Characterization of Gcn5 histone acetyltransferase in Schizosaccharomyces pombe

(1)

Thesis for doctoral degree (Ph.D.) 2009

Characterization of Gcn5 Histone acetyltransferase in Schizosaccharomyces pombe

Anna Johnsson

Thesis for doctoral degree (Ph.D.) 2009Characterization of Gcn5 Histone acetyltransferase in Schizosaccharomyces pombe

(2)

From the DEPARTMENT OF BIOSCIENCES AND NUTRITION Karolinska Institutet, Stockholm, Sweden

CHARACTERIZATION OF GCN5 HISTONE ACETYLTRANSFERASE IN

SCHIZOSACCHAROMYCES POMBE

Anna Johnsson

Stockholm 2009

(3)

All previously published papers were reproduced with permission from the publisher.

Published by Karolinska Institutet. Printed by Larserics Digital AB, Sundbyberg, Sweden.

(4)

Till Bianca 

(5)

(6)

ABSTRACT

The organisation of eukaryotic DNA into chromatin provides a natural barrier that prevents full access to the DNA thereby inhibiting events such as transcription, replication and repair. In order for these DNA-related events to occur, the chromatin structure needs to be modified by ATP-dependent chromatin remodelling or, by reversible post-translational modifications. Histone acetylation is such a modification and is essential of numerous DNA related events. The enzymes involved in this event are conserved throughout evolution, underscoring their importance.

This thesis describes the role of the conserved histone acetyltransferase (HAT) Gcn5 in transcriptional regulation in Schizosaccharomyces pombe. Here we show that Gcn5 plays an important role in stress response. We map genome-wide Gcn5 occupancy and show that Gcn5 is predominantly localized to coding regions of highly transcribed genes. We also map H3K14 acetylation during salt stress and show that Gcn5 collaborates antagonistically with the class-II histone deacetylase, Clr3, to modulate H3K14ac levels and transcriptional elongation. The interplay between Gcn5 and Clr3 is crucial for the regulation of many stress-response genes. Our findings suggest a new role for Gcn5 during transcriptional elongation, in addition to its known role in transcriptional initiation.

We also investigate the interactions between Gcn5 and other histone deacetylases and acetyltransferases and show overlapping functionality between Gcn5 and another histone acetyltransferase, Mst2, in stress response, regulation of subtelomeric genes and DNA damage repair.

Finally, we show that the role of Gcn5 in stress response is mediated by its catalytic activity and that its function in stress response is conserved among yeast species.

(7)

LIST OF PUBLICATIONS

I. Johnsson A., Xue-Franzén Y., Lundin M. and Wright A.P. Stress-specific role of the fission yeast Gcn5 histone acetyltransferase in programming a subset of stress-response genes. Euk Cell. 2006, (5), 1337-46.

II. Johnsson A., Durand-Dubief M., Xue-Franzén Y., Rönnerblad M., Ekwall K.

and Wright A.P. HAT-HDAC interplay modulates global H3K14 acetylation in gene coding regions during stress. EMBO Rep. 2009, Jul 24 (Epub ahead of print).

III. Nugent R., Johnsson A., Fleharthy B., Gogol M., Xue.Franzén Y., Seidel C., Wright A. and Forsburg S. Expression profiling of S. pombe acetyltransferases idientifies redundant pathways of gene regulation. 2009. Manuscript submitted to PLOS Gen.

IV. Xue-Franzén Y., Johnsson A., Brodin D., Henriksson J., Bürglin T. and Wright A.P. Functional aspects of the Gcn5 histone acetyltransferase in the stress responses of evolutionary diverged yeast species. 2009. Manuscript submitted to BMC Genomics.

(8)

LIST OF ABBREVIATIONS

HAT Histone acetyltransferase CBP CREB-biding protein

CESR Central environmental stress response ChIP Chromatin immunoprecipitation CTD C-terminal domain

Gcn5 General control non-derepressable 5 GNAT Gcn5-N-acetyltransferase

GTF General transcription factors HDAC Histone deacetylase KAT Lysine acetyltransferase MPA Mycophenolic acid

MYST MOZ-Ybf2/Sas3-Sas2-Tip60 NAD nicotinamide adenine dinucleotide NuA3 Nucelosome Acetylation on H3 NuA4 Nucleosome Acetylation on H4

p300 adenoviral E1A-associated protein of 300kDa PCAF p300/CBP- associated factor

PHD Plant homeodomain

PIC Pre initiation complex RNAP II RNA polymerase II

SAGA Spt-Ada-Gcn5-Acetyltransferase SAHA Suberoylanilide hydroxamic acid TAF TBP associated factor

TBD Tandem bromodomain

TBP TATA-box binding protein TF Transcription factor

TSA Trichstatin A

UAS Upstream activating sequence

(10)

1 INTRODUCTION

Post-translational modifications of histones play a central role in diverse processes involving chromatin reorganization, such as gene expression and repression, DNA replication and DNA repair. Histone acetylation is regulated by histone acetyltransferases (HATs) and their antagonistic counterpart histone deacetylases (HDACs). Misregulation of either enzymatic activity can have severe effects development and genome stability and has been linked to several forms of cancer.

Understanding how these proteins act and interact with other chromatin modifying entities is crucial in defining the cause and treatment of DNA-related diseases.

This thesis describes the well-conserved histone acetyltransferase Gcn5 and its role in regulation of transcription in the unicellular fission yeast Schizosaccharomyces pombe.

As S. pombe has a chromatin structure very similar to higher eukaryotes it has become a tractable model system to study gene regulatory networks.

The first part of this thesis covers some of the fundamentals in chromatin structure modifications and discuss the outcome of different modifications have on transcriptional regulation. It also includes also a brief presentation of events and aspects I find important in this field and that has been relevant for my studies. I also include comments on fission yeast as a model system and a brief description of genome-wide studies using Chromatin immunoprecipitation. This will be followed by the aims and results of my studies.

(11)

2 GENERAL INTRODUCTION TO HISTONE MODIFICATIONS AND GENE REGUALTION

2.1 CHROMATIN

Access to DNA is critical for numerous nuclear events such as gene transcription, DNA repair and DNA replication. However, the structural organization of the eukaryotic genome into chromatin forms a natural barrier that limits the accessibility. DNA can be more or less condensed, heterochromatin is highly condensed and the transcriptionally inactive form whereas the less condensed and transcriptionally active form is referred to as euchromatin. The nucleosome is the fundamental unit of chromatin with 147 base pairs of DNA wrapped around a core octamer of so called histone proteins. This octamer contains two of each histone H2A, H2B, H3 and H4 respectively (Figure 1).

Figure 1. Crystal structure of the nucleosome core particle. Image of the double stranded DNA wrapped around the histone octamer with two copies of each H2A (yellow); H2B (red); H3 (blue) and H4 (green) with the N-terminal tails protruding. The left image shows the nucelosome from the front and the right image from the side. Reprinted by permission from Macmillan Publishers Ltd: Nature, Luger et al, copyright 1997.

Histones are highly conserved proteins found in all eukaryotic cells underscoring their fundamental role in DNA maintenance. Each histone contains a “histone-fold” motif that enables histone-histone contact as well as histone-DNA contact within each nucleosome. Protruding each core histone are the N-terminal and C-terminal tails,

(12)

which can be susceptible to post-translational modifications in order to alter the interaction with the DNA and surrounding proteins and thereby altering the structure of the chromatin to make the DNA more or less accessible. These modifications are reversible and include acetylation, methylation, phosphorylation and ubiquitination (Figure 2). In addition to the post-translational modifications of histones, chromatin remodelling complexes and histone protein variants are also involved in regulating chromatin structure and function.

Figure 2. Post translational modifications of the histone tails. Shown in the figure are the positions of acetylation (ac), methylation (me), phosphorylation (ph) and ubiquitination (ub1) on human histone tails including the C-terminal modifications on H2A and H2B and the globular modifications on H3.

2.2 HISTONE MODIFICATIONS

2.2.1 Acetylation

Acetylation, were an acetyl group is transferred to the ε-amino group of a lysine (K) residue using acetyl-coenzyme A as a coenzyme, was the first histone modification identified, and is associated with a variety of chromatin functions (Kouzarides, 2007;

Lee & Workman, 2007). The enzymes carrying out this reaction are called histone

(13)

acetyltransferases (HATs) but since their substrates have been shown not to be limited to histone proteins they are also referred to lysine acetyltransferases (KATs). In this thesis however, I will refer to these enzymes as HATs. Acetylation is a reversible process where removal of acetyl groups is mediated through proteins called histone deacetylases (HDACs) and thus the interplay between HATs and HDACs regulate the cellular levels of histone acetylation.

While some evidence suggests that acetylation changes association of DNA with the underlying nucleosomes (Shogren-Knaak et al, 2006) it also creates specific binding sites for proteins involved in a variety of DNA transactions (Kouzarides, 2007; Lee &

Workman, 2007).

2.2.2 Other histone modifications

Another frequently occurring modification is methylation which can occur on lysine and arginine residues. Up to three methyl groups can be added to one lysine, while an arginine can be mono- or dimethylated, either symmetrically or asymmetrically.

Depending on the number of methyl groups and the position of the amino acid, the outcome could be more or less accessible DNA. For instance, trimethylation of histone H3K9 is associated with heterochromatin in fission yeast and mammals (Nakayama et al, 2001) whereas all forms of H3K4 methylation is correlated with active transcription (Millar & Grunstein, 2006). Genomewide studies of nucleosome methylation showed that on actively transcribed genes trimethylated H3K4 peaks in the beginning of the coding region, while dimethylation is most enriched in the middle of the genes and monomethylated H3K4 is low in the 5´end and high in the 3´end of the genes (Pokholok et al, 2005).

Phosphorylation is a well-known protein modification, but it is an infrequent event on histone proteins. Most commonly known is phosphorylation of serine 10 on histone H3 (H3S10) that has been linked to transcriptional activation as well as chromatin condensation (along with H3S28 phosphorylation) (Cheung et al, 2000). In budding yeast, H2B serine 10 (H2BS10) is phosphorylated by Ste20 in response to oxidative stress (Fuchs et al, 2009), while H2B phosphorylation in mammals occurs on serine 14 (H2BS14) and is catalyzed by the Ste20 homolog mammalian Ste20-like kinase (Bhaumik et al, 2007).

(14)

2.2.3 Recognition of histone modifications

Histone modifications do not only alter the association between the nucleosome and DNA but also function as recognition sites for transcription factors and co-regulators, both repressors and activators. Acetylated histones are recognised by bromodomain containing proteins and several co-activators including HAT-complexes have subunits with bromodomains. Chromodomains that recognise methylated lysines and arginines are frequent among repressor proteins but are also found in HAT-complexes such as SAGA and NuA4. The heterochromatin protein HP1 has a chromodomain that binds to trimetylated H3K9. WD domains, Tudor domains and Plant homeodomains (PHD) are other examples of motifs that recognise and interact with methylated histones (Lee &

Workman, 2007).

2.3 CONSEQUENCES OF IMBALANCE IN HISTONE ACETYLATION

Maintaining a balance between acetylation and deacetylation is fundamental to normal cell growth as disruption of either results in numerous defects such as developmental defects and cancer. For instance, loss of the Gcn5 HAT causes embryonic lethality in mouse. The same study shows that loss of p300, another HAT, also caused developmental phenotypes by different from those caused by loss of Gcn5. In contrast, loss of PCAF, a Gcn5 homolog, did not cause any developmental phenotypes at all.

This exemplifies the distinct functional roles of the acetyltranferases in mammals.

Traditionally, cancer is caused by mutations that either inactivates tumour suppressor genes, or activates oncogenes. But it is now becoming more and more evident that epigenetics play an important role in the development of cancer as misregulated patterns of chromatin modifications are implicated in several steps of tumour development and progression. Loss of HAT function, such as translocation of HAT genes are found in several types of cancer (Yang, 2004). Viral oncogenes such as adenovirus E1A, interacts with HAT activities to promote oncogenic transformation (Frisch and Mymryk, Nat Rev Mol Cell Biol 2002).

Although the occurrences of aberrations in histone modifications at individual promoters have been found in several types of cancer, they have yet not been linked to

(15)

clinical outcome of cancer progression. Kurdistani and co-workers show that global levels of histone modifications can predict the prognosis of several different forms of cancer (Seligson et al, 2009; Seligson et al, 2005). They used immunohistochemical staining to determine the proportion of cells stained for acetylation and dimethylation of residues on histones H3 and H4. Low proportion of cells staining for acetylated H3K18 and dimethylated H3K4, and in some cases dimethylated H3K9, is associated with a poorer prognosis. Furthermore they show that the levels of different modifications were positively correlated with each other. These findings could have important implications for customized epigenetic therapy, as histone modifications are reversible. Therefore, changing one modification, for example by HDAC inhibitors, could result in changes in the state of other modifications. There are several compounds that have been shown to inhibit HDAC activity, such as trichostatin A (TSA), suberoylanilide hydroxamic acid (SAHA), phentylbutyrate and valporic acid, the three latter have been used in clinical trials as cancer drugs (Nature review Marks et al 2001).

2.4 HISTONE ACETYLATION = TRANSCRIPTIONAL ACTIVATION?

2.4.1 Transcription by RNA polymerase II

Eukaryotic transcription is a highly regulated process, in which gene activation requires collaboration of a number of factors such as the RNA polymerase II and the Mediator together with general transcription factors, sequence-specific DNA-binding activators and additional co-activators. Since the ground state of eukaryotic gene expression is repressive due to the nucleosomes, initiation of gene expression requires co-activators that disrupt the transcriptionally inactive chromatin structure. Upon activation a transcription factor (TF) binds to sequence specific binding sites, so called upstream activator sequence (UAS). Then co-activators, such as HAT-containing complexes, and/or chromatin remodelling complexes are recruited in order to make the promoter and start site accessible to the TATA-box binding protein (TBP). TBP and a number of TBP associated factors (TAFs) forms the TFIID complex, which is one of several general transcription factors (GTFs). The GTFs also include the RNAP II. Together with the other GTFs (TFIIA, TFIIB, TFIIE, TFIIF and TFIIH), TFDII acts as auxiliary components to the RNAP II and forms a co called pre-initiation complex (PIC) that directs the RNAP II to the transcription start site. The Mediator complex is then recruited to RNAP II following phosphorylation of the C-terminal domain (CTD) of RNAP II. This leads to release of several GTFs that can be reused to form new pre-

(16)

initiation complexes. The phosphorylated RNAP II then forms the elongating holoenzyme together with the Elongator complex in order to proceed to transcriptional elongation. As the elongating RNAP II travels along the coding region nucleosomes needs to be removed from the DNA template. The Elongator complex is thought to facilitate the eviction of nucleosomes along with chromatin remodelling complexes.

Histone acetylation plays a crucial role in several steps of transcription as several complexes involved has intrinsic HAT activity; the co-activator SAGA complex (Gcn5), TFIID (TAFII250), the S. cerevisiae Mediator subunit Nut1 (Lorch et al, 2000) and Elongator (Elp3).

2.4.2 New insights

Histone acetylation has long been linked to transcriptional activation, as the histone acetyltransferase enzymes function generally as transcriptional co-activators (Berger, 2002). Hyperacetylated histones have been associated with transcriptionally active genes while hypoacetylated histones are found in repressed or silenced regions. But the opposite is also true. Recent studies have shown that histone deacetylation is required for transcriptional expression. For instance, in budding yeast the Rpd3(S) complex deacetylates the open reading frame (ORF) of actively transcribed genes in order to prevent aberrant cryptic transcripts (Li et al, 2007). Another HDAC, Hos2, has been found to deacetylate histone lysines in the coding region of highly transcribed genes, functioning as a general activator of transcription in both S. cerevisiae and S.pombe (Wang et al, 2002; Wiren et al, 2005). In contrast, acetylation of H4K12 have been linked to transcriptional repression in S. cerevisiae (Braunstein et al, 1996).

Furthermore, the histone acetyltransferase Gcn5 has been shown to repress expression of the arg1 genes in S. cerevisiae and that the repression is HAT activity dependent.

Similarly, in S. pombe, Gcn5 has been shown to repress the ste11 and mei2 genes (Helmlinger et al, 2008). In both cases deletion of Gcn5 resulted in a decrease of H3 acetylation suggesting that the HAT activity of Gcn5 might not be targeted to histones only (see 1.6.1.2 Gcn5 mediated acetylation of non-histone proteins below).

2.5 CROSSTALK BETWEEN DIFFERENT HISTONE MODIFICATIONS There are several examples where one histone modification promotes the etablishment of another. Phosphorylation of histone H3 serine 10 (H3S10) promoting acetylation of H3K14 has been shown in both mammals and yeast (Cheung et al, 2000; Lo et al,

(17)

2001; Lo et al, 2000). In S. cerevisiae, phosphorylation of S10 is mediated by Snf1which stimulates Gcn5-mediated acetylation of K14 (Lo et al, 2001). H3S10 phosphorylation has also been shown to inhibit H3K9 acetylation (Edmondson et al, 2002) and methylation while H3K9 dimethylation but not trimethylation, opposes H3S10 phosphorylation. In addition, H3K9 deacetylation is required for H3K9 methylation and hence heterochromatin assembly and maintenance (Shankaranarayana et al, 2003; Yamada et al, 2005).

Another example of crosstalk between different histone modifications is the Rpd3(S) dependent deacetylation of actively transcribed genes, where the chromodomain of the Eaf3 and the PHD domain of the Roc1 subunits recognise methylated H3K4 and H3K36. Budding yeast Rpd3(S) deacetylates H4 and is thought to be important to restore the acetylation status behind the elongating RNAP II (Li et al, 2007). As mentioned above, all three forms of methylation of H3K4 is associated with active chromatin. Trimethylation of H3K36, which is catalyzed by Set2, is also correlated with transcription activity and is enriched throughout the coding region with a peak at the 3´end (Pokholok et al, 2005). The location of the modifications is important as Set2 dependent methylation of H3K36 in the promoter region has been shown to inhibit activation (Strahl et al, 2002). To avert inhibition of transcription, H3K36 in the promoter regions of actively transcribed genes is acetylated by Gcn5 (Morris et al, 2007).

The examples above illustrate the complexity of histone modifications and the impact on chromatin structure and regulation of transcription. It also emphasizes the importance of studying the interactions between the different modifying enzymes and the composition of the protein complexes.

2.6 HISTONE ACETYLTRANSFERASES- THE PLAYERS AND THEIR COMPLEXES

Histone acetyltransferases can be separated into three different groups, Gcn5-related-N- acetyl transferase (GNAT) family, p300 (adenoviral E1A-associated protein of 300kDa)/CBP (CREB-binding protein) and the MYST family (see 1.6.3). The HATs found in yeast belong to the GNAT or MYST family of proteins (Figure 3).

(18)

±

2.6.1 Gcn5

The first HAT to be discovered was the Tetrahymena thermophilia protein p55, which was shown to be a homologue to the Saccharomyces cerevisiae co-activator Gcn5 (Brownell et al, 1996). This provided a direct link between histone acetylation and transcriptional activation. Gcn5 is the catalytic subunit of several HAT activity complexes, including the SAGA (see 1.6.1.1 below) and ADA complexes. Genome- wide expression analysis using DNA micro arrays have shown that 4-5 per cent of all genes in S. cerevisiae are dependent on Gcn5 (affected by Gcn5 deletion) under normal growth on rich medium (Holstege et al, 1998; Lee et al, 2000). The S. cerevisiae gene pho5 has been used to study transcriptional activation by Gcn5. Deletion of gcn5 has an effect on the basal levels of pho5 promoter activity but has little consequence for the

Figure 3. Members of the (A) GNAT and (B) MYST families.

y, S. cerevisiae; Tet, Tetrahymena thermofilia; h, human; m, mouse;

d, fly. The relative sizes and location of conserved motifs are indicated; AT, acetyltransferase domains; bromo, bromodomains;

PhD, pland homeodomains; Zn, zink finger domains; chromo, chromodomains. Reprinted from Annual Review of Biochemistry, 70, Roth, Denu and Allis, Histone acetyltransferases, p 81-120, copywright 2001.

(19)

steady-state level of activation under inducing conditions (Gregory et al, 1998). Instead, it seems to affect the rate of chromatin remodelling and gene activation (Barbaric et al, 2001). This may exemplify a general principle by which co-activators play a more important role during physiological transitions than under steady-state growth conditions.

Homologues of Gcn5 have been found in diverse eukaryotic organisms (Sterner &

Berger, 2000), suggesting that its function is highly conserved, which we have addressed in this study where we investigate the role of Gcn5 in fission yeast and budding yeast (paper IV).

In a free histone mixture, Gcn5 acetylates primarily lysine 14 (K14) on histone H3 and lysine 8 (K8) on histone H4. As part of the SAGA complex, Gcn5 has also been found to acetylate additional H3 lysines such as K9, K18 and K27 (Grant et al, 1999). The wider range of substrates obtained with an intact SAGA complex suggest that other subunits are important for recognition of substrate and/or the catalytic activity of Gcn5.

2.6.1.1 The SAGA complex

The SAGA complex was first discovered in S. cerevisiae (Grant et al, 1997) but homologs are found in humans (STAGA) and S. pombe (Helmlinger et al, 2008). The S. cerevisiae SAGA complex contains 19 subunits (Daniel & Grant, 2007) that can be divided into distinct functional modules (figure 4); the HAT module consisting of Gcn5, Ada2 and Ada3; Spt3 and Spt8 that interacts with TBP; the Sus1/Sgf11/Ubp8 module that are required for H2B deubiquitination. The Ada1, Spt7 and Spt20 subunits are important for complex integrity while the essential Tra1 subunit, which also exists in the NuA4 complex, mediates the recruitment of SAGA to promoter regions by direct interaction with transcription factors.

While some studies have suggested redundant functionality for SAGA and TFIID in transcriptional regulation (Lee et al, 2000), other studies have reported SAGA to have a role in regulation of stress-regulated genes and TFIID primary acting on housekeeping genes (Huisinga & Pugh, 2004; Zanton & Pugh, 2004). The composition between S.

cerevisiae SAGA and the human counterpart STAGA is very similar (Lee & Workman, 2007). Recently, SAGA was purified in S. pombe and the two yeast species complexes were shown to have identical composition (Helmlinger et al, 2008). This further

(20)

emphasizes the central role for SAGA in transcriptional regulation and in addition enables direct comparisons between the yeasts.

Figure 4. The SAGA complex. Set1 mediated methylation of H3K4 and Snf1 phosphorylation of H3S10 promotes Gcn5 HAT activity. The chromodomain containing Chd1 and the bromodomains of Gcn5 and Spt7 facilitates chromatin interaction along with Sca7 and Taf12. Tra1 mediates recruitment of SAGA by interaction with transcription factors (TF), this interaction is strengthen by the19S proteasome regulatory particle (RP). Spt3 and Spt8 interact with TBP, that together with SAGA strongly interact with the TFIIA and the elongation factor TFIIS. Not shown are the Ada1, Spt20 and the TAF5,6,9 and 10 subunits.

Reprinted from Mutation Research, 618, Daniel and Grant, Multi-tasking on chromatin with the SAGA coactivator complexes, p 135-148 (2007), with permission from Elsevier.

2.6.1.2 Gcn5 mediated acetylation of non-histone proteins

In addition to acetylating lysines on the histone tails, Gcn5 also acetylates non-histone targets. In S. cerevisiae Gcn5 acetylates Rsc4, a subunit of the RSC chromatin remodelling complex (VanDemark et al, 2007). Rsc4 contains an essential tandem bromodomain (TBD) that interacts with acetylated H3K14 (Kasten et al, 2004).

Acetylation of K25 on Rsc4 inhibits binding of one of the bromodomains to acetylated H3K14 as the other bromodomain bind the acetylated K25, suggesting an Gcn5 dependent autoregulatory mechanism of RSC activity (VanDemark et al, 2007).

Gcn5 has also been shown to acetylate other non-histone proteins in other eukaryotes.

The chromatin remodelling ATPase ISWI in Drosophila melanogaster is acetylated by Gcn5 at lysine 753 in vitro and in vivo (Ferreira et al, 2007). The human Gcn5 has been shown to acetylate Cdc6 at three different lysines in the cycling-docking motif of Cdc6 affecting CDK specific phosphorylation of serine 106 and S-phase progression (Paolinelli et al, 2009). As the case with RSC, these studies show that the role of HATs in regulation of transcription is not restricted to histone acetylation and that focusing on

(21)

other non-histone targets will generate a better understanding of the complexity of gene regulation.

2.6.2 Elp3

In 1999 Jesper Svejstrup and colleagues identified the histone acetyltransferase Elp3 and the Elongator complex as a part of the elongating and phosphorylated form of RNAP II in S. cerevisiae (Otero et al, 1999; Wittschieben et al, 1999). Elp3 belongs to the same family of HATs as Gcn5, the GNAT family. The predominant targets for Elp3 mediated acetylation are H3K14 and H4K8 (Winkler et al, 2002). Cells lacking elp3⁺ have slow growth, delay in gene activation and temperature sensitivity. Deletion of both elp3 and gcn5 results in severe growth defects that can be suppressed by simultaneous deletion of two HDACs, hda1, the ortholog of S. pombe clr3, and hos2 (Wittschieben et al, 2000). Phenotypes similar to elp3Δ^gcn5Δ are obtained when deletion of elp3 is combined by a mutation disrupting the SAGA complex indicating overlapping functionality between Elongator and SAGA (Wittschieben et al, 2000). In agreement with this, deletion of gcn5 and elp3 results in severe hypoacetylation in the coding region, transcriptional inhibition (Kristjuhan et al, 2002) and spreading of heterochromatin related Sir3 protein (Kristjuhan et al, 2003). In addition, Gcn5 and Elp3 have been shown to regulate the Hsp70 genes SSA3 and SSA4 by overlapping function regarding H3 acetylation but with different mechanism with regards to transcription (Han et al, 2008).

2.6.3 MYST family

The MYST family is named by its members MOZ-Ybf2/Sas3-Sas2-Tip60. In S.

cerevisiae there are three members, the essential Esa1, homologous to human Tip60, and the two Something-about-silencing proteins Sas2 and Sas3. Esa1 is the catalytic subunit of the NuA4 (Nucleosomal Acetylation of H4) complex that primarily acetylates lysines on histones H4 and H2A and is involved in DNA repair. The S.

pombe ortholog, Mst1 is also essential (Gomez et al, 2005).

The two other S. cerevisiae MYST family members Sas2 and Sas3, were originally isolated with defects in silencing the mating locus (Ehrenhofer-Murray et al, 1997;

Reifsnyder et al, 1996). Sas2 is the catalytic subunit of the budding yeast SAS complex

(22)

and has been shown to antagonise the Sir2 HDAC by acetylating H4K16 and thereby prevent spreading of heterochromatin (Kimura et al, 2002; Suka et al, 2002).

Sas3 is the catalytic subunit of NuA3 (Nucleosomal Acetylation of H3) and has been shown to have overlapping substrate specificity with Gcn5 (Rosaleny et al, 2007) as both proteins have been shown to acetylate H3K14. Deletion of both genes is synthetically lethal (Howe et al, 2001).

In S. pombe there is only one additional MYST member, Mst2. Mst2 shows sequence homology to both Sas2 and Sas3. It has been suggested that Mst2 is the functional homolog to Sas2, as it antagonizes fission yeast HDAC Sir2 in telomere silencing (Gomez et al, 2005). But Mst2 has functions in common with Sas3 as well, as the deletion mutant shows a loss of H3K14 acetylation and sensitivity to various DNA damage reagents (paper III).

2.6.4 Other HATs

In S. cerevisiae, the Hat1 acetyltransferase is found in the cytoplasm and is thought to acetylate newly synthesised histones and thereby allow transport into the nucleus.

There is a Hat1 homolog in S. pombe but it has not yet been characterized. As mentioned above, there are in higher eukaryotes additional HATs such as p300/CBP and PCAF (p300/CBP- associated factor), a homolog to Gcn5.

2.6.5 Histone deacetylases

Acetyl groups are removed by histone deacetylases (HDACs). HDACs can be separated phylogenetically into three classes, the Rpd3 and Hos2-like Class I, the Hda1-like Class II and the Sir2-like Class III also called Sirtuins. Class I and II are often referred to as classical HDACs and differ from the class III Sirtuins in that the Sirtuins are nicotinamide adenine dinucleotide (NAD) dependent. The class I and class II HDACs share sensitivity to Tricostatin A (TSA).

In S. pombe there are three classical HDACs; the class I Clr6 and Hos2, and the class II Clr3, ortholog of budding yeast Hda1. There are also three S. pombe Sirtuins: Sir2 (Shankaranarayana et al, 2003), Hst2 (Durand-Dubief et al, 2007) and Hst4 (Freeman-Cook et al, 1999). Genome-wide studies of HDAC localization and histone

(23)

acetylation revealed different roles for the S. pombe HDACs (Durand-Dubief et al, 2007; Wiren et al, 2005). Hos2 has been targeted to actively transcribed parts of the genome and deletion of hos2 results in increasing H4K16 acetylation in the coding region with loss of transcription as a consequence (Wiren et al, 2005). The role of Hos2 in promoting transcription has also been reported for S. cerevisiae (Wang et al, 2002)suggesting that this function of Hos2 is evolutionary conserved and may exist in metazoans as well. Mutating the essential Clr6, orthologous to S. cerevisiae Rpd3, affects centromeric silencing and chromosome segregation (Grewal et al, 1998) and similarly to Rpd3, Clr6 has broad substrate specificity and acts as a repressor of gene expression (Robyr et al, 2002; Wiren et al, 2005).

The class II HDAC Clr3 (Ekwall and Ruusala, Genetics, 1994) target acetylated lysines on histone H3 with a preference for H3K14 (Bjerling et al, 2002; Wiren et al, 2005) and is required for mating-type and ribosomal DNA (rDNA) silencing (Bjerling et al, 2002). Deacetylation of H3K9 by Clr3 stabilizes H3K9 trimethylation and thereby maintenance of heterochromatin. In addition, it also prevents other modifications associated with active chromatin (eg H3S10pho and H3K14ac) resulting in limiting access of RNAP II (Yamada et al, 2005). But Clr3 has also been shown to affect euchromatic repression, in particular genes located in the subtelomeric regions (Wiren et al, 2005). Further more Clr3 has an overlapping pattern in localization and gene repression with another important heterochromatic HDAC, namely Sir2 (Wiren et al, 2005). Sir2 is important for heterochromatin assembly (Shankaranarayana et al, 2003) in both S. pombe and S. cerevisiae though in S. cerevisiae H4K16 deacetylation is a mark for heterochromatin while in S. pombe and higher eukaryotes the mark for heterochromatin is H3K9 hypoacetylation.

2.7 HAT-HDAC INTERACTIONS

Similarly to the suppression of gcn5∆elp3∆ phenotypes by deletion of hda1 and hos2, the phenotypes caused by gcn5∆ and esa1-ts can be suppressed by deletion of the rpd3 and hda1 HDACs (Vogelauer et al, 2000), suggesting antagonistic functionality between these HATs and HDACs. Specific HAT-HDAC antagonists are also found in S. pombe. In paper II we show that deletion of clr3 specifically suppress gcn5∆

(Johnsson et al, 2009) indicating that specific HAT-HDAC interactions are conserved.

(24)

In other cases HATs and HDACs collaborate. Proper acetylation followed by deacetylation is also important during DNA repair. During repair of double-strand breaks by homologous recombination the histone acetyltransferases Gcn5 and Esa1, and the Histone deacetylases Rpd3, Sir2 and Hst1 are recruited to the site of the break (Tamburini & Tyler, 2005).

A large part of my studies have been focused on the interactions between Gcn5 and other co-regulators, both HDACs and HATs, in order to understand the mechanisms underlining gene regulatory networks.

(25)

3 COMMENTS ON METHODOLOGY

3.1 FISSION YEAST AS MODEL SYSTEM

The fission yeast Schizosaccharomyces pombe is less well known that the distantly related budding yeast Saccharomyces cerevisiae. Both yeasts can be grown and genetically manipulated in a similar way, and have a fully sequenced genome (Goffeau et al, 1996; Wood et al, 2002) thus facilitating cross species comparison. S. pombe has three chromosomes and a genome size around 14.1Mb with 5027 protein coding genes, many of which are highly conserved in other yeasts and multicellular eukaryotes. A S. pombe predominantly is maintained as a haploid, genetic approaches are fairly uncomplicated. In some aspects, S. pombe is more similar to higher metazoans than S. cerevisiae, such as centromere composition, RNAi machinery and heterochromatin maintenance. As far as histone acetylation, there are many similarities between the two yeasts but also some differences, in the number of enzymes and their function making studies on regarding involving chromatin and histone modifications interesting.

3.2 CHROMATIN IMMUNOPRECIPITATION (CHIP) ON CHIP

By combining chromatin immunoprecipitation (ChIP) techniques with microarray technology it is possible to look at localization of specific proteins or modifications genome wide. In yeast, the ChIP-chip assay has primarily been developed by the labs of Michael Grunstein (Kurdistani et al, 2002; Robyr & Grunstein, 2003; Robyr et al, 2002; Suka et al, 2001; Vogelauer et al, 2000) and Richard Young (Pokholok et al, 2005; Robert et al, 2004). Previously, most studies had been conducted focusing on one ore few genes generating information on targeted modifications but not on the untargeted global effects. Genome-wide techniques can enable a global overview of the localization and enzymatic activity to correlate with impact on transcription for one or several proteins. Briefly, ChIP involves cross-linking of proteins to DNA following steps where the DNA-protein is extracted and shredded into shorter fragments. These fragments have then been introduced to antibodies against a specific histone modification such as acetylated H3K14, or the modifying enzymes. After immunoprecipitation, the cross-linking is reversed and the DNA purified. Following PCR amplification and labelling, the amplified DNA fragments are hybridized to DNA microarrays containing either intergenic regions, open reading frames or both.

(26)

4 RESULTS AND DISCUSSION

4.1 AIMS OF THIS STUDY

This study aims to gain further knowledge of the well conserved co-regulator Gcn5 and its role in regulating gene expression, as well as how Gcn5 interact with other histone modifying enzymes to direct DNA related events in the nucleus. The specific aims for each paper were:

- finding physiological conditions where Gcn5 has a phenotype and where the interactions with other co-regulators can be studied (paper I)

- further investigation of the Gcn5-HDAC interactions (paper II)

- determining the level of redundancy between Gcn5 and other HATs (paper III) - investigate the functional conservation of Gcn5 in regulating KCl stress

response (paper IV)

4.2 PAPER I: GCN5 IS INVOLVED IN KCL STRESS RESPONSE

Gcn5 has been the target for many studies as it was first HAT to be identified. Though it is well conserved throughout evolution and exists in diverse species it is not essential for growth in yeast. Contrary, deletion of Spac1952.05, encoding Gcn5 in S. pombe generated a mutant with seemingly wild type phenotype, with normal cell shape and only a slightly prolonged generation time. To look at Gcn5 function in regulating transcription expression profiling using DNA micro arrays was conducted.The results of expression profiling were consistent with the lack of phenotypes, very few genes where differentially regulated. However, as the initial micro arrays were conducted under growth in rich media we would only detect differences in steady state levels, and as mentioned above gcn5-deficient cells had no obvious defects in growth. We therefore looked for physiological conditions where the deletion mutant had a strong phenotype. The observation that the gcn5∆ mutant had impaired growth on plates containing KCl or CaCl2 lead us to further investigate the KCl induced stress response in S. pombe. As a first step we sought to identify the expression profile caused by KCl stress. We found that many of the genes regulated under KCl induced stress had previously been identified as core environmental stress response genes (CESR genes), genes that are differentially expressed in response to several types of stress including heat, DNA damage and oxidative stress (Chen et al, 2003). When looking at the

(27)

expression profile for gcn5∆ under KCl stress conditions significantly more genes were differentially regulated in gcn5∆ compared to wild type. We found that a portion of the genes that were either induced or reduced in the wild type cells under KCl stress required Gcn5 for proper expression. Furthermore we found that many of the Gcn5- dependent KCl response genes contained specific sequences within the promoter region including a a sequence found in the promoter region of man CESR genes.

The main conclusions from this initial characterization of Gcn5 is that, as in S.

cerevisiae, Gcn5 is not required for growth under normal conditions but plays an important role in stress adaptation by regulating a specific set of genes. We also conclude that KCl induced stress can be used as a physiological condition to study co- regulator interaction.

4.3 PAPER II: ANTAGONISTINC ACTIVITIES OF GCN5 AND CLR3 REGULATE H3K14 ACETYLATION LEVELS IN RESPONSE TO KCL INDUCED STRESS

In this study we continued the characterization of Gcn5 by looking at the genomewide localization of myc-tagged Gcn5 using ChIP-on-chip. In contrast to what has previously been show in S. cerevisiae, we found the highest enrichment of Gcn5 in the coding region of highly expressed genes and not in the promoter region. For this study we used high-resolution arrays from Affymetrix that covers the entire genome with a 20 bp resolution. The localization of Gcn5 to the coding region rather than promoter regions was surprising since Gcn5 and SAGA has previously been know to be recruited to the promoter regions by transcription factors such as Gal1 and Gcn4. The localization to the coding region suggested a role in elongation in addition to activation of transcription. Studies in budding yeast have also suggested a function for Gcn5/SAGA in transcriptional elongation on specific genes (Govind et al, 2007).

Sensitivity to MPA, a drug that depletes the internal pool of GTP, has been used to indicate deficiencies in transcriptional elongation, further suggested a role for Gcn5 in elongation. In support of this there‘s also genetic interaction between gcn5 and several Elongator subunits (Roguev et al, 2008).

We show overlapping localization for Gcn5 and several HDACs, suggesting that there might be common targets. We therefore looked for genetic interactions between Gcn5

(28)

and the S. pombe HDACs. We found that deletion of clr3 could suppress the salt stress sensitivity of gcn5∆, while deletion of hos2 had a synthetic effect on growth. The outcome that deletion of clr3 could suppress the gcn5∆ mutant phenotype was in line with results from other studies showing Clr3 to deacetylate H3K14 (Bjerling et al, 2002) and H3K14 as a target of S. cerevisiae Gcn5 enzymatic activity (Grant et al, 1999). The sir2∆ mutant was however not able to suppress the KCl sensitivity, which was somewhat surprising since Clr3 and Sir2 have been shown to have overlapping localization and substrate specificity (Durand-Dubief et al, 2007; Wiren et al, 2005).

Next, we wanted to see if specific histone marks were important in the response to KCl induced stressed by looking at the phenotype of strains carrying point mutations that changed lysine 9 or lysine 14 on histone H3, either to an uncharged alanine (A) or a changed arginine (R). Replacement of a lysine with an alanine mimics the net changed of an acetylated lysine whereas the replacement to arginine mimics the net charge of a hypoacetylated lysine. However in both cases the loss of a lysine inhibits any interaction with bromodomains regardless of charge. Interestingly only the H3K14R mutant was sensitive to salt stress. We also found synthetic growth defects with H3K14A and gcn5∆ (H3K14R was not tested) but not with gcn5∆ in combination with either K9 mutant.

To look at H3K14 levels globally, we performed ChIP-on chip using antibodies against acetylated H3K14 and the C-terminus of H3 (as a control for nucleosome density). We found a severe reduction of H3K14 acetylation in gcn5∆ cells compared to wild type.

By deletion of clr3 could partly restore the loss of H3K14, especially on highly expressed genes. This suggests that Clr3 might have a specific role in the regulation of genes that require Gcn5 for efficient elongation. An analogous role for the Sir2 HDAC in having a specific role in repression of several low-expressed genes has previously been reported (Durand-Dubief et al, 2007).

In summary, Gcn5 is likely to have a role in transcriptional elongation in addition to activation as it is localized to the coding regions of highly transcribed genes and is sensitive to drugs affecting elongation. We also show that acetylation of H3K14 is important for KCl stress response and that H3K14 is a substrate of Gcn5 enzymatic activity as loss of gcn5 significantly reduces H3K14ac levels globally. Furthermore,

(29)

Gcn5 and Clr3 specifically have opposing roles in H3K14 acetylation levels but their activity is likely to affect other targets, such as non-histone proteins.

4.4 PAPER III: REDUNDANT AND SPECIFIC FUNTIONS OF S. POMBE HATS

The lack of phenotypes for gcn5∆ on normal growth suggests redundant functions with other HATs. The results from paper II show that Gcn5 is important for H3K14 acetylation under KCl stress conditions. To continue investigating Gcn5 we addressed the possibility of redundancy with other HATs, with focus on those likely to target the same substrates, histone H3 lysines. We addressed this by using genetic approach combined with expression profiling and determining bulk levels of H3 acetylation.

As Elp3 had previously not been characterized in S. pombe, so the first task was to perform and initial characterisation. We found that elp3∆ cells had an elongated cell shape and displayed a slow growing phenotype with delayed entry to the cell cycle and premature exit. We then tested single deletion mutants of gcn5, mst2 and elp3, as well as double and triple mutants, for growth under various conditions. Surprisingly, and in contrast to S. cerevisiae, neither of the double mutant combinations nor the triple mutant was synthetically lethal.

The phenotypic analysis suggests a role for Gcn5 and Mst2 in DNA-damage repair and salt stress response while Elp3 and Mst2 in many cases seems have an antagonistic function. Expression profiling suggests that Gcn5, Elp3 and Mst2 either are redundant or not involved in transcriptional regulation as few genes were differentially regulated in the single mutants. However, of the few genes regulated by Gcn5 and Mst2, a significant amount are found within 150 kb ends of chromosomes I and II, suggesting that they are involved in transcriptional regulation of genes near the subtelomeric region.

Deletion of gcn5 had the most severe effect on H3 acetylation as it severely reduced K9, K14 and K18 acetylation and seems to be the HAT responsible for K9 and K18 acetylation in S. pombe. However both Gcn5 and Mst2 are involved in H3K14 acetylation. This might be the reason that gcn5∆mst2∆ has an even stronger growth defect on KCl containing media than gcn5∆ as acetylation of H3K14 was shown to play a central role in KCl stress adaptation (paper II).

(30)

In summary, this study compares HATs from to different families and examines the overlap in their functions in S. pombe. This is the first study that looks at transcriptional regulation by HAT enzymes from different families. We created mutants containing single, double or triple deletions of HAT enzymes within the cell. With these mutants we found redundancies in transcriptional regulation, salt response, DNA damage response, and enzymatic activity. This works suggests that the overlap in function between different HATs is extensive and include many different cellular functions.

4.5 PAPER IV: COMPARISON OF GCN5-DEPENDENT STRESS RESPONSE IN DIFFERENT YEAST SPECIES

In paper I we studied the Gcn5 dependent KCl stress response in S. pombe and defined a set of genes as Gcn5 dependent KCl response genes. In this study we compare the Gcn5 dependent stress response in budding yeast and fission yeast and discuss this from an evolutionary standpoint as S. cerevisiae and S. pombe are highly diverged yeast species. We find that in both yeasts, as well as the budding yeast related Saccharomyces kluyveri, gcn5∆ cells are sensitive to KCl and CaCl2 stress. This indicates that the role of Gcn5 in stress response is conserved between the yeast species despite differences in stress response. We also show that the dependence of budding yeast Gcn5 during KCl induced stress response is HAT-activity dependent.

Next we aimed to determine whether the dependence of Gcn5 is mediated through the same set of genes in both yeast species. As a first step we defined Gcn5 dependent KCl regulated genes similarly to paper I. Consistent with the results of the previous studies in this thesis, and contradictory to the general belief of Gcn5 primarily a co-activator, we find that a substantial number of genes are upregulated in the absence of Gcn5 indicating a repressive role in addition to an activating role. In parallel with expression profiling, we looked at the localization of Gcn5 genome wide using ChIP-chip. By combining the expression profiling with localization data we determined direct and indirect targets.

Gene ontology analysis reveals that the genes dependent on Gcn5 during KCl stress are to some extent involved in the same biological processes. To determine whether Gcn5 is involved in regulation of the same genes on an individual gene level we compared the expression of orthologous pairs of genes but found no significant overlap. This

(31)

suggests that though Gcn5 is involved in the stress response in both yeast species the mechanism behind the requirement of Gcn5 is different.

When comparing the localization pattern of Gcn5 under normal and stress induced conditions we observed a shift in the distribution of Gcn5 within genes to the coding regions during KCl stress adaptation in S. cerevisiae. Under normal conditions Gcn5 is primarily localized near the promoter region and 5´ end of the genes. However, under KCl stress Gcn5 becomes redistributed and is found in the coding region. This altered distribution during KCl stress is not observed in S. pombe as the distribution of Gcn5 in the coding region under normal conditions (paper II) remains. We conclude that Gcn5 has a conserved role in KCl stress response but that the response is mediated through divergent sets of Gcn5 dependent genes and that the targeting of Gcn5 to these genes differs between S. pombe and S. cerevisiae.

(32)

5 CONCLUDING REMARKS

The collective action of both HATs and HDACs determine the acetylation status and consequently the ability of effector proteins to recognise and bind nucleosomal DNA that in turn regulate gene activity. It is becoming more and more evident that to understand the occurrence of cancer and other various diseases originating from defects in cell cycle regulation and transcription related events, we need to improve our knowledge of co-regulators and their dynamic interplay.

There are seemingly many proteins that could in theory do the same thing, yet keeping multiple proteins with the same enzymatic activities is conserved in eukaryotic system.

Initially many of these enzymes seem redundant and dispensable for cell survival but the example in mouse development suggest that HATs indeed have very specific functions. During growth under normal conditions yeast cells manage very well without the highly conserved Gcn5 protein. But if conservation is an indication of importance, then not having Gcn5 should be a disadvantage. But indeed it is, at specific conditions. By studying chromatin structure related proteins under conditions where the chromatin is facing significant conformal changes, as is the case in stress response, we can learn far more that just looking at healthy growing cells.

The discovery that HATs target other proteins than histones underlines the relevance of further understanding how these enzymes act and interact. In human p300 and CPB target several proteins that have a central role in cellular programming, such as the transcription factor E2F and the tumour suppressor p53. Acetylation of these factors affect their DNA-binding properties and as a consequence, their ability in regulating transcription.

Though Gcn5 and Clr3 regulate H3K14 levels indeed, they may also regulate non- histone proteins. Further investigation will show if there is a similar autoregulatory function of Gcn5 in S. pombe as has been shown for S. cerevisiae Gcn5 in controlling RSC function (VanDemark et al, 2007).

(33)

(34)

6 ACKNOWLEDGEMENTS

I have crossed paths with a lot of people during these past year and may of them have in different ways made my PhD studies memorable.

First, I would like to thank my supervisor Tony, for his valuable advice. These past years has truly a growing experience. I’d also like to thank my co-supervisor for her support during the first years of my studies.

To ALL of the AWR and KEK group members, past and present, for the fun times in and outside the lab. I have enjoyed it a lot! A special Thank You! to Helmi for always being so thoughtful and helpful, you are the heart of the AWR. Marianna, my dear friend, we’ve had great fun together (and lots of wine) and you have been missed ever since you went to S F. Azadeh, for your friendship and support. Julian and Fredrik, thank you for all the help when I was stuck in the ChIP-träsk, you guys are great!

Many, many thanks to Mickaël, for coming into my science life and saving the day!

Thank you for your support and belief in my data and our paper. Yongtao, you have always been very helpful, friendly and supportive. The same goes for Ingela, Indranil and Carolina, I’m happy for the lab-time we’ve shared. Michelle, thank you for helping out with the paper, you made a difference. Monica, I’m so glad I didn’t have to go through the last year alone. We did it! All the best in the future! Rachel, thank you for all your help with the manuscripts, for your efforts to organize the group meetings and for being positive and encouraging. A special thank you to Karl Ekwall for being a pombe-wizard and for always taking the time to answer my questions.

Sam, Johan and Asim, thank you for your valuable help with printing posters and just general helping out with computers and soft ware.

Agneta, du har varit en mentor (och en extra mamma ibland)

Cissi, vad ska jag säga….din vänskap har tagit mig igenom tuffa tider. Vet inte vad jag hade gjort utan dig, ditt stöd (och våra dagliga kaffestunder) har varit ovärdeligt. Jag kommer att sakna dig så stanna inte bort för länge……

Ellinor, vilken tur att du läste biokemi och jag undervisade….. Vilken tur att vi fann varann, din vänskap vill jag aldrig vara utan. Tack för alla pratstunder, god råd och all kärlek, du är en sann och älskad vän.

Min familj, mamma, pappa, ni har gett mig en inre styrka, utan vilken jag aldrig hade klarat av det här. Tack för att ni alltid finns där. Maria och Sara, mina fantastiska underbara systrar, jag hoppas att ni vet vad ni betyder för mig. Jesper, det had inte gått utan dig, jag är så glad och tacksam att jag haft dig.

Bianca, min underbara, vackra , älskade du, jag föddes på nytt när du kom in i våra liv.

Du är min sol och min kraft. Jag älskar dig oändligt……

(35)

7 REFERENCES

Barbaric S, Walker J, Schmid A, Svejstrup JQ, Horz W (2001) Increasing the rate of chromatin remodeling and gene activation--a novel role for the histone

acetyltransferase Gcn5. EMBO J 20(17): 4944-4951

Berger SL (2002) Histone modifications in transcriptional regulation. Curr Opin Genet Dev 12(2): 142-148

Bhaumik SR, Smith E, Shilatifard A (2007) Covalent modifications of histones during development and disease pathogenesis. Nat Struct Mol Biol 14(11): 1008-1016 Bjerling P, Silverstein RA, Thon G, Caudy A, Grewal S, Ekwall K (2002) Functional divergence between histone deacetylases in fission yeast by distinct cellular localization and in vivo specificity. Mol Cell Biol 22(7): 2170-2181

Braunstein M, Sobel RE, Allis CD, Turner BM, Broach JR (1996) Efficient transcriptional silencing in Saccharomyces cerevisiae requires a heterochromatin histone acetylation pattern. Mol Cell Biol 16(8): 4349-4356

Brownell JE, Zhou J, Ranalli T, Kobayashi R, Edmondson DG, Roth SY, Allis CD (1996) Tetrahymena histone acetyltransferase A: a homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell 84(6): 843-851

Chen D, Toone WM, Mata J, Lyne R, Burns G, Kivinen K, Brazma A, Jones N, Bahler J (2003) Global transcriptional responses of fission yeast to environmental stress. Mol Biol Cell 14(1): 214-229

Cheung P, Allis CD, Sassone-Corsi P (2000) Signaling to chromatin through histone modifications. Cell 103(2): 263-271

Daniel JA, Grant PA (2007) Multi-tasking on chromatin with the SAGA coactivator complexes. Mutat Res 618(1-2): 135-148

Durand-Dubief M, Sinha I, Fagerstrom-Billai F, Bonilla C, Wright A, Grunstein M, Ekwall K (2007) Specific functions for the fission yeast Sirtuins Hst2 and Hst4 in gene regulation and retrotransposon silencing. EMBO J 26(10): 2477-2488

Edmondson DG, Davie JK, Zhou J, Mirnikjoo B, Tatchell K, Dent SY (2002) Site- specific loss of acetylation upon phosphorylation of histone H3. J Biol Chem 277(33):

29496-29502

Ehrenhofer-Murray AE, Rivier DH, Rine J (1997) The role of Sas2, an

acetyltransferase homologue of Saccharomyces cerevisiae, in silencing and ORC function. Genetics 145(4): 923-934

(36)

Ferreira R, Eberharter A, Bonaldi T, Chioda M, Imhof A, Becker PB (2007) Site- specific acetylation of ISWI by GCN5. BMC Mol Biol 8: 73

Freeman-Cook LL, Sherman JM, Brachmann CB, Allshire RC, Boeke JD, Pillus L (1999) The Schizosaccharomyces pombe hst4(+) gene is a SIR2 homologue with silencing and centromeric functions. Mol Biol Cell 10(10): 3171-3186

Fuchs SM, Laribee RN, Strahl BD (2009) Protein modifications in transcription elongation. Biochim Biophys Acta 1789(1): 26-36

Goffeau A, Barrell BG, Bussey H, Davis RW, Dujon B, Feldmann H, Galibert F, Hoheisel JD, Jacq C, Johnston M, Louis EJ, Mewes HW, Murakami Y, Philippsen P, Tettelin H, Oliver SG (1996) Life with 6000 genes. Science 274(5287): 546, 563-547 Gomez EB, Espinosa JM, Forsburg SL (2005) Schizosaccharomyces pombe mst2+

encodes a MYST family histone acetyltransferase that negatively regulates telomere silencing. Mol Cell Biol 25(20): 8887-8903

Govind CK, Zhang F, Qiu H, Hofmeyer K, Hinnebusch AG (2007) Gcn5 promotes acetylation, eviction, and methylation of nucleosomes in transcribed coding regions.

Mol Cell 25(1): 31-42

Grant PA, Duggan L, Cote J, Roberts SM, Brownell JE, Candau R, Ohba R, Owen- Hughes T, Allis CD, Winston F, Berger SL, Workman JL (1997) Yeast Gcn5 functions in two multisubunit complexes to acetylate nucleosomal histones: characterization of an Ada complex and the SAGA (Spt/Ada) complex. Genes Dev 11(13): 1640-1650 Grant PA, Eberharter A, John S, Cook RG, Turner BM, Workman JL (1999) Expanded lysine acetylation specificity of Gcn5 in native complexes. J Biol Chem 274(9): 5895- 5900

Gregory PD, Schmid A, Zavari M, Lui L, Berger SL, Horz W (1998) Absence of Gcn5 HAT activity defines a novel state in the opening of chromatin at the PHO5 promoter in yeast. Mol Cell 1(4): 495-505

Grewal SI, Bonaduce MJ, Klar AJ (1998) Histone deacetylase homologs regulate epigenetic inheritance of transcriptional silencing and chromosome segregation in fission yeast. Genetics 150(2): 563-576

Han Q, Lu J, Duan J, Su D, Hou X, Li F, Wang X, Huang B (2008) Gcn5- and Elp3- induced histone H3 acetylation regulates hsp70 gene transcription in yeast. Biochem J 409(3): 779-788

Helmlinger D, Marguerat S, Villen J, Gygi SP, Bahler J, Winston F (2008) The S.

pombe SAGA complex controls the switch from proliferation to sexual differentiation through the opposing roles of its subunits Gcn5 and Spt8. Genes Dev 22(22): 3184- 3195

Holstege FC, Jennings EG, Wyrick JJ, Lee TI, Hengartner CJ, Green MR, Golub TR, Lander ES, Young RA (1998) Dissecting the regulatory circuitry of a eukaryotic genome. Cell 95(5): 717-728

(37)

Howe L, Auston D, Grant P, John S, Cook RG, Workman JL, Pillus L (2001) Histone H3 specific acetyltransferases are essential for cell cycle progression. Genes Dev 15(23): 3144-3154

Huisinga KL, Pugh BF (2004) A genome-wide housekeeping role for TFIID and a highly regulated stress-related role for SAGA in Saccharomyces cerevisiae. Mol Cell 13(4): 573-585

Johnsson A, Durand-Dubief M, Xue-Franzen Y, Ronnerblad M, Ekwall K, Wright A (2009) HAT-HDAC interplay modulates global histone H3K14 acetylation in gene- coding regions during stress. EMBO Rep

Kasten M, Szerlong H, Erdjument-Bromage H, Tempst P, Werner M, Cairns BR (2004) Tandem bromodomains in the chromatin remodeler RSC recognize acetylated histone H3 Lys14. EMBO J 23(6): 1348-1359

Kimura A, Umehara T, Horikoshi M (2002) Chromosomal gradient of histone acetylation established by Sas2p and Sir2p functions as a shield against gene silencing.

Nat Genet 32(3): 370-377

Kouzarides T (2007) Chromatin modifications and their function. Cell 128(4): 693-705 Kristjuhan A, Walker J, Suka N, Grunstein M, Roberts D, Cairns BR, Svejstrup JQ (2002) Transcriptional inhibition of genes with severe histone h3 hypoacetylation in the coding region. Mol Cell 10(4): 925-933

Kristjuhan A, Wittschieben BO, Walker J, Roberts D, Cairns BR, Svejstrup JQ (2003) Spreading of Sir3 protein in cells with severe histone H3 hypoacetylation. Proc Natl Acad Sci U S A 100(13): 7551-7556

Kurdistani SK, Robyr D, Tavazoie S, Grunstein M (2002) Genome-wide binding map of the histone deacetylase Rpd3 in yeast. Nat Genet 31(3): 248-254

Lee KK, Workman JL (2007) Histone acetyltransferase complexes: one size doesn't fit all. Nat Rev Mol Cell Biol 8(4): 284-295

Lee TI, Causton HC, Holstege FC, Shen WC, Hannett N, Jennings EG, Winston F, Green MR, Young RA (2000) Redundant roles for the TFIID and SAGA complexes in global transcription. Nature 405(6787): 701-704

Li B, Gogol M, Carey M, Pattenden SG, Seidel C, Workman JL (2007) Infrequently transcribed long genes depend on the Set2/Rpd3S pathway for accurate transcription.

Genes Dev 21(11): 1422-1430

Lo WS, Duggan L, Emre NC, Belotserkovskya R, Lane WS, Shiekhattar R, Berger SL (2001) Snf1--a histone kinase that works in concert with the histone acetyltransferase Gcn5 to regulate transcription. Science 293(5532): 1142-1146

Lo WS, Trievel RC, Rojas JR, Duggan L, Hsu JY, Allis CD, Marmorstein R, Berger SL (2000) Phosphorylation of serine 10 in histone H3 is functionally linked in vitro and in vivo to Gcn5-mediated acetylation at lysine 14. Mol Cell 5(6): 917-926

(38)

Lorch Y, Beve J, Gustafsson CM, Myers LC, Kornberg RD (2000) Mediator- nucleosome interaction. Mol Cell 6(1): 197-201

Millar CB, Grunstein M (2006) Genome-wide patterns of histone modifications in yeast. Nat Rev Mol Cell Biol 7(9): 657-666

Morris SA, Rao B, Garcia BA, Hake SB, Diaz RL, Shabanowitz J, Hunt DF, Allis CD, Lieb JD, Strahl BD (2007) Identification of histone H3 lysine 36 acetylation as a highly conserved histone modification. J Biol Chem 282(10): 7632-7640

Nakayama J, Rice JC, Strahl BD, Allis CD, Grewal SI (2001) Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science 292(5514):

110-113

Otero G, Fellows J, Li Y, de Bizemont T, Dirac AM, Gustafsson CM, Erdjument- Bromage H, Tempst P, Svejstrup JQ (1999) Elongator, a multisubunit component of a novel RNA polymerase II holoenzyme for transcriptional elongation. Mol Cell 3(1):

109-118

Paolinelli R, Mendoza-Maldonado R, Cereseto A, Giacca M (2009) Acetylation by GCN5 regulates CDC6 phosphorylation in the S phase of the cell cycle. Nat Struct Mol Biol 16(4): 412-420

Pokholok DK, Harbison CT, Levine S, Cole M, Hannett NM, Lee TI, Bell GW, Walker K, Rolfe PA, Herbolsheimer E, Zeitlinger J, Lewitter F, Gifford DK, Young RA (2005) Genome-wide map of nucleosome acetylation and methylation in yeast. Cell 122(4):

517-527

Reifsnyder C, Lowell J, Clarke A, Pillus L (1996) Yeast SAS silencing genes and human genes associated with AML and HIV-1 Tat interactions are homologous with acetyltransferases. Nat Genet 14(1): 42-49

Robert F, Pokholok DK, Hannett NM, Rinaldi NJ, Chandy M, Rolfe A, Workman JL, Gifford DK, Young RA (2004) Global position and recruitment of HATs and HDACs in the yeast genome. Mol Cell 16(2): 199-209

Robyr D, Grunstein M (2003) Genomewide histone acetylation microarrays. Methods 31(1): 83-89

Robyr D, Suka Y, Xenarios I, Kurdistani SK, Wang A, Suka N, Grunstein M (2002) Microarray deacetylation maps determine genome-wide functions for yeast histone deacetylases. Cell 109(4): 437-446

Roguev A, Bandyopadhyay S, Zofall M, Zhang K, Fischer T, Collins SR, Qu H, Shales M, Park HO, Hayles J, Hoe KL, Kim DU, Ideker T, Grewal SI, Weissman JS, Krogan NJ (2008) Conservation and rewiring of functional modules revealed by an epistasis map in fission yeast. Science 322(5900): 405-410

Rosaleny LE, Ruiz-Garcia AB, Garcia-Martinez J, Perez-Ortin JE, Tordera V (2007) The Sas3p and Gcn5p histone acetyltransferases are recruited to similar genes. Genome Biol 8(6): R119

Characterization of Gcn5 histone acetyltransferase in Schizosaccharomyces pombe

Characterization of Gcn5 Histone acetyltransferase in Schizosaccharomyces pombe

Anna Johnsson

From the DEPARTMENT OF BIOSCIENCES AND NUTRITION Karolinska Institutet, Stockholm, Sweden

CHARACTERIZATION OF GCN5 HISTONE ACETYLTRANSFERASE IN

SCHIZOSACCHAROMYCES POMBE

Anna Johnsson

Stockholm 2009

Till Bianca

ABSTRACT

LIST OF PUBLICATIONS

TABLE OF CONTENTS

LIST OF ABBREVIATIONS

1 INTRODUCTION

2 GENERAL INTRODUCTION TO HISTONE MODIFICATIONS AND GENE REGUALTION

3 COMMENTS ON METHODOLOGY

4 RESULTS AND DISCUSSION

5 CONCLUDING REMARKS

6 ACKNOWLEDGEMENTS

7 REFERENCES

Till Bianca 