PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5I08 KlProj/AocstlelRCIDateDuejndd

 

ROLE OF ADA2b AND GCNS IN COR GENE EXPRESSION DURING COLD
ACCLIMATION IN ARABIDOPSIS

By

Kanchan Amol Pavangadkar

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Genetics Program

2008

ABSTRACT

ROLE OF ADA2b AND GCNS IN COR GENE EXPRESSION DURING COLD
ACCLIMATION IN ARABIDOPSIS

By
Kanchan Amol Pavangadkar

Covalent modiﬁcations of histones play important roles in the regulation of
transcription. Acetylation of lysine residues on the amino-terminal tails of histones is
associated with transcriptionally active genes and is catalyzed by histone
' acetyltransferases (HAT). GCNS (a HAT) and ADA2 are components of coactivator
complexes such as SAGA in yeast. The Arabidopsis genome encodes one homologue of

GCNS and two homologues of ADA2 (ADA2a and ADA2b).

Arabidopsis ADA2 and GCNS physically interact with the transcriptional
activator CBFl in vitro which activates the expression of cold-regulated (COR) genes
during cold acclimation. Cold acclimation is the process by which plants increase
freezing tolerance upon exposure to low non-freezing temperatures. CBFl binds to the
cold/dehydration responsive element (CRT/DRE) present in COR gene promoters. ada2b
and gcn5 mutants show a delay in activation and a reduction in expression of COR genes
during cold acclimation. Chromatin immunoprecipitation (ChIP) assays show that the
acetylation of histone H3 at the COR promoters increases upon cold acclimation. Thus
we hypothesized that CBF recruits GCNS and ADA2b to the COR gene promoters to
acetylate histones and thus help activate COR genes. ChIP assays on plants

overexpressing CBF 1 show increased acetylation of histone H3 even without cold stress

and the acetylation levels increase further upon cold acclimation. Thus, CBF is suﬁicient

to bring about an increase in acetylation of histone H3 at the COR gene promoters.

The acetylation levels of histone H3 in ada2b and gcn5 mutants are similar to
wild type plants upon cold acclimation and thus ADA2b and GCNS are not required for
histone H3 acetylation at the COR gene promoters. The Arabidopsis genome encodes 12
potential HATS and the HAC l , HACS, HAC12 and TAP l HATS were tested for their
ability to regulate COR gene expression. None of these HATS were required for the
acetylation of histone H3 at the COR gene promoters as determined by ChIP assays. Thus
we conclude that the acetylation of histone H3 at the COR gene promoters is not solely

dependent on any of the HATS tested.

This thesis is dedicated to the three most important people in my life Aai, Baba and
- Amol.

iv

TABLE OF CONTENTS

List of Tables ............................................................................... viii
List of Figures ............................................................................... ix
List of Abbreviations ....................................................................... xi
Chapter I:
Histone acetyltransferases in plants ................................................ 1-23
1.1 Introduction .............................................................................. 1
1.2 Histone acetylation in plants ........................................................... 4
1.2.1 Histone acetylation and plant pathogens ................................ 5
1.2.2 Histone acetylation and C4 photosynthesis ............................. 5
1.2.3 Histone acetylation in Arabidopsis during ﬂowering ................. 6
1.2.4 Histone acetylation and root patterning ................................. 7
1.2.5 Histone acetylation and seed germination .............................. 7
1.3 Histone acetyltransferases in plants ................................................... 9
1.3.1 GCNS ...................................................................... 10
1.3.2 p300/CBP .................................................................. 15
1.3.3 TAF 1 ....................................................................... 19
1.4 Novel histone acetyltransferases in Arabidopsis. ................................. 20
1.5 Hypothesis and experimental design ................................................ 22
Chapter II:
Increase in acetylation of histone H3 corresponds to activation of
COR genes during cold acclimation ............................................... 24-65
2.1 Introduction ............................................................................. 24
2.2 Materials and methods ................................................................ 26
2.2.1 Plants used and growth conditions ....................................... 26
2.2.2 Cold acclimation ............................................................ 26
2. 2. 3 Tissue harvesting and storage ............................................. 27
2. 2. 4 Chromatin immunoprecipitation assay .................................. 27
2.2.5 Antibodies ................................................................... 30
2.2.6 PCR conditions and data analysis ........................................ 30
2.3 Results ................................................................................... 31
2.3.1 Acetylation of histone H3 at the COR gene promoters increases
during cold acclimation ........................................................... 31

2.3.2 The level of total histone H3 at some COR gene promoters
decreases upon cold acclimation ................................................ 33
2.3.3 Levels of acetylated histone H3 and total histone H3 are

restored to non-acclimated levels within 24 hours of de-acclimation. . . ...34
2.3.4 CBF is sufﬁcient for the acetylation and removal of histone H3

at the COR gene promoters ...................................................... 39
2.3.5 The activation domain of CBF is not necessary to drive

acetylation of histone H3 upon cold acclimation but is required for the
removal of histone H3 from the COR gene promoters ....................... 42
2.3.6 GCN5 is not required for the acetylation of histone H3 at the

COR gene promoters but is required for the removal of histone H3

from the COR gene promoters during cold acclimation ..................... 50
2.3.7 ADA2b is not necessary for the acetylation of histone H3 at the
COR gene promoters but is required for the removal of histone H3 ﬁ'om

the COR gene promoters during cold acclimation ............................ 56
2.4 Discussion .............................................................................. 59
Chapter III: ,
The activation of COR genes by CBF occurs in an ADA2b and GCN5
independent manner .................................................................. 66-81
3.1 Introduction ............................................................................. 66
3.2 Materials and methods ................................................................ 68
3.2.1 Construction of ADA2b-I +/- and GCN5-1 +/- lines
containing the 4X-CRT-GUS promoter-reporter gene ....................... 68
3.2.2 Creating wild type, adaZb-I and gcn5-I plants
overexpressing CBF2. ., ........................................................... 69
3.2.3 RNA isolation and gene expression analysis ........................... 69
3.3 Results ................................................................................... 71
3.3.1 The 4X-CRT driven expression of the reporter gene GUS
by CBF is neither dependent on ADA2b nor GCN5 ......................... 71
3.3.2 ADA2b and GCN5 are not required for the CBF dependent
expression of the COR genes .................................................... 76
3.4 Discussion .............................................................................. 78
Chapter IV:
Histone acetyltransferases and their effect on COR gene regulation
during cold acclimation in Arabidopsis .......................................... 82-109
4.1 Introduction ............................................................................ 82
4.2 Materials and methods ................................................................ 84
4.2.1 Plants used and growth conditions ....................................... 84

vi

4.2.2 RNA isolation and gene expression analysis ........................... 84

4.2.3 Chromatin immunoprecipitation assay .................................. 84
4.3 Results ................................................................................... 84

4.3.1 HAC gene family ........................................................... 84

4.3.2 TAF 1 gene family .......................................................... 96
4.4 Discussion ............................................................................. 106
Chapter V:
Discussion ........................................................................... 1 10-1 14
5.1 Histone acetylation and COR gene expression ................................... 110
5.2 Histone H3 acetylation and CBF ................................................... 110
5.3 Histone acetyltransferases and histone acetylation of COR genes ............ 112
5.4 ADA2b and GCN5 regulation of COR gene expression ........................ 114
6.1 Bibliography .................................................................................................. 1 18

vii

LIST OF TABLES

Table l. Oligonucleotides used in quantitative Real-time PCR analysis of chromatin
immunoprecipitation assays ...................................................................... 115

Table 2. Oligonucleotides used in quantitative Real-time PCR for gene expression
analysis .............................................................................................. 116

Table 3. Oligonucleotides used in the generation of the 4X-CRT-GUS plasmid... ......1 17

viii

LIST OF FIGURES

Figure 1: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ............................................. 32

Figure 2: Level of total histone H3 at COR gene promoters decreases upon
cold acclimation ............................................................................ 35

Figure 3: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation and is restored upon
de-acclimation .............................................................................. 37

Figure 4: Levels of total histone H3 at some COR gene promoters decreases
upon cold acclimation and is restored upon de-acclimation .......................... 38

Figure 5: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ............................................. 41

Figure 6: Level of total histone H3 at COR gene promoters decreases upon
cold acclimation ............................................................................ 43

Figure 7: Level of acetylated histone H3 at COR gene promoters increases
upon cold acclimation ..................................................................... 45

Figure 8: Level of total histone H3 at COR gene promoters remains
unchanged upon cold acclimation in CBFZAC plants ................................. 47

Figure 9: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ............................................. 51

Figure 10: Level of total histone H3 at COR gene promoters decreases upon
cold acclimation in wild type plants but not in gcn5-I plants ........................ 54

Figure 11: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ............................................. 57

Figure 12: Level of total histone H3 at COR gene promoters decreases upon
cold acclimation in wild type plants but not in ada2b-I plants ...................... 60

Figure 13: The expression of the reporter gene GUS ﬁised to 4X-CRT is
similar in wild type and ada2b-I plants ................................................. 73

Figure 14: The expression of the reporter gene GUS fused to 4X-CRT is
similar in wild type and gcn5-I plants ................................................... 75

Figure 15: The expression of CBFZ and GoIS is similar in wild type and

ix

adaZb-I plants overexpressing CBF 2 ................................................... 77

Figure 16: The expression of CBF 2 and G018 is similar in wild type and
gcn5-1 plants overexpressing CBF 2 ..................................................... 79

Figure 17: The expression of CBF 2 and COR genes increases in wild type
Col-0 and hacI-Z plants upon cold acclimation ....................................... 86

Figure 18: The expression of CBF2 and COR genes increases in wild type
Col-0 and hac5-1 plants upon cold acclimation ....................................... 89

Figure 19: The expression of CBF 2 and COR genes increases in wild type
Col-0 and hathacIZ plants upon cold acclimation ................................... 92

Figure 20: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ............................................. 95

Figure 21: Level of total histone H3 at some COR gene promoters decreases
upon cold acclimation ...................................................................... 97

Figure 22: The expression of CBF2 and COR genes increases in wild type Ws
and taﬂ-I plants upon cold acclimation ................................................ 99

Figure 23: Ratio of acetylated histone H3 to total histone H3 at COR gene
promoters increases upon cold acclimation ........................................... 102

Figure 24: Level of total histone H3 at COR gene promoters decreases upon
cold acclimation ........................................................................... 103

Figure 25: Ratio of acetylated histone H4 to total histone H3 at COR gene
promoters increases upon cold acclimation ........................................... 105

LIST OF ABBREVIATIONS

ChIP chromatin immunoprecipitation

CRT/DRES cold/dehydration responsive elements
HAT histone acetyltransferase

HDAC histone deacetylase

IP immunoprecipitation

PIC pre-initiation complex

TSA trichostatin A

xi

Chapter I
Histone acetyltransferases in plants
1.1 Introduction

Assembly of the general transcription factors to form a pre-initiation complex
(PIC) is a requirement for genes transcribed by RNA Polymerase II [1]. The eukaryotic
genome is packed into a tight chromatin structure and hence the ground state of
transcription in eukaryotes is restrictive [2]. Various factors are required to overcome this
restriction, facilitate formation of the PIC and activate gene transcription. Transcriptional
activators can facilitate PIC assembly by recruiting several coactivators that help
counteract the nucleosome barrier and activate transcription [3]. Transcriptional
coactivator complexes can be divided into two broad categories. The ﬁrst category
includes coactivators that interact with or are a part of the RNA polymerase II
transcription machinery. The second type of coactivators are chromatin modifying
complexes that modify chromatin to overcome the nucleosomal barrier to transcription
[4].

Chromatin is made of DNA wrapped around nucleosomes. Each nucleosome
particle consists of a histone octamer comprising two copies each of histones H2A, H2B,
H3 and H4 [5, 6]. Adjacent nucleosomes are bound by the linker histone H1 and the
binding of the linker histones leads to further compaction of the chromatin into a ~30 nm
ﬁber. Two models have been proposed for the structure of the ~30 nm chromatin
structure. The ﬁrst is the one-start helix or the solenoid model in which consecutive

nucleosomes are placed adjacent to each other bound by linker DNA to form a left-

handed Single start helix. The second model proposes that nucleosomes are arranged in a
zigzag manner such that alternate nucleosomes are stacked on top of each other. The
adjacent nucleosomes are connected by linker DNA [7-9]. This compact structure of
chromatin needs to be opened in order for processes like DNA replication and
transcription to proceed.

Chromatin modifying coactivator complexes can be divided into two categories:
ATP-dependent chromatin modifying complexes (like SWI/SNF and ISWI) and
complexes that covalently modify histones (such as SAGA and ADA). ATP-dependent
chromatin modifying complexes use energy from ATP hydrolysis to destabilize the
chromatin structure by displacement or removal of histones [10, 11]. Coactivator
complexes that can covalently modify histones can bring about various modiﬁcations of
histones such as acetylation, methylation, ubiquitination, and phosphorylation. Some
modiﬁcations of histones act as transcriptional repressors while others serve to activate
the process of transcription. The ‘histone code’ hypothesis proposes that covalent
modiﬁcation of the histone tails direct the recruitment of chromatin modifying complexes
that direct other covalent modiﬁcations and help the activation of transcription [12, 13].
Recent work has also shown that some of the histone covalent modiﬁcations in plants
differ considerably from other eukaryotes [14, 15].

Histone acetylation is usually associated with the activation of genes. The
enzymes responsible for histone acetylation are known as histone acetyltransferases
(HATS) and are often a part of large coactivator complexes. HATS acetylate lysine
re$idues in the amino terminal tails of histones. Acetylation of lysine residues neutralizes

the charge on the lysine residue and is thought to weaken the histone-DNA interaction.

Acetylation of lysine residues can also alter internucleosomal histone-histone contacts
and affect interactions between histones and regulatory proteins [16]. The acetylation of
lysine residues on the N-terminal tails of histone has been correlated with transcriptional
activation [17, 18]. Some transcription factors also show an increased afﬁnity towards
acetylated histones [19, 20].

HATS can be divided into two types; nuclear HATS or HAT A and cytoplasmic or
type B HATS [21]. Type A HATS acetylate nucleosomal histones during transcription
whereas Type B HATS function in the acetylation of newly synthesized histones [22, 23].
Type B HATS acetylate lysine 5 and 12 on histone H4 before the histones are assembled
into newly replicated chromatin [24, 25]. Nuclear HATS are associated with transcription
and this review will speciﬁcally focus on nuclear HATS and their function in plants.

In addition to acetylation of histones, nuclear HATS are also known to acetylate
other non-histone proteins and modulate their activity. p3 OO/CBP acetylates the
transcription factor p53 and acetylated p53 has a higher binding afﬁnity to its target
promoters [26, 27]. PCAF can acetylate the transcription factor E2Fl that acts as an
activator of gene expression during the S-phase of the cell cycle [28]. In addition to
acetylating transcription factors HATs can also acetylate members of the general
transcription machinery TFIIEB, TFIIF [29]. Due to this function of HATS it has been
suggested that HATS can play a role in signal transduction [30]. HATS have been
extensively studied in eukaryotes like yeast, ﬂies, mice and humans [16, 17, 31-33].
Nuclear HATS can be further divided into four families, GNAT-MYST, p300/CBP,

TAF 11250 and the nuclear receptor coactivator family. The nuclear receptor family is not

present in plants [34]. The HAT families in plants are further discussed in section 2.

Acetylation of histones is reversible and the enzymes that can deacetylate histones
are known as histone deacetylases (HDAC). Arabidopsis has three HDAC families with a
total of 18 members [34]. Members belonging to RPD3/HDA1 family are present in all
eukaryotes. Members of the Silent Information Regulator 2 (SIR2) family share no
structural similarity with other HDACS and use NAD as a cofactor. The HD2 family of
HDACS is unique to plants. Histone deacetylases are known to deacetylate histones and
repress gene expression. Disruption of the HDAC HDl in Arabidopsis causes a change in
expression of 7% of the total genes. The changes in acetylation are found mostly at the
promoter sites which causes repression of speciﬁc genes [35].

This review will focus on the type A HATS in plants and their function during the
various cellular, biochemical and physiological aspects of plant growth and development.
Plant HATS have some features that are distinctively different from the animal HATS and
this may contribute to some plant speciﬁc functions of these HATS. In this review we will
also bring out these differences and predict how these differences might contribute to
plants Speciﬁc functions in addition to transcription activation.

1.2 Histone acetylation in plants

Plants are subject to a variety of environmental and biological insults due to their
sessile nature. Changes in histone acetylation of target genes in response to abiotic and
biotic stress, biosynthesis, plant growth and development occur in plants [36, 37]. The
HATS that are responsible for the increase in acetylation at these target genes have not
been identiﬁed. Investigating the mode of action of HATS in controlling gene expression

by histone or factor acetylation in response to various stimuli is the next step.

1.2.1 Histone acetylation and plant pathogens

Hyperacetylation of histones has been observed in maize in response to various
biotic and abiotic stresses [3 8, 39]. For example, histones H3 and H4 are hyperacetylated
in response to the host selective HC toxin produced by the fungal pathogen Cochliobolus
carbonum. This is thought to occur via the inhibition of histone deacetylases in maize.
The pathogen has no effect on the acetylation status of histones H2A and H28. The
hyperacetylation of histones H3 and H4 due to the inhibition of histone deacetylase might
interfere with the expression of the maize defense genes and thus give an edge to the
pathogen [3 8]. Plant hormones ethylene, jasmonic acid (JA) and salicylic acid (SA) are
involved in defense responses upon infection by plant pathogens. J A was shown to
induce the expression of the deacetylases HAD6 and HAD19 [40, 41]. Overexpression of
H4019 was responsible for the induction of the J A dependent plant defense genes [41].
HDA19 most likely induces the expression of the defense response genes by inhibiting a
repressor of these genes. HDA6 was found to physically interact with the F-box protein
COIl in viva. It is possible that the F-box protein COIl can target HDA6 for
ubiquitination and control the expression of the defense genes induced by J A. Whether
COIl targets HDA6 for ubiquitination has not been determined [40]. It will be interesting
to ﬁnd out whether the regulation of defense response genes by the HDACS occurs
directly or indirectly. Gene expression of IA induced defense genes seems to be regulated
by histone acetylation and deacetylation in plants.
1.2.2 Histone acetylation and C4 photosynthesis

Maize is a C4 plant where the ﬁxation of C02 occurs in two different types of

photosynthetic plant cells, mesophyll and bundle sheath cells. C4 speciﬁc

phosphoenolpyruvate carboxylase (C4-PEPC) is involved in the ﬁxation of C02 in
mesophyll cells. The expression of C4-PEPC is differentially regulated in the mesophyll
and the bundle sheath cells in the presence of the same external light stimulus. The
expression of C4-PEPC is low in etiolated plants but is highly induced in green plants.
Histone H4 hyperacetylation occurs at the C4-PEPC promoters in green plants although
the hyperacetylation of histone H4 was not sufﬁcient for the transcription of the C4-PEPC
gene [34]. Light signal was shown to be necessary and sufﬁcient for the hyperaceylation
of histone H4 at the C4-PEPC promoter. Thus, histone acetylation may be involved in the
signal transduction pathway that leads to the induction of genes involved in C4
photosynthesis in maize.
1.2.3 Histone acetylation in Arabidopsis during ﬂowering

The ﬂowering in Arabidopis is controlled by four important pathways;
autonomous, vemalization, photoperiod and gibberellins [42]. The autonomous and
vemalization pathways induce ﬂowering by repressing F LC which acts as a repressor of
ﬂowering. The autonomous pathway consists of Six genes, one of which (F LD) is
homologous to a human protein that is a part of the histone deacetylase 1, 2 co-repressor
complex [43, 44]. The ﬂd mutant ﬂowers later than the wild type plants. As the F LD
homologue in humans is a part of a co-repressor complex, it was thought that F LD might
act by controlling the acetylation of histones at the FLC promoters. Indeed, the FLC
promoters showed hyperacetylated histone H4 in the ﬂd mutants as compared to the wild
type plants. Thus, FLD might be a part of a histone deacetylase co-repressor complex in

Arabidopsis and might repress F LC expression by deacetylating histones.

1.2.4 Histone acetylation and root patterning

Root patterning in Arabidopsis occurs by the differentiation of epidermal root
cells into hair and non-hair cells. The cellular differentiation depends on the correct
spatial expression of six key genes. In order to understand the role of histone acetylation
during root patteming the root cells were treated with trichostatin A (TSA), an inhibitor
of histone deacetylases [45]. Treatment of root cells with TSA disrupts the cellular
patterning and increases expression levels of the key genes by inhibiting HDACS.
Investigation of the promoters of these genes showed that the levels of acetylated
histones H3 and H4 were higher than in the plants without the TSA treatment [46]. Thus
the balance between HDACS and HATS controls the spatial expression of the key genes
during root patterning. HDA18 is the main HDAC involved but the identity of the HAT
or HATS involved is yet to be determined.
1.2.5 Histone acetylation and seed germination

Seed germination in Arabidopsis marks the end of the embryonic stage and the
beginning of the vegetative phase. This transition is manifested by a decrease in
expression of genes involved in embryonic development. In order to identify whether
acetylation of histones plays a key role in this switch, the seeds were treated with TSA,
resulting in the upregulation of 45 genes and down regulatibn of 27 genes during the
process of seed germination. Eight of these genes (four of which were late embryonic
genes) were highly expressed in dry seeds but imbibition of seeds caused a rapid
repression of these genes that continued till three days after imbibition [47]. This lead to
deacetylation of histone H4 at the promoters and coding regions of these genes one day

after imbibition but acetylation was restored to basal levels after three days. Treatment

with TSA did not affect the onset of repression and there was no difference in gene
expression observed one day after imbibition in TSA treated and untreated samples. The
eXpression of these genes was higher after three and six days after imbibition in the
presence of TSA than in untreated seeds suggesting that the rate of repression may be
slower in these plants but the onset of repression is not. However, TSA-treated plants did
not show a deacetylation of histone H4 at the promoters or coding regions of these eight
genes one day after imbibition as seen in untreated plants. Thus repression of these genes
after imbibition occurs in the TSA-treated seeds even though higher histone H4
acetylation levels are maintained at the target genes. The deacetylation occurring one day
after imbibition at the target genes might play a critical role in the transition from
embryonic stage to vegetative stage.

The phaseolin gene from bean is repressed in vegetative tissues but is activated
during embryogenesis. The activation of the phaseolin gene from its repressed state
occurs in two stages potentiation and activation [35]. Distinct histone modiﬁcations occur
during the two stages. The repressed state is marked by the presence of a nucleosome
over the TATA box and methylation of lysine 20 of histone H4. The potentiation stage
shows a decrease in the occupancy of histones H3 and H4 at the promoters as well as a
signiﬁcant increase in the acetylation of lysine 9 of histone H3 and lysine 12 of histone
H4. The activation stage is marked by an increase in the acetylation of lysine 14 of
histone H3 and lysine 5 of histone H4 along with an increase in the trimethylation of
lysine 4 of histone H3. Thus, the phaseolin promoter serves as a good model to study the

relationship between histone acetylation within and between histones as well as with

other covalent modiﬁcations. Thus plants seem to have their own ‘histone code’ that
Speciﬁes different stages of gene activation.
1.3 Histone acetyltrasferases in plants

Most of the work on histone acetyltransferases in plants has been from
Arabidopsis and maize. In plants, the ﬁrst correlation between increased histone
acetylation and gene activation was seen in the pea plastocyanin gene (PetE). The
promoter region of the PetE showing increased histone H3 and H4 acetylation was also
more accessible to DNaseI and micrococcal nuclease [48, 49]. Since then there have been
numerous reports of HATS and their role in gene expression in plants.

The Arabidopsis genome contains 12 HAT enzyme coding genes. These genes
can be further classiﬁed into three families [34]. The GNAT-MYST superfamily consists
of the GNAT and the MYST families. The GNAT (GCN5 like histone acetyltransferase)
family in Arabidopsis has three members, designated HAGI (GCN5), HAG2 and HAG3.
GCN5 is the most extensively characterized member of this family [37, 50, 51]. The
MYST family contains two members, HAG4 and HAGS. None of the members of the
MYST family have been studied in Arabidopsis as yet. The p300/CBP family contains
ﬁve members HACl, HAC2, HAC4, HACS and HAC12. Some members of this family
play a role in regulating gene expression of the vemalization pathway of ﬂowering [52,
53]. The TAFIIZSO family has two members, HAF 1 and HAF2 (TAF 1). The TAF 1
protein in Arabidopsis has been shown to govern the expression of light-regulated genes
by controlling the acetylation of histones [54, 55].

Histone acetyltransferases are parts of large multisubunit complexes in yeast,

ﬂies, mice and humans [56-5 8]. These coactivator complexes are recruited by various

transcription factors to the promoters of target genes. The HATS act as the catalytic
subunits of these complexes acetylating histones at the promoters and activating
transcription. Various components of such HAT complexes have been genomically
identiﬁed in plants but the complexes have not yet been characterized biochemically [50,
59].

Histone acetylation is involved in the regulation of ﬂowering, seed development,
embryogenesis and cellular patterning in root cells. The enzymes involved in these
pathways have not been characterized. HATS in animal systems are usually parts of large
coactivator complexes that are speciﬁcally recruited by activators. Coactivator complexes
containing HATS have not been identiﬁed in plant systems. Some plant HATS differ
considerably in their structure from the animal HATS and hence it is very likely that the
structure and composition of HAT coactivator complexes in plants would be very
different.

1.3.1 GCN5

One homologue of GCN5 has been identiﬁed in maize. In order to understand the
role of GCN5 in maize gene expression was tested in plants treated with TSA and in
plants expressing the ZmGCN5 antisense transcript. Treatment of maize with TSA caused
hyperacetylation of histones, increase in the transcription of histone mRNA and a
reduction in the expression of GCN5. The expression of antisense GCN5 transcript
caused hypoacetylation of histones, an increase in the transcript levels of histones and a
reduction in the expression of the HDAC RPD3 in maize [60]. Thus, the loss of a HAT or
HDAC in maize causes a shift in the balance of acetylated and unmodiﬁed histones. The

level of histone transcripts increased in both the treatments but did not result in an

10

increase in the protein levels of histones. The shift in the balance between acetylated and
total histones during TSA treatment is probably compensated by the degradation of
modiﬁed histones and replacement with de nova synthesized histones. In plants
expressing the antisense GCN5 transcript the balance is probably restored by the
replacement of unmodiﬁed histones with de nova acetylation of the newly synthesized
histones. The balance between modiﬁed and total histones in plants is maintained by
regulating the expression levels of HATS, HDACS and histones.

ZmGCN5 interacts with the coactivator ZmADA2 in viva as has been observed in
other eukaryotes [51, 61, 62]. ZmADA2 in turn interacts with the acidic activator Zm02
in viva as seen by Forster resonance energy transfer assay. The b-ZIP transcription factor
Zm02 controls the expression of genes involved in the regulation of seed storage proteins
[63-65]. It was hypothesized that the activator Zm02 recruits a GCN5-ADA containing
complex in order to activate target gene expression. In order to test this hypothesis
expression of the GUS reporter gene fused to the promoter of the Zm02 target gene b32
was tested in the presence of GCN 5 and ADA2. Results Showed that the expression of the
reporter gene by Zm02 was repressed in the presence of ZmGCN5 and ZmADA2. This
inhibition was relieved by the expression of a mutant form of ZmGCN5 that is unable to
interact with ZmADA2. Thus Zm02 and ZmGCN5 might compete for the binding of
ZmADA2 which in turn causes repression of the reporter gene in the presence of these
two coactivators. Thus, GCN5 seems to repress gene expression whereas ADA2 activates
it. GCN5 and ADA2 might have overlapping functions but may also work independently
of each other in plants [37]. GCN5 and ADA2 are parts of the same coactivator

complexes in animals but that may not be the case in plants. Determining the structure of

11

such coactivator complex/es will shed more light on the functions of these two coactivaor
proteins in plants.

Arabidapsis contains one homologue of the enzyme GCN5. The structure of
GCN5 can be divided into four domains. The N-terrninal domain of Arabidapsis as well
as maize GCN5 is longer than its yeast counterpart but Shorter than those found in ﬂies
and humans [50]. The function of the N-terrninal domain in Arabidapsis is unknown but
it is thought to contain the nuclear localization signal (N LS). The HAT domain of GCN5
is conserved in the plant and animal systems. The third domain is the ADA2 interaction
domain which facilitates the interaction of GCN5 with the ADA2 proteins in the
formation of the large coactivator complexes described in the animal systems.
Arabidapsis GCN5 can physically interact with the two ADA2 homologues ADA2a and
ADA2b in vitra. The ﬁill length, the N-terminal and middle domains of the ADA2
proteins can bind to the full length GCN5 protein. The middle regions of the ADA2
proteins are sufficient for binding the HAT domain of GCN5. In addition to this
interaction GCN5 can also acetylate the two ADA2 proteins in Arabidapsis in vitra and
may control their biological function [51]. This is the ﬁrst report of GCN5 acetylating the
ADA2 proteins and acting as a factor acetyltransferase (FAT) in Arabidapsis. The C-
terrninal domain of GCN5 is the bromodomain which is known to recognize acetylated
histone proteins in animal systems [66, 67] .

The GCN5 enzyme in Arabidapsis was ﬁrst shown to be required for the
activation of a reporter gene driven by the activator protein CBF in yeast [50]. The CBF
family of transcription factors is involved in controlling the cold-responsive (COR) gene

expression during the cold acclimation process in Arabidapsis [68]. GCN5 can physically

12

interact with the DNA binding domain of CBF in vitra. T-DNA insertion mutation in
GCN5 shows a delay in the activation and overall expression of COR genes activated by
CBF during cold acclimation [37]. CBF might recruit a GCN5-containing complex to the
COR gene promoters in order to activate COR gene expression during cold acclimation. It
is thought that GCN5 may control the expression of COR genes by acetylation of histones
at the COR gene promoters during cold acclimation.

GCN5 has also been shown to acetylate histones at the light-regulated genes
CABZ and RBCS-IA [54]. These genes are activated by the transcription factor HYS that
is involved in the light-Signaling network [69-72]. The expression of the light-regulated
genes is repressed in two HAT mutants taﬂ and gcn5. This effect is further enhanced in
the taﬂgcn5 double mutant. The expression of CABZ and RBCS-IA is controlled by both
GCN5 and TAF 1 but the 1AA3 gene is regulated only by GCN5 but not by TAF 1. Thus
the genes in response to the light stimulus are controlled by the concerted action of the
two HAT enzymes but they also work independently at some genes. Differences are also
observed in their ability to acetylate histones H3 and H4. The acetylation of histone H3 at
lysines 9 and 27 and of histone H4 is controlled by a combination of GCN5 and TAF 1
[54, 55]. GCN5 is solely responsible for the acetylation of histone H3 at lysine 14. Thus,
these two HATS seem partially redundant for the expression of light-responsive genes in
Arabidapsis. Control of gene expression by combination of HATS could be the norm in
Arabidapsis rather than an exception.

GCN 5 is required for root and shoot development in Arabidapsis. The TOPLESS
protein in Arabidapsis is responsible for the normal development of an embryo into shoot

and root [73, 74]. The (pl-1 mutant shows a double root phenotype that develops as a

13

result of the conversion of the apical shoot pole into a root pole. Mutation in the HAT
GCN5 was identiﬁed as a suppressor of the double root phenotype of 1721-1 [73]. GCN5 is
necessary for the formation of the root in place of the apical shoot in the tpl-l mutant but
is not required for the formation of the basal root. Thus the development of the Shoot and
roots in Arabidapsis depends on the interaction of HATS with the activator-repressor
proteins involved in the development of embryos. Whether GCN5 controls the formation
of the apical root by changes in the histone acetylation of these promoters is yet to be
determined.

Recently a GCN5-interacting protein, AtEML, was identiﬁed in Arabidapsis that
is upregulated in response to abiotic stresses like cold and salt [75]. GCN5 is required for
the activation of the reporter gene IacZ by AtEML in yeast. The histones at the promoter
of this reporter gene are hyperacetylated during the activation of the reporter gene. The
expression of AtEML in Arabidapsis is higher in the gcn5-1 loss-of-function mutant.
Thus GCN5 seems to act as a repressor of AtEML. GCN5 was shown to enhance the
AtEML driven transactivation of a reporter gene in yeast. The direct targets of AtEML in
Arabidapsis have not been identiﬁed as yet but it is hypothesized that GCN5 might
regulate the target genes by regulating the activity of the AtEML protein itself [75].

GCN5 has also been shown to be involved in the expression of genes forming the
ﬂoral meristem [76]. A T-DNA insertion mutation in the bromodomain of GCN5 gene in
Arabidapsis produced plants with homeotic mutations in ﬂowers as well as the
production of terminal ﬂowers. The mutant plants had a higher expression and an
expansion of the expression zones of two of the key regulatory genes WUS and AGA [76].

WUS is a direct activator of the AGA gene. An increase in the expression of WUS most

14

likely causes the increased expression of the AGA gene [77-79]. The overall acetylation
levels of histone H3 were lower in the gcn5 mutant as compared to wild type plants.
Whether the regulation of WUS and AGA by GCN5 is direct or indirect remain to be
investigated.

Microarray analysis indicated that a gcn5 mutant in Arabidapsis affected the
expression of 5% of the 8,200 genes involved in the study. Most of the affected genes
were upregulated in the gcn5 mutant suggesting that GCN5 acts to repress most of its
target genes. Whether the repression of these genes by GCN5 is direct or indirect is not
known. In contrast, GCN5 in animal systems is mostly involved in the activation of genes
and only a few examples of gene repression have been reported [80].

The preceding paragraphs indicate that GCN5 controls the expression of genes in
response to external stimuli particularly light and temperature. GCN5 seems to be
involved in gene repression in most of the cases rather than gene activation. Gene
repression by GCN5 seems to be more common in plants than in animal systems. GCN5
can also act as a factor acetyltransferase in addition to histone acetyltrasferases in
Arabidapsis. GCN5 can acetylate coactivator proteins ADA2a andADA2b in
Arabidapsis. ADA2a and ADA2b are found in separate complexes with GCN5 in
Drasaphila which serve distinct functions [81]. Similarly GCN5 in Arabidapsis might be
working with the two ADA2 proteins in different complexes to control the expression of
distinct genes under a variety of external or internal stimuli. Identifying GCN5 containing
coactivator complexes in plants will help us understand the function of GCN5 and the
proteins required for the biological function of GCN5.

1.3.2 p300/CBP

15

Plants are exposed to various abiotic stresses like high salinity, drought, cold, and
UV radiation. The exposure to UV radiation causes DNA damage in plants. The
transcription factor HY5 regulates the UV response in Arabidapsis [82]. Maize ecotypes
from high altitudes show higher adaptation to UV tolerance than ecotypes from the
temperate regions. The ecotypes with higher adaptation express chromatin remodeling
factors to higher levels. Lines that are more tolerant to UV Show an enrichment of
hyperacetylated of histones H3 and H4 after irradiation at the promoter and transcribed
regions of two UV-B upregulated genes [39]. This increase in acetylation is correlated
with an increase in transcript levels of the UV-B upregulated genes as well as loosening
of the chromatin structure as assessed by DNaseI and microccocal accessibility assay.
Thus, covalent histone modiﬁcations and chromatin remodeling seem to be working in
concert to regulate the induction of genes involved in UV-B tolerance. Chromatin
immunoprecipitation assays showed that CBP was associated with the promoters of these
genes. CBP is known to acetylate histones as well as non-histone proteins in animals and
plants [21, 52]. The CBP HAT associates with the promoter region in the UV-B tolerant
lines and is thought to have a direct role in the regulation of acetylation of histones in the
UV-B tolerant lines.

p300/CBP HATS have been identiﬁed in higher eukaryotes like ﬂies, mice and
humans. No members of the p300/CBP members have been identiﬁed in the yeast
Saccharamyces cerevisiae [21]. The ﬁve members of the p300/CBP family identiﬁed in
Arabidapsis known as HACl, HAC2, HAC4, HAC5 and HAC12 [83]. The primary
structures of the members of the HAC family have some regions of sequence Similarity

but there are also considerable differences. The members show sequence similarities

l6

between the cysteine/histidine domains 2 and 3 and the acetyltransferase domain.
Surprisingly all the members of the HAC family lack the bromodomain and the CREB
transcription factor binding domain which are hallmarks of p300/CBP family members in
animals [16]. Thus the function of the plant p300/CBP proteins may be different than in
animals.

T—DNA insertion mutations in all the ﬁve HAC members have been studied. The
MC] -1 mutant ﬂowered later than wild type plants [52, 53]. Gene expression analysis
Showed that the expression of FLC, the repressor of ﬂowering, was increased in this
mutant. Double mutants hathac5 and haclhac12 showed a further delay in ﬂowering as
compared to the had single mutant. Thus, the HAC], HAC5 and HAC12 HATS seem to
be at least partially redundant in the control of ﬂowering. The acetylation levels of
histones H3 and H4 at the FLC promoter in the hue mutants was no different than that in
wild type Col plants as tested by chromatin immunoprecipitation assays. The regulation
of FLC by the HACS is probably not through the direct acetylation of the FLC promoters.
Thus, the activation of F LC by HACS might occur via an indirect method by the
repression of a protein that acts as a repressor of FLC. It is possible that the HAC
enzymes regulate the function of the direct activator or repressor of FLC by modulating
the acetylation of those proteins which in turn control FLC expression. The HAC family
of acetyltrasferases is involved in the repression of FLC rather than its activation. This is
one more example of HATS acting to repress genes rather than activate genes in plants.

The biggest difference between p300/CBP enzymes in animal and plants is the
absence of the bromodomain in plants. Bromodomains are known to bind acetylated

histones in animals and help target bromodomain containing proteins to the target

17

promoters/ genes. Members of the other HAT families in Arabidapsis have a
bromodomain and thus the absence of bromodomain in the HAC family members might
provide them with speciﬁc ftmctions [52, 83].

It is possible that the HAC proteins might be the HATS that are responsible for the
ﬁrst round of acetylation of histones during gene expression. HAC members could be the
ﬁrst HATS recruited to promoters of genes that have hypoacetylated histones. The
acetylation of these histones by the HAC proteins might then recruit other HATS that
contain the bromodomain able to recognize acetylated histones. These HATS in turn can
bring about the full extent of acetylation required for gene activation. Thus the HAC
members might ‘poise’ the genes for activation by establishing a ‘histone code’.
Crosstalk between these and other histone modifying enzymes will result in the full
activation of the target genes.

HAC proteins might be part of coactivator complexes than contain bromodomain-
containing proteins. These bromodomain containing proteins might direct the recruitment
of the HAC members to the appropriate target genes. HAC members in turn acetylate
histones at the target gene promoters and thus activate gene expression.

It is also likely that the HAC proteins in plants act primarily as FATS instead of
HATS. In this case the recruitment of the HACS to the target proteins will not require the
presence of a bromodomain. p300/CBP HATS in animals are known to acetylate a vast
array of non-histone proteins and regulate the activity of those proteins. The HAC
members in plants might serve the same purpose.

Arabidapsis is the only eukaryote with ﬁve members in the HAC family. In order

to further our understanding of the biological role of the HAC members it is also

18

important to know whether the ﬁve members act independently to control gene
expression or are functionally redundant.

1.3.3 TAFl

The TAF 1 family in Arabidapsis has two members, HAF 1 and HAF2/TAF1. In contrast,
animals only have a single member in the TAF 1 family [55]. T-DNA insertion mutants in
the two genes HAFI and HAF2 were analyzed in order to gain insight into the function of
these genes in Arabidapsis growth and development. T-DNA insertion in the HAFI gene
did not result in any aberrant phenotype. The haf2 mutants were paler than wild type
plants due to a 50% reduction in their chlorophyll content as compared to wild type
plants. The HAF2/TAF1 protein in Arabidapsis contains an ubiquitin, a HAT and a
bromodomain. Analysis of the taﬂ mutant Showed that TAF l is involved in controlling
the expression of light regulated genes such as RBCSI and CABZ. Analysis of the
acetylation of histones H3 and H4 showed that these histones were hypoacetylated at the
RBCSI and CABZ promoters in the taﬂ mutants as compared to wild type [54]. The
acetylation of histones H3 and H4 was dependent on the activity of GCN5 in addition to
TAF]. TAF 1 was solely responsible for the acetylation of lysine 9 of histone H3 at the
target gene promoters. Thus once again we observe an effect on gene expression which
depends upon the concerted effort of more than a single HAT protein.

Microarray analysis Showed that the expression of about 3% of Arabidapsis genes
were affected in the taﬂ mutant and almost an equal number of genes were up or down
regulated. TAF 1 could be regulating gene expression in Arabidapsis by acetylating
histones at the promoters of target genes. This in turn might govern the recruitment of the

TFIID complex to the gene promoters which then leads to activation of transcription.

19

Loss of TAF 1 results in the repression of light responsive genes and this is
associated with deacetylation of histone H3 and H4. The repression of these genes is
exacerbated by the absence of another HAT GCN5. Members of GCN5 containing
coactivator complexes are known to interact with members of the TFIID complex in
animal systems [84, 85]. It will be interesting to ﬁnd out whether such an interaction
takes place in plants. HY5 controls the expression of the light-regulated genes but
whether it recruits the GCN5 and TAF containing complexes remains to be investigated.

The TAF 1 family in Arabidapsis contains two members designated HAFI and
TAF 1. Whether both the proteins are part of the same TFIID complex in Arabidapsis or
whether they are components of two different complexes is not known. The presence of
two TAF 1 members is a unique feature of the plant TAF 1 HAT family [55, 76]. The
function of the HAFl enzyme in Arabidapsis is unknown and the haﬂ mutant does not
Show any visible phenotype. It is possible that the functions of the HAFl and TAF 1
enzymes are partially redundant. To unravel the biological functions of these two proteins
is an important next step. It will be interesting to ﬁnd out whether the haﬂ taﬁ double
mutant has an effect on the normal growth and deve10pment of Arabidapsis. Answering
these questions would help elucidate the biological functions of these enzymes in plants.
1.4 Novel histone acetyltrasferases in Arabidapsis

The X—ray crystal structure and NMR solution structure of a histone
acetyltransferase belonging to the GNAT superfamily in Arabidapsis was recently
described by Tyler et al [86]. The At1g7750 gene product has an acetyltransferase
domain that is considerably smaller than the other members of the GNAT family. Thus,

this may be a ‘minimal’ acetyltransferase that requires the help of a partner protein for its

20

AT function. The rate of acetylation of histones by this enzyme is lower than for other
members and hence histones may not be the physiological substrates. The function of this
protein in Arabidapsis is unknown.

A nuclear Shuttle protein interactor from Arabidapsis (AtNSI) has been shown to
have histone acetyltransferase activity [87]. This protein interacts with the movement
protein nuclear shuttle protein (N SP) that is required for the movement of virus particles
from the nucleus to the cytoplasm and neighboring cells. This protein, encoded by certain
plant viruses, is required for the establishment of viral infection in plants. NSP interacts
.with the viral coat protein (CP), sequesters the replicated genome and shuttles it to the
cytoplasm. Although AtNSI interacts with NSP and not with CP, it is able to acetylate CP
and is thought to function in modulating the interaction between NSP and CP. The AtN SI
acetyltransferase domain belongs to the GNAT family but is only distantly related to the
well known members like the GCN5 from yeast, ﬂies and Arabidapsis. The residues
essential for the acetytransferase activity were conserved in the AtN SI. The AtNSI was
able to acetylate histones H3, H2A and very weakly histone H4 in vitra. The AtNSI lacks
a bromodomain and does not function as a transcriptional coactivator as observed by in
vitra assays.

Thus, plants encode for a large number of acetyltransferases but only 12 of those
are, to date, recognized as legitimate histone acetyltransferases or as likely homologs of
known HATS. These unclassiﬁed proteins with acetyltransferase activity might
exclusively act as factor acetyltransferases and ﬁmction in specialized pathways in plants.
Determining the proteins they acetylate and the pathways in which they function will

help further our understanding of their biological function in plants.

21

1.5 Hypothesis and experimental design

The main aim of this doctoral thesis work is to investigate the role of the
coactivator proteins ADA2b and GCN5 in cold-regulated (COR) gene expression during
the cold acclimation process in Arabidapsis. Members of the CBF family of activator
proteins (CBFI , CBF2 and CBF3) bind to cis-elements in the promoters of the COR
genes and activate their expression during the cold acclimation process. ADA2b and
GCN5 can physically interact with the DNA binding domain of CBF in vitra. The
expression of COR genes during cold acclimation was delayed and reduced in the ada2b
and gcn5 mutants.

It was hypothesized that CBF recruits ADA2b and/or GCN5 containing complex
to the COR gene promoters during cold acclimation to activate the expression of COR
genes. GCN5 then acetylates histones at the COR gene promoters and activates the
expression of COR genes. Chromatin immunoprecipitation assays were performed using
antibodies that speciﬁcally recognize acetylated histone H3 (at lysine 9 and/or 14) in
order to determine the levels of acetylated histone H3 at the COR gene promoters before
and after cold acclimation. Hyperacetylation of histone H3 at COR gene promoters was
observed in wild type plants upon cold acclimation and CBF was sufﬁcient to drive this
acetylation. Hyperacetylation of histone H3 at COR gene promoters was also observed in
the ada2b and gcn5 mutants. Therefore, GCN5 is not the HAT responsible for acetylation
of histone H3 at the COR gene promoters during cold acclimation. Surprisingly, the
occupancy of histone H3 at the COR gene promoters in the ada2b and gcn5 mutants did
not decrease during cold acclimation as observed in wild type plants. ADA2b and GCN5

are required for the removal of nucleosomes from the COR gene promoters but are

22

dispensable for the histone H3 acetylation. Thus an ADA2b-GCN5containing complex
might be required for the recruitment of a chromatin remodeling complex which clears
nucleosomes off the COR gene promoters.

The Arabidapsis genome encodes ll HAT enzymes in addition to GCN5 and any
of these HATS may be responsible for acetylation of histone H3 at the COR gene
promoters. A few candidate HATS from the HAC and TAF 1 families were tested for their
ability to regulate COR gene expression by acetylating histone H3 at the COR gene
promoters. Results showed that all the HAT enzymes tested were dispensable for the
acetylation of histone H3 at the COR gene promoters. Thus, acetylation of histone H3 at
the COR gene promoters is not dependent on any of the individual HATS tested.

In order to test whether the CBF driven expression of COR genes requires
ADA2b and/or GCN5 two approaches were utilized. The expression of COR genes was
compared in Wt, ada2b and gcn5 plants overexpressing CBF. In the second approach the
expression of the reporter gene GUS fused to four tandem repeats of CRT was compared
in Wt, ada2b and gcn5 plants after cold acclimation. Results Showed that the CBF-
dependent expression of COR genes via CRTs does not require ADA2b and GCN5.

We can conclude from the results that GCN5 is not the HAT responsible for the
acetylation of histone H3 at the COR gene promoters. GCN5 and ADA2b are required for
the removal of nucleosomes from the COR gene promoters during cold acclimation.
GCN5 and ADA2b are not required for the CBF-dependent expression of COR genes and

probably regulate COR gene expression indirectly.

23

Chapter II

Increase in acetylation of histone H3 corresponds to activation of COR genes during

cold acclimation.

2.1 Introduction:

The presence of chromatin in eukaryotes serves as a barrier to transcription [88].
Eukaryotes have devised ways to facilitate gene expression even in the presence of
chromatin. Transcriptional activators that bind to speciﬁc cis-elements present in the
promoters of genes speciﬁcally recruit the general transcription machinery and thus
facilitate transcription of target genes [89, 90]. Activators also recruit proteins known as
chromatin modifying proteins that can aid the process of transcription. There are two
types of chromatin modifying factors. The ATP-dependent coactivator proteins include
SWI/SNF proteins that are able to slide or remove nucleosomes and thus allow
transcription to proceed. The second type of chromatin remodeling proteins are enzymes
that covalently modify histones by acetylation, methylation, ubiquitination or
phosphorylation that are either associated with transcriptional activation or repression or

both .

Cold acclimation is the process by which plants can increase their freezing
tolerance when exposed to low non-freezing temperatures [91]. The CBF family of
transcription factors acts as ‘master regulators’ during the process of cold acclimation.
The transcription of the CBF genes (CBFl, 2 and 3) is induced within 15 minutes of
exposure to cold temperatures [92-94]. The CBF proteins bind speciﬁc cis-elements

known as the cold/dehydration responsive elements (CRT/DRES) in the promoters of

24

their target genes [95-97]. The CBF proteins activate their target genes, including the
cold-regulated (COR) genes, at about 2-4 hours after exposure to cold temperature [98].
Overexpression of CBF results in constitutive expression of the COR genes and these
plants are also constitutively freezing tolerant [93, 94]. CBF l , a transcriptional activator,
can physically interact with at least two co-activator proteins, ADA2b and GCN5 (a
histone acetyltransferase) [3 7, 50, 51]. ADA2 and GCN5 are components of large multi-
subunit co-activator complexes in eukaryotic organisms like yeast, ﬂies and humans but
the function of these enzymes in Arabidapsis is not completely known as yet [61, 99-
103]. ADA2 is able to enhance the ability of GCN5 to acetylate nucleosomal histones
[104, 105]. Arabidapsis GCN5 is able to acetylate histone H3 in vitra speciﬁcally at
lysine 9 and 14 [106]. Homozygous T-DNA insertion mutants of GCN5 (gcn5-1) and
ADA2b(adaZb-1) Show pleiotropic phenotypes with defects in grth and development.
gcn5-1 and adaZb-I mutant plants Show delayed activation and reduced expression of
COR genes during cold acclimation. The freezing tolerance of ada2b-1 and gcn5-1 plants
is similar to wild type plants but their ability to activate COR gene expression is
compromised. Thus we hypothesized that upon cold acclimation CBF recruits to the COR
gene promoters a coactivator complex containing ADA2b and GCN5. This co-activator
complex in turn acetylates histone H3 at the promoters of COR genes and thus facilitates

the activation of COR gene upon cold acclimation.

This chapter will answer three important questions related to the regulation of
COR gene expression during cold acclimation. The ﬁrst question is whether acetylation
of histone H3 increases at the COR gene promoters during cold acclimation. The second

question that will be addressed is if CBF is necessary and/or sufﬁcient to drive histone

25

H3 acetylation at the COR gene promoters. The last question that will be addressed is
whether GCN5 and/or ADA2b are necessary to bring about acetylation of histone H3 at

the COR gene promoters.

To answer these questions chromatin immunoprecipitation assays were carried
out using antibodies against acetylated histone H3 (acetylated lysine 9, lysine 14) and an
antibody that recognizes the C-terminal region of histone H3 irrespective of the
modiﬁcations on the N—terminal tail.

2.2 Materials and methods:
2.2.1 Plants used and growth conditions

Wild type plants of the Wassilewskija-2 (WS) ecotype were used in this study.
ada2b-I and gcn5-1 plants obtained from the Wisconsin T-DNA mutant collection and
characterized in the Triezenberg lab were used [3 7] . Seeds for the CBFI over-expressing
plants and plants expressing a C-terrninal truncation mutant, CBF2AC, in the Ws
background were obtained from the Thomashow lab (Gilmour, unpublished data). Seeds
were sterilized using a 40 % bleach solution containing 1 % Triton-X 100 for 10 minutes
at room temperature. The seeds were washed 3 times with sterilized MilliQ water for 10
minutes each. The seeds were stratiﬁed at 4°C for 4 days before plating out. Plants were
grown on Garnborg’s basal medium with sucrose and 1 % phyta agar [107, 108]. Plants
were grown in a growth chamber at a constant temperature of 22°C and continuous lights

at 120 umol/mzsec.

2.2.2 Cold acclimation

26

Cold treatment of plants was carried out in chambers maintained at a temperature
of 4°C and continuous lights at 120 umol/mzsec. Plants were treated for either 4 h or 24 h.

2.2.3 Tissue harvesting and storage

Whole seedlings (0.65 g tissue per IP) were harvested from plates using tweezers
and transferred to a ﬂask containing a solution of 1 % formaldehyde. Aﬁer cross linking
plants were stored in 50 ml tubes, ﬂash frozen in liquid nitrogen and stored at -80°C until

further use.

2.2.4 Chromatin immunoprecipitation assay

The chromatin immunoprecipitation assays were adapted from previously

published methods [109, 110]

15-day old seedlings (before bolting) grown on plates were harvested in 1 %
formaldehyde and placed under vacuum for 25 minutes. Glycine was added to ﬁnal
concentration of 0.125 M to quench the reaction and placed under vacuum for 5 minutes.

The tissue was rinsed with water twice, frozen in liquid nitrogen and stored at —80°C.

The tissue was ground in liquid nitrogen (to keep it cold) using mortar and pestle.
The tissue was resuspended in Solution I (0.4 M sucrose, 10 mM Tris pH 8 and 10 mM
MgC12) and ground again. The tissue was then ﬁltered through four layers of cheesecloth
and two layers of Miracloth into Oakridge tubes and spun in an 8834 rotor at 10,000 rpm
for 10 minutes. The supernatant was discarded; the pellet was resuspended in Solution 11
(0.4 M sucrose, 10 mM Tris pH 8, 10 mM Mng and 1% Triton X-100) and spun at
10,000 rpm for 10 minutes. The supernatant was discarded; the pellet was homogenized
(hand dounced using a homogenizer) in solution III (1.7 M sucrose, 10 mM Tris pH 8, 2

27

mM MgClz and 0.15% Triton X-100) and the homogenate was layered over equal volume
of Solution III. The tube was centrifuged at 15,000 rpm for 30 minutes after which the
homogenate was poured off. A protease inhibitor solution containing 5 mM [3-
mercaptoethanol, 0.1 mM phenylmethanesulphonylﬂuoride and EDTA-free complete
protease inhibitor tablets ﬁom Roche was added to solutions I, II and 111 just before use.

The pellet was resuspended in sonication buffer (1% sodium dodecyl sulfate, 10
mM EDTA and 50 mM Tris pH 8.1) using, 150 [1.1 buffer per 0.65 g of original tissue,
and kept on ice 5-15 minutes. A protease inhibitor solution containing 0.5 mM
phenylmethanesulphonylﬂuoride, 100 ug/ml leupeptin and 100 ug/ml aprotinin was
added to the sonication buffer just before use. The solution was sonicated 5 times for 20
sec (with a Microtip sonicator, on continuous cycle, output control 3), rested on ice for 5
min between pulses and spun in microfuge at full speed for 10 min at 4°C. 100 pl aliquot
of the supernatant was taken as size sample (to run on gel to assess shear size). VP16
viral plasmid DNA (to assess non-speciﬁc binding of antibody) was added to the rest to a
ﬁnal concentration of 2.6 pg/ul. An aliquot (30 ul) of this input DNA was taken and the
Size and the input aliquots were stored at -20°C.

Protein G beads (50% slurry in Triton-X lysis buffer with 10 uyml salmon sperm
DNA) were added using 60 ul protein G / 1 ml extract and incubated at 4°C for 2 hours
on rotator. Sample was centrifuged at 5000 rpm for 5 min; and 900 pl Triton-X lysis
buffer (50 mM Tris-HCl pH 7.4, 150 mM sodium chloride, 1% Triton X—100 and 1%
bovine serum albtunin) was added per 150 pl supernatant for each sample. Antibodies
were added to appropriate samples and incubated for 8 hours at 4°C on rotator. 40 111 of

protein G beads was added and samples incubated overnight at 4°C on rotator.

28

Samples were centrifuged at 5000 rpm for 5 min at 4°C and beads were then
washed with buffers (Triton wash buffer - 50 mM Tris-HCl pH 7.4, 150 mM sodium
chloride and 1% Triton X-100, Lysis buffer 500 - 0.1% sodium deoxycholate, 1 mM
EDTA, 50 mM HEPES pH 7.5, 500 mM sodium chloride, LiCl/detergent — 0.5% sodium
deoxycholate, 1 mM EDTA, 250 mM LiCl, 0.5% IGEPAL CA-630 and 10 mM Tris pH
8.0, TB pH 8.0) for 10 minutes on rotator at RT followed by 2 minute spin at 5000 rpm.
100 pl elution buffer (1.0 mM EDTA, 1% SDS and 50 mM Tris pH 8) was added to beads
and incubated at 65°C for 10 minutes (inverting several times during incubation). The
samples were centrifuged 2 min at 5000 rpm and the supernatant was transferred to fresh
tubes. 150 pl TE/0.67 % SDS was added to beads, inverted and centrifuged at top speed
for 1 min and then combined with ﬁrst elution. The ﬁnal volume was 250 pl for each IP,
input (30 pl input + 220 pl water) and size (100 pl size sample + 150 pl water) Samples.
To each tube 10 pl of 5M NaCl and 1 pl RNaseA was added and incubated at 65°C
ovemight.

DNA was precipitated with 2 volumes of ethanol (incubated at —20°C for 15
minutes, centrifuged at 4°C in microfuge at full speed for 15 minutes) and the DNA
pellet was resuspended in 100 pl TE pH 7.5, 25 pl proteinase K buffer (1.25% SDS, 50
mM TrSi-HCl pH 8) and 1 pl proteinase K. Samples were incubated at 42°C for 2 hours
and the DNA was extracted with 25:24:] (v/v) phenol: chloroform: isoarnyl alcohol
(equal volume, centrifuged in microfuge at full speed for 3 minutes). The upper phase
was transferred to a fresh tube and DNA was precipitated with 1/10 volume NaOAc pH
5.2 and 2 voltnnes 100% ethanol (incubated at —-20°C for 15 minutes, spun at 4°C in

microfuge at full speed for 30 minutes). The pellet was washed with 250 pl 75% ethanol

29

once and resuspended in 80 pl nuclease free water (except size sample in 10 pl to run on

gel). 3 pl of DNA samples were used per PCR reaction of 30 p1.

2.2.5 Antibodies

Commercially available antibodies were used for chromatin immunoprecipitation
assays. The anti-acetyl-H3 antibody (rabbit polyclonal IgG, #06-599) was purchased
from Upstate Biotechnology and the anti-histone H3 antibody (rabbit polyclonal IgG,
abl791) was purchased from Abcam Inc. 2 pl of the anti-acetyl-H3 antibody and 1 pl of
the anti-histone H3 antibody was used per 0.65 g of plant tissue.
2.2.6 PCR conditions and data analysis

Real-time PCR was carried out using 10X buffer containing SYBR Green, 0.25
pM of each primer, 3 mM MgC12, 12.4 mM dNTPs and 0.15 pl Ampli Taq Gold DNA
polymerase enzyme (all reagents were obtained ﬁ'om ABI) on the puriﬁed DNA samples.
Input samples were diluted and used as a standard curve. The amount of DNA pulled
down by the acetylated H3, the C-terminal H3 antibodies and the no-antibody control
were calculated using the input standard curve. PCRS were carried out using primers that
span the CRT region of the COR gene promoters (COR6. 6, COR15a, COR4 7, COR 78
and G013). The ACTZ/ 7 promoter region was used as a control. The % input values
obtained for the no-antibody control were subtracted from the values obtained for the
anti-acetylated histone H3 IP and the anti-histone H3 IP. The % input values thus
obtained for the COR gene promoters were normalized to the values obtained for the
ACTZ/ 7 promoter. The values were expressed as a ratio of normalized acetylated histone
H3 to normalized total histone H3 or as level of normalized total histone H3 at the COR

gene promoters. Values for the ACTZ/ 7 promoter were not normalized and were

30

expressed as a ratio of acetylated histone H3 to total histone H3 or as total histone H3.
Data were analyzed using single-factor ANOVA using Microsoft Excel 2007 software.
To determine whether the difference between two values is signiﬁcant, the least
Signiﬁcant difference (LSD) method at p S 0.05 was used.

2.3 Results:

2.3.1 Acetylation of histone H3 at the COR gene promoters increases during cold
acclimation.

Activation of genes is sometimes also associated with the removal of histones in
addition to histone modiﬁcations and thus when analyzing the data it is also important to
monitor the changes in the level of total histone H3 during COR gene expression upon
cold acclimation. To test if the acetylation levels of histone H3 at the COR gene promoter
change upon cold acclimation, chromatin immunoprecipitation (ChIP) assays were
carried out in Ws wild type plants. Plants were cold treated at 4°C for 4 or 24 hours.
Plants grown at 22°C were used as controls. ChIP assays were carried out using two
different antibodies. One antibody recognizes histone H3 acetylated at lysines 9, 14 or
both. The second antibody recognizes the C-terminal region of histone H3 irrespective of
the modiﬁcations on the N-terminal tail. The changes in the acetylation of histone H3 at
the COR gene promoters are represented as a ratio of acetylated histone H3 to total
histone H3 to account for changes occurring in the total level of histone H3 during COR

gene expression.

Results (Figure 1) Show that the acetylation of histone H3 at the COR gene

promoters increased at least two-fold upon cold acclimation in wild type Ws plants as

31

 

 

   

 

 

 

 

 

 

   

2 ] ACT2/7 promoter m 1.2 COR6.6 promoter
.4 a 3
g E 0.8 - b b
M 0
% <2
< E 0 4 ‘ a
'3
E 0.0 -
o 4 24 8 0 4 24
Time in h at 4°C Time in h at 4°C
«4 0-3 ' com 5a promoter .., 1 2 C0R47 13mm"tcr
$0.6 - b § ° b b
20.4 - E 0.8 '
'° 1:
§ 0.2 - a N 0.4 -
a ‘3
€00 ' E 0.0 .
i: 0 . . 4 24 a 0 . .4 24
T1me m h at 4°C T1me 1n h at 4°C
COR78 romoter _ GolS romoter
m 1.2 1 P .., 2-5 p b/c
E b b § 2.0 -
r£0.13 - m 1 5 , b
o o .
<1 <2 _
80.4 - . g 1'0
.E a u— 0.5 "' a
'8 '3
goo 4 E00 -
8 o 4 24 8 0 4 24
Time in h at 4°C Time in h at 4°C

Figure 1: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type plants cold
acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc for acetylated histone
H3 (K9, K14) and the C-terminal of histone H3. Samples were analyzed using real-time
PCR to quantify promoter DNA enriched for acetylated histone H3 and total histone H3
at various COR gene promoters. Ct values were normalized to values obtained for
ACT2/7 promoter. Data were represented as ratio of acetylated histone H3 to total histone
H3 at various COR gene promoters. Values at the AC T2/ 7 promoter were not normalized
and were represented as ratio of acetylated histone H3 to total histone H3. Error bars
represent SE where N (number of experiments) = 5. Data were analyzed using single
factor ANOVA. Values of bar graphs denoted by the same letter are not signiﬁcantly
different as determined by LSD at p S 0.05.

32

compared to non-acclimated plants. The increase in acetylation occurred within 4 hours

of cold treatment at all the COR gene promoters. Histone H3 remained hyperacetylated

even after 24 hours of cold acclimation. The acetylation levels of histone H3 at the ACT2

promoter did not change signiﬁcantly upon cold acclimation.

H3 histones at the COR gene promoters are hypoacetylated in non-acclimated
plants when then COR are inactive. Upon cold acclimation H3 histones are
hyperactetylated and this increase in acetylation at the COR gene promoters corresponds
to the activation of COR genes. We can conclude ﬁom the results that the acetylation of

histone H3 at all COR gene promoters increase during cold acclimation.

2.3.2 The level of total histone H3 at some COR gene promoters decreases upon cold

acclimation.

Activation of gene expression involves chromatin remodeling in addition to
covalent histone modiﬁcations. Chromatin remodeling enzymes are known to slide or
remove nucleosomes from genes in order to facilitate transcription activation. To test if
nucleosome removal occurs at the COR promoters, the changes in level of histone H3
were monitored during cold acclimation. Chromatin immunoprecipitation (ChIP) assays
were carried out in Ws wild type plants that were cold treated at 4°C for 4 or 24 hours
and plants grown at 22°C were used as controls. ChIP assays were carried out using an
antibody that recognizes the C-terminal of histone H3 irrespective of the modiﬁcations on

the N-terrninal tail.

Results (Figure 2) show that the occupancy of histone H3 at COR6. 6, COR 78 and
0015 promoters decreased upon cold acclimation. The occupancy of histone H3 at these

33

COR gene promoters decreased Signiﬁcantly after 24 hours of cold treatment. The level
of histone H3 at the ACT2, COR15a and COR47 promoters remained unchanged after

cold acclimation.

The activation of a subset of COR genes involved removal of nucleosomes as
seen by the decrease in the level of total histone H3 during cold acclimation. This
removal of nucleosomes was only evident after 24 hours of cold acclimation (Figure 2),
whereas the acetylation of histone H3 was evident as early as 4 hours after cold
acclimation. We conclude that histone H3 acetylation at the COR gene promoters
precedes removal or loss of nucleosomes at a subset of COR gene promoters. However,
removal is not a certain consequence of acetylation, since acetylation occurs at promoters
where no histone loss was evident. Moreover, these results suggest that even though the
activation proﬁle of these COR genes was similar, subtle differences may exist in the

manner in which the individual COR genes are regulated.

2.3.3 Levels of acetylated histone H3 and total histone H3 are restored to non-acclimated

levels within 24 hours of de-acclimation.

Activation of COR gene expression is associated with both an increase in histone
H3 acetylation as well as decrease in histone H3. De—acclimation of plants leads to a
decrease in COR gene expression back to non-acclimated levels within 24 hours [1 l 1].
To test whether deacclimation also affects the levels of acetylated histone H3, ChIP
assays were carried out in non-acclimated, cold acclimated and de-acclimated Ws wild
type plants. For de-acclimation, the plants were ﬁrst cold acclimated at 4°C for 24 hours

and then placed at 22°C for 24 hours. ChIP assays were carried out using an antibody that

34

12 ACT2/7 promoter
10 a
E; 8 a a
E 6
o\° 4
2
0
4 24
Time in h at 4°C
,_, 1.5 COR15a promoter
3 a
2‘ a
°\° 1.0 a
1:1
3 o 5
E .
E
c: 0.0
0 4 24
Time in h at 4°C
31.2 COR78 promoter
.3
X 0.8
-o
8
:3 0.4
E
s: 0.0

0 4
Time in h at 4°C

 

24

     

 

... 1.5 COR6.6 promoter

3 a

.E l 0 a

°\° .

1: b
B

'7; 0.5

E

s: 0.0

0 24
Time mill at 4°C

... 1.6 COR47 promoter
3 a
.S 1.2 a
°\°
3 0.8
.5
'5 0.4

E

8 0.0

0 4 24
Time in h at 4°C

...1-6 GolS promoter

8.

.E 1.2 a

 

0 4 24
Time in h at 4°C

Figure 2: Level of total histone H3 at COR gene promoters decreases upon cold
acclimation. ChIP assays were performed on wild-type plants cold acclimated at 4°C for
0, 4 and 24 hours using an antibody speciﬁc for the C-terrninal of histone H3. Samples
were analyzed using real-time PCR to quantify promoter DNA enriched for total histone
H3 at various COR gene promoters. Ct values were normalized to values obtained for
ACT2/7 promoter. Data were represented as normalized % input levels of total histone H3
at various COR gene promoters. Ct values for the ACT2/ 7 promoter were not normalized
and were represented as % input. Error bars represent SE where N (number of
experiments) = 5. Data were analyzed using single factor ANOVA. Values of bar graphs
denoted by the same letter are not signiﬁcantly different as determined by LSD at p S

0.05.

35

recognizes acetylated histone H3 and an antibody that recognizes the C-terrninal region

of histone H3 irrespective of the modiﬁcations on the N-terminal tail.

The results in ﬁgure 3 Show that H3 histones at the COR gene promoters were
hypoacetylated in non-acclimated plants. Within four hours of cold acclimation H3
histones at the COR promoters were hyperacetylated and high levels of acetylated histone
were maintained even after 24 hours of cold acclimation. There was at least a two-fold
increase in acetylation of histone H3 at the COR gene promoters in the acclimated plants
as compared to the non-acclimated plants. Upon de-acclimation the H3 histones were

deacetylated and acetylation levels were restored to non-acclimated levels after 24 h.

Thus, activation of COR genes in wild type plants is associated with an increase
in acetylation of histone H3 whereas the repression of COR genes during de-acclimation

is linked to a decrease in the acetylation of histone H3 at the COR gene promoters.

The histone H3 occupancy at COR gene promoters during de-acclimation was
also tested. Results (Figure 4) show that the occupancy of histone H3 decreased
signiﬁcantly at the COR6. 6, COR 78 and GalS promoters upon 24 hours of cold
acclimation. When plants were de-acclimated the levels of histone H3 were restored to
non-acclimated levels at these promoters. The levels of histone H3 occupancy at the
ACT2/ 7, COR! 5a and COR4 7 did not decrease upon cold acclimation or de-acclimation
as compared to non-acclimated plants. Thus, activation of some COR genes in wild type
plants is associated with a decrease in histone H3 occupancy whereas the repression of
COR genes during de-acclimation leads to the restoration of histone H3 at the target COR

gene promoters.

36

 

    

 

 

 

3 ' ACT2/7 promoter m 1'0 ' COR6'6 promoter
.4 a g b
a 2 ‘ a g b
m a g 0.5 d l
o l q a
< a

'3 a
0 ‘ EO'O .._- If I
0 4 24 24-24 ‘= 0 4 24 2424

Time in h at 4°C Time in h at 4°C

    

 

 

 

 

   

0.8 - .
g CORl 5a promoter 3 1 5 COR47 prorgroter
E b $1.0 -
o o b
<0.4 - b <2
'0
E E 0'5 1 a a
'3 a a '3
§o.o - goo -
° 0 , 4 , 24 24-24 = 0 . 4 . 24 24-24
T1me mhat 4°C T1me mhat 4°C

3 1-5 ' COR78 promoter 3 3.0 - GolS promoter
E 1.0 - E
O 0
<1 <

. 'U
E 0 5
'3
g 0.0 ~ E
8 0 4 24 24-24 8 0 4 24 24-24

Time in h at 4°C Time in h at 4°C

Figure 3: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation and is restored upon de-acclimation. ChIP assays were
performed on wild-type plants cold acclimated at 4°C for 0, 4 and 24 hours using
antibodies speciﬁc for acetylated histone H3 (K9, K14) and the C-terrninal of histone H3.
One set of plants was acclimated at 4°C for 24 hours and then de-acclimated by placing
them at 22°C for 24 hours. Samples were analyzed using real-time PCR to quantify
promoter DNA enriched for acetylated histone H3 and total histone H3 at various COR
gene promoters. Ct values for the COR promoters were normalized to the Ct values of
AC T2/ 7 promoter. Data were represented as ratio of acetylated histone H3 to total histone
H3 at various COR gene promoters. Ct values obtained for the AC T 2/ 7 promoter were not
normalized and were represented as a ration of acetylated histone H3 to total histone H3.
Error bars represent SE where N (number of experiments) = 2, except COR 78 where N =
1. Data were analyzed using single factor AN OVA. Values of bar graphs denoted by the
same letter are not signiﬁcantly different as determined by LSD at p S 0.05.

37

  

 

 

 

 

          

 

 

 

 

   

a COR6.6 promoter
a
a b
.E
°\°
0 4 24 24.24 0 4 24 2424
Time in h at 4°C Time in h at 4°C
.5 2-0 ‘ CORISa promoter a .5 2-0 ‘ COR47 promoter
a
E; 1.5 - a 31.5 - a 3
0° °\° a
g 1.0 - a '8 1.0 -
.E ..‘S
a 0.5 - 1g 0.5 .
g 0.0 - g 0.0 '
0 4 24 24-24 0 4 , 24 2424
Time in h at 4°C Time in h at 4°C
21.2 “ COR78 promoter 451.6 T 6018 promoter
.5 0 8 _ 21.2 -
o\° ' o\°
B v50.8 "
.s 0.4 - .§
'3 760.4 -
50.0 -‘ $0.0 '
o 4 24 24-24 0 4 24 2424
Time in h at 4°C Time in h at 4°C

Figure 4: Levels of total histone H3 at some COR gene promoters decreases upon cold
acclimation and is restored upon de-acclimation. ChIP assays were performed on wild-
type plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc for the
C-terminal of histone H3. One set of plants was acclimated at 4°C for 24 hours and then
de-acclimated by placing them at 22°C for 24 hours. Samples were analyzed using real-
time PCR to quantify promoter DNA enriched for acetylated histone H3 and total histone
H3 at various COR gene promoters. Ct values were normalized to values obtained for
ACT 2/ 7 promoter. Data represented as levels of total histone H3 at various COR gene
promoters. The Ct values for the ACT2/ 7 promoter were not normalized and represented
as % input values for total histone H3. Error bars represent SE where N (number of
experiments) = 2, except for the COR 78 promoter where N=l. Data were analyzed using
single factor AN OVA. Values of bar graphs denoted by the same letter are not
signiﬁcantly different as determined by LSD at p 50.05.

38

One unexplained discrepancy was observed in the experiments using de-
acclimated plants. The level of total histone H3 at the ACT 2/7 promoter was signiﬁcantly
diﬁ'erent in the acclimated (24 h sample) and de-acclimated samples, compared to the 4 h
acclimated sample and non-acclimated control. This effect was not observed in the
comparable experiment shown in Figure 2. Compared to the previous experiment (ﬁgure
2) the value of total histone H3 in the non-acclimated control was twofold lower. The
values of total histone H3 in the non-acclimated and the 4 h acclimated samples in the
two independent experiments (ﬁgure 4) were identical and the samples showed no
variation. This could account for the difference between these samples and the 24 h

acclimated and de-acclimated samples.

We conclude that the activation of COR genes involves both an increase in the
acetylation of histone H3 as well as a removal of nucleosomes (as observed by the
removal of histone H3) during cold acclimation. Both these marks of activation are
transient and are restored to basal levels once the expression of COR genes decreases (as

observed by the de-acclimation experiment).

2.3.4 CBF is sufﬁcient for the acetylation and removal of histone H3 at the COR gene

promoters.

Overexpression of CBF in wild type plant leads to constitutive expression of COR
genes. These plants are also constitutively freezing tolerant [94]. Results in section 2.3.1
show that the expression of COR genes in wild type plants upon cold acclimation is
associated with hyperacetylation of histone H3. Results in section 2.3.2 show that

expression of some COR genes is also associated with a loss of histone H3 occupancy. It

39

is possible that these marks of gene activation occur at the COR gene promoters in the
CBF overexpressing plants even without cold acclimation. To test whether CBF is
sufﬁcient to drive the acetylation and/or removal of histone H3 at the COR gene
promoters, ChIP assays were carried out using plants that overexpress CBFl. Plants
overexpressing CBF] were cold acclimated at 4°C for 4 or 24 hours. Plants grown at 22°C
were used as controls. ChIP assays were performed using the anti-acetylated histone H3

antibody and the anti-histone H3 antibody.

The levels of acetylated histone H3 and total histone H3 for the plants
overexpressing CBF 1 were compared to the results obtained for Ws wild type plants
(Figures 1 and 2). Results shown in Figure 5 indicate that the acetylation levels of histone
H3 at COR gene promoters in plants overexpressing CBF] under non-acclimated
conditions were signiﬁcantly higher than in the wild type plants. The acetylation levels of
histone H3 in the non-acclimated CBF] overexpressing plants approached the acetylation
levels observed for the cold acclimated wild type plants. When the CBF] overexpressing

plants were cold acclimated, the acetylation levels of histone H3 increased further.

We conclude that CBFl is sufﬁcient to drive the acetylation of histone H3 at the
COR gene promoters even without cold acclimation. Although overexpression of CBF is
sufﬁcient to drive acetylation of histone H3 at the COR gene promoters, it is not
sufﬁcient for the further increase in acetylation of histone H3 upon cold acclimation. This
increase could be due to the activation of endogenous CBFs or some other transcription

factors upon cold acclimation.

40

    
 
 
     

 

 

   

   
 

 

2 ACT2/7 promoter M 3.2 COR6.6 promoter I Wt
M I Wt § 2_4 * I CBFlox
§ CBFlox ‘ . a
:1: 1 . .
2

O

0 4 24
Time in h at 4°C Time in h at 4°C
m 1'2 CORl5a promoter I Wt m 2.4 COR47 promoter *
§ I 4 m CBFlox E18 'Wt
% 0.8 ”‘ ‘2 "14* E '
< ' . :5 1 2
B 0.4 y B
.5 .33 0.6
E 0 0 TE 0 0
o ' o '
E 0 4 24 c: 0 4
Time in h at 4°C Time in h at 4°C

to 4'0 COR78 promoter 4 M 2'5 G018 promoter ":1
II: 1" '7 a 1,5 CBFlox 4,;-
it, 2.0 l CBFlox ; i < 13;}; 4'; -.
1: “U 1'0 23f? . i.
B 1-0 .is’ 0.5 ' '
... '8 ,
g 0.0 E 0.0
8 0 8 0 24

4 24 4
Time in h at 4°C Time in h at 4°C

Figure 5: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type and CBF]
over-expressing plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies
speciﬁc for acetylated histone H3 (K9, K14) and the C-terminal of histone H3. Samples
were analyzed using real-time PCR to quantify promoter DNA enriched for acetylated
histone H3 and total histone H3. Ct values obtained were normalized to values obtained
for ACT2/ 7 promoter. Data were represented as ratio of acetylated histone H3 to total
histone H3 at various COR gene promoters. The Ct values for the ACT2/ 7 promoter were
not normalized and represented as ratio of acetylated histone H3 to total histone H3.
Results for the wild type plants were taken from ﬁgure 1. Error bars represent SE where
N (number of experiments) = 3, except N=1 for GolS. Comparison between wild type and
CBFlox plants was carried out using the t-test. * represents values of CBFlox plants
signiﬁcantly different than wild type plants at p S 0.05 at that particular time point.

41

The levels of histone H3 occupancy at COR genes were also examined. The
results shown in ﬁgure 6 reveal that the occupancy of histone H3 at the COR6. 6, COR15a
and COR 78 promoters in CBF 1 overexpressing plants under non-acclimated conditions
were signiﬁcantly lower as compared to the level of histone H3 in Ws wild type plants.
These levels were not further depleted upon cold-acclimation. The levels of histone H3 at
these promoters were similar to the levels of histone H3 after 24 h of cold acclimation in

wild type plants.

Thus, CBFl is sufﬁcient to decrease the occupancy of histone H3 at the COR6. 6,

COR 78 and GolS promoters without cold acclimation.

2.3.5 The activation domain of CBF is not necessary to drive acetylation of histone H3
upon cold acclimation but is required for the removal of histone H3 from the COR gene

promoters.

The previous section indicated that CBF overexpression was sufﬁcient to drive
histone H3 acetylation and depletion of nucleosomes at the COR gene promoters. To ﬁnd
out whether the activation domain of CBF is necessary to drive the acetylation of histone
H3 at the COR gene promoters, a mutant form of CBF was used. This mutant form of
CBF (CBFZAC) contains an intact DNA-binding domain but lacks the C-terminal
activation domain because of a nonsense mutation in the CBF 2 gene. has a stop codon
just before the activation domain and thus lacks the C-terminal activation domain. The
phenotype of plants overexpressing CBF2AC is similar to wild type plants. Plants
overexpressing CBFZAC are neither constitutively freezing tolerant nor do they show

constitutive expression of COR genes. The freezing tolerance of plants overexpressing

42

 
 
   
  
    
   
   

 

 
    

 

16 H3 at ACT2/7 promoter ,5, 1.5 H3 at COR6.6 promoter
12 I Wt & I Wt
:- CBFlox g 1.0 CBFlox
13
d)
L3 0.5
a:
E
8 0.0
0 4 24 O 4 24
Time in h at 4°C Time in h at 4°C
... 1.5 H3 at COR15a promoter ,_, 1.6 H3 at COR47 promoter
2‘ l Wt §‘1 2 I Wt
$1.0 CBFlox 1,3 ' V CBFIOX
E E 0.8
E 0.5 :2 0.4
8 0.0 8 0.0
O 4 24 0 4 24
Time in h at 4°C Time in h at 4°C
5- 1-2 H3 at COR78 promoter ... 12 H3 at GolS promoter
.30 8 é- I Wt
\° ' "‘ 0.8
% °\° ‘91 CBFlox
3 0 4 "8
E ' g 0.4
8 0-0 E 0.0
0 4 24
Time in h at 4°C Time in h at 4°C

Figure 6: Level of total histone H3 at COR gene promoters decreases upon cold
acclimation. ChIP assays were performed on wild-type and CBFl overexpressing plants
cold acclimated at 4°C for O, 4 and 24 hours using antibodies speciﬁc for acetylated
histone H3 (K9, K14) and the C-terminal of histone H3. Samples were analyzed using
real-time PCR to quantify promoter DNA enriched for acetylated histone H3 and total
histone H3. Ct values obtained were normalized to values obtained for ACT2/ 7 promoter.
Data were represented as normalized levels of total histone H3 at various COR gene
promoters. Ct values for the AC T2/ 7 promoter were not normalized and were represented
as % input. The wild type results were taken from ﬁgure 2. Error bars represent SE where
N (number of experiments) = 3 except GolS where N = 1. Comparison between wild type
and CBFlox plants was carried out using the t-test. * represents values of CBFlox plants
signiﬁcantly different than wild type plants at p S 0.05 at that particular time point.

43

CBF2AC is similar to wild type plants but the expression of COR genes is compromised
upon cold acclimation (Gilmour, unpublished data and Doherty, unpublished data).
CBF2AC is thought to act as a dominant negative where it is able to occupy most of the
CBF binding sites and thus cause a reduction of COR gene expression in plants

overexpressing CBF2A’C.

To determine whether the activation domain of CBF is necessary to drive
acetylation at the COR gene promoters, ChIP assays were carried out on plants
overexpressing CBF2AC and cold treated at 4°C for 4 and 24 hours. Non-acclimated

plants were used as controls.

Results obtained for the CBF2AC overexpressing plants were compared with
results previously obtained for Ws wild type plants (Figures 1 and 2). Results (Figure 7)
show that the acetylation of histone H3 at the COR gene promoters increased upon cold
acclimation in plants overexpressing CBF2AC. The non-acclimated CBFZAC plants did
not show hyperacetylation of histone H3 as observed in non-acclimated CBFlox plants.
Figure 8 shows that the level of histone H3 in plants overexpressing CBF2AC did not
decrease upon cold acclimation as observed in wild type plants. The level of histone H3
before and after cold acclimation was similar in the CBFZAC overexpressing plants.
Thus, the activation domain of CBF is not necessary for the acetylation of histone H3 at
the COR gene promoters during cold acclimation. However, in the absence of the
activation domain non-acclimated plants do not show hyperacetylation of histone H3 as
observed in the plants overexpressing CBFlox. Thus even though the activation domain
is not necessary for the cold induced increase in acetylation histone H3 it seems to be

required for the acetylation of histone H3 in some way. The activation domain of CBF is

44

 

 

 

     

 

 

 

 

ACT2/7 promoter in CBFZAC 3 ' ACT2/7 promoter in Ws plants
' plants
:0 b M 2 '
§ 2 - ,. -. b1 5 a
.‘I: a in U
o . . 1 ‘
< 1 [~13 ‘ _ <
0 g .- at. ' . - I -'-_. 7 I 0 .
0 . . 4 24 0 4 24
Time m h at 4°C Time in h at 4°C
COR6.6 promoter in CBFZAC M 1.2 ' COR66 promoter in WS plants
m 1.2 - plants E b
EB 0.8 ‘ I b it)
o ,,. :' _ 4.. U
E 0.0 ‘JJ '3 I l I w ﬁ g 0.0 '
o o 4 24 ‘= o 4 24
a Time in h at 4°C Time in h at 4°C

Figure 7: Level of acetylated histone H3 at COR gene promoters increases upon cold
acclimation. ChIP assays were performed on CBFZAC plants cold acclimated at 4°C for
O, 4 and 24 hours using antibodies speciﬁc for acetylated histone H3 and C-terminal of
histone H3. Samples were analyzed using real-time PCR to quantify promoter DNA
enriched for acetylated and total histone H3. Ct values were normalized to values
obtained for ACTZ/ 7 promoter. Data were represented as ratio of normalized levels of
acetylated histone H3 to normalized levels of total histone H3 at various COR gene
promoters. Ct values for ACT2/7 were not normalized and represented as ratio of
acetylated histone H3 to total histone H3. Data for wild type plants were taken from
ﬁgure 1. Error bars represent SE where N (number of experiments) = 3. Data were
analyzed using single factor ANOVA. The trend of change of histone H3 acetylation in
plants overexpressing CBFZAC was compared with the trend observed in Ws wild type
plants. Values of bar graphs denoted by the same letter are not signiﬁcantly different as
determined by LSD at p S 0.05.

45

Figure 7 continued

99999
ON-AONOO

I L l
n)
4‘ er
a

normalized AcH3/H3

t—Iu—IN
OOsNOO-h

normalized AcH3/H3
9 9

.O .9 t"
A on N

O

normalized AcH3/H3
b

.. COR15a promoter in CBFZAC

 

 

lants
p b
I *4 * E
a .
_I&I I I 444 I
0 4 24
Time in h at 4°C

COR47 promoter in CBFZAC

 

 

 

 

 

[1
O 4 24
Time in h at 4°C
- COR78 promoter in CBFZAC
plants b
“ b .4174
a ‘. .‘
4714: . . 5:5 5
O 4 24
Time in h at 4°C

J

 

 

GolS promoter in CBFZAC plants

 

b
4 ‘21
3.“: 4:43.744
. "i :2;
4 "5
I ‘ I 54;:
4 24
T1me in h at 4°C

N

u—s
OOﬁNOO-b

normalized AcH3/H3
o Q ~—

o 1° 1"
00 N

in

O

normalized AcH3/H3
'o

46

COR15a promoter in Ws plants

 

   

O 24
Time in h at 4°C

4 COR47 promoter in Ws plants

 

 

0 4

Time in h at 4°C

24

“ COR78 promoter in Ws plants

 

 

0

4
Time in h at 4°C

24

l 6018 promoter in Ws plants

 

O 4

Time in h at 4°C

24

 

 

 

12 ACT2/7 promoter in CBFZAC 12 ACT2 promoter in Ws plants
plants a
‘5' 8 ‘5' 8
,3 a a .3
.\° 4 I I 3 .\° 4
0 I I I I i I O
O 4 24 0 4 24
Time in h at 4°C Time in h at 4°C
... 1_ 5 COR6.6 promoter in CBF2AC ... 1_5 COR6.6 promoter in Ws plants
3. a plants a a
.E 1 0 .E a
°\° . o\° 1.0 b
s s
.5 0.5 .E 0.5
'3 '3
E 0 0 E
z: - g 0.0
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C

Figure 8: Level of total histone H3 at COR gene promoters remains unchanged upon cold
acclimation in CBFZAC plants. ChIP assays were performed on CBFZAC plants cold
acclimated at 4°C for 0, 4 and 24 hours using an antibody speciﬁc for the C-terminal of
histone H3. Samples were analyzed using real-time PCR to quantify promoter DNA
enriched for total histone H3. Ct values were normalized to values obtained for AC T2/ 7
promoter. Data represented as normalized levels of total histone H3 at various COR gene
promoters. Ct values or the ACT2/7 promoter were not normalized and represented as %
input. Error bars represent SE where N (number of experiments) = 3. Data for wild type
plants were obtained from ﬁgure 2. Data were analyzed using single factor AN OVA. The
trend of change of histone H3 acetylation in plants overexpressing CBFZAC was
compared with the trend observed in Ws wild type plants. Values of bar graphs denoted
by the same letter are not signiﬁcantly different as determined by LSD at p S 0.05.

47

Figure 8 continued

_ COR] 5a promoter in CBFZAC

 

 

‘5 1.5 plants
.3

g 1.0

E

a 0.5

E

a 0.0

0 4 24
Time in h at 4’C

 

 

 

 

 

 

 

       

2.4 1 COR47 promoter in CBFZAC
S lants
a. P
.5 16 d
g .
1:
$0.8 -
Eon
a .
0 4 24
Time in h at 4°C
3 2 COR78 promoter in CBFZAC
*5 ' ' plants
a.
.S 2.4 - a a
°\° ..... ....
B 1.6 - “I”
.F‘ O 8 " a '1 I’. l
8 00 -r—"—" . " . L---=—.r'
O 4 24
Time in h at 4°C
4.. 1.6 . GolS promoter in CBFZAC plants
:3
31.2 - a a a
.,\°
30.8 "
20.4 -
g 0.0 -
0 4 24
Time in h at 4°C

 

         

... 1,5 - COR15a promoter in Ws plants
é a
- a
°\o 1.0 "
B
.E -
,3 0.5
E
a 0.0 -
O 4 24
Time in h at 4°C
2-4 ' COR47 promoter in Ws plants

y—n
O\
J

 

normalized % input
o
00

 

.0
o
l

 

0 4 24

Time in h at 4°C

1:
w
ix.)

COR78 promoter in Ws plants

 

 

     

 

Time in h at 4°C
.5 1-6 ‘ GolS promoter in Ws plants
9..
.S 1.2 4 a
°\°
E 0.8 - a
'"d 0.4 -
E
‘3 0.0 "
0 4 24
Time in h at 4°C

48

also required for the loss of histone H3 occupancy at the COR6. 6, COR 78 and GolS

promoters during cold acclimation.

These experiments reveal three interesting aspects of CBF activity:

1.

The histone H3 acetylation proﬁles of plants over-expressing CBFZAC were
similar to wild type plants. These plants did not show constitutively high levels of
acetylated histone H3 like plants over-expressing CBF. Thus, the activation
domain of CBF is necessary to drive acetylation of histone H3 without cold
acclimation.

The COR gene expression levels in the CBF2AC over-expressing plants upon cold
acclimation were lower as compared to wild type plants, but the acetylation of
histone H3 did not seem to be affected in these plants and was similar to wild type
plants. Thus, the DNA-binding domain of CBF is sufﬁcient to recruit chromatin
modifying factors necessary to carry out acetylation of histone H3 at the COR
gene promoters but is not able to activate transcription. It is also possible that
other cold induced transcription factors bring about the increase in histone H3
acetylation at COR gene promoters but CBFZAC blocks the trans activation of

COR genes during cold acclimation.

. The levels of total histone H3 did not decrease at the COR6. 6, COR 78 and 0018

promoters in the CBF2AC plants upon cold acclimation. Thus the activation
domain of CBF is necessary to recruit chromatin remodelers that can clear histone

H3 from the COR gene promoters and the DNA-binding domain lacks this ability.

Thus, overall we can conclude that the activation domain of CBF is necessary to drive

acetylation of histone H3 and the removal of histone H3 at the COR gene promoters in

49

the absence of cold acclimation. The activation domain of CBF is not necessary to drive
acetylation of histone H3 at the COR gene promoters upon cold acclimation as
overexpression of the DNA-binding domain is sufficient to increase acetylation of
histone H3 upon cold acclimation. The activation domain of CBF seems necessary for the
removal of histone H3 at the COR gene promoters during cold acclimation as
overexpressing the DNA-binding domain by itself is unable to bring about this change.
2.3.6 GCN5 is not required for the acetylation of histone H3 at the COR gene promoters
but is required for the removal of histone H3 from the COR gene promoters during cold

acclimation.

Arabidopsis GCN5, a histone H3 speciﬁc HAT, can physically interact with the
activator CBF in vitro [106]. CBFs bind to the CRT/DREs cis-elements in the promoters
of COR genes and activate COR gene expression upon cold acclimation [68, 98]. gcn5-I
mutants show a delay in the activation and a reduction in the overall expression of COR
genes during cold acclimation [3 7]. Thus we hypothesized that, upon cold acclimation,
CBF recruits a GCN5 containing complex to the COR gene promoters. This complex in
turn acetylates histone H3 at the COR gene promoters and facilitates the activation of
COR genes. According to this model, the acetylation of histone H3 at the COR gene
promoters in gcn5-I plants would be hampered, resulting in a delay in the activation and
reduction in the overall expression COR genes. To test this hypothesis, ChIP assays were
carried out in gcn5-I plants that were cold acclimated at 4°C for 4 or 24 hours. Plants

grown at 22°C were used as controls.

Results obtained for the gcn5-1 plants were compared with results previously

obtained for Ws wild type plants (Figures 1 and 2). Figure 9 shows that the acetylation of

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 - ACT2/7 promoter in gcn5-1 plants 2 ' ACT2/ 7 promoter in WS plants
a
a
m a «1
§ 1 a ? T g 1 . a
a 1 1 1 3
< <2
0 I I I O "
0 . . 4 24 O 4 24
Tune in h at 4°C Time in h at 4°C
COR6.6 t ' 5-1
m 1.2 - p rgﬁgér in gen 3 1-2 ‘ COR6.6 promoter in Ws plants
E
gos- b b @03‘ b b
2 T <
304- 1 E04
.5 a '
E00 "'. . . 300
8 0 4 M = 0 4 M
Time in h at 4°C Time in h at 4°C

Figure 9: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type and gcn5 -1
plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc for
acetylated histone H3 (K9, K14) and the C—terminal of histone H3. Samples were
analyzed using real-time PCR to quantify promoter DNA enriched for acetylated histone
H3 and total histone H3. Ct values obtained were normalized to values obtained for

AC T2/ 7 promoter. Data were represented as ratio of acetylated histone H3 to total histone
H3 at various COR gene promoters. Ct values for the ACT2/ 7 promoter were not
normalized and represented as ratio of acetylated histone H3 to total histone H3. Data for
wild type plants were obtained from ﬁgure 1. Error bars represent SE where N (number
of experiments) = 3. The trend of change of histone H3 levels upon cold acclimation over
time in wild and gcn5-I plants was analyzed using single-factor AN OVA. Values of bar
graphs denoted by the same letter are not signiﬁcantly different as determined by LSD at
pSODi

51

Figure 9 continued

('1

E

m

o
<
B
.5.
E"

O
G

normalized AcH3/HB

normalized AcH3/H3

normalized AcH3/H3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

COR15a promoter in gcn5-1
0.8 1
plants
0. -
6 b
0.4 a b I
0.2 - a
0.0 I I I
O 4 24
Time in h at 4°C
15 .COR47 promoter in gcn5-I plants
b .3—
1.0 -
0.5 - $
1
0.0 : . .
0 4 24
Time in h at 4°C
1.6 - COR78 promoter in gcn5-1 plants
b
1.2 -
l
0.8 - b .l.
0.4 " a H
0.0 [+1 I I I
0 4 24
Time in h at 4°C
25 .GolS promoter in gcn5-1 plants
2.0 -
b
1.5 - b
1.0 - T
0.5 - a ' I
0.0 "I“ . . .
O 4 24
Time in h at 4°C

9.9.953
toe-moo

I l 1
O a:
a o-
c

ngrmalized AcH3/H3
'o

p—t
M

p—a
O

.O
U.

C

normalized AcH3/H3
'o

OOI—AI—t
hOONON

I J l
o m
6
ﬂ cr-

rmalized AcH3/I-I3

no
.9
o

52

COR15a promoter in Ws plants

 

24
Time in h at 4°C

COR47 promoter in Ws plants

 

a:
c"
G

 

0

4
Time in h at 4°C

24

COR78 promoter in Ws plants

 

Time in‘li at 4°C

l GolS promoter in Ws plants

. c
b

 

a

 

O

4
Time in h at 4°C

24

histone H3 at the COR gene promoters increased upon cold acclimation in gcn5-I plants
to a degree similar to that observed for Ws wild type plants. Low levels of acetylated
histone H3 were observed at all the COR gene promoters in non-acclimated control plants
in the gcn5-1 mutants. The acetylation levels of histone H3 at the COR gene promoters
increased at least two-fold after 4 h of cold acclimation and high levels of acetylated
histones were maintained after 24 h. Interestingly, results in ﬁgure 10 show that the
decrease in the occupancy of histone H3 at the COR6. 6, COR 78 and GolS promoters in
gcn5-1 plants was not signiﬁcant upon cold acclimation as observed in wild type plants.
The histone occupancy at the COR 78 and GoIS promoters did not decrease after 24 h of
cold acclimation as observed in wild type plants. Surprisingly, loss of histone H3 was
observed at the COR15a promoter after 24 h of cold acclimation in the gcn5-I plants.
Such a decrease in histone H3 occupancy was not observed in wild type plants. The level
of histone H3 at the CORISa promoter in non-acclimated gcn5-1 plants was higher than
in non-acclimated wild type plants. The levels of histone H3 in the gcn5-I and wild type
plants were similar after cold acclimation. Thus the decrease in the occupancy of histone
H3 at the COR15a promoter in gcn5-1 plants could be because of the higher level of

histone H3 in non-acclimated plants.

We can conclude from the results that GCN5 is not required for acetylation of
histone H3 at the COR gene promoters during cold acclimation. However, GCN5 seems
to be required for the loss of histone H3 occupancy at the COR6. 6, COR 78 and 6018
promoters during cold acclimation. Nevertheless, the expression levels of COR genes are
delayed and lowered in these mutants. Thus GCN5 might regulate the expression of COR

genes independent of its HAT activity.

53

12 - ACT2/7 promoter in gcn5-1 plants 12 - ACT2/7 promoter in Ws plants

 

 

 

 

 

 

 

 

 

 

 

 

     

 

 

 

 

 

 

   

 

 

 

 

 

 

a a
S 8 4 a +5 8 " a
a I l g a
o\° 4 - J. °\° 4 .
O I I 1 O _
0 4 24 o 4 24
Time in h at 4°C Time in h at 4°C
“ 2.0 . COR6'6 proino‘t: in gcn5-1 4.. 2,0 1 COR6.6 promoter in Ws plants
:s p an :s
.3315 - la a 31.5 - 3
8° 1 T o\°
B 1'0 ‘ 1 a 3 1.0 -
T
.5 J, .5
"g 0.5 - 20.5 .
8 0.0 . - . 8 0.0 a
o 4 24 o 4 24
Time in h at 4°C Time in h at 4°C

Figure 10: Level of total histone H3 at COR gene promoters decreases upon cold
acclimation in wild type plants but not in gcn5-1 plants. ChIP assays were performed on
wild-type and gcn5-I plants cold acclimated at 4°C for 0, 4 and 24 hours using an
antibody speciﬁc for the C-terminal of histone H3. Samples were analyzed using real-
time PCR to quantify promoter DNA enriched for total histone ~H3. Ct values obtained
were normalized to values obtained for AC T 1N2 promoter. Data were represented as
normalized levels of total histone H3 at various COR gene promoters. Ct values for the
AC T2/ 7 promoter were not normalized and represented as % input. Data for wild type
plants were taken from ﬁgure 2. Error bars represent SE where N (number of
experiments) = 3. The trend of change of histone H3 levels upon cold acclimation over
time in wild and gcn5-I plants was analyzed using single-factor AN OVA. Values of bar
graphs denoted by the same letter are not signiﬁcantly different as determined by LSD at
pSODi

54

Figure 10 continued

COR15a promoter in gcn5-I

 

 

 

 

 

 

 

 

 

 

 

 

 

1.6 - 1,6 - COR15a promoter in Ws plants
‘5 plants ‘5 a
ca. 9.
.5 1.2 "‘ ? a .5 1.2 ' a
of 1 I b o\° a
“U 0 8 "' T '8 0.8 .
.53.

3'04- 1 “a 0.4-
8 0.0 I I I 8 00 .-

0 4 24 0 4 24

Time in h at 4°C Time in h at 4°C

COR47 promoter in gcn5-1 plants

t—n
0‘
I

ormaltz
o o
o 4:
L
Hm
Film
normalized % input
O F’ 9 7‘
o A 00 N
l L l l
m
N
93

COR47 promoter in Ws plants

a
¢\° J T

 

 

 

 

 

 

 

 

 

 

 

 

 

0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,_, 1.2 1C0R78 promoter in gcnS-I plants ,_, 1.2 - COR78 promoter in Ws plants
.°\o 0.8 - 'I' 12} a 30.8
T e
E 04- 1 f7 1 £04
E" ' E '
8 00 . T . 300
O 4 24 4 24
Time in h at 4°C Time in h at 4°C
H 1.6 . GolS promoter in gcn5-1 plants 45 1-6 ' GolS promoter in Ws plants
:1 a.
.g‘ 1.2 . .§ 1.2 " a
s 3 °\
g 0.8 - I a 2f 30.8 ' a
=3 0.4 - 1 1 1 E304 - b
E 0 0 8
8: . I I 0.0 .-
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C

55

2.3.7 ADA2b is not necessary for the acetylation of histone H3 at the COR gene
promoters but is required for the removal of histone H3 from the COR gene promoters

during cold acclimation.

Arabidopsis GCN5 is a histone H3 speciﬁc histone acetyltransferase and its
ability to acetylate nucleosomal histones is enhanced by the presence of ADA2b [51,
106]. ADA2b can physically interact with GCN5 and CBF in vitro. ada2b-1 plants show
delayed activation and an overall reduction in the expression of the COR genes activated
by CBF. Thus we hypothesized that upon cold acclimation CBF recruits a GCN5-ADA2b
containing complex at the COR gene promoters. This complex in turn acetylates histone
H3 at the COR gene premoters and facilitates the activation of COR genes. According to
this model, acetylation of histone H3 at the COR gene promoters in the adaZb-I plants
would be hampered resulting in a delay in the activation and reduction in the overall
expression COR genes. In order to test this hypothesis, ChIP assays were carried out in
ada2b-1 plants cold acclimated at 4°C for 4 or 24 hours. Non-acclimated plants were

used as controls.

Results obtained for the ada2b-I plants were compared with results previously
obtained for Ws wild type plants (Figures 1 and 2). Results (Figure 11) show that the
acetylation of histone H3 at all the COR gene promoters except the COR4 7 promoter
increased upon cold acclimation in ada2b-I plants. The increase in acetylation of histone
H3 at the C OR47 did not increase signiﬁcantly at p S 0.05 as determined by ANOVA.

The large variation observed in the 24 h cold acclimated sample was probably the reason

56

2 ,ACT2/7 promoter in ada2b-1 2 ‘ ACT2/7 promoter in Ws plants

 

 

 

 

 

 

   

 

 

 

   

plants a
m a m
E a a § 3
E 1 l m 1 ‘
.. + °
0 '—' I". I i I 0 '
O . . 4 24 0 4 24
T1me in h at 4°C Time in h at 4°C
R . t ' d 2b-1 .
CO 6 6 promo er in a a m 1-2 ‘ COR6.6 promoter in Ws plants
12 . plants E
E M b
”0.8 . :5 0.8 - b
:5 <
<‘04 - b b E 04 -
t . m m -~ .
E00 Hm ' i ' '1 ﬂ E00 -
t:
g 0 , , 4 24 0 4 24
T1me in h at 4°C Time in h at 4°C

Figure 11: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type and adaZb-I
plants cold acclimated at 4°C for O, 4 and 24 hours using antibodies speciﬁc for
acetylated histone H3 (K9, K14) and the C-terminal of histone H3. Samples were
analyzed using real-time PCR to quantify promoter DNA enriched for acetylated histone
H3 and total histone H3. Ct values obtained were normalized to values obtained for
ACT 2/ 7 promoter. Data were represented as ratio of acetylated histone H3 to total histone
H3 at various COR gene promoters. Ct values for AC T2/ 7 promoter were not normalized
and were represented as ratio of acetylated histone H3 to total histone H3. Data for wild
type plants were taken from ﬁgure 1. Error bars represent SE where N (number of
experiments) = 3. The trend of change of histone H3 levels upon cold acclimation over
time in wild and adaZb-I plants was analyzed using single-factor AN OVA. Values of bar
graphs denoted by the same letter are not signiﬁcantly different as determined by LSD at
pSOﬂi

57

Figure 11 continued

 

COR15a promoter in ada2b-1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

30.8 7 plants
m0.6 T
‘5
0.4 .
4 2
. 0.2 - a .41.. 4*
"a .
goo [2'11] 4 ' 4 . .
‘3 O 4 24
Timeinhat 4°C
COR47 promoter in ada2b-1
M 25 - plants a
E 2.0 4
M
=5 1.5 4
§ 1.0 - a
.5 0.5 ~ r-I—
'8
E 0.0 I """ 1
8 0 4 24
Timeinhat 4°C
1 2 _ COR78 promoter in ada2b-1
‘2 ° plants
E.
m) 0.8 - Ib a
a ‘ g
a . .
8
0 4 24
Time in h at 4°C
GolS promoter in ada2b-1 plants
M 2'5 -
§ 2.0 -
E, 1.5 -
'8 1.0 '
g 0.5 -
E 0.0
O
G O 4 24
Timeinhat 4°C

«1 0'8 ' COR15a promoter in Ws plants

 

orrnalizedA
O .O S3
o N A
I I l
m
c‘
c

0 4 24
Time in h at 4°C

COR47 promoter in Ws plants

 

 

0 4
Time in h at 4°C

("3
§ b b
$0.8 4
<
"O
3’04 -
7.; a
Eon -
G
24

24

p—n
N

COR78 promoter in Ws plants

 

0
Time int at 4°C

_ GolS promoter in Ws plants

 

 

rmalizedAcH3
‘3 .9 5" I" P
C U’I O U1 0
I I4 I
.9?
O

O 4
Time in h at 4°C

24

58

for this.

We can conclude from the results that ADA2b is not required for the acetylation

of histone H3 at the COR gene promoters during cold acclimation.

Results in Figure 12 show that the occupancy of histone H3 in ada2b-1 plants did
not decrease at the COR6. 6, COR 78 and 6018 promoters even after 24 h of cold
acclimation as observed in wild type plants. We conclude that ADA2b is required for the
decrease in histone H3 occupancy at the COR6. 6, COR 78 and GolS promoters during

cold acclimation.
2.4 Discussion:

Any valid model for mechanisms of inducing eukaryotic gene regulation in
response to external signals must account for the barrier imposed by chromatin. The
standard mechanism as developed largely in yeast experimental models involves both
histone-modifying enzymes (especially HATS) and remodeling enzymes. This chapter
reports evidence that several features of this standard model apply to the activation of
COR genes upon cold acclimation. My results reveal that histone H3 at the COR gene
promoters is hypoacetylated in non-acclimated plants but the acetylation of histone H3 at
the COR gene promoters increases upon cold acclimation in wild type plants. The
occupancy of histone H3 decreases at some COR gene promoters (COR6. 6, COR 78 and
0015') after 24 h of cold acclimation. Acetylation of histone H3 seems to precede removal
of histone H3 at these COR gene promoters. The increase in acetylation is a dynamic

change occurring within 4 hours of cold stress and stable levels of acetylation are

59

   

 

12 ACT2/7 promoter in Ws plants
a
‘5 S 8
.E‘ E
o\° o\° 4
0
4 24 0 4 24
Time in h at 4°C Time in h at 4°C
COR6 6 promoter in “1‘72 b", ... 1.6 COR6.6 promoter in Ws plants
1. 6 p anltsa/b a a
‘5 .S 1.2 a
212 2° b
°\o 0 8 '8 0.8
T, .E
a E
E 0. 0 1 8 0.0
8 o 4 24
Time in 4h at 4°C Time in h at 4°C

Figure 12: Level of total histone H3 at COR gene promoters decreases upon cold
acclimation in wild type plants but not in ada2b-I plants. ChIP assays were performed on
wild-type and ada2b-1 plants cold acclimated at 4°C for O, 4 and 24 hours using an
antibody speciﬁc for the C-terminal of histone H3. Samples were analyzed using real-
time PCR to quantify promoter DNA enriched for total histone H3. Ct values obtained
were normalized to values obtained for ACT2/ 7 promoter. Data were represented as
normalized levels of total histone H3 at various COR gene promoters. Ct values for
ACT2/7 were not normalized and were represented as % input. Data for wild type plants
was obtained from ﬁgure 2. Error bars represent SE where N (number of experiments) =
3. The trend of change of histone H3 levels upon cold acclimation over time in wild and
ada2b-1 plants was analyzed using single-factor ANOVA. Values of bar graphs denoted
by the same letter are not signiﬁcantly different as determined by LSD at p S 0.05.

60

Figure 12 continued

 

 

 

 

 

 

 

 

   

 

 

 

 

 

COR15a promoter in ada2b-I 1. 6 COR15a promoter in Ws plants
3 1.6 l plants 2: T a
31.2 .. a 1.2 ‘ a
o\° O 8 3: O 8 ' a
".g 0.4 E 0.4 ~
§0.0 g 0.0
a 4
Time in h at 4°C Time in h at 4°C
2.5 - COR47 promoter m ada2b-1 .. 2-5 ' COR47 promoter in Ws plants
4.. plants :3
“2.0 .
‘ a
4
Time in h at 4°C Time in h at 4°C
,5 1.6 4 COR7 8 promoter in ada2b-1 ‘5’ 1.6 - COR78 promoter in Ws plants
.S‘ 1.2 l .g‘ 1.2 -
X 0 8 a"
. “U 0.8 .
E .24
'8 0.4 '3 0.4 .
2 2
a 0-0 a 0.0 -
O 4 24
Time in h at 4°C Time in 4h at 4°C2
4.. 1.5 . GolS promoter in adaZb-I plants .5 1.6 - GolS promoter in Ws plants
:3
fa" 3
.,\°
E a
=3 b
E
s:
O 4 24 4
Time in h at 4°C Time in h at 4°C

61

maintained even after 24 hours. Upon de-acclimation the acetylation level of histone H3
decreases to non-acclimated levels in 24 hours and the levels of total histone H3 are

restored to levels similar to non-acclimated plants. Acetylation of histones and depletion
of nucleosomes seem to be hallmarks of the COR gene activation process in Arabidopsis

as in othereukaryotes [110, 112, 113].

The COR genes are a set of cold-regulated genes whose transcription is stimulated
by CBF. Other transcription factors also bind to COR genes and so it was important to
dissect whether CBF itself was responsible for the chromatin modifications at the COR
gene promoters. Results show that overexpression of CBFl is sufﬁcient to drive the
acetylation and depletion of histone H3 from the COR gene promoters in the absence of
cold acclimation. The level of acetylated histone H3 further increased when the plants
were cold-acclimated, which can probably be attributed to the function of the endogenous
CBF proteins or to the activity of other transcription factors induced during cold
acclimation. To determine whether the activation domain of CBF is necessary to drive the
acetylation of histone H3 at the COR gene promoters, ChIP assays were carried out in
plants overexpressing a mutant form of CBF, lacking the C-terminal activation domain
(CBF2AC). Plants overexpressing CBFZAC showed lower COR gene expression as
compared to wild type plants. ChIP assays revealed that the activation domain of CBF is
not necessary to drive acetylation of histone H3 upon cold acclimation but is required for
the removal of histone H3 from some COR gene promoters. However, overexpression of
CBF2AC did not lead to increased acetylation of histone H3 in non-acclimated plants as
observed in plants overexpressing CBF. Thus, the activation domain might play a role in

driving acetylation of histone H3 in the absence of cold signal. The activation and DNA-

62

binding domains of CBF seem to control different aspects of chromatin remodeling
during cold acclimation. The DNA-binding domain of CBF might recruit chromatin
modifying proteins that can acetylate histone H3 whereas the activation domain might
recruit chromatin remodeling enzymes. The GCN5 and ADA2 proteins in Arabidopsis
physically interact with the DNA binding domain of CBF in vitro which is unique given
that most activator proteins bind coactivators via their activation domain. Although this is
an unusual finding in Arabidopsis examples of coactivator proteins binding to DNA-
binding domains have been observed in eukaryotes [114]. Thus CBF might be able to
recruit ADA2b-GCN5 containing chromatin modifying complexes to the COR gene

promoters via interaction with its DNA binding domain.

Activation domains of transcription factors are known to bind members of
coactivator complexes. Mutation of an activation domain can lead to loss of binding of
these coactivator complexes and also reduce their activation potential [89, 115]. The
CBF2AC form of CBF retains the ability to acetylate histone H3 after cold acclimation
but cannot activate COR gene expression to wild type levels. However, the CBF2AC
form of CBF is unable to deplete nucleosomes from the COR gene promoters. The
activation domain might recruit chromatin remodeling factors that have the ability to
remove nucleosomes off promoters and regulate COR gene expression. Deletion of the
activation domain of CBF seems to affect its activation potential but not its ability to
bring about histone modiﬁcations. Thus, acetylation of histone H3 seems to precede gene

activation and acetylation at lysines 9/14 of histone H3 is not sufficient for transcription.

GCN5, a histone acetyltransferase, and its likely interaction partner protein,

ADA2b, were prominent candidates for the role of carrying out histone H3 acetylation at

63

the COR gene promoters. The two proteins can interact in vitro not only with each other
but also with CBF in vitro. Homozygous mutants of GCN5 and ADA2b show a delay in
activation and two-fold lower expression of COR genes upon cold acclimation. However,
ChIP assay results show that the increase in acetylation of histone H3 in the gcn5-1
mutant is similar to the increase in wild type plants upon cold acclimation. Thus GCN5 is
not required for the acetylation of histone H3 (lysines 9/14) at the COR gene promoters.
Nonetheless, GCN5 is required for the removal of histone H3 from the COR gene
promoters, as the occupancy of histone H3 did not decrease in the gcn5-1 mutants upon
cold acclimation. Therefore, a GCN5-containing complex might be directly required for
the histone H3 removal from the COR gene promoters. Alternatively, the function of such
a complex might be essential for the recruitment and removal of histone H3 by an ATP-
dependent chromatin remodeling complex. In addition to functioning as a histone
acetyltransferase, GCN5 can also act as an adaptor protein facilitating the binding of the
general transcription factors and the activator at the promoter of target genes. Thus
deleting GCN5 might delay the recruitment of the general transcription factors at the

COR gene promoters and in turn cause a delay in the activation of transcription.

The acetylation of histone H3 at the COR gene promoters increases in the ada2b-
] mutants upon cold acclimation. Thus, ADA2b is not required to bring about acetylation
of histone H3 at the COR gene promoters upon cold acclimation. ADA2b is required for
the loss of histone H3 occupancy at the COR6. 6, COR 78 and G018 promoters. The human
ADA2]3 protein can interact in vitro with the chromatin remodeling enzyme BRGl and
activate gene expression [61]. The Arabidopsis ADA2b might be able to interact with

chromatin remodeling complexes and regulate COR gene expression.

64

ADA2b and GCN5 are known to act as adaptors facilitating the interaction of
activators with the general transcription factors and thus activating transcription [84].
GCN5 and ADA2b are parts of large multi-subunit complexes that act to recruit TBP to
promoters and thus facilitate transcriptional activation [116]. Absence of either GCN5 or
ADA2b might disrupt such a complex or de-stabilize the complex which in turn causes
delay in the activation of target genes and lower transcription. In addition to the HAT
domain, GCN5 also contains a bromodomain that is known to bind acetylated histones as
well as other transcription factors [66]. The bromodomain of GCN5 may be required for
the recruitment of the chromatin remodeling enzymes that remove nucleosomes from the

COR gene promoters.

Recent studies show that the expression of COR genes is not solely CBF-
dependent but is a result of the interplay of various transcription factors (Doherty,
unpublished data). Thus, GCN5-ADA2b might be regulating the expression of COR
genes in a CBF-independent manner. It will be interesting to ﬁnd out whether GCN5 and
/or ADA2b interact with transcription factors that regulate COR gene expression in a

CBF-independent manner.

65

Chapter III

The activation of COR genes by CBF occurs in an ADA2b and GCN5 independent

manner.
3.1 Introduction:

Cold acclimation is the process by which exposure to low non-freezing
temperatures increases the ability of some plants to survive freezing stress [117]. The
CBF family of transcription activators acts as ‘master regulators’ during the cold
acclimation pathway in Arabidopsis [118]. In wild type plants, the expression of the CBF
family of activators is induced within 15 minutes of cold stress, which in turn leads to the
transcriptional activation of the cold-regulated (COR) genes [68]. Overexpression of the
CBF proteins is sufﬁcient to drive COR gene expression and causes the plants to be
constitutively freezing tolerant [93, 94]. CBF can bind to speciﬁc cis-elements in the
promoters of these genes known as the cold/dehydration responsive elements
(CRT/DREs) and activate their expression [98]. Studies in yeast have shown that the
Arabidopsis CBF l, CBF2 and CBF3 proteins activate the expression of a reporter gene
ﬁised to the CRT promoter elements. One-hybrid study conducted in yeast showed that
the activation of the reporter gene fused to the CRT by CBF depends upon the presence
of two co-activator proteins ADA2 and GCN5 [50].

Arabidopsis encodes two homologues of ADA2 (ADA2a and ADA2b) and one
homologue of GCN5. Homozygous null mutants ada2b-I and gcn5-1 show pleiotropic
phenotypes whereas the adaZa-I mutant is phenotypically similar to wild type plants

[37]. The ada2b-I and gcn5-I mutants show a delay in activation and a two-fold

66

reduction in overall expression of the CBF regulated cold regulated (COR) genes [37].
The ability of gcn5-1 plants to withstand freezing temperatures is similar to wild type
plants whereas ada2b-1 plants are constitutively freezing tolerant. This constitutive
freezing tolerance of the ada2b-1 plants is thought to occur in a CBF-independent
manner because the expression of CBF or its target genes is not elevated in these mutants.
ADA2b and GCN5 can physically interact with each other as well as with CBF [50, 51].
Arabidopsis GCN5 can acetylate histone H3 in vitro [51].

Results from the previous chapter indicate that the acetylation of histone H3
increases at the COR gene promoters upon cold acclimation. GCN5 is not required for
the acetylation of histone H3 at the COR gene promoters but is required for the depletion
of histone H3 from some COR gene promoters. ADA2b is required for driving the
increase in acetylation at some COR gene promoters. ADA2b and GCN5 can physically
interact with CBF in vitro. Whether they interact with CBF in vivo and regulate the
expression of COR genes via CBF is not known. It is possible that ADA2b and GCN5
control COR gene expression independently of CBF.

To determine whether the CBF-dependent activation of COR genes requires
ADA2b and/or GCN5, two approaches were used. In the ﬁrst approach, a promoter-
reporter plasmid was constructed which contained four tandem repeats of the CRT fused
to the reporter gene GUS. The expression of the reporter gene in wild type plants was
compared with the expression in the adaZb-I and gcn5-1 plants upon cold acclimation.
The rationale behind this approach was that CBF regulates the expression of the COR
genes via the CRTs. The promoters of the COR genes contain various other cis-elements

that could be potentially be regulated by non-CBF transcription factors [119]. Thus

67

fusing a reporter gene speciﬁcally to the CRTs would render that promoter-reporter gene
solely under the regulation by CBFs. Any difference between the expression of the
reporter gene between wild type and mutant plants would be due to a regulation of the
reporter gene via a CBF dependent pathway. Thus by using a CRT driven promoter we
are eliminating the possibility that the reporter gene could be regulated by non-CBF
transcription factors. In the second approach, the CBF2 gene was overexpressed in the
ada2b-1 and gcn5-1 background. The expression of the COR genes in these plants was
compared with plants that overexpressed CBF2 in the wild type background without cold
acclimation. Overexpression of CBF leads to constitutive expression of COR genes
without the requirement of cold stress. Thus any difference in COR gene expression in
between wild type and mutant plants overexpressing CBF must likely occur in a CBF
dependent manner.

3.2 Materials and methods:

3.2.1 Construction of ADA2b-1+/- and GCN5-I +/- lines containing the 4X-CRT-GUS
promoter-reporter gene.

To make these transgenic lines, a promoter-reporter plasmid was constructed by
fusing four tandem repeats of the CRT and a 353 minimal promoter to the GUS gene.
The 358 minimal promoter fused to the GUS reporter gene taken from pBIlOl .1 plus
(V an Buskirk, unpublished data) was inserted into the pCAMBIA33OO vector (Carnbia,
Canberra, Australia) that encodes kanamycin resistance (for selection in bacteria) and
glufosinate (Basta) resistance for selection in plants. A double-stranded oligonucleotide
bearing four c0pies of CRT was then inserted between the XbaI and BamHI sites of the

pCAMBIA33OO plasmid. The plasmid containing the 4X-CRT-353minimal-GUS

68

sequence in the pCAMBIA33OO vector is now referred to as 4X-CRT-GUS. This 4X-
CRT-GUS clone was then transformed into electrocompetent Agrobacterium tumifaciens
GV3101 (pMP90) cells. The Agrobacterium cells transformed with the plasmid were then
used to transform ADA2b-I +/- and GCN5-1 +/- plants. The seeds from the transformed
plants were repeatedly selected for Basta resistance through three generations (T3) to
homozygosity. Two lines containing 4X-CRT-GUS in the ADA2b-1 +/- background
(lines # land 14) and six lines in the GCN5-1 +/- background (lines # 2, 4, 5, 6, 9 and 10)
were selected for gene expression analysis.
3.2.2 Creating wild type, ada2b-1 and gcn5-I plants overexpressing CBF2

ADAZb-I +/- and GCN5—I +/- plants were transformed with Agrobacterium cells
transformed with a pCAMBIA33OO plasmid containing the CBF2 gene driven by the 35S
promoter (Sarah Gilmour, unpublished data). ADA2b-I +/- and GCN5-I +/- plants
containing the transgenic CBF2 were selected on Basta plates and conﬁrmed by
genotyping. Six lines containing the transgenic CBF2 in ADA2b-1 +/- background (lines
1, 5, 16, 22, 41 and 107) and three lines containing the transgenic CBF2 in GCN5-I +/-
background (lines 1, 3 and 182) were selected for gene expression analysis.
3.2.3 RNA isolation and gene expression analysis

Six to eight 15 day old whole seedlings were harvested in eppendorf tubes and
ﬂash frozen in liquid nitrogen. The tissue was stored at -80°C until ﬁu'ther use. For RNA
extraction the frozen tissue was ground in liquid nitrogen using a pestle. RNA extraction
was carried out using 0.5 ml of cold (4°C) Plant RNA Reagent from Invitrogen per 0.1
gram tissue. The mix was brieﬂy vortexed and then the tube was incubated by laying it

horizontally at room temperature for 5 minutes. The solution was clariﬁed by

69

centriﬁrging for 2 minutes at 12,000 x g in a microcentrifuge at room temperature. The
supernatant was transferred to an RNase-free tube and then 0.1 ml of 5 M NaCl was
added to the clariﬁed extract. The solution was mixed and then 0.3 ml of chloroform
added and the tube was inverted to mix the solution. The sample was centrifuged at 4°C
for 10 minutes at 12,000 x g to separate the phases. The top aqueous phase was
transferred to an RNase-free tube and an equal volume of isopropyl alcohol was added to
it. The solution was mixed and incubated at room temperature for 10 minutes. The
sample was centrifuged at 4°C for 10 minutes at 12,000 x g. The supernatant was
discarded and the pellet was washed with 1 ml of 75% ethanol. The pellet was dried by
removing all the residual ethanol and then dissolved in 50 pl RNase-free water to

dissolve the RNA. The RNA was stored at -20°C.

DNase treatment was carried out by adding 5 pl 10X buffer and 1 pl DNase
(10U) to 50 pl of RNA solution. The mix was incubated for 2 hours at 37°C. Aﬁer the
incubation, 50 pl water was added to the mix to increase volume. Samples were extracted
using 1 volume (100 pl) phenol, chloroform, isoamyl alcohol (25:24:1 v/v). The mix was
vortexed and then centrifuged at top speed in a microcentrifuge for 5 minutes. The upper
aqueous phase was transferred to a new tube and RNA was precipitated by adding 1/2
volume 7.5 M ammonium acetate (50 pl) and 2 volumes 100% ethanol (200 pl). The
resulting RNA pellet was washed with 500 pl of 7 5% ethanol. The puriﬁed RNA pellet

was resuspended in 50 pl nuclease-free water.

Levels of speciﬁc mRNAs were assayed using reverse transcription and real-time
PCR. 1 pg of total RNA was used to make cDNA using the Reverse Transcriptase kit

from Promega. The ﬁnal reaction volume was 20 pl. After the reaction the samples were

70

diluted 10 times in nuclease-free water. Real-time PCR was carried out using 10X buffer
containing SYBR Green, 0.25 pM of each primer, 3 mM MgC12, 12.4 mM dNTPs and
0.15 pl Ampli Taq Gold DNA polymerase enzyme (all reagents were obtained from ABI)
on the cDNA samples. 3 pl of cDNA sample was used in a 30 pl reaction. PCRS were
carried out by using primers that detect the open reading frames of the COR genes
(COR6. 6, COR15a, COR4 7, COR 78 and GolS) and the AC T3 open reading frame as a
control. The Ct values thus obtained for the COR genes were normalized to the values
obtained for the AC T 3 promoter. The values were expressed as ACT3- normalized
expression levels of the target genes. The Ct values for the ACT 3 gene were not
normalized and were represented as average Ct. t—tests were conducted to compare the
expression of the target genes between wild type and mutants plants to determine whether

the difference was signiﬁcant at p S 0.05.

3.3 Results:

3.3.1 The 4X-CRT driven expression of the reporter gene GUS by CBF is neither

dependent on ADA2b nor GCN5

In order to determine whether the CBF dependent expression of COR genes via
CRTs requires ADA2b and GCN5, expression of the reporter gene GUS fused to 4X-
CRT was compared in wild type, ada2b-1 and gcn5-1 plants. ADA2b-I +/- and GCN5-
] +/- plants were transformed with Agrobacterium cells containing the 4X-CRT—GUS
plasmid by the ﬂoral dip method. T3 lines homozygous for the 4X-CRT-GUS reporter
gene were selected and screened for the cold-induced GUS expression. Lines showing at

least a two-fold induction of GUS after cold acclimation at 4°C for 24 hours as compared

71

to non-acclimated plants were selected. Seeds from six lines containing the reporter gene
in the GCN5-1 +/— background and two lines containing the reporter gene in the ADA2b-
1+/- background were sown on Garnborg media plates containing Basta in order to select
for the reporter gene. Seeds from the ADA2b-1 +/- plants generated Wt, ADA2b-1 +/- and
adaZb-I -/- plants. Seeds from the GCN5-1+l- plants generated Wt, GCN5-1+l- and
gcn5-1 -/- plants. The plants heterozygous for either ADA2b or GCN5 are phenotypically
wild type whereas mutant plants have a dwarf phenotype. Plants heterozygous for ADA2b
or GCN5 have no known effect on COR gene expression during cold acclimation and
hence this was not thought to have any effect on the expression of the reporter gene GUS.
Wild type or mutant plants were collected before or after cold treatment at 4°C for 24
hours. Gene expression analysis was performed to compare the expression of the GUS

reporter gene in the Wt and the mutant (gcn5 or ada2b) background.

Figure 13 compares the expression of the reporter gene GUS between Wt and
ada2b-I plants from two independent lines. Results are represented as expression levels
of GUS relative to ACT3 from two independent biological experiments for each line. The
expression level of GUS between the two experiments showed a lot of variance in the
wild type plants. In the ﬁrst experiment the expression of GUS in wild type plants was
higher as compared to adaZb-I plants but in the second experiment the expression levels
were similar. To determine whether the difference observed in the GUS expression was
signiﬁcantly different between the wild type and ada2b-1 plants, the data were analyzed
using a paired t-test, which showed that the expression of the reporter gene GUS after
cold acclimation was not signiﬁcantly different in wild type and ada2b-I plants at p S

0.05.

72

GUS expression in Wt and ada2b-1 plants

100.0 1
g . tha)
g 10.0 1 0ada2b(1)
g I IWt(l4)
2 0 Dada2b(l4)
0
.a ,
7. 1.0
E
a D ‘ g

0.1 -

 

Wt ada2b Wt ada2b

Experiment 1 Experiment 2

Figure 13: The expression of the reporter gene GUS fused to 4X-CRT is similar in wild
type and ada2b-1 plants. The expression of the reporter gene GUS fused to the promoter
element 4X-CRT was compared in wild type and ada2b-1 plants in two independent lines
(lines 1 and 14). Gene expression analysis was performed in Ws wild type and ada2b-1
plants after cold treatment at 4°C for 0 and 24 hours. Real-time RT PCR was performed
using primers for the ACT3 and GUS genes. The Ct values for the GUS gene were
normalized to the Ct values for the ACT3 gene. Results were represented as normalized
expression levels of GUS after cold treatment at 4°C for 24 hours using the DCt method.
The results shown are from two independent experiments.

73

The expression of the GUS reporter gene in wild type and ada2b-I plants
measured in the two independent experiments show different results. In the ﬁrst
experiment the expression of GUS is lower in the ada2b-1 plants as compared to wild
type plants whereas the expression of GUS in the second experiment is similar in ada2b-1
and wild type plants. Even though the paired t-test shows that the expression of GUS is
similar in the wild type and ada2b-1 plants at p S 0.05 we have to state that the
experiment is inconclusive. We cannot conclude whether the CBF regulation of COR

genes via CRTs requires ADA2b.

The expression of the 4X-CRT—GUS reporter gene was also compared between
Wt and gcn5-1 plants from six independent lines. Figure 14 shows the expression levels
of GUS normalized to AC T3 from two independent biological experiments. The
expression level of GUS was similar in the wild type and gcn5-1 plants within an
experiment, but the expression levels between the experiments showed a lot of variation.
Expression of the GUS reporter gene in four out of the six independent lines were slightly
higher in wild type plants as compared to gcn5-1 mutants in the ﬁrst experiment. In the
second experiment expression of the GUS reporter gene was slightly higher in the gcn5—1
mutants as compared to wild type plants. To determine whether the differences observed
were signiﬁcant, the data were analyzed using a paired t—test. Analysis of the data by
paired t-test show that the difference between expression of the reporter gene GUS after

cold acclimation is not signiﬁcantly different in wild type and gcn5-I plants at p S 0.05.

Since the CRT-driven expression of the reporter gene GUS during cold
acclimation was similar in wild type and gcn5-1 plants, we conclude that the CBF driven
expression of the reporter gene GUS fused to the 4X-CRT does not require GCN5. Thus,

74

GUS expression in Wt and gcn5-1 plants

 

-

r: _ O
.9.
8 10.0 ~ + -
E. I
:5 g n
"3 I
D 1 0 _
CD 0 9
1: O
0
E ,
a
E 0.1 ~ :
g A

Wt gcn5 Wt gcn5

0.0 - Experimentl Experiment 2

Figure 14: The expression of the reporter gene GUS fused to 4X-CRT is similar in wild
type and gcn5-1 plants. The expression of the reporter gene GUS fused to the promoter
element 4X-CRT was compared in wild type and gcn5-1 plants in six independent lines.
Gene expression analysis was performed in Ws wild type and gcn5-1 plants after cold
treatment at 4°C for 0 and 24 hours. Real-time RT PCR was performed using primers for
the ACT3 and GUS genes. The Ct values for the GUS gene were normalized to the Ct
values for the ACT3 gene. Results were represented as normalized expression levels of
GUS after cold treatment at 4°C for 24 hours using the DCt method. The results shown
are from two independent experiments.

75

the CBF-dependent activation of COR gene expression via CRT elements does not

require GCN5.

3.3.2 ADA2b and GCN5 are not required for the CBF-dependent expression of the COR
genes.

Overexpression of CBF is sufficient to drive the expression of COR genes in non-
cold-acclimated plants. Thus, the expression of COR genes in these plants is solely CBF-
dependent. Thus any difference in COR gene expression in wild type, adaZb-I and gcn5-
] plants overexpressing CBF would be through the CBF-dependent pathway of COR
gene regulation. To test whether the CBF-dependent expression of COR genes requires
ADA2b and/or GCN5, the expression of the CBF2 and GolS gene was compared in wild
type, adaZb-I and gcn5-1 plants overexpressing CBF2. Seeds from ADA2b-1 +/- and
GCN5-1 +/- plants overexpressing CBF 2 were sown on Gamborg plates containing Basta
to select for plants containing the CBF 2 transgene. Six independent lines overexpressing
CBF2 in the ADA2b-I +/- background and three independent lines overexpressing CBF 2
in the GCN5-I +/- background were chosen. Wt-like and mutant plants were identiﬁed
phenotypically. Tissue from each line was harvested and used for gene expression
analysis. The expression of CBF2 and the COR gene GolS in the wild type and mutant
backgrounds was assayed using real-time RT-PCR. Results in ﬁgures 15 and 16 are
represented as the normalized expression levels of CBF 2 and GolS obtained in two

biological replicate experiments.

The overexpression levels of CBF2 varied widely in the wildtype plants derived

from transformation of heterozygous ada2b plants (Fig. 15). However, the loss of

76

CBF2 expression in Wt and ada2b-1 plants

 

1000.0 7
5
E u
3 100.0 - '1' O
o +
a 1:1 I- g
E 10.0 ~ * $ ‘ +
E =
2 ° 0

1.0 - Wt ada2b Wt ada2b

Experiment 1 Experiment 2

[GolS expression in Wt and ada2b-1 plants

 

30.0 1
= D
g 25.0 ~ .
2
g' 20.0 - O D
L”.
5 15.0 - i O
E 10.0 -
g + i
Q 5.0 q A + _ +
:1
0.0 J *' ‘3' ‘ 8
Wt ada2b Wt ada2b
Experiment 1 Experiment 2

Figure 15: The expression of CBF2 and GolS is similar in wild type and adaZb-I plants
overexpressing CBF2. The expression of CBF 2 and GolS was compared in wild type and
ada2b-1 plants in six independent lines. Gene expression analysis was performed on non
cold acclimated Ws wild type and adaZb-I plants. Real-time RT PCR was performed
using primers for the ACT 3, CBF 2 and GolS genes. The Ct values for the CBF2 and GolS
genes were normalized to the Ct values for the ACT 3 gene. Results were represented as
normalized expression levels of CBF 2 and GolS using the DCt method. The results
shown are from two biological replicate experiments.

77

ADA2b did not have a signiﬁcant effect on the overexpression levels of CBF2 as seen by
t-test at p S 0.05. Also, the variation between the two biological replicate experiments
was not large. The expression of GolS in the six independent wild type lines showed a
large variation between lines. This was most likely due to the variation in CBF2 levels
between different lines. Again the variation between the two biological experiments was
very small. The expression of GolS in the adaZb-I plants showed a lot of variation
between the six independent lines. Results from the t-test showed that the loss of ADA2b
did not signiﬁcantly affect the expression levels of GolS at p S 0.05.

Figure 16 shows that the overexpression levels of CBF 2 varied widely in the three
independent Ws wild type lines. The expression of CBF2 in the gcn5-1 mutants was
higher as compared to wild type plants but this difference was not signiﬁcant at p S 0.05
as analyzed by the paired t-test. The expression of GolS in the three independent wild
type lines also showed a large variation between lines. This was probably due to the
variation in CBF2 levels between different lines. The expression of GolS in the gcn5-I
plants showed a lot of variation between the three independent lines. Results from the t-
test showed that the loss of GCN5 did not signiﬁcantly affect the expression levels of
GoIS at p S 0.05.

From the results we can conclude that the CBF-dependent expression of COR
genes does not require ADA2b and GCN5.

3.4 Discussion

CBF is a major activator of COR gene expression during cold acclimation.

Overexpression of CBF is sufﬁcient to drive the expression of COR genes in non-cold-

acclimated plants. CBF is able to interact with coactivators ADA2b and GCN5 in vitro

78

CBF2 expression in Wt and gcn5-I plants
1000.0 7

100.0 - A

A
10.0 — a O a

normalized CBF2 expression

 

A

1.0 J I l
O

0.1 - Wt gcn5 Wt gcn5
Experiment 1 . Experiment 2

GoIS expression in Wt and gcn5-1 plants

 

10.0 -
O
S O 0
E, _ A
5 1.0 D D
m
:3 e A l A
'3
g 0.1 ~ I
E
e
:1
Wt gcn5 Wt gcn5
0-0 " Experiment 1 Experiment 2

Figure 16: The expression of CBF 2 and GolS is similar in wild type and gcn5-1 plants
overexpressing CBF2. The expression of CBF 2 and GolS was compared in wild type and
gcn5-I plants in three independent lines. Gene expression analysis was performed on non
cold acclimated Ws wild type and gcn5-1 plants. Real-time RT PCR was performed using
primers for the ACT3, CBF 2 and GolS genes. The Ct values for the CBF 2 and GolS genes
were normalized to the Ct values for the ACT3 gene. Results were represented as
normalized expression levels of CBF2 and GolS using the DCt method. The results
shown are from two biological replicate experiments.

79

and these coactivators are known to regulate COR gene expression. Hence we proposed
that the coactivators ADA2b and GCN5 are involved in the CBF-dependent activation of
COR genes.

However, the results of the experiments presented here show that ADA2b and
GCN5 are not involved in the CBF-dependent expression of COR genes. The CBF-driven
activation of the 4X-CRT-GUS reporter gene was not affected by the loss of GCN5
during cold acclimation. Also, the loss of ADA2b and GCN5 had no effect on the
expression of COR genes in CBF overexpressing plants. Hence, the regulation of COR
gene expression by ADA2b and GCN5 most likely occurs in a CBF-independent manner.

The COR gene promoters contain several other cis-elements that are potentially
regulated by non-CBF transcription factors [120]. The expression of COR genes could be
controlled by transcription factors other than CBF either synergistically with CBF or
totally independent of CBF. Thus, ADA2b and GCN5 might regulate the expression of
the COR genes through other cis-elements.

ADA2b and GCN5 might regulate the expression of COR genes in a CBF-
independent manner through another transcription activator. It is also possible that the
regulation of COR genes by ADA2b and GCN5 occurs indirectly. Thus, ADA2b and
GCN5 might be regulating the expression of a transcription factor which in turn regulates
COR gene expression.

The expression of COR genes in ada2b-1 and gcn5-1 mutants is delayed and two-
fold reduced as compared to wild type plants but not completely repressed. Thus ADA2b
and GCN5 might regulate the COR gene expression in a subtle way. Overexpression of

CBF 2 by passes the requirement of a cold stimulus for the induction of COR genes.

80

Overexpression of CBF2 may also bypass the requirement for ADA2b and GCN5 for
COR gene induction and regulation as a large amount of CBF2 might be capable of
recruiting the transcription machinery required for COR gene activation. Hence we are
unable to see any effect of ADA2b and GCN5 in the CBF dependent pathway of COR
gene expression in plants overexpressing CBF.

We conclude from the results that the regulation of COR genes by CBF does not
require ADA2b and GCN5. The CBF driven activation of COR genes via the CRTs is
independent of the function of GCN5. Thus, the regulation of COR genes by ADA2b and
GCN5 may occur through a non-CBF transcription factor binding to cis-elements other

than the CRTs or it may occur indirectly.

81

Chapter IV

Histone acetyltransferases and their effect on COR gene regulation during cold

acclimation in Arabidopsis.

4.1 Introduction

The results described in previous chapters have revealed that the acetylation of
histone H3 at the COR gene promoters increases in response to cold acclimation. The
histone acetyltransferase GCN5 regulates the expression of COR genes but is not required
for acetylation of histone H3 at the COR gene promoters during cold acclimation. To
identify the HAT (or HATS) responsible for acetylation of histone H3 at the COR gene
promoters during cold acclimation, several candidate genes were investigated in the

current chapter.

The Arabidopsis genome includes 12 HAT encoding genes belonging to four different
families. The GNAT family contains members that have a HAT domain characterized by
four motifs (AD) [16]. The GNAT family in Arabidopsis has three members, designated
GCN5, ELP3 and HATl (or HAG], HAG2 and HAG3 respectively) [34]. The MYST
family members have HAT domains similar to those of the GNAT family but comprising
only the A motif. The MYST family in Arabidopsis contains two members, designated
HAG4 and HAGS. The Arabidopsis genome contains ﬁve members of the CBP/p300
family, designated HAC], HAC2, HAC4, HAC5 and HAC12. The CBP family members
in Arabidopsis differ from their animal counterparts in that they lack a bromodomain,
which is known to bind acetylated histones [66]. The Arabidopsis genome encodes two

members of the TAF 1 family, designated HAF 1 and HAF2/TAF1. The TAFl protein is

82

part of the general transcription factor TFIID [121-123]. Other eukaryotes such as yeast,

ﬂies and humans all have one single member in the TAP 1 family [106, 124].

GCN5 is the most extensively characterized HAT in Arabidopsis. Microarray
studies showed that GCN5 regulates the expression of 5% of the Arabidopsis genes [3 7].
GCN5 has been shown to physically interact in vitro with the transcriptional activator
CBF and also with the coactivator protein ADA2 [51]. GCN5 regulates the. activity of
cold-regulated (COR) genes during cold acclimation in Arabidopsis [3 7]. In addition,
GCN5 regulates the ﬂoral meristem activity through the WUS/AG pathway [76]. GCN5
along with another HAT protein, TAF 1 , also regulates the expression of light-responsive
genes CAB2 and RBCS-lA that are bound by the transcriptional activator HY5 [54, 55].
Some members of the CBP family are involved in the regulation of ﬂowering by
regulating the ﬂowering repressor FLC [52, 53]. Homozygous mutants of hacI show
delayed ﬂowering due to an increase in the expression of FLC. This phenotype is further
enhanced in haclhac5 and hathacIZ double mutants. Members of the HAC family thus
act as repressors of F LC rather than as activators. Chromatin immunoprecipitation assays
performed using antibodies against acetylated H3 and H4 show that the acetylation
proﬁles at the FLC promoter are not different in the mutants than in wild type plants [52,
53, 83]. Thus the HAC genes regulate FLC in a manner that is independent of histone
acetylation at the FLC promoter. The manner in which they regulate the expression of

FLC is not known as yet.

To ﬁnd the HAT (or HATS) responsible for acetylating histone H3 at the COR
gene promoters during cold acclimation, several members of the HAC and TAF 1 families

were tested for their role in COR gene regulation during cold acclimation.

83

4.2 Materials and methods:
4.2.1 Plants used and growth conditions:

Seeds for the W5 and Col-0 plants were obtained from the Thomashow lab. T-
DNA insertion lines were obtained from the ABRC stock center for the hacI-l, hac5-1
and hac12-1 homozygous mutants. Seeds for the had hac12 double mutant were obtained
from the Noh lab in Korea. All of these T-DNA insertion lines were constructed in the
Col-0 background. Seeds for the taﬂ -1 homozygous mutants were obtained from the
Zhou lab in France. This T-DNA line was in the Ws background. Plant grth conditions

were as described in 2.2.1

4.2.2 RNA isolation and gene expression analysis:

RNA isolation and gene expression studies were carried out as described in
section 3.2.3. Genes expression analysis of the CBF2 and COR genes in wild type and
mutant plants were compared by ANOVA using SAS.

4.2.3 Chromatin immunoprecipitation assay

Chromatin immunoprecipitation assays were performed on Col-0, hathacIZ,

Ws-Wt and taﬂ -1 plants as described in sections 2.2.2 through 2.2.6.
4.3 Results:
4.3.1 HAC gene family:

We ﬁrst asked whether members of the HAC gene family of histone
acetyltransferase play a role in regulating the expression of COR genes upon cold

acclimation.

84

Three genes from the HAC family, HAC], HAC5 and HAC12, were chosen for
this study. The HAC2 gene was excluded because the encoded protein does not have a
functional HAT domain [52, 83]. To determine whether any of the HATS from the HAC
gene family play a role in regulating COR gene expression during cold acclimation, wild
type Col-O and mutant plants were treated at 4°C for various times and gene expression
analysis was carried out. The hac12-l homozygous mutant seeds failed to germinate and
hence this mutant was not included in this study. The ability of the hacI-Z, hac5-1 and
hathacIZ mutants to activate COR gene expression upon cold acclimation was
compared with wild type Col-0 plants. Plants were cold treated at 4°C for O, l, 2, 4, 6, 8,
10, 12 and 24 hours. Gene expression analysis was performed using real-time RT-PCR.
The ﬁgures represent an average of two biological experiments for the hacI-Z and hac5-1

mutants and an average of four biological experiments for the hathac12 double mutant.

Figure 17 shows the expression of ACT 3, CBF 2 and COR genes in wild type and
hac1-2 plants upon cold acclimation. The ACT 3 expression levels did not change with
time upon cold acclimation and were similar in the wild type and hacI-Z plants. The
expression of CBF 2 in hac1-2 plants was induced upon cold acclimation, peaked between
two and four hours and then decreased in concert with wild type plants. The expression of
all COR genes was induced upon cold acclimation in wild type and hacI-2 plants. The
expression of COR47 and GoIS was similar in hacI-Z and wild type plants. The
expression of COR6. 6, COR15a and GolS genes was lower in hacI-Z plants than in wild
type plants. This difference was not signiﬁcant at p S 0.05 as analyzed by ANOVA,
likely because the expression of these COR genes in wild type plants showed a large

variation in the two independent experiments. We can conclude that our evidence

85

     

 

 

 

 

 

 

 

35 - ACT3 expression 1200 a CBF2 expressron
<— 30 - g +C01
L; i i E €800 ‘ -I-hacl-2
on
g 25 r E
3;: E400 -
20 ‘ +Col In
-I-hac1-2
15 l 1 I 1 1 r r 1 l 0 "1
012468101224 012468101224
Time in h at 4°C Time in h at 4°C
500 - COR6.6 expression 16000 ~ COR15a expressron
400 ~
:1 |=I12000 ~
g +001 ~§ +Col
3300 q +hacl-2 .g 3000 _ -I-hacl-2
DE 200 .1 CE
3:". E
Q G
@100 _ [a 4000 a
0 - 0 ~
012468101224 012468101224
Time in h at 4°C Time in h at 4°C

Figure 17: The expression of CBF2 and COR genes increases in wild type Col-0 and
had-2 plants upon cold acclimation. Gene expression analysis was performed in Col-0 '
and hacI-Z plants after cold treatment at 4°C for O, 1, 2, 4, 6, 8, 10, 12 and 24 hours.
Real-time PCR was performed using primers for the AC T3, CBF2 and COR genes and
the Ct values obtained for COR genes were normalized to the Ct values obtained for
ACT3. Ct values for the ACT 3 promoter were not normalized. Fold induction of CBF 2
and COR genes after cold treatment were calculated by normalizing the cold treated
plants with the untreated sample. Error bars represent standard error where N=2.

86

Figure 17 continued

150 _ COR47 expression

Fold induction
8
O

 

0

Fold induction
‘5
O
O

 

 

+Col
-I-hac l -2

 

 

1 2 4 6 8 1012 24
Timeinhat4°C

GolS expression

 
  
 

+Col
+hacl -2

  

Fold induction
A
o
o

 

012468101224
Time in h at 4°C

800 -

 

87

  

COR78 expression

+Col
+hac l -2

 

T —l

012468101224
Timeinhat4°C

does not support the hypothesis that HAC] regulates the expression of COR genes during

cold acclimation.

Figure 18 shows the expression of AC T3, CBF2 and COR genes in wild type and
hac5-1 plants upon cold acclimation. Data for the wild type plants were obtained from
ﬁgure 17. The ACT3 expression levels did not change with time upon cold acclimation
and were similar in the wild type and hac5-1 plants. The expression of CBF2 in hac5-1
plants was induced upon cold acclimation, peaked at two hours and then decreased in
concert with wild type plants. The expression of CORI5, COR 78 and GolS upon cold
acclimation was Similar in hac5-I and wild type plants. COR6. 6 expression was modestly
lower in hac5-I plants at 10, 12 and 24 hours after cold acclimation as compared to wild
type plants. The expression of COR47 in hac5-I plants was lower at 8 and 12 hours after
cold acclimation as compared to wild type plants. Both the differences were not
signiﬁcant at p S 0.05 as analyzed by AN OVA using SAS. Thus, our evidence does not
support the hypothesis that HAC5 regulates the expression of COR genes during cold

acclimation.

The had -1 single mutant was late ﬂowering, likely due to an increase in the
expression of F LC, a ﬂowering repressor. This phenotype was enhanced in the
hathacIZ mutant, suggesting that the HACl and HAC12 HATS together regulate the
expression of FLC [52, 53, 83] . To determine whether the HACl and HAC12 HATS act
synergistically to control the expression of CBF 2 and/or COR genes the haclhac12

double mutant was used in this study.

88

 

 

 

 

 

 

 

 

35 1 ACT3 expression 1200 - CBF2 expressron
,, 30 ~ .2 -0-Col
g i i é £3800 4 -I-hac5-l
60
g 25 - . E
3: g 400
20 A +Col “‘
-I-hacS-l
15 l T T l l T l 1 I O
012468101224 012468101224
Time in h at 4°C Time in h at 4°C
500 - COR6.6 expression 16000 1 COR15a expressron
g 400 ‘1
g +Col § 12000 4 +001
,3 300 ‘ -I-hac5-1 g +hac5-1
“U
5200 .5 800° 1
"" '5
g 100 . E 4000 -
O 4 0 s
012468101224 012468101224
Time in h at 4°C Time in h at 4°C

Figure 18: The expression of CBF2 and COR genes increases in wild type Col-0 and
hac5-1 plants upon cold acclimation. Gene expression analysis was performed in Col-0
and hac5-I plants after cold treatment at 4°C for 0, 1, 2, 4, 6, 8, 10, 12 and 24 hours.
Real-time PCR was performed using primers for the ACT3, CBF2 and COR genes and
the Ct values obtained for COR genes were normalized to the Ct values obtained for
ACT 3. Ct values for the ACT3 gene were not normalized. Data for the wild type plants
were obtained from ﬁgure 17. Fold induction of CBF 2 and COR genes aﬁer cold
treatment were calculated by normalizing the cold treated plants with the untreated
sample. Error bars represent standard error where N=2.

89

Figure 18 continued

75-

50-

25

Fold induction

 

Fold induction

COR47 expression

012468101224
Timeinhat4°C

J

 

 

 
   

+Col
-I-hac5-l

  

 

GolS expression

+Col
-I-hac5-l

  

 

012468101224
Timeinhat4°C

1600 lCOR78 expression

1200 4 Col

+hac5-1

Fold induction

.b 0°
C O
O O
i

 

O
m

012468101224
Timeinhat4°C

90

Figure 19 shows the expression of ACT 3, CBF2 and COR genes in wild type and
hathacIZ plants upon cold acclimation. The ACT3 expression levels did not change
with time upon cold acclimation and were similar in the wild type and haclhac12 plants.
The expression of CBF 2 in hacI hac12 plants was induced upon cold acclimation, peaked
at two hours and then decreased in concert with wild type plants. Although the peak level
of CBF 2 expression was lower in the haclhac12 mutants as compared to wild type plants
this difference was not signiﬁcant at p S 0.05. COR6.6 and COR15a expression levels in
haclhac12 plants were lower than in wild type plants. The expression of these COR
genes in wild type plants showed a large variation in the two independent experiments,
and the difference observed between mutant and wild type plants was not signiﬁcant at p
S 0.05 as analyzed by ANOVA using SAS. The expression of the COR47 gene was lower
in hathacIZ as compared to wild type plants. The expression of COR47 was lower at 8,
1'0, 12 and 24 hours after cold acclimation. This difference was signiﬁcant at p S 0.05 as
analyzed by ANOVA using SAS. The evidence presented in these results suggests that
HAC] and HAC12 may be involved in the regulation of COR4 7 gene expression during

cold acclimation but none of the other COR genes or CBF 2.

The induction of CBF2 and COR genes upon cold acclimation in the hacI-Z and
hac5-I mutants was Similar to the wild type Col-0 plants. Thus HACl and HAC5 genes
do not seem to regulate the expression of COR genes upon cold acclimation. In the
haclhacIZ double mutant, expression of the COR4 7 gene during cold acclimation was
signiﬁcantly lower than in the wild type plants but none of the other COR genes are
signiﬁcantly affected. Hence, HAC] and HAC12 are involved in controlling the

expression of COR47 upon cold acclimation. As the hacI-Z single does not affect the

91

 

 

 

 

 

   

:3 1 ACT3 expression 2500 _ C F2 expressron
2000 -

,_, 27 . E +CO1
2 26 1 31500 ‘ -I-h 1h 12
go 25 q :2 ac ac
g 24 ' E
< 23 4 -0-Col g

22 I -I-haclhac12

21 1“" ‘l""’t" t’ﬂ—r—r—Wﬁ—t

012468101224 012468101224
Time in h at 4°C Time in h at 4°C

400 " COR6.6 expression 10000 ' COR15a expression
,8 300 - 8 7500 —
5 +Col “*3 --O-Col
5 5
g 200 ~ -I-haclhac12 E 5000 ~+haclhac12
E 2
£ 100 a 2500

0 0
012468101224 012468101224
Time in h at 4°C Time in h at 4°C

Figure 19: The expression of CBF2 and COR genes increases in wild type Col-O and
hathac12 plants upon cold acclimation. Gene expression analysis was performed in
Col-0 and hathac12 plants after cold treatment at 4°C for 0, 1, 2, 4, 6, 8, 10, 12 and 24
hours. Real-time PCR was performed using primers for the ACT3, CBF2 and COR genes
and the Ct values obtained for COR genes were normalized to the Ct values obtained for
ACT3. Ct values for the ACT3 gene were not normalized. Fold induction of CBF2 and
COR genes after cold treatment were calculated by normalizing the cold treated plants
with the untreated sample. Error bars represent standard error where N=4.

92

Figure 19 continued

60 7 COR47 expression

20-

Fold induction

 

0

7500 -

l

g 6000
.O

E
'1, 3000 —

F01

1500 a

 

40 * +Col

-I-haclhac12 H

O J,“

_ +Col

   
    

Fold induction

 
  
 

 

T l l l I l j

1 2 4 6 8 101224
Time in h at 4°C

GolS expression

  

-I-hac1hac12

 

012468101224
Timeinhat4°C

93

800

COR78 expression

 
  
 

+Col
+haclhac12

J

  

 

 

012468101224
Timeinhat4°C

expression of COR genes, the effect of the haclhac12 double mutant on COR47 gene

expression is most likely due to the loss of the HAC12 gene.

Since the expression of the C 0R4 7 was affected in the hathac12 mutant we
checked whether the acetylation of histone H3 at the C OR47 promoter was also affected
inthe double mutant. Results in chapter I showed that the acetylation of histone H3 at all
the COR gene promoters increased upon cold induction and corresponded with the
induction of COR genes. HAC] and HAC12 together may acetylate histone H3 at the
COR4 7 promoter speciﬁcally during cold acclimation and control its gene expression. If
HACl and HAC12 are responsible for the acetylation of histone H3 at the COR4 7
promoter we would expect to see no increase in the acetylation at the COR47 promoter in
the haclhac12 mutant upon cold acclimation. The acetylation of histone H3 at the other
COR gene promoters will be similar to wild type plants and will not be affected in the

hathacI 2 mutant.

ChIP assays were conducted on Col-0 and hathacIZ plants cold treated at 4°C
for 4 and 24 hours. Non-acclimated plants were used as controls. ChIP assays were
performed using the anti-acetylated histone H3 and the anti-histone H3 antibodies. Figure
20 shows that the acetylation of histone H3 at the COR47 and the other COR gene
promoters increased upon cold acclimation in hathacIZ mutant plants, in a manner and
to a degree similar to that observed for the Col-0 wild type plants. Exceptions were
observed at the COR 78 and GolS promoters where the acetylation of histone H3 in wild
type plants was signiﬁcantly higher than that observed for the hathac12 mutant. The
level of histone H3 at the ACT2/7 promoter was unaffected in the hathacIZ and wild

type plants upon cold acclimation. Hyperacetylation of histone H3 and depletion of

94

ACT2/7 promoter 2-0 COR6.6 promotflr
I t

   

     
   
   
   

 

 
   
   

1.0 "M g 1-5 Ih 1h 12
a -haclhac12 E , ac ac
*5 1 0
E ‘ r 3
< z ‘ g 0.5
. . . _ E
.. . .._.z. i-
“T‘” ‘1'— g
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C
m 1-0 ’ COR15a promoter 2-0 COR47 promoter
é I Wt . a ' Wt
E haclhac12 I % 'haclhacl2
< 0.5 - <
s ' s
a ‘ N
g 00 .l 7.. _-_Z., r. E
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C
M 3-5 COR78 promoter 3 4-5 GolS promoter
§ 2.8 I Wt m I Wt
3 2,1 lhaclhac12 E 3'0 lhaclhac12
E 1.4 E 1.5
T: 0.7 T:
g 0.0 g 0.0
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C

Figure 20: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type Col-0 and
hathacIZ plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc
for acetylated histone H3 (K9, K14) and the C-terminal of histone H3. Samples were
analyzed using real-time PCR to detect acetylated histone H3 and total histone H3 at
various COR gene promoters. Ct values obtained were normalized to values obtained for
AC T2/ 7 promoter. Data were represented as ratio of acetylated histone H3 to total histone
H3. Ct values for the AC T2/7 promoter were not normalized. Error bars represent SE
where N (number of experiments) = 2.

95

histone H3 occurs at some COR gene promoters during cold acclimation. The occupancy
of the COR gene promoters was tested in hathacIZ and wild type plants. Figure 21
shows that the occupancy of histone H3 at the COR 78 and GolS promoters decreased
during cold acclimation in wild type plants. The histone H3 occupancy at the COR6. 6,
COR15a and COR47 promoters is unchanged upon cold acclimation in wild type plants.
The effect of hathac12 mutant on histone H3 occupancy was also tested. Figure 21
shows that the level of total histone H3 at the ACT2/ 7 promoter was lower in the non-
acclimated control hathacIZ mutants as compared to the wild type plants. This most
likely accounts for the increased histone H3 occupancy observed at all the COR gene
promoters in the non-acclimated hathacIZ plants. Data analysis by ANOVA indicated
that the level of histone H3 at the C 0R4 7 promoter remained unchanged after cold
acclimation in Col-O and hathacIZ plants. Also, the decrease in histone H3 at the
COR 78 and G015 promoters in the hathacIZ mutants was similar to the decrease

observed in Col-0 plants.

Given these results, we can conclude that HAC] and HAC12 histone
acetyltransferases together contribute to the regulation of COR47 gene expression but do
so without affecting the acetylation or the occupancy of histone H3 at the COR47 gene

promoter.
4.3.2 TAF] gene family:

The TAF 1 family of histone acetyltransferases in Arabidopsis consists of two members,
designated HAFl and HAF2/T AF 1 [34, 54, 55]. The TAFl protein is a part of the general

transcription factor complex TFIID [121]. The HAF2/T AF 1 protein has been shown to

96

 

 

     

 

   

 

 

50 1 ACT2/7 promoter .5 2-0 ‘ COR6.6 promoter
.. 4o « I Wt g- I Wt
g. 30 - Ihaclhach ,\° Ihaclhac12
' 1:
°\° o
é.
E
2
O 4 24
Time in h at 4°C Time in h at 4°C
.5 2-0 7 COR15a promoter .5 2.0 ' COR47 promoter
.E' 1'6 ' "W .E‘ 1.5 ~ IWt
e\° 1-2 lhaclhac12 e\° lhaclhac12
“g 0.8 '8
‘7. 0.4 ‘5
go. g
0 4 24 0 4 24
Time in h at 4°C Time in h at 4°C
1-5 7 COR78 promoter 2-5 1 GolS promoter

IWt
Ihaclhac12

2.0 r I Wt

1.5 - Ihaclhac12
1.0 -

  

 

 

 

normalized % input
normalized % input
Q
ﬁll

 

O 4 24 O 4 24
Time in h at 4°C Time in h at 4°C

Figure 21: Level of total histone H3 at some COR gene promoters decreases upon cold
acclimation. ChIP assays were performed on wild-type and hue] hac12 plants cold
acclimated at 4°C for 0, 4 and 24 hours using an antibody speciﬁc for the C-terminal of
histone H3. Samples were analyzed using real-time PCR to detect total histone H3 at
various COR gene promoters. Ct values obtained were normalized to values obtained for
ACT 2/ 7 promoter. Data were represented as normalized levels of total histone H3. Ct
values for the AC T2/ 7 promoter were not normalized and were represented as % input
values. Error bars represent SE where N (number of experiments) = 2.

97

regulate light-responsive genes and homozygous haf2/taﬂ -1 mutant plants show
decreased acetylation of histones H3 and H4 at the target gene promoters. Microarray
studies showed that a homozygous null mutation in TAFI affects the expression of 9% of
the total genes in Arabidopsis. A homozygous null mutant of HAFI showed no visible

phenotype and hence this gene was not used in this study [54, 55].

We tested whether the histone acetyltransferase HAF 2/T AF 1 regulates the
expression of COR genes during cold acclimation. The expression of CBF 2 and COR
genes was compared in taﬂ -1 mutant and wild type Ws plants after treating the plants at
4°C for O, 1, 2, 4, 6, 8, 10, 12 and 24 hours. RNA was extracted and gene expression was

analyzed using real-time RT-PCR, as described previously.

The results (Figure 22) show that the expression of the AC T3 gene did not change
with time and was similar in the wild type W3 and the taﬂ -1 plants. The expression of
CBF2 in the taﬂ -1 plants was induced within one hour of cold acclimation, peaked after
two hours and then decreased in concert with wild type Ws plants. The expression of all
the COR genes increased upon cold acclimation in wild type plants. The expression
COR6. 6, COR15a and GolS genes in the taﬂ-I mutant plants was signiﬁcantly lower
than in the wild type Ws plants after 8 hours of cold acclimation (at p5 0.05). A similar
trend was observed for the COR47 and COR 78 genes but this difference was not
signiﬁcant at p S 0.05 as analyzed by AN OVA using SAS. We conclude that the TAF 1
histone acetyltransferase controls the expression of COR6. 6, COR15a and GolS genes
during cold acclimation but is not required for the expression of COR4 7 and COR78
genes. Thus, TAF 1 controls the regulation of a subset of COR genes during cold

acclimation.

98

l

ACT3 expression 400
35 “ +Ws

CBF2 expression

La)

O

O
1

Average Ct
w
o
Fold induction
N
o
o

 

 

 

 

 

 

 

 

 

100 ~
25 O ' I l l 1 I I r F 1
012468101224 012468101224
Time in h at 4°C Time in h at 4°C
250 w COR6.6 expression 3000 1COR15a expression 1'
:1 200 4 g 2500 4
@150 +WS '5 2000 - +Ws
g +tafl ..g 1500 ‘ +taf1
.5100 _ .5
:3 g 1000 -
h 50 ‘ [a 500 ~
0 - O —
012468101224 012468101224
Time in h at 4°C Time in h at 4°C

Figure 22: The expression of CBF2 and COR genes increases in wild type Ws and taﬂ-I
plants upon cold acclimation. Gene expression analysis was performed in Ws and ta]? -1
plants aﬁer cold treatment at 4°C for 0, l, 2, 4, 6, 8, 10, 12 and 24 hours. Real-time PCR
was performed using primers for the ACT3, CBF2 and COR genes. Ct values obtained for
COR genes were normalized to the Ct values obtained for ACT 3 . Fold induction of CBF 2
and COR genes after cold treatment were calculated by normalizing the cold treated
plants with the untreated sample. Ct values for ACT 3 were not normalized and were
represented as average Ct. Error bars represent standard error where N=4.

99

Figure 22 continued

Fold induction

Fold induction

402

30'

20

10

 

O4

5000 -

4000
3000
2000
1 000

COR47 expression

 

O

J

J

l

 

+Ws
-I-taf1

 

1 2 4 6 8101224
Timeinhat4°C

GolS expression

 
 
 
 

+Ws
-I-tafl

   

 

I ‘1

012468101224
Timeinhat4°C

500 _ COR78 expression
g 400

1

 

 

5 +Ws
3300 ‘ -I-taf1
.5
E 200 -
O
F“ 100 .

O a

 

012468101224
Timeinhat4°C

100

Given the result that the disruption of T AF 1 resulted in diminished cold-induced
expression of some COR genes we tested whether the acetylation or occupancy of histone
H3 at these COR gene promoters was also affected in the taﬂ -1 mutant. Chromatin
immunoprecipitation assays were carried out on wild type W8 and ta]? -1 plants cold
acclimated at 4°C for 4 and 24 hours using the anti-acetylated histone H3, anti-tetra-
acetylated histone H4 and anti-histone H3 antibodies. Non-acclimated plants were used
as controls. Real-time PCR was performed by using primers spanning 100-150 bp of the

COR gene promoters. Data were analyzed using two-factor ANOVA.

The results shown in Figure 23 reveal that the increase in acetylation of histone
H3 at the COR gene promoters in the tafI-I mutants was similar to the increase in wild
type plants. The increase in acetylation in the taﬂ -1 mutants occurred at all the COR gene
promoters irrespective of whether the expression of that COR gene was affected in the
taﬂ-I mutant. The only difference observed was that the increase in acetylation after 24
hours of cold treatment at the COR 78 promoter was higher in wild type plants as
compared to the taﬂ-I mutant. However, the expression of COR 78 was not affected in
the taﬂ -1 mutant upon cold acclimation. We also tested whether histone H3 occupancy at
the COR gene promoters was affected by the taﬂ -1 mutation. Figure 24 shows that the
level of total histone H3 at the ACT2/ 7 promoter was lower in the 24 hour cold
acclimated taﬂ -1 mutant as compared to the wild type plants. So any difference between
wild type and taﬂ -1 plants at the 24 hour time point could be the result of higher levels of
histone H3 at the ACT 2/ 7 promoter in wild type plants. Results indicate that the level of
total histone H3 at the COR 78 and GolS promoters decreased after 24 h of cold treatment

at 4°C in the taﬂ -1 mutants in a manner and to a degree similar to wild type plants. A

101

1 5 ACT2/7 promoter I Wt 1 2 COR6. 6 promoter I Wt

      

     

 

T tafl 0% l ‘ tafl
g 1.0 +1~E l
M U
m ‘t 1
0 0.5 ~ 1: «
0.0 i a - --
24 g 0 4 24
0Time in 4h at 4°C Time in h at 4°C
1.6 COR15a promoter 20 .. COR47 promoter
5 I Wt a | I Wt
m m 1-5 ‘1 3 tafl
E E 2
< < 1.0 1
B E
:E ..‘S 0.5 4
a T: L
E E 0.0
S 0 4 24 S 0 4 24
Time in h at 4°C Time in h at 4°C
I Wt
30 COR78 promoter m 2.5 GolS promoter ltafl
E 2-5 ' W E 2.0
a - E 1.5
< 1.5 <
3 1'0 3 1.0
E 0.5 g 0.5
a a
E 0.0 E 0.0
g 0 4 24 g 0 4 24
Time in h at 4°C Time in h at 4°C

Figure 23: Ratio of acetylated histone H3 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type Ws and taﬂ-
] plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc for
acetylated histone H3 (K9, K14) and the C-terminal of histone H3. Samples were
analyzed using real-time PCR to detect acetylated histone H3 and total histone H3 at
various COR gene promoters. Ct values obtained were normalized to values obtained for
AC T2/ 7 promoter. Data represented as ratio of acetylated histone H3 to total histone H3
at various COR gene promoters. Ct values for the AC T2/ 7 promoter were not normalized.
Error bars represent SE where N (number of experiments) = 3.

102

       

     

   

30 ACT2/7 promoter 1 2 COR6.6 promoter
I Wt E. ' l I Wt
*- l tafl F M tafl
= 20 2
5“ °\ :21,
0 E
O
0 4 24 = 0 4 24
Time in h at 4°C Time in h at 4°C
1 2 COR15a promoter 1 6 COR47 promoter
‘5 ' I Wt ‘5 ' I Wt
2' 2' 1.2 I taf]
-- 0.8 --
°\° e\°
E B
E 0.4 E
E E
O 0.0 c
n o 4 24 = 0 4 24
Time in h at 4°C Time in h at 4°C
._ 1.5 COR78 promoter 15 GolS promoter
a "W i
.E ' tafl .E
e\° e\°
ﬂ '6
0 0
a a
a GI
E E
Time in h at 4°C Time in h at 4°C

Figure 24: Level of total histone H3 at COR gene promoters decreases upon cold
acclimation. ChIP assays were performed on wild-type Ws and taﬂ -1 plants cold
acclimated at 4°C for 0, 4 and 24 hours using an antibody speciﬁc for the C-terminal of
histone H3. Samples were analyzed using real-time PCR to detect total histone H3 at
various COR gene promoters. Ct values obtained were normalized to values obtained for
AC T2/ 7 promoter. Data represented as normalized levels of total histone H3 at various
COR gene promoters. Ct values for the AC T2/ 7 promoter were not normalized and were
represented as % input. Error bars represent SE where N (number of experiments) = 3.

103

decrease in histone H3 occupancy was also observed at the COR6. 6 promoter but the
level of histone H3 in wild type plants alter 24 hours of cold acclimation was lower as
compared to the taﬂ -1 mutant. This could be due to the higher level of histone H3
observed at the AC T 2/ 7 promoter observed in wild type plants. There was no decreased in
histone H3 occupancy at the COR15a and COR4 7 promoters upon cold acclimation in the
wild type or mutant plants. Thus, TAF 1 is not required for the acetylation or depletion of

histone H3 from the COR gene promoters upon cold acclimation.

TAF] histone acetyltransferase might be controlling the expression of COR genes
by regulating the acetylation of histone H4 at the COR gene promoters [31, 125]. Hence,
ChIP assays were carried out in wild type Ws and taﬂ-I mutant plants using anti-tetra-
acetylated histone H4 antibodies. The results are represented as a ratio of acetylated
histone H4 to total histone H3. Histones H3 and H4 exist as dimers in nucleosomes and
thus the level of change in total histone H3 upon cold acclimation should reﬂect the
changes occurring with histone H4. We know that the level of histone H3 decreases at
some COR gene promoters during cold acclimation and hence representing the level of
acetylated H4 as a ratio accurately reﬂects the actual change occurring at the COR gene
promoters during cold acclimation. Results in ﬁgure 25 show that the acetylation of
histone H4 at most of the COR gene promoters did not increase upon cold acclimation in
wild type plants and the level of acetylated histone H4 in the taﬂ -1 plants was similar to
the levels observed in wild type plants. Acetylation of histone H4 increased at the 0018
promoter upon cold acclimation but there was no difference in the acetylation levels
between wild type and taﬂ-I plants. Acetylation of histone H4 might be involved in the

regulation of GolS expression but none of the other COR genes. We can conclude from

104

0.16 ACT2/7 promoter 1.6 COR6.6 promoter

        

  
  

 

 

I Wt a I Wt
E 0.12 , tafl E 1.2 I tafl
E 0.08 I I 2 0.8
O ' ‘ "I '5
< 004 . ‘ .5 0.4
0.00 ' ‘ ‘ ‘ — E 0.0
0 4 24 S 0 4 24
Time in h at 4°C Time in h at 4°C
1.6 COR15a promoter 12 - COR47 promoter
3 I Wt a l I Wt
v 1.2 tafl v r '
I I
3 0.8 :5
3 E
E 0.4 E
E 0.0 ”E’
a 0 4 24 g 0 4 24
Time in h at 4°C Time in h at 4°C
1.0 C0R73 promoter 12 GolS promoter
"SI nan E
e 05 5
<1 - <
'5 '6
0 0
E E
a S
g 0.0 E
‘= 0 4 24 2 0 4 24
Time in h at 4°C Time in h at 4°C

Figure 25: Ratio of acetylated histone H4 to total histone H3 at COR gene promoters
increases upon cold acclimation. ChIP assays were performed on wild-type W5 and tafl-
] plants cold acclimated at 4°C for 0, 4 and 24 hours using antibodies speciﬁc for tetra-
acetylated histone H4 and the C-terminal of histone H3. Samples were analyzed using
real-time PCR to detect tetra-acetylated histone H4 and total histone H3 at various COR
gene promoters. Ct values obtained were normalized to values obtained for ACT2/7
promoter. Data represented as ratio of acetylated histone H4 to total histone H3 at various
COR gene promoters. Ct values for the ACT2/ 7 promoter were not normalized. Error bars
represent SE where N (number of experiments) = 3.

105

the results that the regulation of COR gene expression by TAFl occurs independently of

changes in histone acetylation at the target COR gene promoters.

The results presented above indicate that the TAFl histone acetyltransferase
controls the expression of C 0R6. 6, COR15a and GolS genes during cold acclimation but
has no effect on the expression of COR4 7, COR 78 or CBF 2 genes. The set of genes
affected by TAFl is different from the set controlled by the HACl and HAC12 proteins.
Although TAF 1 histone acetyltransferase regulates expression of a subset of COR genes
the level of histone H3 and H4 acetylation and histone H3 depletion during cold
acclimation was similar in the wild type and taﬂ -1 plants. Thus, TAF 1 does not control
the expression of COR genes by regulating the acetylation of histones H3 and H4 or

nucleosome occupancy.
4.4 Discussion

Studies carried out in chapter 2 show that GCN5 regulates COR gene expression
but this activity is independent of its ability to act as a HAT. GCN5 seems to affect the
occupancy of histone H3 and thus nucleosomes at the COR6. 6, COR 78 and 6015
genepromoters. In order to ﬁnd the HAT responsible for carrying out the acetylation of
histone H3 at the COR gene promoters, several HAT genes were studied for their
possible role in regulation of COR gene expression during cold acclimation. Results show
that the HACl and HAC12 proteins together regulate the expression of the COR4 7 gene
but none of the other COR genes. Chromatin immunoprecipitation assays show the

HACl and HAC12 are not required for the acetylation of histone H3 at the COR4 7

106

promoter. The HACl and HAC12 genes also do not affect the occupancy of histone H3 at

the COR4 7 promoter.

The HAC genes may be controlling the regulation of C OR47 by regulating the
acetylation of histone H4, by acetylation of a transcription factor or their function might
be totally independent of their-acetyltransferase activity. The members of the HAC
family are broad speciﬁcity enzymes and it is possible that they regulate COR47 gene
expression by affecting the acetylation of histone H4 during cold acclimation [106].
Studies were conducted on wild type Wt plants to determine whether the acetylation of
histone H4 increases at COR gene promoters dm'ing cold acclimation. Results show that
the acetylation of histone H4 at the COR gene promoters (except for the GoIS promoter)
does not increase upon cold acclimation in wild type Ws plants (Fig.25). Hence the
regulation of COR4 7 by HACl and HAC12 is independent of their histone
acetyltransferase activity at the COR4 7 promoter. CBP and p3 00 in other eukaryotes are
known to acetylate transcription factors and regulate the expression of target genes [26,
27]. Thus the HAC genes might control the expression of COR4 7 by acetylating a
transcription factor that in turn controls the expression of COR4 7. The ability of GCN5 to
regulate COR4 7 expression is also independent of its HAT activity. It is possible that in
the absence of GCN5, HAC] and HAC12 might acetylate histone H3 at the COR47
promoter and thus we fail to see an effect of GCN5 on histone H3 acetylation. Similarly,
in the absence of HACl and HAC12, GCN5 may sufﬁce for acetylation of histone H3 at
the COR4 7 promoter. The next step would be to create a had hac12gcn5 triple mutant
and to test its ability to activate COR4 7 expression and acetylate histone H3 at the

COR4 7 promoter. If the above hypothesis is true then the acetylation of histone H3 at the

107

COR4 7 promoter in the triple mutant should be lower as compared to wild type plants

upon cold acclimation.

One more potential candidate studied was the TAP] protein which is also a
member of the transcription factor IID. Data revealed that TAF 1 controls the regulation
of only a subset of COR genes, namely, COR6. 6, COR15a and GoIS but does not affect
the expression of COR4 7, COR 78 or CBF 2 during cold acclimation. The regulation of the
COR genes by TAFl is also independent of its HAT activity as seen by the chromatin
immunoprecipitation assays. TAF proteins can act as adapters that connect the activators
with the general transcription machinery and TAFl could be acting as an adapter at its
target COR genes and thus regulate their expression. TAF l and GCN5 together control
the regulation of light-responsive genes in Arabidopsis by affecting acetylation of
histones H3 and H4 at the target gene promoters. This could also be true for the
regulation of COR genes. The next step would be to test the effect of the taflgcn5 double
mutant on COR gene expression and acetylation of histone H3 at the COR gene
promoters. If the hypothesis is true then the effect of the double mutant on COR gene
expression would be greater than any of the single mutant. We should also be able to see
a diﬁ‘erence in acetylation of histone H3 at the COR promoters between wild type and the

double mutant.

Thus, it seems no single family of HATS is predominantly responsible for the
acetylation of histone H3 at the COR gene promoters. Mutating any single HAT protein
has an effect on the expression of COR genes but not on the acetylation of histone H3 per
se. It is possible that in the absence of one HAT the other HATS are sufﬁcient for the

acetylation of histone H3 at the COR gene promoters. All HATS are known to acetylate

108

proteins other than histones and thus these HATS might be controlling the activity of a
factor upstream of COR genes and hence indirectly regulate the expression of the COR
genes. It is also interesting to note that GCN5 affects all the ﬁve COR genes used in this
study but the other HATS control only a subset of these genes. GCN5 could be the main
HAT responsible for acetylation of histone H3 but in its absence other HATS might be
sufficient for the acetylation of histone H3 during cold acclimation. Thus, testing the
ability of double or triple HAT mutants for their ability to efﬁciently acetylate histone H3
and activate COR gene expression might provide more detailed insight into the regulation

of COR genes by histone H3 acetylation during cold acclimation.

109

Chapter V

Discussion

5.1 Histone acetylation and COR gene expression:

The results of my study reveal that acetylation of histone H3 increases at all the
COR gene promoters tested during cold acclimation. The increase in acetylation of
histone H3 is evident within 4 hours of cold acclimation. In addition to acetylation of
histone H3, a subset of COR genes (COR6. 6, COR 78 and 0015') also Show a decrease in
nucleosome occupancy upon cold acclimation. This decrease in nucleosome occupancy is
evident after 24 hours of cold acclimation. Acetylation of histone H3 seems to precede
loss of nucleosome occupancy at some COR gene promoters. The results also reveal
subtle differences in the manner in which COR genes are regulated during cold
acclimation. The hallmarks of COR gene regulation, histone H3 acetylation and loss of
histone H3, are transient, as deacclimation of plants results in decrease in histone H3
acetylation back to non-acclimated levels as well as restoration of histone H3 to non-

acclirnated levels at the COR gene promoters.

5.2 Histone H3 acetylation and CBF:

The CBF family of activators (CBFl, CBF2 and CBF 3) act as master regulators ‘
of the cold acclimation pathway in Arabidopsis. Overexpression of CBF leads to
constitutive expression of COR genes which in turn leads to constitutive freezing
tolerance of the CBF overexpressing plants [93, 94]. In the experiments reported here we
found that overexpression of CBF was sufﬁcient for the increase in histone H3

acetylation as well as loss of histone H3 at the COR gene promoters in the absence of

110

cold acclimation. Cold acclimation of plants overexpressing CBF led to a further increase
in the acetylation of histone H3 at the COR gene promoters. This increase in histone H3
acetylation might be driven by the endogenous CBF proteins or by other transcription

factors that are activated during cold acclimation.

The CBF activator is composed of a DNA-binding domain and an activation
domain. In order to test whether the activation domain of CBF is sufﬁcient to drive the
acetylation of histone H3 as well as removal of histone H3 at the COR gene promoters, a
mutant form of CBF lacking the activation domain was used. My experiments showed
that the mutant form of CBF lacking the activation domain was unable to bring about
increase in acetylation of histone H3 and the removal of histone H3 in the absence of cold
acclimation. However, acetylation of histone H3 increased upon cold acclimation in
plants overexpressing this mutant form of CBF. I conclude that the activation domain is
required for driving histone H3 acetylation in the absence of the cold Signal but is
dispensable in the presence of cold acclimation. Also, the mutant lacking the activation
domain of CBF is able to act as a dominant negative with respect to histone removal but
not histone H3 acetylation. The increase of histone H3 acetylation observed upon cold
acclimation might be due to a non-CBF transcription factor or due to the recruitment of
chromatin modifying machinery by the DNA binding domain of CBF. Previous studies
showed that the DNA domain of CBF can physically interact with the coactivator
proteins ADA2a and ADA2b in vitro [51]. Therefore it is possible that chromatin
modifying factors that bring about histone H3 acetylation are recruited by the DNA
binding domain of CBF upon cold acclimation. In contrast, the recruitment of chromatin

remodeling factors bringing about histone H3 removal are probably recruited by the

111

activation domain of CBF. Thus, the two domains of the CBF proteins seem to play

distinct roles with respect to chromatin modiﬁcations during COR gene regulation.
5.3 Histone acetyltransferases and histone acetylation of COR genes:

The coactivator proteins ADA2 can interact in vitro with GCN5, a histone
acetyltransferase, in Arabidopsis [51]. T-DNA insertion mutants ada2b and gcn5 Show a
delay in activation and two-fold reduction of COR gene expression upon cold
acclimation, leading to the hypothesis that CBF recruits an ADA2b-GCN5 containing
complex at the COR gene promoters during cold acclimation [3 7]. The ADA2b-GCN5
complex might in turn acetylate histones at the COR gene promoters and aid the CBF-
driven activation of COR genes. However, in opposition to this hypothesis, experiments
revealed that neither ADA2b nor GCN5 are required for the acetylation of histone H3 at
the COR gene promoters during cold acclimation. It is possible that GCN5 brings about
histone H3 acetylation at the COR gene promoters but in the absence of GCN5 other
HATS might be capable of bringing about acetylation of histone H3 at the COR gene
promoters during cold acclimation. However, both ADA2b and GCN5 seem to be
necessary for the removal of histone H3 at the COR gene promoters. ADA2b is known to
interact with the chromatin remodeling enzyme BRGl in humans and the absence of
GCN5 hampers histone removal of some genes in yeast [61, 126, 127] . Thus, ADA2b
and GCN5 might be required for the recruitment of a chromatin remodeling enzyme at

the COR gene promoters to bring about histone removal.

To test the possibility that other HATS may be responsible for acetylation of

histone H3 during cold acclimation, members of the HAC and TAF 1 gene families of

112

HATS were tested for their role in COR gene expression and acetylation of histone H3 at
the COR gene promoters. The haclhacIZ double mutant showed lower expression of
COR47 as compared to wild type plants during cold acclimation. However, the

had hac12 mutant had no effect on the histone H3 acetylation nor on histone H3 removal

at the COR4 7 promoter.

Loss of taﬂ affected the expression of COR6. 6, COR15a and GolS genes during
cold acclimation, but the taﬂ mutant did not affect the acetylation or removal of histone
H3 at the target COR gene promoters. TAFl has been shown to regulate histone H4
acetylation in Arabidopsis and hence the ability of taﬂ mutants to bring about histone H4
acetylation at the COR gene promoters was tested [54]. Chromatin immunoprecipitation
assay Showed that there was no signiﬁcant increase in the histone H4 acetylation upon
cold acclimation at the COR gene promoters in wild type and taﬂ-1 plants, except for the
GoIS promoter. TAFl was not required to drive histone H4 acetylation at the GolS
promoter during cold acclimation. TAF 1 is part of the large coactivator complex TFIID
and hence loss of TAF 1 might interfere with the stability or recruitment of TFIID
affecting COR gene expression during cold acclimation [122]. Results so far indicate that
functional redundancy might exist between HATS controlling the histone H3 acetylation
at the COR gene promoters. To test this hypothesis the effect of gcn5taﬂ,
gcn5hac1hac12 and taﬂhaclhach mutants on histone H3 acetylation at the COR gene

promoters needs to be tested.

5.4 ADA2b and GCN5 regulation of COR gene expression:

113

CBFS bind Speciﬁc ciS-elements known as cold/dehydration responsive elements
(CRT/DRES) present in the promoters of their target genes and activate their expression
[68, 128]. ADA2b and GCN5 regulate the expression of COR genes but whether they do
so in a CBF-dependent manner is not known [3 7]. In order to test this possibility two
approaches were used. In the ﬁrst approach, the expression of the reporter gene GUS
driven by four tandem repeats of CRT fused to a minimal 3SS promoter was compared in
wild type and gcn5-I plants. AS the reporter gene GUS is driven by CRTs, the induction
of GUS is solely dependent on CBF. Experiments Showed that GCN5 is not required for
driving the CBF-dependent expression of GUS during cold acclimation. In the second
approach the expression of COR genes was compared in wild type, ada2b-1 and gcn5-1
plants overexpressing CBF. Overexpression of CBF leads to constitutive expression of
COR genes and any effect on COR expression in the mutant plants as compared to wild
type plants will be the result of their regulation of the CBF-driven COR gene expression.
However, experiments revealed that the expression of COR genes is not affected in the
mutants as compared to wild type plants. Thus, ADA2b and GCN5 do not regulate the
CBF-dependent COR gene regulation. Recent studies Show that the expression of COR
genes might be regulated by transcription factors other than CBF (Doherty, unpublished
data). Thus it is possible that the ADA2b and GCN5 regulate the expression of COR

genes via some other non-CBF transcription factor(s).

114

Table l: Oligonucleotides used in quantitative Real-time PCR analysis of chromatin

immunoprecipitation assays. Primers were designed to detect promoters of various cold-

regulated genes. Primer locations are in relation to the translation start site +1.

 

 

1D# Gene Gene sequence Position
KPl COR15a(F) 5’-GGCTTGCAAACCCAAAT'I'AAC-3 ’ -507 to -487
KP2 COR15a(R) 5’-AAGAGGGTCGTTCTCATTTCC-3 ’ -420 to -400
KP3 COR6.6(F) 5’-AAACGACACGTGATGTCTTGA-3’ -224 to -204
KP4 COR6.6(R) 5’-TGCCACGTGTAATCTGAAACC-3 ’ -l40 to -120
KPS COR47(F) 5’-GACTTCAAGAAATAAGAGGGGTCA-3’ -266 to -243
KP6 COR47(R) S’-AGAAGGGTCCAGGCCAAC-3’ -164 to -l47
KP7 COR78(F) 5’-ATGGGCCAATAGACATGGAC-3 ’ -371 to -3 52
KP8 COR78(R) 5’-GCTTTITGGAACTCATGTCG—3’ -244 to -225
KP9 ACT2/7(F) 5 ’-CGTTTCGCT'I‘TCC'ITAGTG'I‘TAGCT-3’ -156 to -132
KPlO ACT2/7(R) 5 ’-AGCGAACGGATCTAGAGACTCACCTI‘G-3 ’ -49 to -23
681 GolS(F) 5’—TGACGTTAGGACAAAAGAAGAAA-3 ’ -528 to -506
682 GolS(R) 5’-CAAAGCTAATTTGCTGCTCGT-3’ -425 to -405

 

 

115

 

Table 2: Oligonucleotides used in quantitative Real-time PCR for gene expression

analysis. Primers were designed to detect open-reading frames (ORFS) of various cold-

regulated genes. Primer locations are in relation to the translation start site +1.

 

 

ID# Gene Gene sequence Position
KPl3 COR15a(F) 5’-ATGGCTTCITCTTTCCACAGC-3’ l - 21

KP 14 COR15a(R) 5 ’-GAAGCTTTCTTTGTGGCCTC-3 ’ 1 5 l — 170
KP17 COR6.6(F) 5’-CTGGCAAAGCTGAGGAGAAG-3 ’ 44 — 63
KPl 8 COR6.6(R) 5’-ACTGCCGCATCCGATATACT-3’ 136 — 155
KP19 COR47(F) 5’-CGGTACCAGTGTCGGAGAGT-3 ’ 497 — 516
KP20 COR47(R) 5’-ACAGCTGGTGAATCCTCTGC-3 ’ 604 - 623
KP21 COR78(F) 5 ’-GAAAGGAGGAGGAGGAATGG-3 ’ 1908 — 1927
KP22 COR78(R) 5’-AACCAGCCAGATGAT'ITTGG-3 ’ 2008 — 2027
KP23 ACT3 (F) 5’-GGTCGTACTACTGGTATTGTGCT-3 ’ 442 - 464
KP24 ACT3(R) 5’-TGACAATTTCACGCTCAGCT-3 ’ 621 - 640
KPl 1 GolS(F) 5’-GGAGTGGTI‘GGTCTGGCTAA-3 ’ 91 — 1 10
KPlZ GolS(R) 5’-TTGGTTATCCGGTGGGTAAA-3 ’ 230 - 249
KP15 CBF2(R) 5’-GGATGCTCATGGTCTTGACATG-3 ’ 431 - 452
KPl6 CBF2(R) 5’-TCTI‘CATCCATATAAAACGCATCTTG—3’ 495 - 520

 

 

116

 

 

Table 3: Oligonucleotides used in the generation of the 4X—CRT-GUS plasmid.

 

ID# Gene sequence

 

ST 1 103 5’-CTAGATACCGACATTACCGACAT

TACCGACATTACCGACATG — 3 ’

ST 1104 5’ — GATCCATGTCGGTAATGTCGGTA

ATGTCGGTAATGTCGGTAT — 3 ’

 

 

117

 

111

[21

[31

[4]

[5]

[61

[7]

[3]

[91

[101

[11]

[121

[13]

[14]

[15]

6.1 BIBLIOGRAPHY

S.G. Roberts, Mechanisms of action of transcription activation and repression
domains, Cell Mol Life Sci 57 (2000) 1149-1160.

R.D. Kornberg, Eukaryotic transcriptional control, Trends Cell Biol 9 (1999)
M46-49.

R.G. Roeder, Transcriptional regulation and the role of diverse coactivators in
animal cells, FEBS Lett 579 (2005) 909-915.

A. M. Naar, B. D. Lemon, R. Tjian, Transcriptional coactivator complexes, Annu
Rev Biochem 70 (2001) 475- 501.

K. Luger, A.W. Mader, R.K. Richmond, D.F. Sargent, T.J. Richmond, Crystal
structure of the nucleosome core particle at 2.8 A resolution, Nature 389 (1997)
251-260.

K. Luger, T.J. Rechsteiner, A.J. F lauS, M.M. Waye, T.J. Richmond,
Characterization of nucleosome core particles containing histone proteins made in
bacteria, J Mol Biol 272 (1997) 301-311.

P.J. Robinson, D. Rhodes, Structure of the '30 nm' chromatin ﬁbre: a key role for
the linker histone, Curr Opin Struct Biol 16 (2006) 336-343.

D.J. Tremethick, Higher-order structures of chromatin: the elusive 30 nm ﬁber,
Cell 128 (2007) 651-654.

J. Zlatanova, S.H. Leuba, Chromatin ﬁbers, one-at-a-time, J Mol Biol 331 (2003)
1-19.

R.D. Kornberg, Y. Lorch, Chromatin-modifying and -remodeling complexes,
Curr Opin Genet Dev 9 (1999) 148-151.

J .L. Workman, R.E. Kingston, Alteration of nucleosome structure as a mechanism
of transcriptional regulation, Annu Rev Biochem 67 (1998) 545-579.

T. Jenuwein, C.D. Allis, Translating the histone code, Science 293 (2001) 1074-
1080.

S.L. Berger, Histone modiﬁcations in transcriptional regulation, Curr Opin Genet
Dev 12 (2002) 142-148.

K. Zhang, V.V. Sridhar, J. Zhu, A. Kapoor, J .K. Zhu, Distinctive core histone
post-translational modiﬁcation patterns in Arabidopsis thaliana, PLoS ONE 2
(2007)e1210.

L. Johnson, S. Mollah, B.A. Garcia, T.L. Muratore, J. Shabanowitz, D.F. Hunt,
S.E. Jacobsen, Mass spectrometry analysis of Arabidopsis histone H3 reveals

118

 

[16]

[171

[18]

[19]

[20]

[21]

[22]

[231

[241

[25]

[26]

[27]

[28]

distinct combinations of post-translational modiﬁcations, Nucleic Acids Res 32
(2004) 651 1-6518.

S.Y. Roth, J .M. Denu, C.D. Allis, Histone acetyltransferases, Annu Rev Biochem
70 (2001) 81-120.

P. Loidl, Histone acetylation: facts and questions, Chromosoma 103 (1994) 441-
449.

B.M. Turner, L.P. O'Neill, Histone acetylation in chromatin and chromosomes,
Semin Cell Biol 6 (1995) 229-236.

D.Y. Lee, J .J . Hayes, D. Pruss, A.P. Wolffe, A positive role for histone
acetylation in transcription factor access to nucleosomal DNA, Cell 72 (1993) 73-
84.

M. Vettese-Dadey, P. Walter, H. Chen, L.J. Juan, J .L. Workman, Role of the
histone amino termini in facilitated binding of a transcription factor, GAL4-AH,
to nucleosome cores, Mol Cell Biol 14 (1994) 970-981.

D.E. Sterner, S.L. Berger, Acetylation of histones and transcription-related
factors, Microbiol Mol Biol Rev 64 (2000) 435-459.

J .E. Brownell, C.D. Allis, Special HATS for special occasions: linking histone
acetylation to chromatin assembly and gene activation, Curr Opin Genet Dev 6
(1996) 176-184.

A. Lusser, A. Eberharter, A. Loidl, M. Goralik-Schramel, M. Homgacher, H.
Haas, P. Loidl, Analysis of the histone acetyltransferase B complex of maize
embryos, Nucleic Acids Res 27 (1999) 4427-4435.

A.B. Ruiz-Garcia, R. Sendra, M. Galiana, M. Pamblanco, J .E. Perez-Ortin, V.
Tordera, HAT] and HAT2 proteins are components of a yeast nuclear histone
acetyltransferase enzyme speciﬁc for free histone H4, J Biol Chem 273 (1998)
12599-12605.

D. Kolle, B. Sarg, H. Lindner, P. Loidl, Substrate and sequential Site Speciﬁcity of
cytoplasmic histone acetyltransferases of maize and rat liver, F EBS Lett 421
(1998) 109-114.

S.R. Grossman, p300/CBP/p53 interaction and regulation of the p53 response, Eur
J Biochem 268 (2001) 2773-2778.

W. Gu, R.G. Roeder, Activation of p53 sequence-speciﬁc DNA binding by
acetylation of the p53 C-terrninal domain, Cell 90 (1997) 595-606.

M.A. Martinez-Balbas, U.M. Bauer, S.J. Nielsen, A. Brehm, T. Kouzarides,
Regulation of E2F1 activity by acetylation, EMBO J 19 (2000) 662-671.

119

 

[29]

[30]

[31]

[321

[33]

[34]

[35]

[361

[371

[38]

[39]

[40]

A. Imhof, X.J. Yang, V.V. Ogryzko, Y. Nakatani, A.P. Wolffe, H. Ge,
Acetylation of general transcription factors by histone acetyltransferases, Curr
Biol 7 (1997) 689-692.

T. Kouzarides, Acetylation: a regulatory modiﬁcation to rival phosphorylation?,
EMBO J 19 (2000) 1176-1179.

M.H. Kuo, C.D. Allis, Roles of histone acetyltransferases and deacetylases in
gene regulation, Bioessays 20 (1998) 615-626.

S.K. Kurdistani, M. Grunstein, Histone acetylation and deacetylation in yeast, Nat
Rev Mol Cell Biol 4 (2003) 276-284.

L. Verdone, M. Caserta, E. Di Mauro, Role of histone acetylation in the control of
gene expression, Biochem Cell Biol 83 (2005) 344-353.

R. Pandey, A. Muller, C.A. Napoli, D.A. Selinger, C.S. Pikaard, EJ. Richards, J.
Bender, D.W. Mount, R.A. Jorgensen, Analysis of histone acetyltransferase and
histone deacetylase families of Arabidopsis thaliana suggests functional

diversiﬁcation of chromatin modiﬁcation among multicellular eukaryotes,
Nucleic Acids Res 30 (2002) 5036-5055.

D.W. Ng, M.B. Chandrasekharan, T.C. Hall, Ordered histone modiﬁcations are
associated with transcriptional poising and activation of the phaseolin promoter,
Plant Cell 18 (2006) 119-132.

J .M. Hernandez, A. Feller, K. Morohashi, K. Frame, E. Grotewold, The basic
helix loop helix domain of maize R links transcriptional regulation and histone
modiﬁcations by recruitment of an EMSY-related factor, Proc Natl Acad Sci U S
A 104 (2007) 17222-17227. .

K.E. Vlachonasios, M.F. Thomashow, S.J. Triezenberg, Disruption mutations of
ADA2b and GCN5 transcriptional adaptor genes dramatically affect Arabidopsis
growth, development, and gene expression, Plant Cell 15 (2003) 626-638.

R.F. Ransom, J .D. Walton, Histone Hyperacetylation in Maize in Response to
Treatment with HC-Toxin or Infection by the Filamentous Fungus Cochliobolus
carbonum, Plant Physiol 115 ( 1997) 1021-1027.

P. Casati, M. Campi, F. Chu, N. Suzuki, D. Maltby, S. Guan, A.L. Burlingame, V.
Walbot, Histone Acetylation and Chromatin Remodeling Are Required for UV-B-
Dependent Transcriptional Activation of Regulated Genes in Maize, Plant Cell
(2008)

A. Devoto, M. Nieto-Rostro, D. Xie, C. Ellis, R. Hannston, E. Patrick, J. Davis,
L. Sherratt, M. Coleman, J .G. Turner, COI] links jasmonate Signalling and
fertility to the SCF ubiquitin-ligase complex in Arabidopsis, Plant J 32 (2002)
457-466.

120

 

[41]

[42]

[43]

[441

[45]

[46]

[471

[481

[49]

[50]

[51]

[52]

C. Zhou, L. Zhang, J. Duan, B. Miki, K. Wu, HISTONE DEACETYLASE19 is
involved in jasmonic acid and ethylene signaling of pathogen response in
Arabidopsis, Plant Cell 17(2005) 1196-1204.

Y. He, R.M. Amasino, Role of chromatin modiﬁcation in ﬂowering-time control,
Trends Plant Sci 10 (2005) 30-35.

G.W. Humphrey, Y. Wang, V.R. Russanova, T. Hirai, J. Qin, Y. Nakatani, B.H.
Howard, Stable histone deacetylase complexes distinguished by the presence of
SANT domain proteins CoREST/kiaa007l and Mta-Ll , J Biol Chem 276 (2001)
6817-6824.

M.A. Hakimi, Y. Dong, W.S. Lane, D.W. Speicher, R. Shiekhattar, A candidate
X-linked mental retardation gene is a component of a new family of histone
deacetylase-containing complexes, J Biol Chem 278 (2003) 7234-7239.

M. Yoshida, M. Kij ima, M. Akita, T. Beppu, Potent and Speciﬁc inhibition of
mammalian histone deacetylase both in vivo and in vitro by trichostatin A, J Biol
Chem 265 (1990) 17174-17179.

C.R. Xu, C. Liu, Y.L. Wang, L.C. Li, W.Q. Chen, Z.H. Xu, S.N. Bai, Histone
acetylation affects expression of cellular patterning genes in the Arabidopsis root
epidermis, Proc Natl Acad Sci U S A 102 (2005) 14469-14474.

H.H. Tai, G.C. Tai, T. Beardmore, Dynamic histone acetylation of late embryonic
genes during seed germination, Plant Mol Biol 59 (2005) 909-925.

Y.L. Chua, A.P. Brown, J .C. Gray, Targeted histone acetylation and altered

nuclease accessibility over short regions of the pea plastocyanin gene, Plant Cell
13 (2001) 599-612.

Y.L. Chua, L.A. Watson, J .C. Gray, The transcriptional enhancer of the pea
plastocyanin gene associates with the nuclear matrix and regulates gene
expression through histone acetylation, Plant Cell 15 (2003) 1468-1479.

E.J. Stockinger, Y. Mao, M.K. Regier, S.J. Triezenberg, M.F. Thomashow,
Transcriptional adaptor and histone acetyltransferase proteins in Arabidopsis and
their interactions with CBF] , a transcriptional activator involved in cold-regulated
gene expression, Nucleic Acids Res 29 (2001) 1524-1533.

Y. Mao, K.A. Pavangadkar, M.F. Thomashow, S.J. Triezenberg, Physical and
functional interactions of Arabidopsis ADA2 transcriptional coactivator proteins
with the acetyltransferase GCN5 and with the cold-induced transcription factor
CBF], Biochim BiophyS Acta 1759 (2006) 69-79.

S.K. Han, J .D. Song, Y.S. Noh, B. Noh, Role of plant CBP/p300-like genes in the
regulation of ﬂowering time, Plant J 49 (2007) 103-114.

121

 

[53]

[54]

[55]

[56]

[57]

[581

[591

[601

[61]

[62]

[63]

[641

W. Deng, C. Liu, Y. Pei, X. Deng, L. Niu, X. Cao, Involvement of the histone
acetyltransferase AtHACl in the regulation of ﬂowering time via repression of
FLOWERING LOCUS C in Arabidopsis, Plant Physiol 143 (2007) 1660-1668.

M. Benhamed, C. Bertrand, C. Servet, D.X. Zhou, Arabidopsis GCN5, HD] , and
TAF 1/HAF 2 interact to regulate histone acetylation required for light-responsive
gene expression, Plant Cell 18 (2006) 2893-2903.

C. Bertrand, M. Benhamed, Y.F. Li, M. Ayadi, G. Lemonnier, J .P. Renou, M.
Delarue, D.X. Zhou, Arabidopsis HAF2 gene encoding TATA-binding protein
(TBP)-associated factor TAF 1 , is required to integrate light signals to regulate
gene expression and growth, J Biol Chem 280 (2005) 1465-1473.

R. Marmorstein, S.Y. Roth, Histone acetyltransferases: function, structure, and
catalysis, Curr Opin Genet Dev 11 (200]) 155-161.

A. Eberharter, D.E. Sterner, D. Schieltz, A. Hassan, J .R. Yates, 3rd, S.L. Berger,
J .L. Workman, The ADA complex is a distinct histone acetyltransferase complex
in Saccharomyces cerevisiae, Mol Cell Biol 19 (1999) 6621-663 ].

P.A. Grant, L. Duggan, J. Cote, S.M. Roberts, J .E. Brownell, R. Candau, R. Ohba,
T. Owen-Hughes, C.D. Allis, F. Winston, S.L. Berger, J .L. Workman, Yeast Gcn5
functions in two multisubunit complexes to acetylate nucleosomal histones:
characterization of an Ada complex and the SAGA (Spt/Ada) complex, Genes
Dev 1] (1997) 1640-1650.

Z.J. Chen, L. Tian, Roles of dynamic and reversible histone acetylation in plant
development and polyploidy, Biochim Biophys Acta 1769 (2007) 295-307.

R.A. Bhat, M. Riehl, G. Santandrea, R. Velasco, S. Slocombe, G. Donn, H.H.
Steinbiss, R.D. Thompson, H.A. Becker, Alteration of GCN5 levels in maize

reveals dynamic responses to manipulating histone acetylation, Plant J 33 (2003)
455-469.

N.A. Barlev, A.V. Emelyanov, P. Castagnino, P. Zegerman, A.J. Bannister, M.A.
Sepulveda, F. Robert, L. Tora, T. Kouzarides, B.K. Birshtein, S.L. Berger, A
novel human Ada2 homologue functions with GcnS or Brg] to coactivate
transcription, Mol Cell Biol 23 (2003) 6944-6957.

A. Barrios, W. Selleck, B. Hnatkovich, R. Kramer, D. Sermwittayawong, S. Tan,
Expression and puriﬁcation of recombinant yeast Ada2/Ada3/Gcn5 and Piccolo
NuA4 histone acetyltransferase complexes, Methods 41 (2007) 271-277.

R.A. Bhat, J .W. Borst, M. Riehl, R.D. Thompson, Interaction of maize Opaque-2
and the transcriptional co-activators GCN5 and ADA2, in the modulation of
transcriptional activity, Plant Mol Biol 55 (2004) 239-252.

R. Kodrzycki, R.S. Boston, B.A. Larkins, The opaque-2 mutation of maize
differentially reduces zein gene transcription, Plant Cell 1 (1989) 105-114.

122

 

 

[65]

[66]

[67]

[68]

[691

[70]

[711

[77-1

[731

[74]

[75]

[76]

S. Lohmer, M. Maddaloni, M. Motto, N. Di Fonzo, H. Hartings, F. Salarnini, R.D.
Thompson, The maize regulatory locus Opaque-2 encodes a DNA-binding protein
which activates the transcription of the b-32 gene, EMBO J 10 (1991) 617-624.

F. Jeanmougin, J .M. Wurtz, B. Le Douarin, P. Chambon, R. Losson, The
bromodomain revisited, Trends Biochem Sci 22 (1997) 151-153.

C. Dhalluin, J.E. Carlson, L. Zeng, C. He, A.K. Aggarwal, M.M. Zhou, Structure
and ligand of a histone acetyltransferase bromodomain, Nature 399 (1999) 491-
496.

S.J. Gihnour, D.G. Zarka, E.J. Stockinger, M.P. Salazar, J .M. Houghton, M.F.
Thomashow, Low temperature regulation of the Arabidopsis CBF family of AP2

transcriptional activators as an early step in cold-induced COR gene expression,
Plant J 16 (1998) 433-442.

S. Chattopadhyay, L.H. Ang, P. Puente, X.W. Deng, N. Wei, Arabidopsis bZIP
protein HY5 directly interacts with light-responsive promoters in mediating light
control of gene expression, Plant Cell 10 (1998) 673-683.

V. Yadav, S. Kundu, D. Chattopadhyay, P. Negi, N. Wei, X.W. Deng, S.
Chattopadhyay, Light regulated modulation of Z-box containing promoters by
photoreceptors and downstream regulatory components, COP] and HY5, in
Arabidopsis, Plant J 31 (2002) 741-753.

J. Lee, K. He, V. Stolc, H. Lee, P. Figueroa, Y. Gao, W. Tongprasit, H. Zhao, 1.
Lee, X.W. Deng, Analysis of transcription factor HY5 genomic binding Sites

revealed its hierarchical role in light regulation of development, Plant Cell 19
(2007) 731-749.

Y. Jiao, O.S. Lau, X.W. Deng, Light-regulated transcriptional networks in higher
plants, Nat Rev Genet 8 (2007) 217-230.

J .A. Long, C. Ohno, Z.R. Smith, E.M. Meyerowitz, TOPLESS regulates apical
embryonic fate in Arabidopsis, Science 312 (2006) 1520-1523.

J .A. Long, S. Woody, S. Poethig, E.M. Meyerowitz, M.K. Barton, Transformation
of shoots into roots in Arabidopsis embryos mutant at the TOPLESS locus,
Development 129 (2002) 2797-2806.

M.J. Gao, D.D. Hegedus, A.G. Sharpe, S.J. Robinson, D.J. Lydiate, A. Hannoufa,
Isolation and characterization of a GCN5-interacting protein from Arabidopsis
thaliana, Planta 225 (2007) 1367-1379.

C. Bertrand, C. Bergounioux, S. Domenichini, M. Delarue, D.X. Zhou,
Arabidopsis histone acetyltransferase AtGCNS regulates the ﬂoral meristem
activity through the WUSCHEL/AGAMOUS pathway, J Biol Chem 278 (2003)
28246-2825].

123

 

 

[77]

[73]

[79]

[80]

[81]

[321

[83]

[34]

[85]

[36]

[37]

[83]

[39]

I. Baurle, T. Laux, Regulation of WUSCHEL transcription in the stem cell niche
of the Arabidopsis shoot meristem, Plant Cell 17 (2005) 2271-2280.

R. Gross-Hardt, M. Lenhard, T. Laux, WUSCHEL signaling functions in
interregional communication during Arabidopsis ovule development, Genes Dev
16 (2002) 1129-1138.

T. Laux, K.F. Mayer, J. Berger, G. Jurgens, The WUSCHEL gene is required for
shoot and ﬂoral meristem integrity in Arabidopsis, Development 122 (1996) 87-
96.

A.R. Ricci, J. Genereaux, C.J. Brand], Components of the SAGA histone
acetyltransferase complex are required for repressed transcription of ARG] in rich
medium, Mol Cell Biol 22 (2002) 4033-4042.

L.A. Lebedeva, S.G. Georgieva, E.N. Nabirochkina, Study of the properties of
two homologues of yeast ADA2 in Drosophila melanogaster, Dokl Biochem
Biophys 398 (2004) 297-299.

R. Ulrn, F. Nagy, Signalling and gene regulation in response to ultraviolet light,
Curr Opin Plant Biol 8 (2005) 477-482.

L. Bordoli, M. Netsch, U. Luthi, W. Lutz, R. Eckner, Plant orthologs of
p300/CBP: conservation of a core domain in metazoan p300/CBP

acetyltransferase-related proteins, Nucleic Acids Res 29 (2001) 589-597.

N.A. Barlev, R. Candau, L. Wang, P. Darpino, N. Silverman, S.L. Berger,
Characterization of physical interactions of the putative transcriptional adaptor,
ADA2, with acidic activation domains and TATA-binding protein, J Biol Chem
270 (1995) 19337-19344.

D. Sermwittayawong, S. Tan, SAGA binds TBP via its Spt8 subunit in
competition with DNA: implications for TBP recruitment, EMBO J 25 (2006)
3791-3800.

R.C. Tyler, E. Bitto, C.E. Bemdsen, C.A. Bingrnan, S. Singh, M.S. Lee, G.E.
Wesenberg, J .M. Denu, G.N. Phillips, Jr., J .L. Markley, Structure of Arabidopsis
thaliana At1g77540 protein, a minimal acetyltransferase ﬁ'om the COG23 88
family, Biochemistry 45 (2006) 14325-14336.

R.C. McGarry, Y.D. Barron, M.F. Carvalho, J.E. Hill, D. Gold, E. Cheung, W.L.
Kraus, S.G. Lazarowitz, A novel Arabidopsis acetyltransferase interacts with the
geminivirus movement protein NSP, Plant Cell 15 (2003) 1605-1618.

G.J. Narlikar, H.Y. Fan, R.E. Kingston, Cooperation between complexes that
regulate chromatin structure and transcription, Cell 108 (2002) 475-487.

S.J. Triezenberg, Structure and function of transcriptional activation domains,
Curr Opin Genet Dev 5 (1995) 190-196.

124

 

 

[901

[91]

[921

[931

[94]

[951

[961

[97]

[93]

[99]

[100]

[101]

T.I. Lee, R.A. Young, Transcription of eukaryotic protein-coding genes, Annu
Rev Genet 34 (2000) 77-137.

S.J. Gilmour, R.K. Hajela, M.F. Thomashow, Cold Acclimation in Arabidopsis
thaliana, Plant Physiol 87 (1988) 745-750.

S.J. Gilmour, S.G. Fowler, M.F. Thomashow, Arabidopsis transcriptional
activators CBF], CBF2, and CBF3 have matching functional activities, Plant Mol
Biol 54 (2004) 767-781.

S.J. Gilmour, A.M. Sebolt, M.P. Salazar, J .D. Everard, M.F. Thomashow,
Overexpression of the Arabidopsis CBF 3 transcriptional activator mimics

multiple biochemical changes associated with cold acclimation, Plant Physiol 124
(2000) 1854-1865.

K.R. Jaglo-Ottosen, S.J. Gihnour, D.G. Zarka, O. Schabenberger, M.F.
Thomashow, Arabidopsis CBF] overexpression induces COR genes and enhances
freezing tolerance, Science 280 (1998) 104-106.

K. Yamaguchi-Shinozaki, K. Shinozaki, A novel ciS-acting element in an
Arabidopsis gene is involved in responsiveness to drought, low-temperature, or
high-salt stress, Plant Cell 6 (1994) 251-264.

S.S. Baker, K.S. Wilhehn, M.F. Thomashow, The 5'-region of Arabidopsis
thaliana cor15a has cis-acting elements that confer cold-, drought- and ABA-
regulated gene expression, Plant Mol Biol 24 (1994) 701-713.

K. Nordin, T. Vahala, E.T. Palva, Differential expression of two related, low-
temperature-induced genes in Arabidopsis thaliana (L.) Heynh, Plant Mol Biol 21
(1993) 641-653.

E.J. Stockinger, S.J. Gilmour, M.F. Thomashow, Arabidopsis thaliana CBF]
encodes an AP2 domain-containing transcriptional activator that binds to the C-
repeat/DRE, a ciS-acting DNA regulatory element that stimulates transcription in
response to low temperature and water deﬁcit, Proc Natl Acad Sci U S A 94

(1997) 1035-1040.

S.P. Baker, P.A. Grant, The SAGA continues: expanding the cellular role of a
transcriptional co-activator complex, Oncogene 26 (2007) 5329-5340.

T. Pankotai, O. Komonyi, L. Bodai, Z. Ujfaludi, S. Muratoglu, A. Ciurciu, L.
Tora, J. Szabad, I. Boros, The homologous Drosophila transcriptional adaptors
ADA2a and ADA2b are both required for normal development but have different
functions, Mol Cell Biol 25 (2005) 8215-8227.

J. Horiuchi, N. Silverman, B. Pina, G.A. Marcus, L. Guarente, ADA], a novel
component of the ADA/GCN5 complex, has broader effects than GCN5, ADA2,
or ADA3, Mol Cell Biol 17 (1997) 3220-3228.

125

 

[102]

[103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113]

[114]

RA. Grant, D.E. Sterner, L.J. Duggan, J .L. Workman, S.L. Berger, The SAGA
unfolds: convergence of transcription regulators in chromatin-modifying
complexes, Trends Cell Biol 8 (1998) 193-197.

D.E. Sterner, R. Belotserkovskaya, S.L. Berger, SALSA, a variant of yeast
SAGA, contains truncated Spt7, which correlates with activated transcription,
Proc Natl Acad Sci U S A 99 (2002) 11622-11627.

D.E. Sterner, X. Wang, M.H. Bloom, G.M. Simon, S.L. Berger, The SANT
domain of AdaZ is required for normal acetylation of histones by the yeast SAGA
complex, J Biol Chem 277 (2002) 8178-8186.

A. Ciurciu, O. Komonyi, T. Pankotai, I.M. Boros, The Drosophila histone
acetyltransferase Gcn5 and transcriptional adaptor Ada2a are involved in
nucleosomal histone H4 acetylation, Mol Cell Biol 26 (2006) 9413-9423.

K.W. Barley, M.S. Shook, B. Brower-Toland, L. Hicks, C.S. Pikaard, In vitro
speciﬁcities of Arabidopsis co-activator histone acetyltransferases: implications

for histone hyperacetylation in gene activation, Plant J 52 (2007) 615-626.

O.L. Gamborg, D.E. Eveleigh, Culture methods and detection of glucanases in
suspension cultures of wheat and barley, Can J Biochem 46 (1968) 417-421.

O.L. Gamborg, R.A. Miller, K. Ojima, Nutrient requirements of suspension
cultures of soybean root cells, Exp Cell Res 50 (1968) 151-158.

R. Ascenzi, J .S. Gantt, Subnuclear distribution of the entire complement of linker
histone variants in Arabidopsis thaliana, Chromosoma 108 (1999) 345-355.

F.J. Herrera, S.J. Triezenberg, VP16-dependent association of chromatin-
modifying coactivators and underrepresentation of histones at immediate-early
gene promoters during herpes Simplex virus infection, J Virol 78 (2004) 9689-
9696.

R.K. Hajela, D.P. Horvath, S.J. Gilmour, M.F. Thomashow, Molecular Cloning
and Expression of cor (Cold-Regulated) Genes in Arabidopsis thaliana, Plant
Physio] 93 (1990) 1246-1252.

H. Reinke, W. Horz, Histones are ﬁrst hyperacetylated and then lose contact with
the activated PHOS promoter, Mo] Cell 1] (2003) 1599-1607.

H. Boeger, J. Griesenbeck, J .S. Strattan, R.D. Kornberg, Nucleosomes unfold
completely at a transcriptionally active promoter, Mol Cell 1] (2003) 1587-1598.

T. Suzuki, A. Kimura, R. Nagai, M. Horikoshi, Regulation of interaction of the
acetyltransferase region of p300 and the DNA-binding domain of Sp] on and
through DNA binding, Genes Cells 5 (2000) 29-41.

126

__L

 

 

[115]

[116]

[117]

[118]

[119]

[120]

[121]

[122]

[123]

[124]

[125]

[126]

[127]

[123]

G]. Ingles, M. Shales, W.D. Cress, SJ. Triezenberg, J. Greenblatt, Reduced
binding of TFIID to transcriptionally compromised mutants of VP16, Nature 351
(1991) 588-590.

S.R. Bhaumik, M.R. Green, Differential requirement of SAGA components for
recruitment of TATA-box-binding protein to promoters in vivo, Mol Cell Biol 22
(2002) 7365-7371.

M.F. Thomashow, PLANT COLD ACCLIMATION: Freezing Tolerance Genes
and Regulatory Mechanisms, Annu Rev Plant Physiol Plant Mol Biol 50 (1999)
571-599.

M.F. Thomashow, So what's new in the ﬁeld of plant cold acclimation? Lots],
Plant Physiol 125 (2001) 89-93.

J .T. Voge], D.G. Zarka, H.A. Van Buskirk, S.G. Fowler, M.F. Thomashow, Roles
of the CBF2 and ZAT12 transcription factors in conﬁguring the low temperature
transcriptome of Arabidopsis, Plant J 4] (2005) 195-211.

K. Shinozaki, K. Yamaguchi-Shinozaki, M. Seki, Regulatory network of gene
expression in the drought and cold stress responses, Curr Opin Plant Biol 6 (2003)
410-417.

D.A. Wassarman, F. Sauer, TAF(II)250: a transcription toolbox, J Cell Sci 114
(2001) 2895-2902.

R. Auty, H. Steen, L.C. Myers, J. Persinger, B. Bartholomew, S.P. Gygi, S.
Buratowski, Puriﬁcation of active TFIID from Saccharomyces cerevisiae.

Extensive promoter contacts and co-activator ftmction, J Biol Chem 279 (2004)
49973-49981.

0. Matangkasombut, R. Auty, S. Buratowski, Structure and ﬁmction of the TFIID
complex, Adv Protein Chem 67 (2004) 67-92.

J .D. Irvin, B.F. Pugh, Genome-wide transcriptional dependence on TAF]
functional domains, J Biol Chem 281 (2006) 6404-6412.

P.A. Wade, D. Pruss, A.P. Wolffe, Histone acetylation: chromatin in action,
Trends Biochem Sci 22 (1997) 128-132.

P. Syntichaki, I. Topalidou, G. Thireos, The Gcn5 bromodomain co-ordinates
nucleosome remodelling, Nature 404 (2000) 414-417.

C.K. Govind, F. Zhang, H. Qiu, K. Hoﬁneyer, A.G. Hinnebusch, Gcn5 promotes
acetylation, eviction, and methylation of nucleosomes in transcribed coding
regions, Mol Cell 25 (2007) 31-42.

Y. Sakuma, Q. Liu, J .G. Dubouzet, H. Abe, K. Shinozaki, K. Yamaguchi-
Shinozaki, DNA-binding speciﬁcity of the ERF/AP2 domain of Arabidopsis

127

 

 

DREBS, transcription factors involved in dehydration- and cold-inducible gene
expression, Biochem Biophys Res Commun 290 (2002) 998-1009.

128