providing an indication of

87

whether a region of interest is protected within a nucleosome . In standard MNase experiments,
chromatin accessibility is determined in a qualitative manner by Southern blot using
radioactive-labeled probes against the DNA of interest . To obtain a more quantitative picture
of nucleosome structure, MNase digested chromatin can be analyzed by quantitative PCR using
overlapping primers. This technique has been used to determine chromatin structure in yeast,
as well as tissue-culture cells [50-52]. Current studies are also analyzing chromatin from
multicellular organisms, but no protocols are available for application of MNase mapping in
embryos. Here, we describe a protocol we developed and optimized for use with the Drosophila
blastoderm embryo, providing a method to study chromatin structure in a developmental
setting.
Collection of Drosophila embryos and formaldehyde crosslinking
To ensure reproducibility, it is critical to standardize the amount of starting materials. For each
MNase experiment, we start with approximately 500 mg of 2-4 hr old embryos collected from
10 laying bottles. Embryos are rinsed free of yeast paste and any adults by washing through
coarse nylon mesh into a large collection basket, dechorinated by bleaching for 2.5 min, rinsed
in stream of tap water for 1 min and blotted dry from below using Kimwipes.
Embryos are then transferred into a 50 ml Corning tube and crosslinked by vigorously shaking
for 30 min in 10 ml crosslinking buffer (3% formaldehyde in 50 mM HEPES, pH 7.6, 1 mM EDTA,
0.5 mM EGTA, 100 mM NaCl) and 30 ml heptane. Heptane is added to the crosslinking mixture
to increase the permeability of the embryos. Embryos are centrifuged at 1000 rpm (167 g) for 1
min in a clinical centrifuge. The top organic layer and the crosslinking buffer are carefully
88

removed so as not to lose too many embryos, and the reaction is quenched by addition of 25 ml
crosslinking stop buffer (0.125 M glycine in PBS (137mM NaCl, 4.3mM Na 2HPO4, 1.4mM
NaH2PO4) -Triton-X-100 0.01%). After gently mixing at room temperature for 30 min, the
mixture is centrifuged at 1000 rpm (167 g) for 1 min, and the supernatant is discarded. At this
point, the embryos can be flash frozen by immersing the Corning tube in liquid nitrogen for
later processing, or the material can be directly processed as described below.

Preparation of nuclei and MNase digestion
Embryos are washed three times in 10 ml PBS-Triton-X-100 (0.01%) with gentle rocking for 5
min. The crosslinked embryos are resuspended in 5 ml homogenization buffer (10mM HEPES,
pH 7.6, 0.3M sucrose, 10mM KCl, 1.5mM MgCl2, 0.5mM EGTA, 1mM DTT, 1 pellet of protease
inhibitor (cOmplete mini, Roche, Catalog No. 04 693 124 001) per 10 ml) per 0.5 gram of
embryos and disrupted in a Dounce homogenizer 10 times with a loose pestle, and 15 times
with a tight pestle. To collect the nuclei, the lysate is transferred to a 15 ml Corning tube and
centrifuged at 1000 rpm (167 g) for 10 min at 4⁰C in a clinical centrifuge. The supernatant is
removed and the pellet is resuspended in 1 ml MNase digestion buffer (10mM Tris-HCl, 15mM
NaCl, 60mM KCl, 0.15mM spermine, 0.5mM spermidine). After transferring to a 1.7 ml
microcentrifuge tube, the material is centrifuged at 14,000 rpm (16,000 g) for 10 min at 4⁰C.
The supernatant is removed and discarded, and the pellet is flash frozen until ready for use.
Nuclei from 500 mg embryos are suspended in 1ml of MNase digestion buffer in a
microcentrifuge tube. Half of the volume is reserved to serve as the undigested control. 100 U
MNase (USB, Catalog No. 70196Y) is added to the rest of the sample and incubated at 37⁰C for
89

30 min, long enough to digest most of the chromosomal DNA into mononucleosomal DNA. The
sample is chilled on ice for 10 min and then EDTA is added to the final concentration of 10 mM
to quench the digestion. The digested nuclei are collected by centrifuging at 14,000 rpm
(16,000 g) for 10 min at 4⁰C in a microcentrifuge.

Dissolving the nuclei and reversing crosslinking and DNA purification
Digested nuclei are washed twice in 1 ml sonication buffer (10mM HEPES pH 7.6, 1mM EDTA,
0.5mM EGTA, 0.1% sodium dioxycholate). For each wash, the pellet is completely resuspended,
and then recollected by centrifuging at 14,000 rpm (16,000 g) for 10 min at 4⁰C using a
microcentrifuge. Nuclei are then suspended by addition of 500 µl sonication buffer and
sonicating three times at output 4, 60% duty cycle and 20 pulses with a Branson sonicator,
using a microtip. The material is then centrifuged at 14,000 rpm (16,000 g) for 20 min to pellet
the nuclear membrane; the supernatant contains the mononucleosomes.

The crosslinking of both the digested sample and the undigested control is reversed by addition
of SDS to a final concentration of 1%, and NaCl is added to a final concentration of 0.2 M. The
samples are heated at 65⁰C overnight to reverse the crosslinking. RNA within the samples is
removed by incubating tubes at 37°C for 30 min with 10 µg RNAse (Roche, Catalog No.
10109142001) per reaction. To remove protein, the solution is adjusted to 10 mM EDTA, 40mM
Tris-Cl pH 6.5, and 20 µg Proteinase K (Roche, Catalog No. 03115836001) is added to the
reaction. Samples are incubated at 42°C for 2 hours. DNA is purified by extraction once with
buffered phenol-chloroform. To precipitate DNA, 400 µl of supernatant is transferred into a
90

microcentrifuge tube (avoid sucking up the interface), and incubated with 1 µl of GlycoBlue
(Ambion, AM9515), 44 µl NaOAc 0.3 M, and 444 µl of room temperature isopropanol for at
least 30 min. DNA is then collected by centrifugation at 14,000 rpm (16,000 g) for 15 min at
room temperature, washed once with 70% EtOH, and dried in a Speedvac for 10 min at 65°C or
in air overnight. The pellet is dissolved in 100 µl water; typically 1 µl of this solution is sufficient
for real-time PCR analysis.

Analysis of digested nucleosomes
To obtain a detailed map of chromatin structure, the resulting materials (both the digested and
undigested) are analyzed by real-time PCR with an Applied Biosystem 7500 thermocycler using
Power Syber Master Mix (Catalog No. 4367659). For a detailed mapping of nucleosome
structure, primer pairs that generate short-overlapping PCR products are normally used
{Sekinger, 2005 #56}. If the purpose of the experiment is to identify changes in nucleosome
density, less densely spaced, non-overlapping primer pairs can be used. Primers were designed
using Primer Express software (Applied Biosystems) with lengths of 100-150bp and Tms 57-60°C.
Nucleosome density of a given region is defined by the ratio of digested DNA versus undigested
control DNA.
If a nucleosome is reproducibly positioned on a length of DNA in all cells, the DNA will be
protected from MNase digestion and an amplicon from a primer pair located totally within this
nucleosome will yield a digested/undigested ratio of 1. On the other hand, if part of the
amplicon is not protected by a nucleosome, MNase digestion will reduce the amount of intact
genomic DNA for this amplification, yielding a digested/undigested DNA ratio of as little as 0.
91

Thus, a positioned nucleosome array will generate peaks and valleys, with the valleys
corresponding to linker regions between nucleosomes. If only a fraction of cells contain a
positioned nucleosome, the ratio of digested/undigested DNA will lie between 0-1.

Analysis of chromatin structure at early Drosophila zygotic genes using MNase mapping
During early Drosophila development, zygotic gap, pair-rule and segment polarity genes are
expressed in a spatially dynamic manner representing a well-studied transcriptional cascade [1].
A set of transcriptional repressors encoded by knirps, Krϋppel and giant gap genes dictate
critical positional information in the blastoderm embryos, but little is known of these proteins’
biochemical activities. We applied this MNase mapping method to assess the effect of the
short-range repressor Knirps on the chromatin structure of a cis-regulatory region located 9.5
kbp downstream of the hunchback (hb) transcription start site, and compared it to the
chromatin structure of adam, a non-target locus. Knirps protein was uniformly expressed in 2-4
hr blastoderm embryos to repress target genes using a heatshock driver [25]. We observed a
significant increase in the resistance to MNase digestion across the 400 bp hb enhancer,
whereas the digestion pattern remained unchanged at the nonspecific adam gene, as well as an
intergenic regions tested on the 3rd chromosome (Fig. 1 and data not shown). We have
observed similar increases in nuclease resistance at specific even-skipped enhancers targeted
by Knirps (Li and Arnosti, submitted). These results suggest that chromatin structure is
significantly altered during transcriptional repression by this factor, and that such effects can be
observed using the whole Drosophila embryos. This protocol is suitable for local analysis of

92

individual regulatory elements controlled by gap gene repressors, and may provide useful
insight into chromatin alterations mapped by more global approaches [53].

93

Figure A-1. Chromatin sensitivity to MN ase digestion in the Drosophila embryo, before and
after induction of Knirps repressor. (A) Chromatin accessibility of adam promoter region
before (solid line) and after (dashed line) induction of Knirps repressor. (B) Increase in
resistance to MN ase digestion at hb enhancer located at +9.5 kbp after repression of the gene
by Knirps. Data represent averages of 3 biological replicates; error bars indicate standard errors;
points in the graph represent the centers of the amplicons used in real-time PCR analysis.

94

Figure A-1 continued

Figure A-1. Chromatin sensitivity to MN ase digestion in the Drosophila embryo, before and
after induction of Knirps repressor. (A) Chromatin accessibility of adam promoter region
before (solid line) and after (dashed line) induction of Knirps repressor. (B) Increase in
resistance to MN ase digestion at hb enhancer located at +9.5 kbp after repression of the gene
by Knirps. Data represent averages of 3 biological replicates; error bars indicate standard errors;
points in the graph represent the centers of the amplicons used in real-time PCR analysis.

95

References

96

References
1.

Rivera-Pomar, R., and Jackle, H. (1996). From gradients to stripes in Drosophila
embryogenesis: filling in the gaps. Trends Genet 12, 478-483.

2.

Small, S., and Levine, M. (1991). The initiation of pair-rule stripes in the Drosophila
blastoderm. Curr Opin Genet Dev 1, 255-260.

3.

Cai, H.N., Arnosti, D.N., and Levine, M. (1996). Long-range repression in the Drosophila
embryo. Proc Natl Acad Sci U S A 93, 9309-9314.

4.

Gray, S., Szymanski, P., and Levine, M. (1994). Short-range repression permits multiple
enhancers to function autonomously within a complex promoter. Genes Dev 8, 18291838.

5.

Small, S., Arnosti, D.N., and Levine, M. (1993). Spacing ensures autonomous expression
of different stripe enhancers in the even-skipped promoter. Development 119, 762-772.

6.

Paroush, Z., Finley, R.L., Jr., Kidd, T., Wainwright, S.M., Ingham, P.W., Brent, R., and IshHorowicz, D. (1994). Groucho is required for Drosophila neurogenesis, segmentation,
and sex determination and interacts directly with hairy-related bHLH proteins. Cell 79,
805-815.

7.

Jimenez, G., Paroush, Z., and Ish-Horowicz, D. (1997). Groucho acts as a corepressor for
a subset of negative regulators, including Hairy and Engrailed. Genes Dev 11, 3072-3082.

8.

Song, H., Hasson, P., Paroush, Z., and Courey, A.J. (2004). Groucho oligomerization is
required for repression in vivo. Mol Cell Biol 24, 4341-4350.

9.

Chen, G., Fernandez, J., Mische, S., and Courey, A.J. (1999). A functional interaction
between the histone deacetylase Rpd3 and the corepressor groucho in Drosophila
development. Genes Dev 13, 2218-2230.

10.

Flores-Saaib, R.D., and Courey, A.J. (2000). Analysis of Groucho-histone interactions
suggests mechanistic similarities between Groucho- and Tup1-mediated repression.
Nucleic Acids Res 28, 4189-4196.

11.

Courey, A.J., and Jia, S. (2001). Transcriptional repression: the long and the short of it.
Genes Dev 15, 2786-2796.

12.

Martinez, C.A., and Arnosti, D.N. (2008). Spreading of a corepressor linked to action of
long-range repressor hairy. Mol Cell Biol 28, 2792-2802.

97

13.

Nibu, Y., Zhang, H., Bajor, E., Barolo, S., Small, S., and Levine, M. (1998). dCtBP mediates
transcriptional repression by Knirps, Kruppel and Snail in the Drosophila embryo. EMBO
J 17, 7009-7020.

14.

Subramanian, T., and Chinnadurai, G. (2003). Association of class I histone deacetylases
with transcriptional corepressor CtBP. FEBS Lett 540, 255-258.

15.

Shi, Y.J., Matson, C., Lan, F., Iwase, S., Baba, T., and Shi, Y. (2005). Regulation of LSD1
histone demethylase activity by its associated factors. Mol Cell 19, 857-864.

16.

Payankaulam, S., and Arnosti, D.N. (2009). Groucho corepressor functions as a cofactor
for the Knirps short-range transcriptional repressor. Proc Natl Acad Sci U S A 106, 1731417319.

17.

Arnosti, D.N., Gray, S., Barolo, S., Zhou, J., and Levine, M. (1996). The gap protein knirps
mediates both quenching and direct repression in the Drosophila embryo. EMBO J 15,
3659-3666.

18.

Ish-Horowicz, D., and Pinchin, S.M. (1987). Pattern abnormalities induced by ectopic
expression of the Drosophila gene hairy are associated with repression of ftz
transcription. Cell 51, 405-415.

19.

Yu, Y., and Pick, L. (1995). Non-periodic cues generate seven ftz stripes in the Drosophila
embryo. Mech Dev 50, 163-175.

20.

Calhoun, V.C., and Levine, M. (2003). Long-range enhancer-promoter interactions in the
Scr-Antp interval of the Drosophila Antennapedia complex. Proc Natl Acad Sci U S A 100,
9878-9883.

21.

Hiromi, Y., Kuroiwa, A., and Gehring, W.J. (1985). Control elements of the Drosophila
segmentation gene fushi tarazu. Cell 43, 603-613.

22.

Macarthur, S., Li, X.Y., Li, J., Brown, J.B., Chu, H.C., Zeng, L., Grondona, B.P., Hechmer, A.,
Simirenko, L., Keranen, S.V., et al. (2009). Developmental roles of 21 Drosophila
transcription factors are determined by quantitative differences in binding to an
overlapping set of thousands of genomic regions. Genome Biol 10, R80.

23.

Fujioka, M., Emi-Sarker, Y., Yusibova, G.L., Goto, T., and Jaynes, J.B. (1999). Analysis of
an even-skipped rescue transgene reveals both composite and discrete neuronal and
early blastoderm enhancers, and multi-stripe positioning by gap gene repressor
gradients. Development 126, 2527-2538.

98

24.

Clyde, D.E., Corado, M.S., Wu, X., Pare, A., Papatsenko, D., and Small, S. (2003). A selforganizing system of repressor gradients establishes segmental complexity in Drosophila.
Nature 426, 849-853.

25.

Struffi, P., Corado, M., Kulkarni, M., and Arnosti, D.N. (2004). Quantitative contributions
of CtBP-dependent and -independent repression activities of Knirps. Development 131,
2419-2429.

26.

Wang, X., Lee, C., Gilmour, D.S., and Gergen, J.P. (2007). Transcription elongation
controls cell fate specification in the Drosophila embryo. Genes Dev 21, 1031-1036.

27.

Schubeler, D., MacAlpine, D.M., Scalzo, D., Wirbelauer, C., Kooperberg, C., van Leeuwen,
F., Gottschling, D.E., O'Neill, L.P., Turner, B.M., Delrow, J., et al. (2004). The histone
modification pattern of active genes revealed through genome-wide chromatin analysis
of a higher eukaryote. Genes Dev 18, 1263-1271.

28.

Struffi, P., and Arnosti, D.N. (2005). Functional interaction between the Drosophila
knirps short range transcriptional repressor and RPD3 histone deacetylase. J Biol Chem
280, 40757-40765.

29.

Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P., Jones, R.S., and
Zhang, Y. (2002). Role of histone H3 lysine 27 methylation in Polycomb-group silencing.
Science 298, 1039-1043.

30.

Czermin, B., Melfi, R., McCabe, D., Seitz, V., Imhof, A., and Pirrotta, V. (2002). Drosophila
enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that
marks chromosomal Polycomb sites. Cell 111, 185-196.

31.

Dearolf, C.R., Topol, J., and Parker, C.S. (1989). The caudal gene product is a direct
activator of fushi tarazu transcription during Drosophila embryogenesis. Nature 341,
340-343.

32.

Li, X.Y., MacArthur, S., Bourgon, R., Nix, D., Pollard, D.A., Iyer, V.N., Hechmer, A.,
Simirenko, L., Stapleton, M., Luengo Hendriks, C.L., et al. (2008). Transcription factors
bind thousands of active and inactive regions in the Drosophila blastoderm. PLoS Biol 6,
e27.

33.

Janssens, H., Hou, S., Jaeger, J., Kim, A.R., Myasnikova, E., Sharp, D., and Reinitz, J.
(2006). Quantitative and predictive model of transcriptional control of the Drosophila
melanogaster even skipped gene. Nat Genet 38, 1159-1165.

34.

Pankratz, M.J., Busch, M., Hoch, M., Seifert, E., and Jackle, H. (1992). Spatial control of
the gap gene knirps in the Drosophila embryo by posterior morphogen system. Science
255, 986-989.
99

35.

Zeitlinger, J., Stark, A., Kellis, M., Hong, J.W., Nechaev, S., Adelman, K., Levine, M., and
Young, R.A. (2007). RNA polymerase stalling at developmental control genes in the
Drosophila melanogaster embryo. Nat Genet 39, 1512-1516.

36.

Ardehali, M.B., Yao, J., Adelman, K., Fuda, N.J., Petesch, S.J., Webb, W.W., and Lis, J.T.
(2009). Spt6 enhances the elongation rate of RNA polymerase II in vivo. EMBO J 28,
1067-1077.

37.

Adelman, K., Wei, W., Ardehali, M.B., Werner, J., Zhu, B., Reinberg, D., and Lis, J.T.
(2006). Drosophila Paf1 modulates chromatin structure at actively transcribed genes.
Mol Cell Biol 26, 250-260.

38.

Strahl-Bolsinger, S., Hecht, A., Luo, K., and Grunstein, M. (1997). SIR2 and SIR4
interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev
11, 83-93.

39.

Papp, B., and Muller, J. (2006). Histone trimethylation and the maintenance of
transcriptional ON and OFF states by trxG and PcG proteins. Genes Dev 20, 2041-2054.

40.

Fakhouri, W.D., Ay, A., Sayal, R., Dresch, J., Dayringer, E., and Arnosti, D.N. Deciphering a
transcriptional regulatory code: modeling short-range repression in the Drosophila
embryo. Mol Syst Biol 6, 341.

41.

Payankaulam, S., Li, L.M., and Arnosti, D.N. Transcriptional repression: conserved and
evolved features. Curr Biol 20, R764-771.

42.

Li, L.M., and Arnosti, D.N. Fine mapping of chromatin structure in Drosophila
melanogaster embryos using micrococcal nuclease. Fly (Austin) 4.

43.

Lorch, Y., LaPointe, J.W., and Kornberg, R.D. (1987). Nucleosomes inhibit the initiation of
transcription but allow chain elongation with the displacement of histones. Cell 49, 203210.

44.

Li, B., Carey, M., and Workman, J.L. (2007). The role of chromatin during transcription.
Cell 128, 707-719.

45.

Saha, A., Wittmeyer, J., and Cairns, B.R. (2006). Mechanisms for nucleosome movement
by ATP-dependent chromatin remodeling complexes. Results Probl Cell Differ 41, 127148.

46.

Kouzarides, T. (2007). Chromatin modifications and their function. Cell 128, 693-705.

47.

Jenuwein, T., and Allis, C.D. (2001). Translating the histone code. Science 293, 10741080.
100

48.

Das, P.M., Ramachandran, K., vanWert, J., and Singal, R. (2004). Chromatin
immunoprecipitation assay. Biotechniques 37, 961-969.

49.

Giresi, P.G., and Lieb, J.D. (2006). How to find an opening (or lots of them). Nat Methods
3, 501-502.

50.

Ozsolak, F., Song, J.S., Liu, X.S., and Fisher, D.E. (2007). High-throughput mapping of the
chromatin structure of human promoters. Nat Biotechnol 25, 244-248.

51.

Sekinger, E.A., Moqtaderi, Z., and Struhl, K. (2005). Intrinsic histone-DNA interactions
and low nucleosome density are important for preferential accessibility of promoter
regions in yeast. Mol Cell 18, 735-748.

52.

Petesch, S.J., and Lis, J.T. (2008). Rapid, transcription-independent loss of nucleosomes
over a large chromatin domain at Hsp70 loci. Cell 134, 74-84.

53.

Dorschner, M.O., Hawrylycz, M., Humbert, R., Wallace, J.C., Shafer, A., Kawamoto, J.,
Mack, J., Hall, R., Goldy, J., Sabo, P.J., et al. (2004). High-throughput localization of
functional elements by quantitative chromatin profiling. Nat Methods 1, 219-225.

101

Chapter III
Analysis of Hairy-mediated Long-range Transcriptional Repression in a Reporter
System
Abstract
Endogenous targets allow for studying repressor mechanisms in the native setting of the
repressors. However, the complexity of gene regulatory circuits and the inhomogeneity of cell
populations in an intact organism complicate the interpretation of the results from molecular
studies. In this chapter, reporter systems with defined positions of activator and repressor
binding sites are used to study the repression mechanisms of the long-range repressor Hairy.
One set of reporter constructs allows for placement of a constitutively expressed activator on
the promoter and induction of a LexA-Hairy fusion protein in a facultative manner, converting
the embryo from an “all on” to “all off” state. This system allows chromatin studies in a
homogenous cell population. Changes of RNA polymerase II occupancy, histone density, histone
acetylation and histone methylation at the reporter upon Hairy repression resemble the
observations made with endogenous targets. Another reporter system contains the
endogenous twist enhancer close to the basal promoter, with Hairy binding sites introduced at
various distances, which was used to study the distance limitation in Hairy-mediated repression.
This section describes a series of preliminary experiments that led to the development of the
system described in Chapter II.

102

Introduction
Studies on endogenous targets have limitations
To obtain physiologically meaningful information, it is important to study long- and short-range
repression in the context of endogenous target genes. However, such studies have several
technical limitations. First, endogenous targets for Hairy and Knirps are expressed in a spatially
and temporally limited fashion. Chromatin studies performed on embryos involve a mixture of
cell populations. Second, endogenous genes are often controlled by complex cis-regulatory
modules, which are bound and regulated by multiple transcription factors. Due to the
complexity in the gene circuit during early Drosophila development, transcription factors bound
to the same regulatory modules might cross-regulate each other. Thus, when Hairy or Knirps is
overexpressed, the final effects, either at the chromatin or transcription level, might be also
caused by other transcriptional factors whose expression levels are regulated by these two
repressors.
A reporter system to study the mechanisms of Hairy-mediated repression
Studies described by Martinez and Arnosti outline a useful approach to bypass these limitations
[1]. In their study, they engineered a gene regulatory system in which the activator is
constitutively expressed and binds to a lacZ reporter promoter, and the LexA-Hairy fusion
repressor is expressed in an inducible manner driven by the hsp70 promoter, converting the
embryo from an “all on” to “all off” state. The lacZ reporter gene contains five binding sites for
the Gal4 activator, flanked by two pairs of LexA binding sties to accommodate the repressors
(Figure 3-1).
103

B

Figure 3-1. Hairy-regulated reporter system (A) The bacterial LexA protein was fused in frame
to a Hairy fragment lacking the bHLH DNA-binding domain (93-337). The resulting LexA-Hairy
construct contains the Orange, CtBP binding and Groucho binding motifs, but not the dSir2
corepressor binding motif. The Hairy regulated reporter system contains three transgenes,
which were combined onto a single chromosome. The lacZ reporter contains Gal4 and LexA
sites. The Gal4 activator is driven by the daughterless (da) enhancer for ubiquitous expression
in the early Drosophila embryo. The LexA-Hairy repressor is driven by the hsp70 promoter to
achieve inducible expression. B) Expression of lacZ mRNA and LexA-Hairy protein upon heat
shock treatment for varying time as indicated on top of the panels. As LexA-Hairy protein
accumulates, the lacZ mRNA decreases. (Figures are adopted from Figure1 of reference [1])

104

As measured by in situ hybridization, lacZ transcription levels in the entire embryo drop
markedly after even a short (5 min) heat shock. Successively longer heat shocks resulted in the
complete loss of lacZ staining. In typical experiments, the percentage of embryos showing
strong staining dropped from almost half to less than 1% 1 hour after the induction of the
repressor, indicating that the repression was effective in the vast majority of nuclei and
embryos.
Previous chromatin studies in the reporter system
After establishing the efficacy of the system, chromatin studies were performed to determine
the chromatin changes associated with Hairy-mediated repression. Overall levels of H3
decreased in the activated state relative to those of the unactivated gene throughout the open
reading frame, and are not much further affected during repression. Strong increases in relative
H3 and H4 acetylation were observed at promoter and +1kb of lacZ reporter with the Gal4
activator present, and this acetylation decreases after the induction of the repressor. No
change within the transcription unit was observed during activation or repression. Activation of
the reporter gene is associated with a decrease in relative H3K27 methylation levels, whereas
no changes in relative H3K27 methylation levels were observed during repression [1].
Unanswered questions
RNA polymerase II occupancy upon LexA-Hairy repression
My studies of the ftz locus suggest that Pol II is excluded from this gene when it is
repressed by Hairy. To determine whether this effect on Pol II is a general property of Hairy
repression, I turned to the lacZ reporter system. Previous studies suggested that upon LexA105

Hairy repression, activator and coactivator binding is not affected on the lacZ reporter, similar
to my observations made on endogenous ftz locus. This implies that repression affects the
cross-talk between activators and the basal transcriptional machinery. Examining Pol II
occupancy at the repressed lacZ reporter will indicate whether inhibition of Pol II recruitment
by Hairy is the general rule or an effect specific to the ftz locus.
Changes of histone H3 lysine4 and lysine 36 methylation marks upon Hairy repression
Besides histone acetylation, histone H3 lysine4 and lysine36 methylation has also been
linked with active transcription. The common theme from genome-wide studies in various
eukaryotes is that high levels of H3K4 trimethylation are associated with the 5’ regions of
almost all active genes, and there is a strong positive correlation between this modification,
transcription rates, active Pol II occupancy, and histone acetylation [2-4].

The function of

H3K4 methylation in transcription regulation is still to be elucidated; however, accumulating
evidence suggests that this modification permits the recruitment of downstream effector
proteins, which in turn affect chromatin modification and remodeling. These effectors include
Chd1 in ATP-dependent chromatin remodeling, the ING proteins in histone acetylation and
deacetylation, and JMJD2A in histone demethylation [5-7].
Histone H3K36 methylation is associated with transcription elongation. Both di- and
trimethylation are enriched at the open reading frame (ORF), however, only trimethylation
displays a positive correlation with transcription rates [3]. The main function of histone H3K36
trimethylation is to recruit the Rpd3S histone deacetylase complex through the interaction with
the chromodomain Eaf3. The Rpd3S complex then produces a hypoacetylated environment in
106

the recently transcribed chromatin region to inhibit spurious transcription and ensure genome
integrity [8, 9].
While the correlation between histone H3 lysine 4 and lysine 36 methylation and active
transcription has been well established, few studies have looked into the changes of these
marks as the target locus changes from an active state to a repressed state. The inducible
repression reporter system provides a unique opportunity to examine the changes of these
active marks during transcription repression.
Histone variants and Hairy-mediated repression
Chromatin structure is not regulated solely by modification of the core histones but also
by the incorporation of histone variants. The histone variant H2A.Z is conserved across
eukaryotes [10]. Genome-wide studies suggest that in yeast H2A.Z marks both active and
inactive promoters, whereas in Drosophila and Human H2A.Z deposition is positively correlated
to transcription level, and marks the promoters of actively transcribed genes [11-14]. An
increased level of H2A.Z is associated with the process of activation, whereas gene silencing is
associated with decreased H2A.Z level [12, 15]. In yeast, repression by Tup1, a Groucho
homolog, also deposits H2A.Z at the targeted promoter to mark it for future reactivation [16]. It
will be interesting to investigate whether Hairy, a repressor that heavily replies on Groucho,
also deposits H2A.Z at the promoter upon repression.

107

Repressing a distal enhancer
One advantage of working with the synthetic enhancers is the ease of manipulation of
transcription factor binding sites. The Lex-Hairy repressor binds close to both the Gal4 activator
and the transcription start site in the transcription system used in Martinez and Arnosti. This
transcription system can be modified to study Hairy-mediated repression in a long-range
context, by putting LexA-Hairy repressor binding sites away from the activator binding sites and
the transcription start site at the lacZ reporter.
The studies in this chapter first summarize the changes of RNA polymerase II occupancy,
histone H3 lysine4 and lysine36 methylation marks, and histone variant H2A.Z level upon
repression when LexA-Hairy repressor binds close to the activators and the basal promoter.
Then, I discuss efforts to study Hairy-mediated repression in a long-range context.

108

Materials and Methods
Chromatin immunoprecipitation
Embryo collections and chromatin immunoprecipitations were performed according to a
previously described protocol [1].
Quantitative PCR analysis
The samples from ChIP were analyzed using real-time PCR (Applied Biosystem 7500). Primer
pairs had a Tm in the range of 58-60:C, and amplicons ranged from 50 to 150 bp. Primer
sequences are available upon request. A standard curve was generated by serially diluting
input samples to quantify IP samples. All values used were collected from the linear range of
amplification. The primers used to amplify the different regions of the reporter were promoter
(DA

1780:

5’-TATACAGAAGCTTGCCTGCAGGTCGGA-3’

and

DA

1781:

5’-

TATACAGCCATGCTAGAGTCTCCGCT-3’), +1kb (DA1764: 5’-AGCGTGGTGGTTATGCCG-3’ and
DA1765: 5’- GTTCAACCACCGCACGATAGA-3’), +2kb (DA1751: 5’-AAATGGCTTTCGCTACCTGGA-3’
and DA1752: 5’-ATTTAGCGAAACCGCCAAGA-3’), 3rd chromosome intergenic region (DA1938: 5’CTTTCCCGACATCCTGCAAA-3’ and DA1939: 5’-TCAGCCAACTTTAGCTATTGACAG-3’) and hsp70A
promoter

(DA1705:

5’-GTCACACAGTAAACGGCGCACT-3’

CGCTTGTTTGTTTGCTTAGCTTT-3’).

109

and

DA1706:

5’-

Construction of lacZ reporter plasmids
The lacZ reporter plasmids with LexA sites distal from the 2 X PE were constructed by cloning
SphI-spacer-2 X PE-SphI fragments from pBluescript into the pC2L5U2L used in Martinez and
Arnosti, 2008. The two SphI sites were introduced into pBluescript through KpnI using oligos
DA1214/DA1215

(DA1214:

5’-GTTAGCATGCTACGGTAC-3’

and

DA1215:

5’-

CGTAGCATGCTAACGTAC-3’, KpnI site underlied and SphI site in bold) and SacII sites using oligos
DA1205/DA1206

(DA1205:

5’-GCTTAGCATGCTACCCGC-3’

and

DA1206:

5-

GGGTAGCATGCTAAGCGC-3’, SacII site underlined and KpnI site in bold). The 2XPE enhancer
was digested out of the Twi 105 plasmid and cloned in pBluescript through NotI site. The 1.5
kbp spacer (in p4L2S2PE) consists of two fragments. SpacerI is from nucleotide 1 to 750 bp of
knirps cDNA, which was amplified from phskni1-429 plasmid using DA1212/DA1218 (DA1212:
5’-GTCATCGATATGAACCCAGACATGCAAA-3’

and

DA1218:

5’-

TATAAGCTTCGGGACTAAAGCGGTTCT-3’) and cloned into pBluescript as a ClaI-HindIII fragment.
SpacerII is the nucleotides from 1-740 bp of gfp cDNA, which was amplified from pPelican
plasmid using DA1209/DA1216 (DA1209: 5’-TATAAGCTTCGCCACCATGGTGAGCAA-3’ and
DA1216: 5’-TTATCTAGAGCCGGCCGCTTTACTTGT-3’) and cloned into pBluescript through HindIII
and XbaI sites. The lacZ reporter with 750 bp (p4LS2PE) only contains the gfp spacer. The Hairy
sites

were

cloned

into

the

p4LS2PE

and

p4L2S2PE

using

DA1391

(5’-

AATTCGCGGCACGCGACATGACCCGCGGCACGCGACATG-3’, EcoRI sites underlined and core Hairy
binding motif in bold)/DA1392 (5’-AATTCATGTCGCGTGCCGCGGGTCATGTCGCGTGCCGCG-3’,
EcoRI sites underlined and core Hairy binding motif in bold).

110

In situ hybridization of Drosophila embryos
Embryos were fixed for hybridization and stained using a digoxigenin-UTP labeled antisense
RNA probe to lacZ as described [17].
Results and Discussion:
Gal4 activator and LexA-Hairy repressor occupancy at the lacZ reporter measured by Realtime PCR
Previous studies established that on a lacZ reporter integrated into the fly genome, the Gal4
activator occupies the promoter when the reporter is active, and continue to occupy the
promoter even when the LexA-Hairy represses the lacZ reporter. This previous study used
conventional PCR and gel electrophoresis analysis following ChIP to monitor protein occupancy
[1]. To obtain a more quantitative picture of changes on the lacZ reporter, I analyzed the
precipitated materials using real-time PCR. This quantitative analysis yielded similar trends of
activator and repressor occupancies at the lacZ reporter compared to the observations made
using densitometry analysis of gels, and we calculated the fold changes of the activator and
repressor occupancy at the reporter (Figure 3-2). Activation of the lacZ reporter resulted in at
least 10 fold increase of Gal4 activator at the promoter. Upon repression, no significant change
of Gal4 occupancy was observed. LexA-Hairy repressor occupancy increases by around 10 fold
upon heat shock induction. This level of quantitation is especially useful for determining the
changes of histone modifications associated with different states of the lacZ reporter as
described below.

111

2.0

Percentage of Input (%)

A

Gal4

1.5

1.0

0.5

0.0
1

Promoter

2

+1kb

3

Intergenic

3.5

Percentage of Input (%)

3.0

B

LexA-Hairy

2.5
2.0
1.5
1.0
0.5
0.0
1

Promoter

2

+1kb

3

Intergenic

Figure 3-2. Promoter occupancy by Gal4 activator and LexA-Hairy repressor was measured by
chromatin immunoprecipitation followed by Real-time PCR analysis. (A) Background level of
Gal4 and LexA was observed at the lacZ using materials prepared from embryos containing only
the reporter (black bar). Strong Gal4 signal was observed at the promoter from the embryos
containing both the activator and the reporter (light grey bar). The activator occupancy was not
significantly affected upon binding of the repressor, consistent with previous reporter (grey bar).
(B) In the embryos contain all three transgenes, heatshock induction increased LexA-Hairy
repressor level at the promoter (grey bar).

112

Hairy repression inhibits RNA polymerase II recruitment to the targeted reporter
Upon Gal4 mediated activation, RNA polymerase II is recruited to the promoter and body of the
lacZ reporter. Upon induction of LexA-Hairy, RNA polymerase II occupancy decreases, similar to
the effect observed when Hairy represses the ftz locus (Figure 3-3). The loss of polymerase
occupancy suggests that Hairy-mediated repression generally inhibits RNA polymerase II
recruitment on all of its targets, unless there are specific features shared by the ftz and lacZ
reporter genes. A genome-wide analysis of all Hairy mediated repression events will clarify this
question.
Activation marks increase upon activation by Gal4 activator and decrease upon repression by
LexA-Hairy
I observed that H3K4 mono-, di- and tri-methylation all increased upon transcription activation
by the Gal4 activator, consistent with the earlier report. Upon repression, H3K4 methylation
levels decreased throughout the transcribed region of the lacZ reporter. Changes of H3K36 trimethylation resembled the pattern of H3K4 methylation[18-20] (Figure 3-4).
Decreased activation marks upon Hairy-mediated repression may be a result of active histone
demethylation, mediated by enzymes in the Hairy corepressor complex. Alternatively, when the
reporter gene is switched off, histone methyltransferases can no longer be recruited by
activators or the polymerase, baseline demethylation continues at a steady rate, and overall
levels of methylation drop.

113

2.0

Percentage of Input (%)

A
1.5

1.0

0.5

0.0
1

2

3

4

5

6

Position

Figure3-3. Repression of lacZ reporter expression by LexA-Hairy results in inhibition of RNA
polymerase II recruitment measured by chromatin immunoprecipiation using antibodies
again (A) unphosphorylated, (B) ser-5 phosphorylated, (C) ser-5 and ser-2 phosphorylated
forms of RNA polymerase II. Background levels of Pol II were detected with chromatin
prepared from embryos containing only the lacZ reporter (black bar). Increased Pol II signals
were observed from the embryos containing both the reporter and Gal4 activators (light grey
bar). Induction of LexA-Hairy repressor using heat shock decreased Pol II occupancy at loci
tested out on the reporter gene (gray bar). Numbers on the Y-axis represent the precipitated
materials relative to the percentage of inputs. Numbers on the X-axis represent the positions
tested out. Specifically, 1: promoter; 2: +1kb; 3: +2kb; 4: +4kb; 5: one intergenic region on the
3rd chromosome used as negative control; 6: the hsp70 promoter used as a positive control. All
the embryos have been subjected to heat shock treatment in order to eliminate non-specific
effects caused by this treatment.

114

Figure 3-3 continued

Percentage of Input (%)

2.0

B
1.5

1.0

0.5

0.0
1

2

3

4

5

6

5

6

Position
Percentage of Input (%)

2.0

C
1.5

1.0

0.5

0.0
1

2

3

4

Position
Figure3-3. Repression of lacZ reporter expression by LexA-Hairy results in inhibition of RNA
polymerase II recruitment measured by chromatin immunoprecipiation using antibodies
again (A) unphosphorylated, (B) ser-5 phosphorylated, (C) ser-5 and ser-2 phosphorylated
forms of RNA polymerase II. Background levels of Pol II were detected with chromatin
prepared from embryos containing only the lacZ reporter (black bar). Increased Pol II signals
were observed from the embryos containing both the reporter and Gal4 activators (light grey
bar). Induction of LexA-Hairy repressor using heat shock decreased Pol II occupancy at loci
tested out on the reporter gene (gray bar). Numbers on the Y-axis represent the precipitated
materials relative to the percentage of inputs. Numbers on the X-axis represent the positions
tested out. Specifically, 1: promoter; 2: +1kb; 3: +2kb; 4: +4kb; 5: one intergenic region on the
3rd chromosome used as negative control; 6: the hsp70 promoter used as a positive control. All
the embryos have been subjected to heat shock treatment in order to eliminate non-specific
effects caused by this treatment.

115

3.0

Percentage of Input (%)

2.5

A

2.0
1.5
1.0
0.5
0.0
1

1
Promoter

2

2
+1kb

3

3
+2kb

4

6

Percentage of Input (%)

5

B

4
3
2
1
0
0.5

1.0
1.5
Promoter

2.0
+1kb 2.5

3.0
+2kb 3.5

Figure 3-4. Histone H3 lysine 4 and lysine 36 methylation marks track with transcription
activity of the lacZ reporter measured by chromatin immunoprecipitation using antibodies
specific to (B) Histone H3 Lysine 4 and (A) Lysine 36 methylation. A) Gal4 activation of the lacZ
reporter gene (light grey bar) resulted in a significant increase of H3K36 tri-methylation at +1kb
and +2kb of the lacZ reporter, but not the promoter compared to the inactivated state (black
bar). LexA-Hairy repression decreases H3K36 methylation downstream of the lacZ reporter at
around +2kb (grey bar). B) Gal4 activation (light grey bar) causes increased H3K4 methylation
level at all regions tested for the lacZ reporter compared to before activation (black bar). LexAHairy repression resulted in decreases of H3K4 methylation level throughout (grey bar).

116

Density of H2A.Z histone variant tracks with H3 level
Upon both activation and repression of the lacZ reporter, no significant change in the level of
H2A.Z was observed at the lacZ reporter, although H2A.Z levels at the promoter are enriched
compared to an intergenic region (Figure 3-5, and data not shown). H2A.Z may simply mark a
promoter for future activation, but Groucho-mediated repression does not deposit more H2A.Z
at the promoter, unlike its homolog Tup1 in yeast.
Study of Hairy-mediated long-range repression using the reporter systems
The long-range effect of Hairy was first identified using transgenic embryo assays, in which
Hairy was shown to repress the rhomboid NEE enhancer and the twist 2XPE enhancer when
bound 1kb from the nearest activators within these elements [21]. To test whether the LexAHairy fusion protein also functions as a long-range repressor, the twist 2XPE enhancer was
inserted between the binding sites for LexA-Hairy and the lacZ reporter. Two spacer DNA
sequences derived from the knirps and gfp ORFs, or a single spacer from gfp ORF, were inserted
between the LexA sites and the twist 2XPE enhancer to separate the repressor binding sites 1.5
kbp or around 750 bp from the enhancer and the lacZ reporter (Figure 3-6, A). In the absence of
the LexA-Hairy repressor, the 2XPE enhancer drives the expression of lacZ at the ventral region
of the blastoderm embryos (Figure 3-6, A).
To test the ability of LexA-Hairy to repress the 2XPE enhancer-activated reporter in transgenic
embryos, we expressed LexA-Hairy in a central stripe in embryos using a Kruppel driver. LexAHairy protein is able to repress a rhomboid enhancer-driven lacZ reporter, when bound in
proximity of the enhancer (Figure 3-6, A) [1]. LexA-Hairy fusion protein driven by the Kruppel

117

2.0

Percentage of Input (%)

A
1.5

1.0

0.5

0.0
0.5

1.0

1.5

+1kb

1.0

1.5

+1kb

Promoter

2.0

2.5

2.0

2.5

2.5

B
2.0

H2A/H3

1.5
1.0
0.5
0.0
0.5

Promoter

Figure 3-5. Histone variant H2A.Z tracks with histone H3 during activation and repression of
the lacZ reporter measured by immunoprecipitation using antibody against H2A.Z. (A) Upon
activation of the reporter (light grey bar), H2A.Z level decreases compared to the inactivated
state of the reporter (black bar). Induction of LexA-Hairy repressor increases the level of H2A.Z
(grey bar). This trend is similar to that of histone H3. (B) When normalized to the H3 chromatin
IP signals, no significant change was observed upon activation and repression.

118

enhancer failed to repress lacZ expression in the corresponding region of the embryos, whether
it was situated from either 1.5 kbp or 750 bp upstream of the 2XPE enhancer (Figure 3-6, A).
This result was not expected, as the native Hairy protein can regulate enhancers over similar
distances.
Two possibilities may account for the inability of the LexA-Hairy fusion protein to repress the
2XPE. The Kruppel driver used to express LexA-Hairy protein expression may not produce
sufficient levels of the transgene, or alternatively, the removal of the Hairy DNA binding domain
(DBD) abolishes interaction with the dSir2 corepressor, which may be important for Hairy’s
long-range function.
To increase expression levels of the LexA-Hairy protein, I used the hsp70A promoter to drive the
ubiquitous expression of LexA-Hairy protein. Upon a 20 minute heat shock, LexA-Hairy protein
partially (around 50% of the embryos in the blastoderm stage) repressed the lacZ reporter
expression (Figure 3-6, C). This repression is less efficient, compared to the situation when
LexA-Hairy binds close to the activators and the transcription start site, in which after 20
minute heat shock lacZ expression was repressed in most embryos.
To compare this repression efficacy with the effectiveness of the endogenous Hairy protein,
and to test whether dSir2 binding is important for Hairy-mediated long-range repression, a lacZ
reporter with six Hairy binding sites 1.5kbp upstream of the 2XPE enhancer was used. Only a
limited number of blastoderm embryos exhibited repression by endogenous Hairy protein,
suggesting that Hairy, for reasons we do not yet understand, is not able to interfere with the
2XPE enhancer when bound 1.5 kbp away 5’ of this enhancer (Figure 3-6, B).
119

A.

B.

Figure3-6. Repression of twist 2 X PE enhancer by LexA-Hairy or endogenous Hairy protein
from a distance. A) LexA-Hairy protein driven by Kruppel enhancer fails to repress the 2XPE
enhancer from 1.5 kbp or 750 bp. Top panel, lacZ reporter driven by rhomboid (rho) and twist
(twi) enhancers with 2 LexA sites nearby is repressed by LexA-Hairy protein driven by a Kruppel
driver (The figures were adapted from figure 1 of Martinez and Arnosti, 2008). Lower two
panels, 2 X PE enhancer drives lacZ reporter express in a single stripe across the ventral region
of the embryos. This enhancer was not repressed by LexA-Hairy driven by Kruppel enhancer
from a distance. (B) Endogenous Hairy protein is able to repress 2XPE driven lacZ reporter from
750bp but not from 1.5 kbp.

120

Figure 3-6 continued

C.

70
60
50
40
30
20
10
0
1

2

Figure 3-6. Repression of twist 2 X PE enhancer by LexA-Hairy or endogenous Hairy protein
from a distance. (C) Quantification of lacZ staining pattern in embryos expressing both the lacZ
reporter and LexA-Hairy protein under a heat shock driver with no heat shock (1) or 20 minute
heat shock (2). After 20 minute heat shock, the number of unstained embryos increased by
three-fold. Black: strong staining; light grey: light staining; grey: no staining; Y-axis: numbers of
embryos in each category; n=100.

121

In light of previous reports demonstrating Hairy’s activity over long distances, it is puzzling that
even the endogenous Hairy protein was unable to function effectively in this transgenic
reporter assay. One explanation involves a “hot chromatin” model for Hairy-mediated longrange repression, in which Hairy can only bind active or ‘open’ enhancers. This means that in
order for Hairy to mediate long-range repression, its binding sites need to localize within an
active enhancer.
Recruitment of LexA-Hairy fusion protein to the LexA sites 1.5 kbp away from the 2XPE
enhancer
In order to test whether the LexA-Hairy fusion protein indeed fails to access the binding sites
when there is no activator at the vicinity, we performed chromatin immunoprecipitation to
determine the occupancy of LexA-Hairy binding. There is indeed a lower level of LexA-Hairy
occupancy at the LexA sites, compared to when there is Gal4 activator binding close by. This
result implies an open chromatin environment is more favorable for Hairy binding (Figure 3-7).
I attempted to measure changes in chromatin structure and the RNA polymerase II machinery
upon LexA-Hairy repression of the reporter. However, little change was observed (data not
shown). This is possibly due to two effects; first, the 2XPE enhancer only drives the expression
of lacZ in the ventral region of the embryo; second, the repression observed is only less than
50%. To set up a reporter system suitable for study of Hairy mediated repression in a longrange context, enhancers that drive ubiquitous expression of the lacZ reporter should be used.
Also, activator binding sites might need to be added to the vicinity of the LexA sites in order to
open up the chromatin.
122

Figure3-7. Closed chromatin environment inhibits LexA-Hairy recruitment. LexA-Hairy fusion
protein can still be recruited to the LexA sites in the absence of closeby activators measured by
chromatin immunoprecipitation, although the occupancy is at a lower level. Input titration 2%,
1%, 0.5%, 0.25%.

123

Conclusion
Studies presented here use reporter systems with well-defined activator and repressor binding
sites to study the effects of Hairy-mediated repression on the transcription machinery and
chromatin structure. When binding to the vicinity of activators and the basal promoter,
repression of the reporter by LexA-Hairy inhibits the recruitment of RNA polymerase II and
decreases the levels of histone H3 lysine 4 and lysine 36 methylation marks. Chromatin studies
in the reporter system not only set the stage for investigation of the endogenous locus
described in Chapter II, but also provide complementary information for Hairy repression
mechanisms. Both the LexA-Hairy and the endogenous Hairy protein are less potent in
repressing a distal twist enhancer, suggesting the recruitment of this repressor and possibly its
corepressor complex requires an open chromatin environment.

124

APPENDICES

125

Appendix B

126

Cofactor contributions to Hairy Repression Activities

Introduction
Hairy and its multiple corepressors
Given the important roles of Hairy protein during various biological processes, many studies
have looked into its repression mechanism. Groucho was the first purified as a Hairy
corepressor recruited by the WRPW motif, and was shown to be important for Hairy function
[22-24]. Hairy also recruits short-range corepressor CtBP through the PLSLV motif. CtBP
maternal depletion suppresses h mutant phenotype, and CtBP is required for Hairy repression
in a cell-type specific manner [25, 26]. In ectopic expression assays, CtBP antagonize Grouchodependent Hairy repression of slx, fkh, hkb and tll [27]. dSir2 protein also interacts physically
and genetically with Hairy. This interaction maps to a highly conserved “RRAR” motif in the
basic region of the Hairy DNA binding domain [28]. Another critical motif for Hairy function is
the Orange domain, which is required for rescue of hb-scute male lethality [29]. DamID
mapping studies suggest that dSir2, CtBP and Groucho corepressors differentially overlap with
Hairy binding, suggesting that Hairy might utilize these different corepressors in a contextdependent manner [30].
Requirement of multiple corepressors is a common theme in repression
It is frequently observed that a single repressor uses multiple corepressors for repression in
eukaryotes. Drosophila repressor Brinker requires both CtBP and Groucho for maximal and
versatile repression during the Dpp signaling. The conclusion is based on analysis of
transcriptional outcomes arising from the genetic removal of these corepressors, and from
127

ectopically expressing Brk variants in the embryo [31]. A new corepressor dNAB for Brinker was
recently indentified, which is required for regulation of dMyc expression [32]. Besides Groucho
binding motif, repressor Runt repressor also has other repression domains [33].
Although previous studies suggest that Hairy is likely to deploy different corepressors in a
context-dependent manner, the specific contexts where particular repressors are used are still
not clear. Also, most studies of Hairy repression mechanism focus on Drosophila embryonic
stage and cell culture system; it is not clear how these corepressors contribute to Hairy
repression during other developmental stages? In order to understand the general rules of
corepressor usage, it is necessary to analyze Hairy repression at a genome-wide level.
We plan to use previous established Hairy ectopic expression assays to investigate the
contribution of different corepressors in Hairy repression, by overexpressing mutant versions of
Hairy that does bind to different corepressors. The purpose is to set up a platform that can be
used for future analysis at a genome-wide level. We focus the assays on the role of Groucho
corepressor in Hairy mediated repression, as previous studies suggested Groucho accounts for
most of Hairy repression activity.

128

Materials and Methods:
Constructs for protein overexpression
Heat shock inducible wild-type Hairy and mutant Hairy that does not bind to Groucho
constructs were generated as described in Chapter II. Heat shock inducible Orange domain
mutant Hairy (NEVSR/AAAAA) was generated through site-directed mutagenesis using primer
DA2248

(5’

-GCCGGATTCGCCGACTGTGTGGCCGCGGCTGCCGCCTTTCCCGGCATCGAGCCCG-

3’)/DA2249 (5’- CGGGCTCGATGCCGGGAAAGGCGGCAGCCGCGGCCACACAGTCGGCGAATCCGGC3’) using wild-type Hairy plasmid as a template. Heat shock inducible Groucho plasmid was
generated using modified pCasper plasmid used for Knirps overexpression in previous studies
[34].

Full

length

groucho

cDNA

was

CGTCTAGAATAAATAACTTCGTAGAC-3’)/DA1579

amplified

using

primer

DA1580

(5’-

(5’-ATATGGTACCATGTATCCCTCACCGGTG-3’)

from Pet17b-Gro used in previous studies [22].
Crude embryo lysate preparation
Approximately 50 mg of dechrionated embryos were resuspended in 1.2 ml of lysis buffer (25
mM HEPES pH 7.9, 150 mM NaCl, 1 mM DDT, 1 mM PMSF, 1 mM Na-metabisulfite, 1 mM
benzamidine, 10 uM pepstatin A) and disrupted by sonication using a Branson Sonifier 250 (2
cycles of 12 pulses each, output 3, duty cycle 60%). After sonication, lysates were centrifuged
for 15 minutes at 14,000 rpm using an Eppendorf centrifuge, and the protein centrifuge of the
supernatant was determined using the Bradford assay, with BSA as the standard.
Western blot analysis

129

Immunoblotting was performed according to standard protocols using a tank transfer system
(Mini Trans-Blot Cell, BioRad). Sequi-BlotTM PVDF membranes (BioRad) were used and
antibody incubation was in TBST(20 mM Tris-HCl, pH 7.5, 120 mM NaCl, 0.1% Tween-20)
supplemented with 5% (w/v) nonfat dry milk as blocking agent. The primary anti Hairy
monoclonal antibody (Abcam, catalog #: ab20165) was used at 1:5,000 dilution incubated at
4 °C for overnight. The primary anti Tubulin antibody (Hybridoma Bank) was used at 1:10,000
dilution. The secondary ImmunoPure Goat Anti-Mouse HRP-conjugated antibody (Pierce) was
used at 1:10,000 dilution incubated at room temperature for 3 hour. The blot is detected using
SuperSignal Western Blotting Kit (Pierce), and visualized using Fuji LAS-1000 imager.
Quantification of mRNA expression levels
hairy, ftz, ac, prd, stg mRNA expression levels were quantified through real time PCR analysis.
Actin5C mRNA expression level, whose expression level is not affected by Hairy overexpression,
was used as an internal control. Primers used for quantification of mRNA expression levels are
as followed: DA2013(5’-TCCAAAACCGACGCATGAA-3’)/DA2014(5’-GTGTAGCCGGCACCACAGT-3’)
for ftz; DA2005 (5’-ACGCCTGGCACCAAGAGTT-3’)/DA2006 (5’-CGGAATCGTCGATGTTGCT-3’) for
ac; DA2049 (5’-GGATCCAGCCACTGTGTACCA-3’)/DA2050 (5’-GGCGATAGATTACTACTGGCCATT-3’)
for

prd;

DA2019

(5’-ACCAACGCAACACATAACACAAC-3’)/DA2020

(5’-

CGGCATAATTACCTGATATTGCAAT-3’) for stg.
In situ staining of mRNA expression pattern
Embryos were fixed for in situ hybridization and stained using anti-digoxigenin-UTP-label RNA
probes for targeted genes as previously described [17]. Template from making ac probe was
130

purchase from Drosophila Genome Resource Center (Stock #: IP01413). The template plasmid
was digested using BamH1, and the digoxigenine-labeled RNA was synthesis using SP6
polymerase (NEW ENGLAND BioLabs, Catalog# M0207S) following standard protocol.
Results and Discussions:
Groucho cofactor and Hairy-mediated repression
Mutant Hairy protein that does not bind to Groucho accumulates to a lower level than wildtype Hairy upon heat shock induction
In order to make direct comparison of the repression activity between wild-type and Grouchobinding deficient Hairy, we first compared the mRNA and protein induction levels of Hairy
protein. Upon 20 minute induction, wild-type and mutant version of Hairy expressed at
compared levels, suggesting positions of the inserted heat shock inducible transgenes do not
affect the transcription levels of these two versions of Hairy proteins (Figure B-1, A). However,
Hairy protein that does not bind to Groucho reproducibly expresses at a lower level when
compared to the wild-type protein. This might suggest that binding to the Groucho cofactor is
important for the stability of the Hairy repressor (Figure B-1, B). Alternatively, the Hairy protein
is measured using an antibody that is raised using a GST-tag full-length Hairy protein. It is
possible that this antibody is less efficient in detecting the mutant Hairy protein.
Hairy repression of fushi tarazu is dependent on Groucho corepressor
As described in Chapter II, Hairy repression of ftz is dependent on Groucho, and that
overexpression of Hairy mutant protein that does not bind to Groucho failed to ectopically

131

repress the seven-stripe expression pattern of ftz. We quantified the mRNA expression of ftz
upon induction of these two versions of Hairy, and found that only overexpression of wild-type
Hairy protein resulted decreased ftz mRNA level, consistent with the results from the in situ
hybridization analysis (Figure B-2).
Other Hairy targets in the blastoderm embryos
To investigate Hairy targets other than fushi tarazu in early embryos, we analyzed expression
patterns of several important early Drosophila genes. DamID studies suggested that Hairy binds
close to acheate, egghead and paired genes in early embryos. Indeed, ectopic expression of
wild-type Hairy protein resulted in complete repression of acheate and egghead, as well as
repression of parasegments of paired. Hairy overexpression also disrupts even skipped
expression pattern. This regulation might be indirect, as endogenous Hairy does not occupy this
locus according to a recent genome-wide ChIP-chip study [35]. On the other hand, Hairy
expression does not affect the expression levels of maternal and gap genes, such as bicoid ,
caudal (Caudal protein level before and after induction of Hairy repression was compared by
western blotting, data not shown) and Kruppel (Figure B-3).
Hairy might use Groucho repressor in a context-dependent manner
We further analyzed the role of Groucho corepressor during Hairy repression of other targets.
Overexpression of Hairy decreased paired expression level by around two-fold. Overexpression
of the mutant version of Hairy also represses paired expression with similar efficiency (Figure B4, A). This indicates that there is Groucho-independent Hairy repression activity.

132

Analysis of acheate expression upon induction of Hairy resulted in similar observation, that
both versions of Hairy repress acheate with similar efficiency, although even overexpression of
wild-type Hairy resulted in a really minor (p =0.15) change in acheate mRNA expression level
(Figure B-4, B). This might be caused by the fact that acheate is only expressed in a few nuclei in
the blastoderm embryos, and the assay used is not sensitive enough to detect the difference
(Figure B-3, 3rd panel, right).
Role of Orange domain in Hairy repression of fushi tarazu
Other than the WRPW motif that recruits Groucho, Orange domain is another motif that is
highly conserved among Hairy and Enhancer of Split family members. A five-amino acid
deletion in the orange domain (ΔNEVSR) rendered Hairy unable to rescue hb-scute lethality [29].
In order to test the role of orange domain in Hairy-mediated transcriptional repression, we
generated flies expressing mutant Hairy with NEVSR motif mutated to Alanines under a heat
shock driver. The Orange domain mutant Hairy can be induced to similar levels as the wild-type
Hairy protein in adult flies after 20 minute heat shock induction (Figure B-5, A). Different from
the mutant Hairy protein that does not bind to Groucho, this Orange domain mutant Hairy can
still disrupt fushi tarazu expression pattern, although less efficient than the wild type. After 20
minute heat shock induction, 39% of the embryos showed normal fushi tarazu overexpression
pattern. Among the embryos that were repressed by Hairy overexpression, most of them have
only stripe 4 and 7 repressed (Figure B-5, B). This result suggests that although Groucho is
indispensable in repression of fushi tarazu expression by Hairy, Orange domain also plays an
important role during this process. Stripe 4 and 7 are the stripes that are most sensitive to
Hairy repression, with Hairy binding likely to be closest to their enhancers. Orange domain
133

mutant Hairy is most efficient in regulating these stripes, suggesting that this mutant is only
deficient in repressing distal enhancers. Although it is quite speculative, it is possible that
Orange domain is required for Groucho to adapt a conformation that is suitable for long-range
repression.

Characterization of heat shock inducible Histidine and Flag double-tagged Groucho protein in
embryos
To provide direct evidence for the involvement of Groucho spreading in Hairy-mediated longrange repression, it is necessary to directly chromatin immunoprecipitate Groucho onto Hairy
endogenous targets. However, attempts to ChIP Groucho using available antibodies have not
been successful (data not shown). One possibility is that the qualities of the antibodies are not
good enough for chromatin immunoprecipitation. We overexpressed a Histidine and Flag
double-taqqed Groucho protein under a heat shock inducible hsp70A promoter, in order to
provide epitopes that can be precipitated by other antibodies.
We first examined the effects of Groucho overexpression on fushi tarazu. To our surprise,
Groucho overexpression did not result in obvious disruption of ftz expression pattern (data not
shown). It is possible that there is not enough endogenous Hairy protein to recruit ectopically
expressed Groucho to the ftz locus. We then overexpressed the Groucho protein together with
wild type Hairy protein, and observed that not only ftz expression was repressed; the
development of the embryo was also disrupted (data not shown). This suggested that once
recruited to the chromatin, the double-tagged Groucho is functional. However, attempts to
chromatin immunoprecipitates Groucho using antibodies against either the Histidine or Flag
epitopes have not been successful (data not shown). The N-terminal of Groucho protein is
134

important for its oligomerization, and the C-terminal WD40 motif is important for Groucho
interaction with the repressors. It is possible that the epitopes at these two ends are masked by
protein-protein interaction.
We also used this overexpression lines to examine the size of Groucho corepressor complex.
Overnight embryos were collected and subjected to 20 minute heat shock. Groucho protein
was first purified over a Ni2+-NTA matrix and then analyzed by gel filtration chromatography.
When fractions were analyzed by immunoblotting, Groucho was observed to migrate in
fractions with size larger than 158 KDa, with more intense signals at fractions bigger than 670
kDa. Previous studies suggested Groucho presents in the form of tetramers, our result implied
that Groucho can also present in smaller size complex in Drosphila embryos.

135

6

A
5
4
3
2
1
0
yw

B

hsh

w/a

yw

hsh

w/a

Figure B-1. Mutant Hairy accumulates at a lower level than the wild-type Hairy protein,
despite that the mRNA levels are comparable. A) mRNA expression level of Hairy in embryos
without transgene (yw), with heat shock inducible wild type hairy transgene (hsh), or mutant
hairy transgene that does not bind to Groucho cofactor (w/a). 20 minute heatshock increases
the hairy mRNA levels by around 4 folds in the embryos containing heat shock inducible
transgenes. Relative mRNA levels were calculated using ΔCt method. The hairy mRNA levels
from different samples were normalized to hairy mRNA levels in the embryos containing no
transgene. B) Hairy protein level in the embryos containing mutant hairy transgene (w/a) is
higher than that in the embryos without transgene (yw), and lower than that in the embryos
containing wild type hairy transgene (hsh). Hairy protein is measured by an antibody that is
raised using a GST-tag full-length Hairy protein. Tubulin level is as an internal control to make
sure equal protein loading.

136

Figure B-2. Abundance of ftz mRNA in embryos with overexpressed wild-type Hairy or mutant
Hairy lacking the Groucho-binding WRPW motif (Hairy w/a) relative to embryos lacking the
Hairy transgne (yw). mRNA was extracted from 2-3 hour embryos that were exposed to 20
minute heat shock with no recovery period. The y axis represents relative levels of ftz mRNA,
which were calculated by ΔCt method.

137

Figure B-3. Expression patterns of early Drosophila genes before and after 20 minute
induction of Hairy protein analyzed by in situ hybridization.

138

1.2

A

1.0
0.8
0.6
0.4
0.2
0.0
yw

hsh

w/a

1.2

B

1.0
0.8
0.6
0.4
0.2
0.0
yw

hsh

w/a

Figure B-4. Abundance of paird and achaete mRNA in embryos with overexpressed wildtype Hairy or mutant Hairy lacking the Groucho-binding WRPW motif (Hairy w/a) relative
to embryos lacking the Hairy transgne (yw). mRNA was extracted from 2-3 hour embryos
that were exposed to 20 minute heat shock with no recovery period. The y axis represents
relative levels of mRNA, which were calculated by ΔCt method. (A) paired mRNA level; (B)
achaete mRNA level.
139

A

B

Figure B-5.Mutation in the Orange domain rendered Hairy less efficient in repression of ftz.
(A)Protein levels of Hairy in adult flies containing no transgene (lane 7), heat shock inducible
wild type Hairy (lanes 5,6), mutant Hairy that does not bind to Grouch (lanes 3,4) and Hairy
bearing Orange domain mutations (lanes 1,2) measured by western blotting using anti-Hairy
monoclonal antibody. All the transgenic lines expressed Hairy protein to a higher level. (B) fushi
tarazu mRNA expression pattern upon overexpression of Hairy bearing Orange domain
mutations in blastoderm embryos analyzed by in situ hybridization. Number below the images
showed the percentage of embryos showing according expression patterns (n=100).

140

Figure B-6. Groucho migrates in molecular masses from around 158 KDa to above 670 KDa.
Peak fractions from Superdex 200 size exclusion experiments with purified Groucho protein
were resolved on SDS-PAGE gels and western blotted with anti-M2 antibody (to detect FLAGtagged purified proteins). Groucho was found in molecular mass that is around 158 KDa,
suggesting that other than tetramers, Groucho can also present in monomeric in a complex
with other proteins.

141

References

142

References
1.

Martinez, C.A., and Arnosti, D.N. (2008). Spreading of a corepressor linked to action of
long-range repressor hairy. Mol Cell Biol 28, 2792-2802.

2.

Bernstein, B.E., Kamal, M., Lindblad-Toh, K., Bekiranov, S., Bailey, D.K., Huebert, D.J.,
McMahon, S., Karlsson, E.K., Kulbokas, E.J., 3rd, Gingeras, T.R., et al. (2005). Genomic
maps and comparative analysis of histone modifications in human and mouse. Cell 120,
169-181.

3.

Pokholok, D.K., Harbison, C.T., Levine, S., Cole, M., Hannett, N.M., Lee, T.I., Bell, G.W.,
Walker, K., Rolfe, P.A., Herbolsheimer, E., et al. (2005). Genome-wide map of
nucleosome acetylation and methylation in yeast. Cell 122, 517-527.

4.

Schneider, R., Bannister, A.J., Myers, F.A., Thorne, A.W., Crane-Robinson, C., and
Kouzarides, T. (2004). Histone H3 lysine 4 methylation patterns in higher eukaryotic
genes. Nat Cell Biol 6, 73-77.

5.

Pray-Grant, M.G., Daniel, J.A., Schieltz, D., Yates, J.R., 3rd, and Grant, P.A. (2005). Chd1
chromodomain links histone H3 methylation with SAGA- and SLIK-dependent
acetylation. Nature 433, 434-438.

6.

Shi, Y., and Whetstine, J.R. (2007). Dynamic regulation of histone lysine methylation by
demethylases. Mol Cell 25, 1-14.

7.

Shi, X., and Gozani, O. (2005). The fellowships of the INGs. J Cell Biochem 96, 1127-1136.

8.

Li, B., Gogol, M., Carey, M., Lee, D., Seidel, C., and Workman, J.L. (2007). Combined
action of PHD and chromo domains directs the Rpd3S HDAC to transcribed chromatin.
Science 316, 1050-1054.

9.

Li, B., Gogol, M., Carey, M., Pattenden, S.G., Seidel, C., and Workman, J.L. (2007).
Infrequently transcribed long genes depend on the Set2/Rpd3S pathway for accurate
transcription. Genes Dev 21, 1422-1430.

10.

Jackson, J.D., and Gorovsky, M.A. (2000). Histone H2A.Z has a conserved function that is
distinct from that of the major H2A sequence variants. Nucleic Acids Res 28, 3811-3816.

11.

Mavrich, T.N., Jiang, C., Ioshikhes, I.P., Li, X., Venters, B.J., Zanton, S.J., Tomsho, L.P., Qi,
J., Glaser, R.L., Schuster, S.C., et al. (2008). Nucleosome organization in the Drosophila
genome. Nature 453, 358-362.

143

12.

Schones, D.E., Cui, K., Cuddapah, S., Roh, T.Y., Barski, A., Wang, Z., Wei, G., and Zhao, K.
(2008). Dynamic regulation of nucleosome positioning in the human genome. Cell 132,
887-898.

13.

Zhang, H., Roberts, D.N., and Cairns, B.R. (2005). Genome-wide dynamics of Htz1, a
histone H2A variant that poises repressed/basal promoters for activation through
histone loss. Cell 123, 219-231.

14.

Guillemette, B., Bataille, A.R., Gevry, N., Adam, M., Blanchette, M., Robert, F., and
Gaudreau, L. (2005). Variant histone H2A.Z is globally localized to the promoters of
inactive yeast genes and regulates nucleosome positioning. PLoS Biol 3, e384.

15.

Leach, T.J., Mazzeo, M., Chotkowski, H.L., Madigan, J.P., Wotring, M.G., and Glaser, R.L.
(2000). Histone H2A.Z is widely but nonrandomly distributed in chromosomes of
Drosophila melanogaster. J Biol Chem 275, 23267-23272.

16.

Gligoris, T., Thireos, G., and Tzamarias, D. (2007). The Tup1 corepressor directs Htz1
deposition at a specific promoter nucleosome marking the GAL1 gene for rapid
activation. Mol Cell Biol 27, 4198-4205.

17.

Small, S., Blair, A., and Levine, M. (1992). Regulation of even-skipped stripe 2 in the
Drosophila embryo. EMBO J 11, 4047-4057.

18.

Carrozza, M.J., Li, B., Florens, L., Suganuma, T., Swanson, S.K., Lee, K.K., Shia, W.J.,
Anderson, S., Yates, J., Washburn, M.P., et al. (2005). Histone H3 methylation by Set2
directs deacetylation of coding regions by Rpd3S to suppress spurious intragenic
transcription. Cell 123, 581-592.

19.

Joshi, A.A., and Struhl, K. (2005). Eaf3 chromodomain interaction with methylated H3K36 links histone deacetylation to Pol II elongation. Mol Cell 20, 971-978.

20.

Keogh, M.C., Kurdistani, S.K., Morris, S.A., Ahn, S.H., Podolny, V., Collins, S.R., Schuldiner,
M., Chin, K., Punna, T., Thompson, N.J., et al. (2005). Cotranscriptional set2 methylation
of histone H3 lysine 36 recruits a repressive Rpd3 complex. Cell 123, 593-605.

21.

Barolo, S., and Levine, M. (1997). hairy mediates dominant repression in the Drosophila
embryo. EMBO J 16, 2883-2891.

22.

Paroush, Z., Finley, R.L., Jr., Kidd, T., Wainwright, S.M., Ingham, P.W., Brent, R., and IshHorowicz, D. (1994). Groucho is required for Drosophila neurogenesis, segmentation,
and sex determination and interacts directly with hairy-related bHLH proteins. Cell 79,
805-815.

144

23.

Fisher, A.L., Ohsako, S., and Caudy, M. (1996). The WRPW motif of the hairy-related
basic helix-loop-helix repressor proteins acts as a 4-amino-acid transcription repression
and protein-protein interaction domain. Mol Cell Biol 16, 2670-2677.

24.

Jimenez, G., Paroush, Z., and Ish-Horowicz, D. (1997). Groucho acts as a corepressor for
a subset of negative regulators, including Hairy and Engrailed. Genes Dev 11, 3072-3082.

25.

Poortinga, G., Watanabe, M., and Parkhurst, S.M. (1998). Drosophila CtBP: a Hairyinteracting protein required for embryonic segmentation and hairy-mediated
transcriptional repression. EMBO J 17, 2067-2078.

26.

Phippen, T.M., Sweigart, A.L., Moniwa, M., Krumm, A., Davie, J.R., and Parkhurst, S.M.
(2000). Drosophila C-terminal binding protein functions as a context-dependent
transcriptional co-factor and interferes with both mad and groucho transcriptional
repression. J Biol Chem 275, 37628-37637.

27.

Zhang, H., and Levine, M. (1999). Groucho and dCtBP mediate separate pathways of
transcriptional repression in the Drosophila embryo. Proc Natl Acad Sci U S A 96, 535540.

28.

Rosenberg, M.I., and Parkhurst, S.M. (2002). Drosophila Sir2 is required for
heterochromatic silencing and by euchromatic Hairy/E(Spl) bHLH repressors in
segmentation and sex determination. Cell 109, 447-458.

29.

Dawson, S.R., Turner, D.L., Weintraub, H., and Parkhurst, S.M. (1995). Specificity for the
hairy/enhancer of split basic helix-loop-helix (bHLH) proteins maps outside the bHLH
domain and suggests two separable modes of transcriptional repression. Mol Cell Biol
15, 6923-6931.

30.

Bianchi-Frias, D., Orian, A., Delrow, J.J., Vazquez, J., Rosales-Nieves, A.E., and Parkhurst,
S.M. (2004). Hairy transcriptional repression targets and cofactor recruitment in
Drosophila. PLoS Biol 2, E178.

31.

Hasson, P., Muller, B., Basler, K., and Paroush, Z. (2001). Brinker requires two
corepressors for maximal and versatile repression in Dpp signalling. EMBO J 20, 57255736.

32.

Ziv, O., Suissa, Y., Neuman, H., Dinur, T., Geuking, P., Rhiner, C., Portela, M., Lolo, F.,
Moreno, E., and Gerlitz, O. (2009). The co-regulator dNAB interacts with Brinker to
eliminate cells with reduced Dpp signaling. Development 136, 1137-1145.

33.

Walrad, P.B., Hang, S., Joseph, G.S., Salas, J., and Gergen, J.P. Distinct contributions of
conserved modules to runt transcription factor activity. Mol Biol Cell 21, 2315-2326.
145

34.

Struffi, P., Corado, M., Kulkarni, M., and Arnosti, D.N. (2004). Quantitative contributions
of CtBP-dependent and -independent repression activities of Knirps. Development 131,
2419-2429.

35.

MacArthur, S., Li, X.Y., Li, J., Brown, J.B., Chu, H.C., Zeng, L., Grondona, B.P., Hechmer, A.,
Simirenko, L., Keranen, S.V., et al. (2009). Developmental roles of 21 Drosophila
transcription factors are determined by quantitative differences in binding to an
overlapping set of thousands of genomic regions. Genome Biol 10, R80.

146

Chapter IV
Conclusions and Future Directions
Conclusions
Studies presented here provide insights into mechanisms mediated by two major classes of
repressors that are conserved throughout the metazoans. Repression mechanisms at the
chromatin level not only confirm that there are indeed two distinct modes of repression during
early Drosophila development, but also suggest two different ways in which repressors bound
to distal enhancers can effect transcription at the basal promoter. Key discoveries of this thesis
are summarized as follows.
Secondary pair-rule genes are driven by stripe-specific enhancers
My studies suggest that Hairy mediates progressive repression on fushi tarazu. This has
important implications for the Hairy repression mechanism and the regulation of secondary
pair-rule gene expression. Secondary pair-rule genes like fushi tarazu have long been
considered to be driven by a single enhancer [1]. The fact that different stripes of ftz can be
repressed in a step-wise manner suggests that there are distinct modular enhancers underlying
the seven-stripe expression pattern, similar to the case of primary pair-rule genes like even
skipped.
Mechanisms of Hairy-mediated long-range repression
The progressive repression of ftz mediated by Hairy suggests that Hairy is not directly targeting
the basal promoter, in which case all the modular enhancers would be switched off at the same
147

time. A short induction of the repressor protein followed by recovery time resulted in complete
repression of the ftz locus. This phenomenon might reflect the kinetics by which Groucho
corepressor spreads to repress the whole ftz locus.
Hairy-mediated long-range repression involves a “spreading” mechanism. Chromatin analysis
suggests that extensive repression of the ftz locus by Hairy resulted in wide-spread histone
deacetylation, and a mild induction of Hairy repression resulted in localized histone
deacetylation around the enhancer region. This implicates that Hairy action first nucleates
around its binding sites, and then extends to the whole locus, probably through Groucho
corepressor spreading. Repression of ftz by Hairy resulted in the inhibition of RNA polymerase II
recruitment, unlike in the case of many other developmental regulated genes having a paused
RNA polymerase around the basal promoter region (Figure 4-1).
On a reporter gene repressed by Hairy, similar changes of RNA polymerase II occupancy and
histone modifications were observed, suggesting these observations are likely to be the general
rules of Hairy repression other than gene-specific effects [2].
Analysis of the reporter system also suggests that the levels of histone H3 lysine 4 and lysine 36
marks track with the transcription levels during repression, whose mechanism is not clear. The
decrease may result from binding of histone demethylase to the Hairy corepressor complex;
alternatively, reduction in activation marks could be just a passive effect of decreased
transcription activity.

148

When the Hairy sites are moved 1.5 kbp away from the closest enhancer. Hairy is a less potent
repressor. This observation supports a “hot chromatin” model in which recruitment of Hairy
and its corepressors is favored by an open chromatin environment.
Mechanisms of Knirps-mediated short-range repression
Knirps-mediated repression resulted in local changes of chromatin structure. Micrococcal
nuclease mapping and chromatin immunoprecipitation reveals that Knirps mediated repression
causes a local increase of histone density. The increased histone density might result from
Knirps inhibiting the disassembly of the nucleosome or Knirps recruiting factors that deposit
nucleosomes around its binding region. These results suggest that in addition to previously
proposed models, short-range repression can happen through chromatin-related mechanisms.
Knirps repression inhibits the recruitment of Caudal but not Bicoid activator, suggesting Knirps’
effects on activators might be target-dependent. Knirps does not affect the recruitment of RNA
polymerase II, though it is still not clear at which downstream step Knirps represses RNA
polymerase II (Figure 4-1).
Future Directions
There are several possible future directions that will increase our knowledge of the mechanisms
of long- and short-range repression.
Changes in chromatin structure upon repression at a genome-wide level
As more and more transcription factor binding patterns have been profiled, one emerging
question is how to define a functional binding event. Elucidation of chromatin features

149

associated with repression would help to define functional repressor binding sites. The
chromatin studies for long- and short-range repression have been performed only on the
canonical targets fushi tarazu and even skipped. To understand whether the observations are

Figure 4-1. Mechanisms of Hairy and Knirps-mediated transcriptional repression. The top
panel is a schematic representation of an active gene. Middle panel: Hairy (H) mediated
repression causes wide-spread histone deacetylation and inhibition of RNA polymerase II (Pol II)
recruitment, without affecting the recruitment of the activator (A). Lower panel: Knirps (K)
locally increases histone density and causes histone deacetylation. Knirps inhibits activator
recruitment in a context-dependent manner, and affects RNA polymerase II at the steps after
recruitment.

150

target-specific or generally representative, it is necessary to perform similar chromatin studies
at a genome-wide scale using techniques like ChIP-chip or ChIP-seq.
How do corepressors dictate context-dependent effects?
Hairy and Knirps recruit similar corepressor complexes including both CtBP and Groucho
corepressors, but mediate distinct repression output and chromatin effects [3, 4]. It would be
interesting to investigate the mechanisms underlying context-dependent effects of these
corepressors. Hairy and Knirps recruit Groucho through different motifs. Because of differences
in interaction surfaces, Groucho may adopt alternative configurations when recruited by
different transcription factors, so that oligomerization and spreading is only possible when
associated with a long-range repressor. This hypothesis would suggest that chimeric repressor
proteins carrying suitable interaction surfaces would adopt short- or long-range activities,
regardless of the rest of the protein’s structure. Alternatively, as noted in previous chapters
that Groucho is phosphorylated in response to signaling pathways, so differently modified
forms of the protein may exhibit different abilities to oligomerize, affecting short- or long-range
activity. Or, as with other corepressor proteins, Groucho may assemble into distinct complexes
with different abilities to mediate spreading, and these complexes may be separately recruited
by short- or long-range repressors. This hypothesis can be tested by determining the forms of
Groucho recruited to different corepressor complexes by chromatin immunoprecipitation
(Figure 4-2) [5].

151

Figure 4-2. Possible mechanisms underlying Groucho-mediated context-dependent repression.
(A) Groucho adopts alternative conformations when associated with different repressors, and
only spreads when recruited by long-range repressors. (B) Groucho is associated with distinct
corepressor complexes that are recruited by different repressors; a form that is capable of
spreading is recruited by long-range repressors.

152

Contributions of distinct corepressors to repression mediated by a single repressor
Similar to many other transcription factors, Hairy and Knirps recruit several different
corepressors. It would be interesting to determine the roles of various distinct corepressors
during the repression process. It is possible that these corepressors could act together to
achieve maximal repression, function redundantly, mediate target-specific or cell-specific
repression, or antagonize the functions of the other corepressors. This question can be
answered by overexpressing mutant versions of the repressors that do not bind to specific
corepressors, and measuring the target expression by quantitative PCR or in situ hybridization.
The wild-type and mutant versions of the repressors can be induced at different developmental
stages to determine the roles of different corepressors at specific developmental time windows.
Transcriptome analysis upon induction of various versions of the repressors can be used to
determine the roles of various corepressors at the genome-wide scale.

153

References

154

References
1.

Hiromi, Y., Kuroiwa, A., and Gehring, W.J. (1985). Control elements of the Drosophila
segmentation gene fushi tarazu. Cell 43, 603-613.

2.

Martinez, C.A., and Arnosti, D.N. (2008). Spreading of a corepressor linked to action of
long-range repressor hairy. Mol Cell Biol 28, 2792-2802.

3.

Barolo, S., and Levine, M. (1997). hairy mediates dominant repression in the Drosophila
embryo. EMBO J 16, 2883-2891.

4.

Arnosti, D.N., Gray, S., Barolo, S., Zhou, J., and Levine, M. (1996). The gap protein knirps
mediates both quenching and direct repression in the Drosophila embryo. EMBO J 15,
3659-3666.

5.

Payankaulam, S., Li, L.M., and Arnosti, D.N. (2010). Transcriptional repression:
conserved and evolved features. Curr Biol 20, R764-771.

155