Thesis for doctoral degree (Ph.D.)2007

Rasheed Khan

Thesis for doctoral degree (Ph.D.) 2007

Rasheed K

han

DNA methylation and 5-azacytidine in myelodysplastic syndromes: pharmacodynamic,

mechanistic and clinical studies

DN

A m

ethylation an

d 5-azacytidine in

myelodysplastic syn

dromes:

pharmacodyn

amic, m

echanistic an

d clinical stu

dies

Division of Hematology, Department of Medicine, Karolinska University Hospital Huddinge, Karolinska Institutet, Stockholm, Sweden.

DNA methylation and 5-azacytidine in myelodysplastic syndromes: pharmacodynamic,

mechanistic and clinical studies

Rasheed Khan

Stockholm 2007

2007

Gårdsvägen 4, 169 70 Solna

Printed by

All previously published papers were reproduced with permission from the publisher. Published by Karolinska Institutet Printed by Reproprint © Rasheed Khan, 2007. ISBN: 978-91-7357-419-8

u :

o ,

,

e n

3

4

LIST OF PUBLICATIONS

This thesis is based on the following papers, which are referred to by their Roman numbers:

I. Rasheed Khan, Anni Aggerholm, Peter Hokland, Moustapha Hassan and Eva Hellström-Lindberg. A pharmacodynamic study of 5-azacytidine in the P39 cell line. Experimental Hematology 2006; 34:35–43.

II. Rasheed Khan, Jan Schmidt-Mende, Mohsen Karimi, Vladimir Gogvadze, Moustapha Hassan, Tomas J. Ekström, Boris Zhivotovsky and Eva Hellström-Lindberg. Hypomethylation and apoptosis in 5-azacytidine treated myeloid cells. 2007. Accepted in Experimental Hematology (2007).

III. Michael Grövdal, Rasheed Khan, Anni Aggerholm, Petar Antunovic, Jan Astermark, Per Bernell, Lena-Maria Engström, Lars Kjeldsen, Olle Linder, Lars Nilsson, Anna Olsson, Jonas Wallvik, Jon Magnus Tangen, Gunnar Öberg, Peter Hokland, Anna Porwit and Eva Hellström-Lindberg. Negative impact of DNA hypermethylation on the outcome of intensive chemotherapy in elderly patients with high risk myelodysplastic syndromes and acute myeloid leukaemia following myelodysplastic syndrome. Clinical Cancer Research (in press, 2007).

IV. Rasheed Khan, Mohsen Karimi, Martin Jäderstan, Tomas J. Ekström and Eva Hellström-Lindberg. Global and gene-specific DNA methylation status in MDS patients. Submitted to Haematologica-THJ (2007).

5

ABSTRACTPromoter DNA hypermethylation and hence silencing of e.g. tumour suppressor genes is considered to be an important step in carcinogenesis and has been associated with poor outcome in patients with myelodysplastic syndromes (MDS). In contrast to many other chemotherapeutic agents, the DNA hypomethylating compound 5-azacytidine has a positive therapeutic effect in patients with high-risk myelodysplastic syndromes (MDS); however, the lack of knowledge about its mechanism of action hampers clinical development of the drug. The first aim of the thesis was to assess the pharmocodynamic properties of 5-azacytidine, and the mechanism of 5-azacytidine-induced apoptosis as well as its link with DNA methylation. Secondly, we investigated the role of gene-specific methylation status for the outcome of induction chemotherapy in a cohort of patients high-risk MDS and MDS-AML, and assessed global and gene-specific (p15INK4B) methylation patterns in another cohort of patients with low and intermediate-risk MDS. 5-azacytidine induced dose-dependent hypomethylation and growth inhibition, as well as a complex pattern of apoptotic mechanisms regulated by multiple and separate pathways. Furthermore, we demonstrated a possible link between apoptosis and hypomethylation. We showed that optimal cellular effects of 5-azacytidine could be achieved by shorter exposure times than expected. Translated to shorter treatment duration in patients, this may eventually lead to superior and more cost-effective and convenient clinical treatment schedules. P15INK4B methylation plus methylation of CDH or HIC gene was more frequent in patients with conventional high-risk features, such as adverse karyotype and high percentage of marrow blasts. DNA hypermethylation of multiple surrogate genes correlated with the percentage of blasts and marrow CD34 expression, but not with the degree of inefficient haematopoiesis. We demonstrated a significant negative impact of promoter methylation status on the outcome of conventional induction chemotherapy in high-risk MDS and MDS-AML, a finding that may have significant impact on the clinical management of patients with high-risk myeloid disease. Global and gene-specific DNA methylation pattern in MDS correlated with each other in the sense that more advanced MDS, characterized by p15INK4B methylation showed higher global DNA methylation than those with unmethylated p15INK4B. Global methylation status did not correlate with conventional risk factors, but was more pronounced in patients with severe erythroid failure. Patients with stable disease maintained their global DNA methylation level over time, and disease progression was associated with an increase in global DNA methylation in only half of the cases. To conclude, these studies provide important knowledge about the cellular mechanisms of 5-azacytidine and role of DNA methylation in patients with MDS and MDS-AML.

6

CONTENTS

1. LIST OF ABBREVIATIONS..............................................................................8

2. INTRODUCTION ..............................................................................................9

1. Epigenetics ..........................................................................................................................................91.1 Epigenetic modulation.................................................................................................................10

1.1.1.2 The Methyl-Binding Domain Proteins ....................................................................121.1.3.1.1 Mechanism of RNAi ............................................................................................151.1.3.1.2 Functional use of RNAi........................................................................................151.1.3.1. Recent developments..............................................................................................16

2.2. Programmed cell death.................................................................................................................202.2.1 Apoptosis..................................................................................................................................202.2.2 5-azacytidine and apoptosis......................................................................................................26

2.3. Myelodysplastic Syndromes .........................................................................................................272.3.1 The historical perspective.........................................................................................................272.3.2 Epidemiology ...........................................................................................................................282.3.3 Clinical and morphological diagnosis ......................................................................................292.3.4 Classifications ..........................................................................................................................292.3.5 Prognosis ..................................................................................................................................302.3.6 Low-risk and high-risk MDS....................................................................................................312.3.7 Pathogenesis of MDS ...............................................................................................................322.3.8 Treatment..................................................................................................................................35

2.3.8.3.2.1 Hypomethylating therapy..................................................................................372.3.8.3.2.2 Mechanisms of hypomethylating drugs in MDS:..............................................38

3. AIM OF THE THESIS .....................................................................................40

4. MATERIALS AND METHODS .......................................................................41

4.1 Sample selections, cell culture and incubation techniques..........................................................414.1.1 Patient (Paper III & IV) ............................................................................................................414.1.2 Bone marrow sampling (Paper III-IV) .....................................................................................414.1.3 Cell line and cell culture (Paper I-II)........................................................................................414.1.4 Drug dilution, Dose and Dosage (Paper I-II)............................................................................414.1.5 Incubation with inhibitors (Paper II) ........................................................................................42

4.2 Methylation assays .........................................................................................................................424.2.1 Measurement of global DNA methylation by LUMA (Paper II-III) ........................................424.2.2 Gene-specific methylation assay by bisulfite-DGGE (Paper I, III-IV).....................................42

4.3 Apoptosis assays .............................................................................................................................434.3.1 Morphological assay of apoptosis (Paper I-II) .........................................................................434.3.2 Antibodies and Western blot experiments (Paper II)................................................................434.3.3 Flow cytometry (Paper I-II)......................................................................................................434.3.4 Enzyme activity assay (Paper II) ..............................................................................................43

4.4 Other assays....................................................................................................................................444.4.1 Cell growth (Paper I-II) ............................................................................................................444.4.2 Cell proliferation (DNA synthesis assay using 3H-TdR) (Paper I) ...........................................444.4.3 Cell cycle analysis (Paper I) .....................................................................................................444.4.4 Sorting apoptotic cells (Paper II)..............................................................................................44

7

5. RESULTS AND DISCUSSION .......................................................................45

5.1 Pharmacodynamic of 5-azacytidine..............................................................................................455.1.1 Effect on cell growth ................................................................................................................455.1.2 Effect on cell proliferation (DNA synthesis assay using 3H-TdR) ...........................................455.1.3 Effect on cell cycle ...................................................................................................................465.1.4 Pharmacodynamic analysis.......................................................................................................46

5.2 Methylation.....................................................................................................................................475.2.1 Global and gene-specific methylation status in patients with MDS .........................................475.2.2 Methylation status in relation to outcome of treatment ............................................................495.2.3 Hypomethylating effects of 5-azacytidine................................................................................50

5.3 Apoptosis.........................................................................................................................................515.3.1 Morphological assay of apoptosis ............................................................................................51

5.4 Overview of 5-azacytidine-induced hypomethylation and apoptosis.........................................56

5.5 Efficacy of 5-azacytidine in MDS when other drugs fail.............................................................57

6. CONCLUSIONS .............................................................................................58

7. FUTURE PERSPECTIVES.............................................................................59

8. ACKNOWLEDGEMENTS...............................................................................60

9. REFERENCES ...............................................................................................63

10. APPENDICES...............................................................................................77

8

1. LIST OF ABBREVIATIONS

AIF apoptosis inducing factor MDR multidrug resistance gene AML acute myeloid leukaemia MDS myelodysplastic syndromes ANT adenine nucleotide translocator MDS-U MDS unclassified

APSCT allogenic peripheral stem cell transplantation MeCP methyl-CpG binding protein

ATG antithymocyte globulin miRNA micro RNA ATP adenosine triphosphate MPR multi resistance-associated protein

Caspases cysteine-rich aspartate-specific proteases MPT mitochondrial permeability

transition CDH E-cadherin mRNA messenger RNA

CMML chronic myelomonocytic leukaemia PARP poly(ADP-ribose) polymerase

CpG cytosine-phosphate-guanine pi-RNA piwi-interacting RNA CR complete remission PR partial remission CSA cyclosporin-A RA refractory anaemia

DD death domain RAEB refractory anaemia with excess blast

DFS disease free survival RARS refractory anaemia with ringed sideroblasts

DIABLO direct IAP binding protein with low pl RCMD refractory cytopenia with

multilineage dysplasia

DISC death inducing signalling complex

RCMD-RS RCMD with ringed sideroblasts

DNA deoxyribonucleic acid RIC reduced intensity conditioning DNMT DNA methyltransferase RISC RNA interfering silencing complex dsRNA double stranded RNA RNA ribonucleic acid EPO erythropoietin RNAi RNA interference ER estrogen receptor RNase ribonuclease FAB French-American-British ROS reactive oxygen species FADD Fas-associated death domain SAM S-adenosyl-methionine

FDA US Food and Drug Administration

SCT allogenic stem cell transplantation

G-CSF granulocyte colony-stimulating factor shRNA short-hairpin RNA

GM-CSF granulocyte-macrophage colony-stimulating factor siRNA short interfering dsRNA

HAT histone acetyltransferase Smac second mitochondrial activator of caspases

HDAC histone deacetylase smRNA small modulatory RNA HIC hypermethylated in cancer tBid truncated Bid HIV human immunodeficiency virus TNF tumour necrosis factor

IAP inhibitors of apoptosis proteins TRADD TNF-receptor-associated death domain

IPSS international prognostic scoring system TRD transcription repression domain

LDH lactate dehydrogenase VDAC voltage dependent anion channel

LRP lung resistance-associated protein WHO World Health Organization

MBD methyl-CpG binding domain protein WPSS WHO-based prognostic scoring

system

9

2. INTRODUCTION 1. Epigenetics

The diversity of genetic constitution explains the dissimilarity between two species, for example between man and mouse. The genetic constitution may also explain the variations between two individuals from the same species. However, genetic variations alone could not explain the dissimilarity between different tissues or cells within a single individual. Although genetically identical, the cells of a single individual have different phenotypes and functions. The muscle, blood, skin and other cells of an individual have the same DNA sequence (genotype), but vary radically in terms of functionality and structure (phenotype). Hence other regulatory systems must be functional for regulating gene expression. Those regulatory systems fall into the broad headings of epigenetics.

The Greek ‘epi-’ prefix of the word ‘epigenetics’ implies features that are ‘on top of’ or ‘in addition to’ genetics. Thus the term epigenetics encompasses events that influence gene function, but is on top of or in addition to the traditional molecular basis for inheritance. Epigenetics has been defined as modifications of DNA or associated factors with information content other than the DNA sequence itself, which are maintained during cell division (Feinberg and Tycko 2004).

The term ‘epigenetics’ was originally coined by the developmental biologist Conrad Waddington (1905-1975) as “the branch of biology which studies the causal interactions between genes and their products which bring the phenotype into being”. However the modern ‘epigenetics’ was re-introduced in 1983 (Feinberg and Tycko 2004) and the topic has become increasingly visible in the field of cell biology and cancer. Development is, by definition, epigenetic (Reik 2007). Differences in the programmes for gene expression, which result in the development of different organs and tissues, occur without changes to the DNA sequence. Subsets of the ~30,000 genes in our genome are active in different tissues and organs, depending on their regulation by different sets or combinations of transcription factors. This indicates that if all the genes of liver-specific transcription factors are activated in a brain cell while inactivating all brain-specific transcription factors, the brain cell would turn into a liver cell. Moreover, cellular differentiation processes (development of an individual from an unstructured egg) rely strongly on epigenetics rather than inherited DNA sequence only (Reik 2007). However epigenetic modulation is not always inherited (Reik 2007) and has been shown to be affected by age (Fraga et al. 2005; Wong et al. 2005), environment (Bird 2007) and early maternal care (Weaver et al.2004).

10

1.1 Epigenetic modulation Epigenetic modulation plays a critical role in embryogenesis and

development, as well as in carcinogenesis and other diseases (Jones and Laird 1999; Jones and Baylin 2002; Baylin and Ohm 2006; Klose and Bird 2006). It is a rapidly expanding field and at present the boundaries of it are redefined quite shortly. At the moment epigenetics could be divided into three broad headings; DNA methylation, histone modifications and RNA regulatory systems.

1.1.1 DNA methylation DNA methylation is the most studied and characterized epigenetic alteration.

Two contrasting changes in methylation have been associated with neoplasia; global hypomethylation of the genome and regional hypermethylation at CpG (cytosine-phosphate-guanine) dinucleotides. The importance of hypomethylation for tumourigenesis is still somewhat unclear, although it has been associated with genomic instability (Jones and Baylin 2002).

DNA methylation, catalyzed by DNA methyltransferases (DNMTs), involves transferring of a methyl group from a methyl donor, S-adenosyl-methionine (SAM), predominantly to the C-5 position of the cytosine ring, resulting in 5-methylcytosine (me-C) (Figure 1) (Bird 2002). In mammals, me-C is found in approximately 4% of genomic DNA, primarily in CpG sequences. Although CpG dinucleotides occur at lower than expected frequencies throughout the mammalian genome, a subset of CpG dinucleotides rich regions, known as CpG islands, are found in or near promoter regions of genes, where transcription is initiated (Herman and Baylin 2003; Baylin 2005). Methylation in a gene promoter region (CpG islands) generally correlates with a silenced gene (Jones and Laird 1999; Bird 2002; Jones and Baylin 2002).

In the bulk of genomic DNA about 80% of CpG dinucleotides that are not associated with CpG islands are heavily methylated. In contrast, dinucleotides in the CpG islands, which account for around 1% of the genome and are especially associated with gene promoters, are usually unmethylated allowing gene expression to occur (Bird 2002). Promoter DNA methylation is used by mammalian cells to maintain a normal expression pattern; for example regulation of imprinted gene

Figure 1. Conversion of cytosine to 5-methylcytosine.

11

expression and X-chromosome inactivation, among others (Csankovszki et al. 2001; Jones and Takai 2001; Kaneda et al. 2004).

It is considered that epigenetic gene inactivation is at least as common as, if not more frequent, than mutational events in cancer (Herman et al. 1994; Merlo et al.1995; Herman 1999; Jones and Baylin 2002; Leone et al. 2002). The role of DNA methylation in initiation and/or progression of cancer are implicated by the large number of genes that are found to undergo hypermethylation in various malignancies. Many of these genes play a role in regulating cell growth, differentiation, signal transduction and DNA repair, while others are involved in tumour metastasis and angiogenesis (Table 1). In addition, tumour suppressor genes are inactivated by hypermethylation in various cancers including haematopoietic malignancies (Table 2).

1.1.1.1 DNA methyltransferase (DNMT) DNMT is a family of enzymes that catalyze transfer of a methyl group from a

methyl donor, SAM, to cytosine residues in DNA. In mammals, three families of DNMTs (cytosine-5) have been identified: DNMT1, DNMT2, DNMT3A and DNMT3B (Bestor 2000). These enzymes can superficially be separated into de novoor maintenance methyltransferase. De novo methyltransferases mediate establishment of new DNA methylation patterns. Maintenance methyltransferases continue to preserve the methylation pattern through cell divisions that has already been established by de novo methyltransferases.

DNMT1 is the most abundant and largest (184 kDa) DNA methyltransferase (Kim et al. 2002). It has high affinity for hemimethylated (when one strand is methylated) CpG dinucleotides strands. Further, inactivation of the mouse DNMT1 gene by gene targeting results in extensive demethylation of all sequences examined (Li et al. 1992; Lei et al. 1996). Although there is no direct evidence, these findings suggest that DNMT1 functions as a major maintenance methyltransferase in vivo(Yoder et al. 1997).

DNMT2 is a much smaller molecule with a molecular size of 45 kDa (Kim et al. 2002). It is the first DNA cytosine methyltransferase to be identified to methylate tRNA in a vertebrate (Goll et al. 2006). The cytoplasmic localization of DNMT2 contrasts with the exclusive nuclear localization of DNMT1 and DNMT3 (Leonhardtet al. 1992; Bachman et al. 2001). Inactivation of DNMT2 gene in mouse embryonic stem cells by gene targeting perturbs neither de novo nor maintenance methylation of proviral DNA (Okano et al. 1998b). However although a biological role is not yet discovered for DNMT2 in any species, a role in some aspects of centromere function has been proposed (Yoder et al. 1997).

DNMT3a and DNMT3b are intermediate in size (100-130 kDa) (Kim et al.2002). Inactivation of DNMT3a and DNMT3b genes by gene targeting blocks de

12

novo methylation in embryonic stem cells and early embryos, but has no effect on maintenance of imprinted methylation patterns (Okano et al. 1999). These results indicate that both DNMT3a and DNMT3b function primarily as de novomethyltransferases.

Interestingly, it has been reported that for translational purposes, cancer cells may be different in that DNMT1 is not responsible alone for maintaining abnormal gene hypermethylation, both DNMT1 and DNMT3b may cooperate on this function (Rhee et al. 2002). The same group also has shown indirectly that methylation activities other than DNMT1 can maintain the methylation of most of the genome (Rhee et al. 2000). Moreover, it has been shown that murine DNMT3 has no preference for hemimethylated versus unmethylated oligonucleotides in vitro (Okanoet al. 1998a). Therefore although there is a tendency to consider each DNA methyltransferases either to be de novo or maintenance methyl transferase, the real situation might not be that simple (Bestor 2000).

1.1.1.2 The Methyl-Binding Domain Proteins

Methyl-CpG binding domain proteins (MBDs) can influence DNA methylation pattern, rather than directly modifying them. They usually bind to methylated sequences in DNA, and thereby influence transcription. Initially two methyl-binding activities were identified within eukaryotic cells, and these identified complexes were termed methyl-CpG binding protein 1 (MeCP1) and MeCP2 (Meehan et al. 1989; Lewis et al. 1992). Later MeCP1 was shown to be a complex of MBD2, MBD3 and the multisubunit protein corepressor NuRD complex (Cross et al.1997; Ng et al. 1999; Feng and Zhang 2001; Feng et al. 2002); whereas MeCP2 is a single protein. From these proteins, a common MBD was identified and used to identify a family of proteins that include MBD1, MBD2, MBD3, MBD4 and MeCP2 (Figure 2) (Hendrich and Bird 1998; Hendrich et al. 1999).

The MBD proteins MBD1, MBD2 and MeCP2 repress transcription from methylated promoters in vitro and in vivo via the association of their transcription repression domain (TRD) with a corepressor complex (Sansom et al. 2007). Furthermore, MBD1 has been reported to cooperate with the DNA damage protein (methylpurine-DNA glycosylase) in transcriptional repression and DNA repair, suggesting a direct role of MBD1 in sensing base damage in chromatin (Watanabe et al. 2003). Of all the MBD proteins, only MBD3 cannot specifically bind methylated DNA (Saito and Ishikawa 2002), and required MBD2 to recruit it to methylated DNA. Unlike the other MBD proteins, there is little evidence of MBD4 to function as transcriptional repressor; instead MBD4 primarily acts as methylation-specific DNA repair protein (Sansom et al. 2003), interacting with the mismatch repair (MMR) family of DNA repair proteins (Bellacosa et al. 1999). Interestingly, Kaiso, a protein

13

which is structurally unrelated to MBD proteins, has been reported to repress transcription in both a methylation-dependent and sequence-specific manner (Prokhortchouk et al. 2001).

The exact relation between the MBD proteins is yet to be discovered. However in addition to gene expression, MBD family of proteins have been linked to multiple and diverse functions in tumourigenesis; several MBD proteins seem to augment tumourigenesis, whilst MBD4 seems to function primarily as a DNA repair protein. The fact that MBD2 knockout mice are viable and strongly resistant to tumourigenesis makes MBD2, of all the MBD proteins, the most ideal and relevant therapeutic target (Sansom et al. 2007).

1.1.2 Histone modifications Histone modifications, and in particular the posttranslational modifications of

their amino-terminal tail domain, are critical for gene expressions (Egger et al. 2004). Histones are highly conserved protein and chief component of chromatin. They act as spools around which DNA is coiled. They are grouped into five major classes. Two copies of H2A, H2B, H3 and H4 bind to about 200 base pairs of DNA to form the repeating structure of chromatin, the nucleosome, with H1 binding to the linker sequence (Figure 3). The H3 and H4 histones have long tails protruding from the nucleosome which can be covalently modified at several places. Modifications of the

Figure 2. Schematic structure of major MBD proteins. (Abbreviations: BTB/POZ, broad complex, tramtrack, bric a brac/pox virus zinc finger; GR, alternate glycin and arginie repeats).

14

tail include acetylation, methylation, phosphorylation, ubiquitination, sumoylation, ADP ribosylation, glycosylation, biotinylation and carbonylation (Strahl and Allis 2000; Turner 2002). These modifications cause chromatin to be distinguished in two categories, as detected by cytological studies. Firstly heterochromatin, which is generally condensed and transcriptionally inactive, hence regarded as ‘silent’ chromatin. Secondly euchromatin, which is not condensed and favourable for gene transcription, thus regarded as ‘open’ chromatin.

1.1.3 RNA regulatory system RNA regulatory system is the latest addition to the network of epigenetic

control functions. This is a highly conserved gene silencing mechanism and plays an important role in regulation of gene expression.

1.1.3.1 RNA interference RNA interference (RNAi): RNAi was first discovered in plants in 1990 by

Rich Jorgenson and his team (Napoli et al. 1990). RNAi is an evolutionally conserved gene silencing mechanism present in a variety of eukaryotic species. RNAi serves as a safeguard for maintaining the genomic integrity. It protects the host from viral infections and invasion by mobile genetic elements by degrading the exogenous

Figure 3. Schematic presentation of histone and its modifications.

15

genomic material (viral RNA). In recent years, RNAi has become a powerful tool to probe gene function and to rationalize drug design.

1.1.3.1.1 Mechanism of RNAi

RNAi uses double stranded RNA (dsRNA) to trigger degradation or translation repression of homologous RNA targets in a sequence-specific manner (Figure 4). dsRNA, produced by endogenous genes, invading viruses, transposons or experimental transgenes, are initially recognised by a dsRNA binding protein (RDE-4/R2D2) (Tabara et al. 2002; Liu et al. 2003). Subsequent mechanism is mediated by activation of two major molecules; the initial activity of major endonuclease Dicer (an RNAse III family enzyme), followed by the activity of the RNA interfering silencing complex (RISC) (Chiu and Rana 2002). Dicer cleaves the long dsRNA in an adenosine triphosphate (ATP) dependent reaction into short interfering double stranded RNAs (siRNAs), 21-23 nucleotides duplex RNAs with overhanging 3´ ends. siRNAs are incorporated into RISC and RISC then unwinds siRNA using a helicase and subsequently binds to the free antisense strand. The antisense strand of the duplex directs RISC to recognise and cleave cognated target RNAs, which undergoes specific base pairing and endonucleolytic cleavage. This leads to the degradation of the unprotected and single-stranded target RNA.

The dicer also cleaves the 60-70 nucleotide long precursor microRNA (pre-miRNA) into miRNA, which are similar size as siRNAs. The pre-miRNAs, whose structures are imperfectly complementary to each strand, are generated from endogenous stem loop precursors or hairpins, named primary-miRNA (pri-miRNA). The pri-miRNA are cleaved by Drosha RNase III in the nucleus (Lee et al. 2003) and exported to cytoplasm for further processing by Dicer. The complex of activated RISC and miRNA binds the 3’UTR of specific miRNAs, which triggers cleavage by perfect base pairing, or translational repression by partial base-pair recognition (Hutvagner and Zamore 2002a; Hutvagner and Zamore 2002b, Llave et al. 2002; Doench et al. 2003).

1.1.3.1.2 Functional use of RNAi

The RNAi pathway is often used in experimental biology to study the function of genes in cell culture and in vivo in model organisms. In this method, dsRNA is synthesized with a sequence complementary to a gene of interest and introduced into cells, where it is recognized as an exogenous genetic material and activates the RNAi pathway. Subsequently, a decrease in the expression of a targeted gene, followed by its physiological role can be studied.

16

The possibility of using RNAi in therapy is promising. At present, the possibility is suggested to the following conditions: inhibition of viral gene expression in cancer cells, knockdown of host receptors and coreceptors for viruses (e.g. HIV), silencing of genes (e.g. Hepatitis A, Hepatitis B), inhibition of viral replication (e.g. measles) and silencing of genes that are differentially regulated in tumours or genes involved in the cell cycle.

RNAi has been used for applications in biotechnology. It has been used to produce lower levels of natural plant toxins (cotton seed), reduced levels of allergens (tomato), decreased the precursors of likely carcinogens in plants (tobacco) and resistance to common plant viruses.

1.1.3.1. Recent developments

In addition to naturally occurring and synthetic siRNAs, there have been recent publications of alternative forms of RNAi; Piwi-interacting RNAs (pi-RNAs), small modulatory RNAs (smRNAs) and short-hairpin RNAs (shRNAs).

pi-RNA are single stranded RNAs (25-31 nucleotides) and shown to associate to Piwi protein (a subclass of Argonaute proteins) and human RecQ1 protein to form

Figure 4. Schematic presentation of RNA interference pathways.

17

a Piwi-interacting RNA complex (piRC). These complexes are thought to regulate the genome within developing sperm cells (Carthew 2006).

smRNA are short, double stranded RNAs which are found in the nucleus of neural stem cells of mice. It has been shown to play a critical role in mediating neuronal differentiation through dsRNA/protein interaction (Kuwabara et al. 2004).

shRNA are synthetically manufactured based on naturally occurring miRNAs structure. They can be expressed by plasmid vectors or endogenously inserted into the genome. They contain inverted repeat of 19-29 nucleotides with the desired sequence separated by 6-9 base pairs (Medema 2004).

1.1.3.2 RNA induced gene transcription Recently there have been reports that small double stranded RNAs can induce

gene transcription (Li et al. 2006; Janowski et al. 2007). It has been shown that by targeting the promoter region of genes by dsRNAs one can induce potent and long-lasting transcriptional activation in a sequence specific manner (Li et al. 2006). Although the exact mechanism is presently unknown, it has been reported that this process required the Argonaute 2 protein and is associated with histone changes linked to gene activation (Li et al. 2006).

1.1.4 Gene reactivation by influencing chromatin configuration Reversible chromatin rearrangements can influence gene expression by two

interdependent mechanisms: (1) DNA methylation and (2) histone modifications (such as, acetylation, methylation, phosphorylation, etc.). They cause changes in chromatin structure and thereby regulate gene expression, e.g. the accessibility of chromatin to transcription regulator proteins. In contrast, RNA regulatory system is believed not to be involved in this process but rather act at the transcriptional level of gene silencing, mostly by binding to antisense strand of DNA or acting through mRNA.

DNA hypermethylation has been shown to promote gene repression by several mechanisms. Firstly, by sterically inhibiting the binding of activating transcription factors to gene promoters, thereby repressing gene activation directly. Secondly, through recruitment of several methyl-binding domain proteins (MBDs) that recognizes methylated DNA. These protein themselves can inhibit gene transcription, or bind chromatin-remodelling proteins and transcription-regulatory complexes which lead to formation of ‘silent’ chromatin and subsequent gene silencing (Bird and Wolfe 1999). Apart from their methylation ability, DNMTs can induce gene silencing by acting as transcriptional repressor by themselves or by serving as binding scaffolds for transcriptional repressors, histone deacetylases and histone methyltransferases. Hence, DNMTs not only maintains DNA methylation but also may directly target, in a heritable manner, transcriptionally repressive chromatin

18

to the genome during DNA replication (Robertson et al. 2000; Rountree et al. 2000; Fuks et al. 2003).

Histone modifications, alone or in combinations, determine the chromatin functional properties (Kouzarides 2002; Martin and Zhang 2005). The best studied histone modification is histone acetylation which is controlled by histone acetyltransferases (HATs) and histone deacetylases (HDACs). Histone acetylation is generally associated with ‘open’ and transcriptionally active chromatin, whereas histone deacetylation correlates with ‘silent’ condensed chromatin and transcriptional repression (Struhl 1998).

DNA methylation and histone acetylation are interconnected in gene silencing. HDACs are recruited by MBDs after binding of these to methylated DNA, resulting in chromatin remodelling and transcriptional silencing (Bird and Wolfe 1999). Moreover, a direct link between DNA methylation and histone acetylation exists by direct interaction between DNMTs and HDACs (Robertson et al. 2000; Rountree et al. 2000). DNMTs have been suggested to repress transcription using deacetylase activity, independent of their methylation capacity. DNA methylation and histone deacetylation are crucial in X chromosome inactivation, genomic imprinting and establishment of tissue specific gene expression (Jaenisch and Bird 2003). However, aberrant epigenetic regulation of gene expression also plays a major role in the development of cancer.

Rb:retinoblastoma; CDKN2A:cyclin-dependent kinase inhibitor 2A (p16); CDKN2B:cyclin-dependent kinase inhibitor 2B (p15); ARF:alternative reading frame; hMLH1:human mutL homolog 1; 06-MGMT:0-6-methylguanine DNA methyltransferase; GST-Pi:glutathinone S-transferase P; BRCA1:breast cancer 1; DAP:death associated protein; VHL:vonHippel-Lindau tumour suppressor; APC:adenomatosis polyposis coli; LKB1:LKB subfamily protein kinase; TIMP-3:tissue inhibitor of metalloproteinase-3; THBS1:thrombospondin-1; ER- :estrogen receptor- ; RAR-beta:retinoic acid receptor-beta; SOCS-1:suppressor of cytokine signaling-1.

1.1.5 Methylation and cancer For more than 20 years, it has been known that promoter methylation pattern

in tumour cells is altered in comparison to normal cells (Feinberg and Vogelstein 1983). Different cellular pathways in cancer are reported to be disrupted by promoter methylation and associated gene silencing (Table I) (Herman and Baylin 2003).

Table I. Disrupted cellular pathways by promoter hypermethylation of genes in cancer. Pathway Genes Altered cell cycle control Rb, p16, p15, p14, p73 Repair of DNA damage MLH1, O6-MGMT, GST-Pi,

BRCA1 Apoptosis DAP kinase, caspase 8 Tumour-cell invasion or tumour architecture

E-cadherin, VHL, APC, LKB1, TIMP-3, THBS1

Growth-factor response ER, RAR-beta, SOCS-1

19

Interestingly, promoter hypermethylation in cancer is accompanied by genome-wide or global DNA hypomethylation (Feinberg and Tycko 2004).

A large number of tumours, including haematological tumours, have been reported to carry a high level of specific promoter methylation in tumour suppressor genes, which are normally unmethylated (Table II). Inactivation of these tumour suppressor genes presumably results in a growth advantage of the tumour cells. Therefore, it is essential to gain more knowledge of the role of specific methylation in tumour development to be able to use it for diagnosis and treatment.

1.1.6 Hypomethylating drugs It is obvious from the type of genes that are inactivated by promoter

methylation in tumours, that reactivation of these genes possible could have an anti-tumour effect. In experimental settings hypomethylating drugs, such as 5-azacytidine and 5-aza-2’-deoxycytidine (Decitabine), have been shown to cause removal of methylation from promoter CpG islands and thus activation of silenced genes (Silverman et al. 1993). These findings make the reversal of gene silencing an attractive target for therapy.

5-azacytidine and Decitabine has been shown to restore gene transcription in cell lines with hypermethylated genes (Daskalakis et al. 2002; Khan et al. 2006). Interestingly, both 5-azacytidine and Decitabine are not new drugs but was synthesized more than 40 years ago by Sorm and colleagues. Their early studies using high doses in acute leukaemia were less impressive (Karon et al. 1973; Steuber et al.1996). The relatively recent discovery of the hypomethylating properties of the drugs in low doses launched a new interest amongst researchers and clinicians. Both 5-azacytidine and Decitabine have showed promising results in MDS patients (Silverman et al. 2002; Kantarjian et al. 2006). The following summary focuses on 5-azacytidine.

Table II. Gene-promoter hypermethylation in cancer.

Gene Tumour type Function Reference VHL Renal clear cell Promotes angiogenesis (Herman et al. 1994) p16INK4A Lymphoma, solid

tumours Cyclin-dependent kinase inhibitor

(Herman et al. 1995)

p15INK4B MDS, AML, lymphoid leukaemia

Cyclin-dependent kinase inhibitor

(Herman et al. 1997; Uchida et al. 1997)

hMLH1 Colon, gastric and endometrial carcinoma

DNA mismatch repair (Esteller et al. 2001)

BRCA1 Breast and ovarian carcinoma

DNA damage repair (Esteller et al. 2001)

20

1.1.6.1 5-azacytidine and methylation 5-azacytidine has been shown to alter the course of MDS disease, especially

for high risk group of patients (Silverman et al. 2002). It also has been shown to prolong time to leukaemic transformation or death (Silverman et al. 2002). Moreover, it improves the quality of life in MDS patients (Kornblith et al. 2002). For these results, it was approved by the Food and Drug Administration (FDA) in 2006 for use in MDS (Kaminskas et al. 2005).

2.2. Programmed cell death Programmed cell death occurs during organ development and plays an

important role in cellular homeostasis (Jaattela 2004). The principal criterion of programmed cell death or active cell death that distinguishes from the accidental necrosis is the participation of active cellular processes. These cellular processes can be inhibited by interfering cellular signalling (Leist and Jaattela 2001). Three types of cell death are acknowledged at present: apoptosis (type I), autophagy (type II) and necrosis (type III) (Figure 5) (Lockshin and Zakeri 2004; Kim 2005).

2.2.1 Apoptosis Apoptosis is a process of programmed killing of a cell or cells in multicellular

organisms. It involves a series of coordinated biological and morphological events resulting in cell death, but without causing inflammation of surrounding tissue; which distinguishes it from necrosis. Apoptosis leads to a safe disposal of the dead cell with its organelles and fragments and is an essential process in embryonic development and organogenesis, as well as a natural endpoint in the life span of cells (Jacobson et

Cell deaths

Physiological or programmed cell death

Necrotic cell death (Type III)

Cytoplasmic type of degradation (Type IIIB)

Non-lysomal degradation(Type IIIA)

Autophagic cell death(Type II)

Apoptotic cell death(Type I)

Other cell death Ubiquination-proteasome

MMPs

Figure 5. Schematic presentation of different types of cell death.

21

al. 1997). Apoptosis is also involved in many pathological conditions, e.g. when a cell is damaged beyond repair by infection (virus, bacteria, etc), mutation, or injury (ionizing radiation, toxic drugs, etc).

2.2.1.1 Definition Apoptosis is an energy-dependent, asynchronous, genetically controlled

process by which unnecessary or damaged single cells self-destruct when the apoptosis genes are activated (Martin 1993; Earnshaw 1995). Apoptotic cells share a number of common features, such as cell shrinkage, membrane blebbing, nuclear condensation, chromatin condensation and cleavage, formation of pyknotic bodies, externalization of phosphatidylserine on the plasma membrane, break down of DNA into smaller fragments and phagocytosis by macrophages.

2.2.1.2 Clinical implications Deregulation of apoptosis has been reported to be associated with a number of

diseases and pathological conditions. A lower than normal apoptotic rate has been observed in the process of carcinogenesis, and in diseases like leukaemia and lymphoma, auto-immunity, viral infections and intimal hyperplasia of blood vessels (Kam and Ferch 2000). Excessive apoptosis has been observed in ischemic heart disease, stroke, neurodegenerative diseases, sepsis, bacillary dysentery (Shigella dysenteriae) and multiple organs dysfunction syndromes (Kam and Ferch 2000). In addition, increased apoptosis of haematopoietic progenitors is a hallmark of low-risk MDS, which will be further discussed.

Tumour development may be explained as an imbalance between cell division and cell death (Hanahan and Weinberg 2000). Defects in the apoptotic machinery or pathways are common findings in tumours leading to growth advantage of tumour cells and to progressive conversion of normal cells to tumour cells (Hanahan and Weinberg 2000).

2.2.1.3 The apoptotic process The apoptotic process starts with an appropriate stimulus, which can be either

extracellular or intracellular. The “initiation” and “decision” phases constitute the genetic check-point of cell death, during which a cell decides whether the injury or need for apoptosis is unavoidable or not. When a decision of apoptosis is taken, the cell enters the “execution’’ phase during which the central apoptotic events including proteolysis and mitochondrial inactivation take place. The process of apoptosis is controlled by a wide variety of cell signals. These signals can originate either extracellularly, through the extrinsic pathway, or intracellularly through the intrinsic pathway (Figure 6) (Hengartner 2000). Initiation of apoptosis following a ligand-receptor mediated reaction is considered as an extrinsic pathway (Schmitz et al.

22

2000). The intrinsic pathway involves mainly triggering, regulation and amplification of apoptotic signals by mitochondria, but also by endoplasmic reticulum, lysosomes and nucleus (Debatin et al. 2002).

2.2.1.3.1 The extrinsic pathway Binding of ligands to their death receptors, for example FasL to

Fas/CD95/Apo1, TRAIL to tumour necrosis factor (TNF) receptor, mediate cell death via the extrinsic pathway. These complexes recruit adaptor proteins containing death domain (DD), such as Fas-associated death domain (FADD) and TNF-receptor-associated death domain (TRADD). The adaptor proteins bind to procaspase-8, resulting in formation of death inducing signalling complex (DISC). This results in cleavage and activation of procaspase-8. Active caspase-8 is released from the complex as a heterotetramer of two small and two large subunits. It serves as an initiator caspase and activates downstream caspases resulting in activation of cellular death substrates, usually by cleaving them. Moreover, caspase-8 can cleave pro-apoptotic Bid to truncated Bid (tBid). Bid is involved in the regulation of intrinsic mitochondrial pathway of apoptosis. Therefore the extrinsic and intrinsic pathways are not independent, but rather interrelated pathways (Hengartner 2000; Krammer 2000).

Figure 6. Intrinsic (mitochondrial) & extrinsic (dealth receptor) apoptotic pathways.

23

2.2.1.3.2 The intrinsic pathway The exact mechanism of induction of the intrinsic pathway is still not fully

revealed but is considered to involve activation of the mitochondrial pathway followed by subsequent activation of downstream caspases. Numerous cell damage pathways converge on mitochondria and induce loss of mitochondrial membrane potential ( m), increased permeabilization of mitochondrial membrane, and release of mitochondrial proteins, such as cytochrome c, apoptosis inducing factor (AIF), second mitochondrial activator of caspases (Smac)/Direct inhibitors of apoptosis proteins (IAP) binding protein with low pl (DIABLO), endonuclease G and Omi/HtrA2 into cytosol (van Loo et al. 2002).

2.2.1.4 Caspases The caspases (cysteine-rich aspartate-specific proteases) comprise a large

family of cysteine-proteases. Caspases are synthesized as inactive proenzymes and located in different intracellular compartments. Activation of caspases is induced by cleavage of the proenzymes (Cohen 1997). Upon activation they get in contact with their substrates and initiate a specific function. Caspase targets involve Bcl-2 proteins, DNA repair proteins (e.g. PARP, DNA-PKcs), cytoskeleton components (e.g. actin, lamin), proteins involved in cell cycle control, cell adhesion, transcription factors, RNA synthesis and splicing, signal transduction, adaptor proteins, membrane receptors and cytokines. Apart from apoptosis, caspases are involved in maturation of pro-inflammatory cytokines (caspase-1, -4 and -5) (Fadeel et al. 2000), differentiation of skeletal muscle (caspase-3) (Fernando et al. 2002), terminal differentiation of keratinocytes (Weil et al. 1999), the lens fibre cells of the eye (Ishizaki et al. 1998), T-cell maturation and proliferation (Sarin et al. 1996; Kennedy et al. 1999) and haematopoiesis (Zermati et al. 2001; De Botton et al. 2002; Carlile et al. 2004).

In mammals, 14 members of the caspase family have been reported. Although over expression of all caspases resulted in cell death, not all of them are involved in apoptosis. Caspase-1, -4, -5, -13 and -14 have limited or yet unknown roles in the apoptotic machinery. Caspase-2, -8, -9 and -10 belong to the ‘initiator’ caspases. These enzymes are usually believed to be located upstream in the apoptotic pathways and are involved in the activation of other caspases. Caspase-3, -6 and -7 belong to the ‘effector’ caspases. They function downstream in the apoptotic pathways by cleaving proteins in the cytoskeleton, cytoplasm and nucleus. The degradation of numerous structural and regulatory proteins during apoptosis is important for morphological and biochemical changes, which can be detected in apoptotic cells like DNA fragmentation, chromatin condensation, membrane blebbing and cell shrinkage (Cohen 1997).

24

2.2.1.5 The Bcl-2 family proteins Bcl-2 protein family plays an important role in the regulatory system of

apoptosis. It consists of a large group of proteins, which have the ability to induce and augment, as well as inhibit apoptosis. Most Bcl-2 family members possess a hydrophobic C-terminal transmembrane domain (membrane anchor), which allows them to localize into membranes such as the mitochondrial membrane, the endoplasmic reticulum and the nuclear membrane (Krajewski et al. 1993; Lithgow et al. 1994). Functionally, Bcl-2 proteins can be divided into anti-apoptotic (Bcl-2, BclXL, Bcl-w, Mcl-1) and pro-apoptotic proteins (Bid, Bax, Bak, Bcl-XS, Bad, Puma, Noxa). The mechanisms regulating the function of Bcl-2 include transcriptional control of gene expression, post-translational events such as phosphorylation, proteolysis and induction of conformational changes (Fadeel et al. 1999). Bcl-2 proteins regulate mitochondrial protein and ion release by influencing the mitochondrial outer membrane permeability (Muchmore et al. 1996; Minn et al.1997; Schendel et al. 1997). Cleavage of Bid to truncated-Bid (tBid) is known to activate its pro-apoptotic function (Chou et al. 1999). The pro-apoptotic proteins, Bax and Bak, facilitate cytochrome c release from mitochondria and thereby play an important role in the intrinsic apoptotic pathway. Upon apoptosis induction, monomeric Bax oligomerizes (Tan et al. 1999) and translocates to the mitochondrial outer membrane (Wolter et al. 1997; Griffiths et al. 1999) to induce release of cytochrome c. Translocation to mitochondria is often needed for the Bcl-2 family proteins (tBid, Bax, Bak, Bcl-XL) to regulate apoptotic function (Fadeel et al. 1999). Transcriptional regulation is important for some pro-apoptotic members, for example Noxa and Puma can be induced by p53 (Villunger et al. 2003). Physical interaction and dimerization with other Bcl-2 family proteins is also a way to regulate their function. The anti-apoptotic members appear to function, at least in part, by antagonizing pro-apoptotic members, and vice versa. The ability to form homo- and hetero-dimers relates to their different function during apoptosis and the balance or ratio of anti- vs. pro-apoptotic proteins may determine the final outcome of a cell (Fadeel et al. 1999).

2.2.1.6 Mitochondria Fundamental metabolic processes, like oxidative phosphorylation of adenine,

the electron transport in the respiratory chain, the -oxidation of fatty acids, citric acid cycle (Krebs cycle) and parts of the urea cycle, occur in mitochondria. Therefore, it is not surprising when it was suggested that mitochondria have a central executioner and/or regulator function in cells undergoing apoptosis (Kroemer et al.1997; Green and Reed 1998; Mignotte and Vayssiere 1998). More and more evidence suggests that many apoptotic signalling pathways converge at the mitochondrial level, where signals are processed through series of molecular events resulting in release of

25

mitochondrial proteins or potent death factors, which can trigger both caspase-dependent and –independent apoptosis (Wang 2001).

Mitochondria are surrounded by a double membrane system. Mitochondrial regulation of apoptosis results from permeabilization of the outer membrane and release of intermembrane proteins into the cytosol. This process is regulated by pro-apoptotic Bcl-2 family members, such as Bax and Bak. Upon activation, these proteins translocate to mitochondria, insert into the outer membrane and form a large enough pore in the outer membrane to release cytochrome c (Eskes et al. 1998; Antonsson et al. 2000; Eskes et al. 2000). This process can also be regulated by mitochondrial permeability transition (MPT) pore complex (Gogvadze et al. 2001) which consists of the Voltage dependent anion channel (VDAC), Adenine nucleotide translocator (ANT) and mitochondrial cyclophilin D (Mayer and Oberbauer 2003). Activators of MPT pore are calcium, cytosolic alkalinization, oxidative stress, low mitochondrial transmembrane potential, high pyridine nucleotide and thiol oxidation (Crompton 2000). The anti-apoptotic proteins, such as Bcl-2 and Bcl-XL, can inhibit release mitochondrial of mitochondrial proteins and thereby impede apoptosis (Shimizu et al. 1999).

Interestingly, mitochondria-medicated apoptosis might be involved in cytotoxic induced DNA double-strand breaks and cell death (Krammer 1999). The link between DNA damage and mitochondria is not yet fully understood but may depend on the activation of caspase-2 within the nucleus and subsequently targeting of mitochondria and release of cytochrome c (Lassus et al. 2002; Robertson et al.2002; Zhivotovsky and Orrenius 2005).

2.2.1.7 Cytochrome c It has been reported that apoptosis mediated by mitochondria requires the

release of cytochrome c into the cytosol (Liu et al. 1996; Goldstein et al. 2000). Cytochrome c is crucial for formation of the apoptosome and subsequent initiation of apoptosis through the intrinsic pathway (Riedl and Salvesen 2007). Presence of cytochrome c and dATP/ATP promotes oligomerization of Apaf-1, which otherwise exists in a monomeric form, and thereby assembly of a wheel-shaped signalling platform known as the apoptosome (Acehan et al. 2002). This large protein complex then binds and activates procaspase-9 (Srinivasula et al. 1998), which in turn leads to activation of procaspase-3 and cell death (Zou et al. 1997).

2.2.1.8 Cytosolic acidification Cytosolic acidification has been linked with apoptosis, especially in drug-

induced apoptosis (Rosato et al. 2003; Ahmad et al. 2004; Kanno et al. 2004; Munoz-Gamez et al. 2005; Schmidt-Mende et al. 2006). This is further supported by the finding that inhibition of inhibition of cytosolic acidification reduces drug-induced

26

apoptosis (Pervaiz and Clement 2002; Kanno et al. 2004). Acidification of the cytosolic compartment has been shown to create an environment facilitating efficient death execution (Pervaiz and Clement 2002). The increase in cytosolic acidification is shown to be due to production of reactive oxygen species (ROS) within the cell. The link between ROS production and apoptosis is still unclear, but may involve intracellular ROS in some steps of the apoptotic execution (Clement et al. 1998; Simizu et al. 1998; Mansat-de Mas et al. 1999; Hirpara et al. 2001; Fleury et al.2002). Moreover, mitochondria have been suggested, either directly or indirectly, to be the prime source of ROS during drug-induced apoptosis (Quillet-Mary et al. 1997; Hirpara et al. 2001; Childs et al. 2002; Fleury et al. 2002). Even so it is unclear whether ROS generation is a critical initial trigger or a downstream effect of caspase-mediated mitochondrial damage.

2.2.1.9 Poly(ADP-ribose) polymerase (PARP) Poly(ADP-ribose) polymerase (PARP) mediates post-translational

modifications of proteins by poly(ADP-ribosyl)ation (Schreiber et al. 2006). So far 17 members of poly(ADP-ribose) polymerase (PARP) have been recognized. PARP-1 is the first protein to be discovered within the PARP family of proteins and also the most extensively studied of all PARP family proteins. PARP-1 is found in nuclei and is involved in cellular response to DNA damage and DNA metabolism (de Murcia et al. 1997). However with the discoveries of new members of the PARP family, it was found that they not only participate in the first line defence against genotoxins but also in the regulation of various cellular processes including genomic stability, transcriptional control, cell death and transformation (Schreiber et al. 2006; Miwa and Masutani 2007).

2.2.2 5-azacytidine and apoptosis 5-azacytidine is known to induce apoptosis (Ueno et al. 2002; Khan et al.

2006; Kiziltepe et al. 2007). However, the role of 5-azacytidine in apoptosis and cell death is not well studied, and yet largely unknown. This is largely explained by the difficult chemical properties of azacytidine leading to poor solubility and stability in vitro. Only few studies have shed light on the subject and even then with contradictory results. There is agreement that 5-azacytidine induce DNA strand breaks (Murakami et al. 1995; Kiziltepe et al. 2007; Morales-Ramirez et al. 2007), which eventually might lead to apoptosis. It has also been suggested that 5-azacytidine may demethylate and thus re-express silenced cell cycle control genes, hence inducing growth inhibition of tumour cells (Murakami et al. 1995; Ueno et al.2002; Khan et al. 2006). However, it has been proposed that 5-azacytidine not only effects DNA (Li et al. 1970; Snyder and Lachmann 1989) but also RNA (Lu and Randerath 1979; Cohen and Glazer 1985) metabolism. In fact, Murakami T et al

27

suggested a preference for RNA at low concentration of 5-azacytidine (Murakami et al. 1995).

Although many studies have suggested 5-azacytidine-induced cell cycle arrest, there are disputes regarding which steps in the cell cycle that are affected. While G0/G1 phase involvement was shown by Ueno et al. (Ueno et al. 2002), Wang et al. showed involvement also of or only of the G2 phase (Wang et al. 1998), and Murakami et al suggested involvement of both S and G2 phase. However, one of the studies included this thesis showed involvement of all G0/G1, S and G2 phases at different doses of 5-azacytidine (Khan et al. 2006).

Additionally, Kilzitepe T et al. (Kiziltepe et al. 2007) has shown that 5-azacytidine-induced apoptosis is both caspase-dependent and independent, at least in multiple myeloma cell lines. They reported elevation of caspase-8 and -9, which could be blocked only partially by a pan-caspase inhibitor, z-VAD-fmk. Moreover, mRNA level of Fas was found to be elevated in 5-azacytidine-induced apoptosis further supporting the activation of the caspase-8 pathway The same study also reported that 5-azacytidine-induced apoptosis is independent of p53 function; however Ueno et al. (Ueno et al. 2002) showed elevation of p53 expression prior to apoptosis in 5-azacytidine-treated cells. Kilzitepe T et al. also suggested that proteins involved in the response to DNA damage (e.g. PARP, Chk-2, ATR, etc) play a role in 5-azacytidine-induced apoptosis. Among Bcl-2 family proteins, a significant increase in the pro-apoptotic proteins (Bax, Puma and Noxa) were evident following azacytidine treatment, while no changes were seen in anti-apoptotic proteins (Bcl-2 and Bcl-XL). In addition 5-azacytidine caused cleavage of the anti-apoptotic protein, Mcl-1 and release of pro-apoptotic proteins, AIF and EndoG into the cytosol. Again, further studies are necessary to support the findings of 5-azacytidine-induced cytotoxicity.

2.3. Myelodysplastic Syndromes Myelodysplastic syndromes (MDS) constitute a heterogeneous group of

neoplastic clonal bone marrow stem cell disorders. Clinically, MDS is characterized by bone marrow failure and a propensity for development towards AML. The bone marrow failure results in ineffective haematopoiesis and subsequent cytopenia, which may range from mild to severe. While some patients present with only mild anaemia, thrombocytopenia or neutropenia, severe pancytopenia is present in other cases. Morphologically, MDS is characterized by the presence of dysplastic morphological changes in one or more cell line in the bone marrow.

2.3.1 The historical perspective Initially, MDS was considered to be a form of anaemia. In 1907, ‘anemia

pseudoaplastica’ was the first term used to describe MDS (Luzzatto 1907). In 1938,

28

100 anaemic cases were defined as refractory anaemia (RA) (Rhoades and Barker 1938). After summarizing these and other cases, Hamilton-Paterson (Hamilton-Paterson 1949) recognized a group of patients with refractory anaemia, who progressed to acute leukaemia, and introduced the term ‘pre-leukaemic anaemia’. In 1953, Block et al. (Block et al. 1953) expanded the definition to include also multilineage cytopenias and further supported the findings of Hamilton-Paterson (Hamilton-Paterson 1949). Later, Björkman (Bjorkman 1956) described the clinical feature of refractory anaemia with ringed sideroblasts. In 1973, first Saarni and Linman (Saarni and Linman 1973) and later on Linman and Bagby (Linman and Bagby 1976) developed the first criteria for the ‘preleukaemic syndrome’. They suggested this syndrome to be a primary bone marrow disease characterized by peripheral cytopenias, marrow hypercellularity, disordered precursor maturation and ultimate transformation to AML. In 1976, Bennett and his colleagues (Bennett et al.1976) proposed the first French-American-British (FAB) classification for AML and MDS (dysmyelopoietic syndromes). This FAB classification was revised for MDS by the same group in 1982 (Bennett et al. 1982). In 1993, Morel et al. (Morel et al.1993) reported that the cytogenetics analysis has a strong independent prognostic value in de novo MDS. The International Prognostic Scoring System (IPSS) was introduced in 1997 following a review of 816 cases of MDS (Greenberg et al. 1997).

2.3.2 Epidemiology Although MDS appear in all age groups, it predominantly affects elderly

individuals with a median age of 70-75 years at diagnosis (Mufti 2004). Depending on the propensity for bone marrow investigation in case of mild cytopenia in elderly individuals, the incidence of MDS varies between countries, but is estimated to 4-5/100000/year (Germing et al. 2007). The incidence increases with age (Aul et al.2001). Due to a longer life expectancy in the Western world, increased awareness of MDS, an increased use of chemotherapy and environmental influence; a raise in the incidence of MDS over time has been suggested, however not confirmed (Aul et al.1998b). The incidence of MDS is slightly higher in men than in women (Aul et al.1998a), while chronic myelomonocytic leukaemia (CMML) shows a male predominance (Ribera et al. 1987).

While approximately 90% of MDS cases are of unknown aetiology, 10% of patients present with a history of previous exposure to chemotherapy and/or irradiation (Rosenfeld and List 2000; Hofmann et al. 2004). Rare cases also present with a family history of myelodysplastic or leukaemic disorders. Differential diagnoses, such as megaloblastic anaemia due to vitamin deficiency, other haematological malignancies, chronic liver disease, hypothyroidism, Human Immunodeficiency Virus (HIV) infection, cytotoxic treatment, as well as intoxication with heavy metals and consistent exposure to benzene should be ruled out.

29

2.3.3 Clinical and morphological diagnosis A proportion of patients with MDS is asymptomatic at the time of initial

diagnosis and may be diagnosed after routine blood sampling. The diagnostic process is often initiated by a finding of unexplained cytopenia. The clinical symptoms of MDS are mostly caused by mild to severe cytopenias and subsequent complications. Anaemia is the most common cytopenia, about 80% of patients present with a haemoglobin concentration below 10g/dl, and transfusion need develops in a majority of these patients. Anaemia leads to weakness, fatigue, reduced quality of life and dyspnoea (Hellstrom-Lindberg et al. 2003). Around 25-30% of individuals of MDS present with leukopenia and neutropenia at the time of diagnosis, which may lead to recurrent infections. In case of thrombocytopenia, patients may have bleeding symptoms, but fewer than 10% present initially with serious bleeding (Hofmann and Koeffler 2005).

A diagnosis of MDS is based on the finding of dysplastic features of bone marrow precursors in combination with peripheral cytopenia. Mild marrow dysplasia can sometimes be seen even in normal individuals (Bain 1996), therefore, dysplastic features should be present in a significant proportion (>10%) of marrow cells to diagnose MDS. At least 200 marrow cells and 20 megakaryocytes should be evaluated (Bowen et al. 2003).

Table III. The WHO classification of MDS. Category Refractory anaemia Refractory anaemia with ring sideroblasts Refractory cytopenia with multilineage dysplasia Refractory anaemia with multilineage dysplasia and ring sideroblasts Refractory anaemia with excess blasts type I and II 5q syndrome Myelodysplastic syndrome - unclassified

2.3.4 Classifications The FAB classification divided MDS into five different subtypes: refractory

anaemia (RA), RA with ringed sideroblasts (RARS), RA with excess of blasts (RAEB), chronic myelomonocytic leukaemia (CMML) and RAEB in transformation (RAEB-t) (Bennett et al. 1982). This classification was mainly based on morphological features in blood and bone marrow, and percentage of myeloblasts and ringed sideroblasts in the bone marrow. The WHO classification, published 2001, incorporated information about the importance of dysplasia, detailed blast counts, and cytogenetics (Vardiman et al. 2002). WHO divides MDS into 8 different groups

30

(Table III). The major changes included the creation of additional categories (e.g., 5q- syndrome), distinction of unilineage from multilineage dysplasias in the refractory anaemias, and subdivision of the refractory anaemias with excess blasts into two categories based on bone marrow blast percentage.

2.3.5 Prognosis MDS is a heterogeneous disease, and it is important to predict the outcome of

an individual patient in order to plan for management and treatment. The overall survival of patients can differ from several years to a few months. Death is usually a consequence of complication to bone marrow failure and cytopenia, or caused by transformation to AML. Over the last 20 years different attempts have been made to estimate the prognosis for individual patients taking into account the number of cytopenias, the percentage of myeloblasts in bone marrow or blood, type of cytogenetics abnormalities, dysplastic features in non-erythroid cells, and serum lactate dehydrogenase (LDH) level (in particular for CMML patients).

Table IV. International Prognostic Scoring System (IPSS) for untreated MDS (Greenberg et al., 1997).

Score value Prognostic variables 0 0.5 1.0 1.5 2.0

BM blasts (%) <5 5-10 - 10-20 21-30 Karyotype* Good Intermediate Poor Cytopenias 0/1 2/3

*Good: N, 5q-, 20q-, -Y; Poor: -7, 7q-, complex ( 3 abnormalities); Intermediate: other

Risk groups Low Score 0 Intermediate-1 Score 0.5-1.0 Intermediate-2 Score 1.0-1.5 High Score 1.5-2.0

Figure 7. Prognosis of survival and evolution to AML for MDS patients depending on IPSS. ("Originally published in Blood. Greenberg et al.1997;89. © the American Society of Hematology").

31

Currently, the most reliable prognostic tool is considered to be the International Prognostic Scoring System (IPSS) (Greenberg et al. 1997), which was developed in close collaboration with groups involved in other MDS scoring systems. IPSS divides MDS into four different risk groups: Low, INT-1, INT-2 and High (Table IV). A significant difference regarding overall survival and propensity for leukaemic transformation is evident between these different risk groups and hence treatment strategies for these groups also differ. This scoring system takes into account the blast percentage, karyotype and number of cytopenias The IPSS is widely accepted and currently the most accurate prognostic assessment tool. However, as the new information regarding the molecular pathogenesis of MDS will be discovered, these classifications and scoring systems will continue to evolve with time. Recently, the WPSS (WHO classification-based prognostic scoring system) scoring system, including also the presence or absence of transfusion need was published (Malcovatiet al. 2007). The WPSS system now needs to be evaluated in independent patient cohorts.

2.3.6 Low-risk and high-risk MDS Even though IPSS divides MDS in four different risk groups, many

investigators and clinicians tend to speak about “low-risk” and “high-risk” MDS. ‘Low-risk MDS’ encompass the IPSS Low and INT-1 risk groups (Low/INT-1), whereas ‘high-risk MDS’ includes the IPSS INT-2 and High-risk groups (INT-2/High). The median survival in low-risk and high-risk MDS is 5.7-3.5 and 1.2-0.4 years, respectively, as seen in Figure 7 (Greenberg et al. 1997). The distinct difference between the low- and high-risk MDS in terms of overall survival, quality of life, evolution to AML, pathologic features of the malignant clone and bone marrow, is relevant since it affects the choice of therapy. As expected, treatment strategies vary considerable between these two groups. The most relevant therapeutic goals for patients with low-risk MDS are to improve erythroid output and reduce cytopenia in general, to reduce MDS-related morbidity, and to maintain or improve quality of life. For patients with high-risk MDS patients, treatment aims at improving survival or, if possible, at cure. Here, more intense therapies, such as anti-leukaemic induction chemotherapy and/or bone marrow stem cell transplantation, are considered. Considering the advanced age of MDS patients, only one third of these high-risk patients qualify for intensive therapy (Bowen et al. 2003; Fenaux 2005; Hofmann and Koeffler 2005), and even if a complete remission (CR) is achieved, the incidence of relapse is close to 100% and overall survival is short (Wattel et al. 1997; Beran 2000). Hence, new therapeutic approaches both for inducing and maintaining remission are warranted. Supportive care may be the only choice of treatment for older patients with a poor performance status.

32

2.3.7 Pathogenesis of MDS MDS is sometimes viewed as a two-phase disorder: early-phase (low-risk

MDS) and late-phase (high-risk MDS) (Fenaux 2005). Deregulation of tumour suppressor genes, DNA repair genes and enzymes, and immunological abnormalities; have been suggested as causes for development towards MDS and further evolution to AML. MDS is a clonal disorder. In a stepwise progression of disease, normal stem cells must acquire an initial genetic and/or epigenetic insult to develop early-phase MDS clones. Then, additional genetic and/or epigenetic insults are required to develop late-phase MDS or AML.

The pathogenesis of early-phase MDS comprises certain chromosomal rearrangements, activation of apoptotic pathways, increased medullary angiogenesis and immune mechanisms, all of which may be a consequence of an initial genetic insult. Indeed, increased apoptosis of bone marrow progenitor cells is a hallmark of early-phase MDS. In addition, accumulation of genetic insults in the early-phase MDS clones may provide a proliferative advantage and a capacity to avoid apoptosis, and eventually lead to late-phase MDS and progression to AML. Evidently, reduced apoptosis and arrested differentiation of myeloid progenitors are characteristic features of late-phase MDS resulting in blast accumulation in bone marrow, pancytopenia and eventually leukaemic transformation.

The events leading to disease progression of MDS and AML evolution are incompletely understood. Several mechanisms are considered to be part of a stepwise disease progression. Firstly, reduced expression of p15INK4B (a cyclin dependent kinase inhibitor gene) by DNA promoter hypermethylation is much more frequent in patients with more 10% bone marrow blast (INT-2/High) compared with patients with less than 10% bone marrow blast (Low/INT-1) (Quesnel et al. 1998). Secondly, when reduced expression of p15INK4B, the gene Hypermethylated in Cancer-1 (HIC1),E-cadherin (CDH1) and estrogen receptor (ER) by DNA promoter hypermethylation were considered, a coexisting hypermethylation of 3 genes was more frequent in late-phase MDS compared with early-phase MDS (Aggerholm et al. 2006). Thirdly, activation of the ras gene (important for cell proliferation) by point mutations has been suggested to be associated with poor outcome, and is found in 10-15% of patients with MDS (Hirai et al. 1988; Hirai 2003). Fourthly, the internal tandem duplication of Flt3 gene, which play role in cell survival, proliferation and differentiation, is found in 5% MDS and is suggested to be linked to poor prognosis (Horiike et al. 1997; Yokota et al. 1997). Fifthly, inactivation of the tumour suppressor gene p53 by mutation or deletion is detected in 5-10% of MDS and is frequent in particular in secondary MDS and MDS with 17p deletion, and is suggested to play a role in leukaemic progression (Sugimoto et al. 1993).

Increasing evidence suggests a difference in the pathogenesis between different subgroup of MDS. More detailed information is required to complete the

33

puzzle, and only then newer treatment regimens could be developed to target the distinct pathologic mechanisms. Pathobiology-based therapies may be essential to increase the response rate – and cure rate - in high-risk MDS.

2.3.7.1 DNA Methylation in MDS Gene inactivation by promoter methylation is suggested to be critical in

myelodysplastic syndromes (MDS) and acute myeloid leukaemia (AML) as in many other haematological malignancies (Table IV). In MDS, p15INK4B hypermethylation occurs in around 50% of MDS patients and is associated with shorter survival, as well as disease and leukaemic progression (Uchida et al. 1997; Quesnel et al. 1998; Tienet al. 2001; Aggerholm et al. 2006). Moreover, recent reports suggested that promoter methylation of p15INK4B and other genes constitute an independent negative prognostic factor for progression free survival and overall survival in MDS (Shen et al. 2005; Aggerholm et al. 2006). Hence, promoter methylation of genes is critical for MDS and should therefore be included in the list as target to treat MDS. Therefore, methylation of more than a single gene or more appropriately changes in global methylation might be considered while evaluating the methylation pattern in tumours. Unfortunately, the lack of sensitive and accurate methods to detect global methylation has restricted the methylation findings to be used in global context.

2.3.7.2 Apoptosis in MDS Increased apoptosis of bone marrow progenitor cells is a hallmark of MDS

(Parker and Mufti 2000; Mundle 2001; Parker and Mufti 2004). It is generally believed that the regulatory and/or pathophysiological pathway of apoptosis is deregulated leading to cytopenias. Increased cell death in bone marrow cells also can explain the typical finding in MDS of peripheral cytopenias regardless of hypercellular bone marrow. The cause for the increased apoptosis of myelodysplastic bone morrow progenitor cells is incompletely understood. Moreover, due to a lack of

Table V. Gene-promoter hypermethylation in MDS and acute leukaemia. Gene Chromosome Function MDS (%) AML (%)

p15INK4B 9p21 Cyclin-dependent kinase inhibitor 0-59 30-90

HIC-1 17p13 Putative tumour suppressor gene 10

E-Cadherin 16q22 Ca2 dependent intracellular adhesion N.K. 32-78

ER- 6q25 Estrogen receptor N.K. 70-90 CALC1 11p15 Ca2 bone reabsorption 25-60 50-90

p16INK41 9p21 Cyclin-dependent kinase inhibitor 0 0

34

reliable marker for clonality in many cases of MDS, it is still not possible to detect whether apoptosis affect mainly clonal cells or also normal haematopoietic progenitor cells.

2.3.7.2.1 The extrinsic pathway Death receptors (Fas, TNF- , TRAIL-R), as well as adapter Fas-associated

death domain (FADD) are reported to be over-expressed and to trigger apoptosis in low-risk or ‘early-stage’ MDS (Fontenay-Roupie et al. 1999; Mundle et al. 1999a; Mundle et al. 1999b; Zang et al. 2001; Claessens et al. 2002). One study reported that Fas-mediated apoptosis is important in MDS with trisomy 8, but not in MDS with other cytogenetics abnormalities (Sloand et al. 2002). However, attempts to block the extrinsic pathway gave contradictory results; one study showed that blocking of apoptosis by inhibition of the FADD function led to decreased caspase activity and reduced cell death in early-stage MDS (Claessens et al. 2005); whereas another study could not inhibit apoptosis using antibodies against Fas-receptor in RARS patients (Hellstrom-Lindberg et al. 2001). Such discrepancies may in fact be explained by differences within the subgroups of MDS patients.

Changes in the level of cytokines, such as IL-3, IL-6, IFN- , TGF- and TNF-, have been shown to be elevated in serum of MDS patients (Rosenfeld and List

2000). Furthermore, inhibition of TNF- has been shown to reduce apoptosis and increase haematopoietic colony growth, but has failed to be effective in the clinical setting (Gersuk et al. 1998; Raza 2000).

2.3.7.2.2 The intrinsic pathway The intrinsic pathway is suggested to be critical for the apoptosis in MDS

(Span et al. 2005). Important players in the intrinsic pathway, like mitochondria and Bcl-2 family members, have been shown to be responsible for initiating and amplifying the apoptotic process. In RARS, the characteristic ringed sideroblasts are due to excess iron bound to aberrant mitochondrial ferritin (MtF), unlike in normal erythropoiesis (Cazzola et al. 2003). It is unclear whether the iron accumulation disturbs or results from the mitochondrial dysfunction. Moreover, mitochondrial DNA mutations (Gattermann 2000) or mutations of nuclear encoded mitochondrial proteins (Craven et al. 2005) have been reported in MDS.

The ratios between pro- and anti-apoptotic Bcl-2 family member proteins are proposed to differ between low-risk and high-risk MDS, and MDS followed by AML. Leukaemic progression is shown to associate with an increased relative expression of the anti-apoptotic family of Bcl-2 members (Parker et al. 1998; Parker and Mufti 2000), and an altered c-myc:Bcl-2 ratio (Rajapaksa et al. 1996).

35

2.3.7.3 The paradox in MDS Genetic or epigenetic alterations in cancer are usually believed to be

associated with gain of survival function (anti-apoptotic) and/or growth advantage (expansion or proliferation) over normal clone. In contrast to that, MDS progenitors show an excess in apoptosis and yet a growth advantage. Hence, the excess of apoptosis at the progenitor level must somehow be linked to a gain of function either in survival or growth at the stem cell level. Moreover, an increase ratio between cell death and proliferation should logically give rise to elimination of progenitor cells and lead to a development of hypocellular bone marrow; which is contradictory to the findings of the usually hypercellular bone marrow in MDS. One hypothesis to explain this dilemma is that the clonal alterations that offer growth advantage for proliferating stem cells, upon initiation of the differentiation programme, leads to inhibition of maturation and excessive cell death.

2.3.8 Treatment A variety of different therapeutic approaches are employed to treat different

subcategories of the disease and a thorough and precise diagnosis is critical for a successful treatment. Even then, selecting the best treatment option could be a challenge. An individual assessment of the patients should be made along with a decision whether to make an attempt to cure the patients, or to aim at improvement of symptoms, improved quality of life and hopefully overall survival. The choice of treatment varies considerably between low-risk and high-risk MDS.

2.3.8.1 Supportive care Supportive care in MDS includes transfusion therapy and treatment and

prevention of infections and bleeding. It also includes regular follow-up and observation of progression of disease. As progressive and symptomatic anaemia is a frequent finding in MDS, transfusions with packed red blood cell are often required (Sanz et al. 1989). However, prolonged transfusion therapy will eventually lead to iron overload, which might necessitate iron chelation therapy.

2.3.8.2 Stem cell transplantation Allogenic stem cell transplantation (SCT) is currently the only curative

therapeutic approach for MDS, and should be considered in patients with high-risk MDS and age <60-65 years. A significant risk of treatment related mortality and morbidity is associated with this treatment. Treatment outcome various between patients and depends on age, morphology and cytogenetics. A better outcome is observed in patients with young age and with shorter disease duration (Appelbaum and Anderson 1998). Patients with less than 10% blast, favourable cytogenetics and primary MDS respond better to SCT (Runde et al. 1998). However, the cure rate in

36

unselected patient cohorts is still less than 50%, primarily due to a high relapse rate, and to high treatment-related mortality (De Witte et al. 1997). Disease-free survival is even lower in high-risk MDS, often below 10-20%. Hence, a very important goal is to find strategies to improve the outcome of SCT in high-risk MDS.

Reduced intensity conditioning (RIC) regimen followed by stem cell transplantation, is another strategy assessed in several clinical trials (Luger and Sacks 2002; Chan et al. 2003; Taussig et al. 2003). As expected, the treatment-related mortality and morbidity was much less than for the conventional transplant regimen. However, disease free survival (DFS) and overall survival are comparable, due to a higher relapse rate. Nevertheless, taking into account that patients selected for this treatment were older and not suitable for conventional allogenic transplantation, this regimen could prove to be helpful at least in a subset of MDS patients.

Autologous transplantation (APSCT) in MDS has been reported to lead to durable remissions in a subset of MDS patients (De Witte et al. 1997; Wattel et al.1999). However it has not clearly been demonstrated that APSCT is better than conventional high-dose chemotherapy, this treatment may be an option for younger patients with normal cytogenetics and without an appropriate donor.

2.3.8.3 Chemotherapy In both MDS and AML, the abnormal clone has usually a survival and growth

advantage over the normal precursors. Currently the mechanism of suppression of normal haematopoiesis is incompletely understood. Chemotherapy aims at eradicating the abnormal clone in bone marrow. In MDS compared to de novo AML, however, recovery of normal haematopoiesis is less probable due to the underlying marrow failure, which may result in prolonged pancytopenia after cytotoxic therapy.

2.3.8.3.1 High-dose or intensive chemotherapy Intensive chemotherapy may be recommended in high-risk MDS (INT-

2/High) with absence of multiple independent adverse risk factors, such as high age, complex karyotype, poor performance status, and long disease duration. (Bowen et al.2003). The most commonly used AML-type regimens are combinations of cytosine arabinoside (ara-C) with any of the anthracyclines. Poor long-term survival, high relapse rate, prolonged drug-induced aplasia and functional drug resistance are the major hindrances to overcome for this treatment strategy (Estey et al. 1997; Wattel et al. 1997; Beran 2000; Ribrag et al. 2003).

2.3.8.3.2 Low-dose or non-intensive chemotherapy Low-dose chemotherapy with ara-C has been reported to improve peripheral

blood values and reduce blast counts in around 30% of MDS, but no studies have

37

shown that this treatment improves overall survival rate (Hellstrom-Lindberg et al.1992; Miller et al. 1992).

2.3.8.3.2.1 Hypomethylating therapy

Recently, several studies have reported a promising effect of low-dose treatment with the hypomethylating drugs 5-azacytidine (Vidaza®) and decitabine (Dacogen®) in MDS. Both 5-azacytidine and decitabine has been shown to prolong time to leukaemic transformation and death (Silverman et al. 2002; Kantarjian et al.2006). A randomized controlled trial comparing 5-azacytidine with supportive care was reported by Silverman et al. (Silverman et al. 2002). Haematological improvement was observed in 5% of control patients, compared to 60% in 5-azacytidine-treated patients (Cheson et al. 2006). Complete and partial response rates were observed in 7% and 16% of the treated patients, respectively. A significant increase in time to death or AML transformation and improved quality of life was seen in 5-azacytidine-treated patients, compared to patients receiving supportive care only, but no significant overall survival advantage was observed in this study (p=0.1). Based on these data, 5-azacytidine was the first drug to be approved by the U.S. Food and Drug Administration (FDA) for the treatment of high-risk MDS. Decitabine was also recently approved by FDA for treatment of MDS based on a randomized study (Kantarjian et al. 2006). Similar to 5-azacytidine, time to death or AML transformation and quality of life was improved in the treated group compared to patients receiving only supportive care, but no significant difference in median survival was observed between these two groups (p=0.6). The schedules for administration of decitabine were more cumbersome and patients were often hospitalized during treatment. For both 5-azacytidine and decitabine, repeated treatment courses were required before a response could be confidently evaluated. The European Agency for the Evaluation of Medical Products (EMEA) required a confirmatory trial, which recently has been preliminary reported. The results from this large randomized study show that azacytidine significantly improves survival compared to conventional treatment consisting of supportive care, low-dose ara-C or induction chemotherapy (0<0.0001) (Fenaux et al. 2007).

Although one can argue that the overall response rate is not very impressive, around 17-23%, these drugs are emerging as the best treatments options for the high-risk MDS patients who are not eligible for bone marrow transplantation. However, it is likely that optimal doses and dosing schedules for azacytidine have not yet been discovered, which is one of the reasons for the present thesis work (Kantarjian et al.2007).

38

2.3.8.3.2.2 Mechanisms of hypomethylating drugs in MDS:

Whether the clinical effect of 5-azacytidine is due to their DNA hypomethylating effects or to cytotoxic effects is still not apparent. The decreased response rate of high-risk MDS to conventional chemotherapy has in part been suggested to be due to increased expression of drug resistance genes, such as MDR1 (multidrug resistance 1), MRP1 (multi resistance-associated protein 1) and LRP (lung resistance-associated protein) (Lepelley et al. 1994; Wattel et al. 1998; Poulain et al.2000; Kurata et al. 2006). It is important to distinguish the role of epigenetic inactivation in the attainment of drug resistance, and methylation has been suggested to be an important event in MDR development during chemotherapy (Glasspool et al.2006; Vidal et al. 2007). Hence, it is theoretically possible that hypomethylating drugs can suppress multidrug resistance phenotype and become a useful tool for the treatment of cancer, including MDS. Understanding of these mechanisms is essential in order to learn how to combine azacytidine and decitabine with other therapies, hence improving treatment strategies.

2.3.8.4 Haematopoietic growth factors Patients with RA, RARS and RAEB-1 respond poorly to chemotherapy. The

myeloid growth factors granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) may increase peripheral neutrophil count, and erythropoietin (EPO) may improve haemoglobin levels in low-risk MDS (Negrin et al. 1990; Hellstrom-Lindberg et al. 1995; Yoshida et al. 1995; Negrin et al.1996). Around 20-30 % of patients with low-risk MDS and anaemia respond to erythropoietin. It is well documented that RA patients respond better than RARS to monotherapy with EPO (21% vs.7.5%) (Hellstrom-Lindberg 1995; Hellstrom-Lindberg et al. 1998). The respond rate to EPO may be improved by the addition of G-CSF, which acts synergistically with EPO on dysplastic erythropoiesis, in particular in RARS (Tehranchi et al. 2003; Howe et al. 2004; Jadersten et al. 2005; Tehranchi et al.2005). Importantly, treatment with growth factors did not increase the risk for leukaemic transformation (Jadersten et al. 2005). Thrombocytopenia in MDS may be treated conservatively, with platelet transfusions in case of bleeding. Recently, thrombopoietic growth factors have emerged, which may add to the therapeutic arsenal in the future (Rice 2006).

2.3.8.5 Immunosuppressive therapies Immunosuppressive treatment with antithymocyte globin (ATG) and

cyclosporine-A (CSA) can induce prolonged response duration in selected patients (Molldrem et al. 1997; Catalano et al. 2000; Molldrem et al. 2002; Broliden et al.2006). Treatment with ATG in younger patients with low-risk MDS may induce

39

responses in 44-64% of patients, while responses in elderly patients are far less common. A recent studies within the Nordic MDS group indicated that patients with higher risk features should not be treated with ATG, since there was an increased risk for AML transformation after treatment (Broliden et al. 2006). Treatment with CSA in low-risk MDS, particularly RA, has been shown to improve in haemoglobin and/or platelet level in half of the patients (Catalano et al. 2000).

2.3.8.5 Novel therapies Lenalidomide (Revlimid®), an orally administered thalidomide analogue, is

an immunomodulating drug recently approved by FDA for the treatment of MDS. The results of the two large phase II trials have been published (List et al. 2006a; Listet al. 2006b) and a randomized phase III trial has recently been closed. Lenalidomide has been shown to be very effective to promote erythropoiesis and induce cytogenetic remission in transfusion-dependent low-risk MDS with del(5q) (5q- syndrome). Two-thirds of the patients become transfusion-independent and the overall erythroid response was approximately 75%. The effect of lenalidomide in non-5q low-risk MDS is less impressive and needs further evaluation (Raza et al. 2007).

Arsenic oxide (ATO) has been shown to induce tri-lineage haematopoietic response in MDS patients (List et al. 2003). In two parallel phase II studies, haematological improvement including sustainable transfusion independency was observed in approximately 30% of patients (Vey 2004). The drug was reported to be well tolerated treatment also in elderly patients.

40

3. AIM OF THE THESIS

The aim of the thesis was to investigate the role of DNA methylation and the therapeutic substance 5-azacytidine in MDS. The specific aims leading to a step-by-step increase in the knowledge and understanding of in particular high-risk MDS were:

I. The study investigated the pharmacodynamics of 5-azacytidine in myeloid cells with the aim of developing better treatment schedules.

II. The study aimed at a deeper understanding of the molecular mechanisms behind 5-azacytidine-induced cytotoxicity, by studying the association between drug-induced apoptosis and DNA methylation status in myeloid cell lines.

III. The study assessed the prognostic role of global and gene specific methylation (p15INK4B) in relation to conventional risk factors in patients with MDS. Moreover, a longitudinal study was performed.

IV. The aim of the study was to assess gene specific promoter DNA methylation status in high-risk MDS and MDS-AML patients and to investigate its role for the outcome of induction chemotherapy

41

4. MATERIALS AND METHODS All information about manufacturers and companies can be found in the

respective papers.

4.1 Sample selections, cell culture and incubation techniques 4.1.1 Patient (Paper III & IV)

Paper III encompassed consecutive patients with high-risk MDS and MDS-AML included in a prospective multicenter clinical within the Nordic MDS Group. Paper IV included consecutive patients included in the MDS Biobank at Karolinska University Hospital Huddinge. For details, please see the respective papers.

4.1.2 Bone marrow sampling (Paper III-IV) MNC were isolated from bone marrow by density gradient technique through

Lymphoprep. If MNC yield was sufficient, CD34 positive cells were purified by magnetic cell sorting using the CD34 Progenitor Cell Isolation Kit according to the manufacturer’s guidelines. For future methylation analyses, the cells were stored as pellet in -80ºC.

4.1.3 Cell line and cell culture (Paper I-II) P39, HL60 and Jurkat cell lines were used in the studies. Cell line experiments

were principally performed with the P39 myeloid cell line, originally derived from a patient with MDS-CMML. Cells were cultured in RPMI 1640 medium with Glutamax-1 supplemented with 10% heat inactivated foetal bovine serum at 37ºC in a humidified 5% CO2 in air. In paper I, penicillin (25 IU/ml) and streptomycin (25 µg/ml) were added in culture medium. Prior to adding azacytidine, cells were cultured for at least 12 hours to reach exponential growth phase.

4.1.4 Drug dilution, Dose and Dosage (Paper I-II) Azacytidine was diluted with sterile H2O and filtered through sterile Star

plastic syringe filter (0.22 m, non-pyrogenic filter) just before adding to the positions. Cells were incubated with final concentrations of 0.1, 0.25, 0.5, 0.75, 1, 2, 3, 4 and 8 M for 1, 2, 4, 8, 12, 24, 48 and 72 hours. In the pharmacodynamic studies (Paper I), cells were washed with PBS every 24 hours during continuous drug exposure, and drug was re-added at the same concentration; in contrast the drug was added once only at the beginning of the experiments in paper II. Moreover, cytosine arabinoside (ara-C) was also diluted in sterile water and used at final concentrations of 0.08, 0.5, 1 and 2 M.

42

4.1.5 Incubation with inhibitors (Paper II) Caspase inhibitors Z-VDVAD-fmk (caspase-2), Z-DEVD-fmk (caspase-3-

like), Z-IETD-fmk (caspase-8), Z-LEHD-fmk (caspase-9) and Z-VAD-fmk (pan-caspase inhibitor) were added at final concentration of 10-40 µM. PARP inhibitors, nicotinamide and 3-aminobenzamide were used in 1 and 10 mM, respectively. The ionophore monensin, an inhibitor of cytosolic acidification, was used at a concentration of 20, 50 and 100 nM. All inhibitors were added 1 hour prior to adding azacytidine.

4.2 Methylation assays 4.2.1 Measurement of global DNA methylation by LUMA (Paper II-III)

Genomic DNA was extracted using GeneElute DNA extraction kit. Restriction enzymes (HpaII, MspI, and EcoRI), were purchased from New England Biolabs. PSQ™ 96 SNP reagents for pyrosequencing were purchased from Biotage. LUMA was run as described elsewhere in detail (Karimi et al. 2006). Briefly, genomic DNA (200-500 ng) was cleaved with HpaII+EcoRI and MspI+EcoRI in two separate reactions in 96-well Pyrosequencing plates. The digestion reactions were run in the PSQ96™MA system. Peak heights were calculated using the PSQ96™MA software. The HpaII/EcoRI and MspI/EcoRI ratios were calculated as (dGTP+dCTP)/dATP for the respective reactions. The DNA methylation assay was defined as HpaII/MspI ratio or more appropriately by (HpaII/EcoRI)/(MspI/EcoRI) ratio. If DNA is completely unmethylated the HpaII/MspI ratio would be 1.0, and if DNA is 100% methylated the HpaII/MspI ratio would approach zero.

4.2.2 Gene-specific methylation assay by bisulfite-DGGE (Paper I, III-IV)

Genomic DNA was treated with sodium bisulfite as described previously (Zeschnigk et al. 1997). PCR specific for the bisulfite reacted E-cadherin, ER andHIC promoters was carried out in a final volume of 25 µl containing 50-100 ng of bisulfite-modified DNA, 1x TEMPase PCR Buffer I with 1.5 mM MgCl2, 0.2 mM cresol red, 12% sucrose, 0.2 mM each dNTP, GC-clamped 0.5 µM each forward and reverse (Aggerholm et al. 2006) primers and 0.75 units of TEMPase HotStart DNA Polymerase. The enzyme was activated by incubation at 95C for 15 min followed by 40 cycles of 95C for 30s, 48C for 30s, 72C for 30s and a final extension at 72C for 5 min. PCR was performed in a PX2 Thermal cycler. PCR products were examined by electrophoresis in a 2.5% agarose gel. Ten µl of the GC-clamped PCR-product were loaded on to a 10% denaturant/6% polyacrylamide – 70% denaturant/12% polyacrylamide double gradient gel. Gels were run at 160 V for 270 minutes in 1x Tris acetate/EDTA buffer kept at a constant temperature of 55C. After

43

electrophoresis, gels were stained in 1x Tris acetate/EDTA buffer containing ethidium bromide (2 µg/ml) and photographed under UV transillumination.

4.3 Apoptosis assays 4.3.1 Morphological assay of apoptosis (Paper I-II)

Cytospin slides were stained with May-Grünwald-Giemsa. Apoptosis was measured by counting apoptotic cells with typical apoptosis-related morphological features (condensed chromatin, fragmented nuclei and shrinkage of cells). Apoptosis was estimated as a percentage of minimum 400 counted cells per slide. All experiments were performed in triplicates.

4.3.2 Antibodies and Western blot experiments (Paper II) Preparation of cytosolic extracts was described elsewhere (Hellstrom-

Lindberg et al. 2001). Subsequently, whole-cell lysates or supernatant and pellet fractions were analyzed by SDS-PAGE at 120 V followed by western blotting. The membranes were probed overnight with primary antibody at 4ºC followed by rinsing and incubation with a horseradish peroxidase-conjugated secondary antibody. After a repeated washing step, bound antibodies were detected using enhanced chemiluminescence according to the manufacturer's instructions. Primary antibodies were cytochrome c and Bax, Bid, Bcl-2, Caspase-2 and G3PDH.

4.3.3 Flow cytometry (Paper I-II) For flow cytometric analysis, 2-3x105 cells were used. Active caspase-3 was

measured using FITC-conjugated anti-active-caspase-3 antibody. After fixation with 2% paraformaldehyde, cells were incubated for 1 hour in dark with the antibody and 0.02% digitonin. The mitochondrial transmembrane potential ( m) was measured using the cationic fluorescent dye tetramethylrhodamine ethyl ester (TMRE, 25 nM), which normally accumulates in mitochondria according to the m. Cytosolic pH was analyzed using the fluorescent dye Snarf-1 (2.5 M). The cells were incubated 30 min at 37ºC in the medium containing the respective dye. Data from the flow cytometry were analyzed with the CELL Quest.

4.3.4 Enzyme activity assay (Paper II) To estimate caspase activity, the caspase fluorogenic peptide substrates

VDVAD-AMC (caspase-2), IETD-AMC (caspase-8) and DEVD-AMC (caspase-3-like) were used. One million cells were harvested, washed in PBS and stored at –80ºC. Dry cell pellets were resuspended in 25 µL PBS and added to a microtiter plate in combination with the appropriate peptide substrate conjugated to AMC. After addition of reaction buffer (100 mM HEPES, 10% sucrose, 5mM dithiothreitol, 0.1% nonidet P-40 and 0.1% 3-[(3-cholamidopropyl) dimethylammonio]-1-

44

propanesulfonate (CHAPS), pH 7.25) cleavage of the fluorogenic peptide substrate was monitored by release of free AMC in a Fluoroscan II plate reader using 355nm excitation and 460nm emission wavelengths. Fluorescence units were converted into the amount of AMC (pmol) using a standard curve generated with free AMC. Data were then analyzed by linear regression.

4.4 Other assays 4.4.1 Cell growth (Paper I-II)

Cell growth was determined daily during experiments, using trypan blue exclusion for viability assessment.

4.4.2 Cell proliferation (DNA synthesis assay using 3H-TdR) (Paper I)

Aliquots of 0.1 x 106 cells/0.2 ml medium were incubated for 6 hours in triplicates on 96-wells microplates with 1 Ci (20 µl of 50 Ci/ml) 3H-labelled thymidine (3H-TdR) at 37ºC. The cells were harvested onto filters with a Cell Harvester. The radioactivity was measured in scintillation liquid using a liquid scintillation counter and results were expressed as disintegrations per minute (DPM).

4.4.3 Cell cycle analysis (Paper I) Cells were washed, fixed in 70% ice-cold ethanol in PBS and stained with

hypotonic propidium iodide (PI) solution (20 g/ml) containing DNase-free RNase (100 µg/ml). The cell cycle analysis were acquired and analyzed on a FACScan flow cytometer using CELL Quest software.

4.4.4 Sorting apoptotic cells (Paper II) P39 cells were treated with 1 M of azacytidine for 24 hours. Apoptotic and

non-apoptotic cells were sorted using FACSDiVa in a sheath fluid pressure of 30 PSI. Since cell shrinkage and nuclear condensation are hallmarks of apoptosis, it was possible to sort apoptotic cells by forward and sideward scatter criteria. The accuracy of cell sorting was controlled by TMRE-staining.

45

5. RESULTS AND DISCUSSION At the commencement of the project in January, 2003, not much was known

about 5-azacytidine. The results of the initial clinical studies had just been published. Interestingly, at the time of the FDA approval of azacytidine, preclinical information about the pharmacokinetic and pharmacodynamic properties of the drug, as well as its effects on cell lines and in animal models were absent or inadequately documented. As expected, there was an emerging interest in the drug leading to an increased number of publications. However, many of those were clinical trials, or studies of the drug’s hypomethylating property. My research project aimed at a deeper understanding of the pharmacodynamic properties of the drug, as well as an understanding of how it actually inhibits the growth of malignant myeloid cells besides inducing hypomethylation. This naturally led on to the question about how DNA methylation status affects outcome of patients with myelodysplastic syndromes and acute leukaemia.

5.1 Pharmacodynamic of 5-azacytidine Pharmacokinetic studies of 5-azacytidine were not feasible, mostly due to

unavailability of suitable and sensitive method, during the time when we initiated our projects, and, in fact, valid methods have yet not been published. The relative unstable nature of 5-azacytidine plays an impeding role in the development of such a method. Hence, after trying to develop pharmacokinetic methods for a number of months, we redirected our aims to study the pharmacodynamics of 5-azacytidine and managed to provide useful clinical information through this project.

5.1.1 Effect on cell growth 5-azacytidine inhibited cell growth of P39 cells in a dose-dependent manner

(Paper I). Cells tended to recover when they were re-incubated in a drug-free medium. Cells exposed to >0.5 M 5-azacytidine recovered more slowly than cells exposed to 0.5 M azacytidine. Interestingly, drug exposure time (24, 48 or 72 hours) had no significant effect on cell growth (Paper I). Therefore, 5-azacytidine induced inhibition of cell growth is dose-dependent but time independent.

5.1.2 Effect on cell proliferation (DNA synthesis assay using 3H-TdR)

Similar to its effects on cell growth, 5-azacytidine inhibited proliferation (DNA synthesis) of P39 cells in a dose-dependent manner (0.1-1.0 M; Paper I). Again, duration of drug exposure (24-48 hours) had no significant effect on cell proliferation (Paper I). In contrast to cell growth, however, cells recovered when they

46

were re-incubated in drug-free medium, irrespective of the concentration used (0.1-1.0 M; Paper I).

5.1.3 Effect on cell cycle The 5-azacytidine-induced dose-dependent increase of apoptosis was also

confirmed by cell cycle studies (Paper I). A dose-dependent decrease in S phase with a similar decrease in G0/G1 and G2/M phases was observed for all the concentrations (Figure 8).

5.1.4 Pharmacodynamic analysis A significant decrease (p<0.01) of cell growth and cell proliferation rate

(DNA synthesis) was observed when P39 cells were incubated for 24 and 48 hours with 0.5 and 1.0 M 5-azacytidine (Paper I). However, significant values were not obtained when concentrations <0.5 M were used (Paper I). Interestingly, when cells were incubated with a total dose of 2 or 3 M, there was no significant difference in either parameters if the dose was given as a single dose for 24 hours or divided over 2 or 3 days (Figure 9) (Paper I). In addition, a nonlinear exponential model and a non-parametric statistical analysis (Friedman test) showed that 3 or 2 M 5-azacytidine given over 24 hours was as effective as 1 M given over 48 or 72 hours in terms of cell growth, proliferation and apoptosis. These findings suggested that a more convenient dosing schedule might be developed to obtain the optimal benefit for the patients undergoing 5-azacytidine treatment.

Figure 8. P39 cells were treated with 5-azacytidine for 72 hours. Cell cycle analysis was performed by flow cytometry using propidium iodide. Values are expressed as mean±SD (Paper I).

47

5.2 Methylation As mentioned earlier, DNA methylation status has been suggested to play an

important role in diseases, including MDS. The focus of this research has almost exclusively been on gene-specific methylation, since adequate methodology for assessing global methylation has not been available. Recent acknowledgement about the role of changes in global methylation during disease development has shed light on another perspective of DNA methylation. Therefore we decided to explore the role of both global and gene specific methylation changes in patients with MDS.

5.2.1 Global and gene-specific methylation status in patients with MDS

Global methylation status was measured by two parameters; ratios for HpaII/EcoRI and HpaII/MspI. Both parameters differed between patients and age-matched healthy controls (p=0.01 and 0.009, respectively) (Figure 10). Interestingly, we found that controls had higher degree of global methylation, measured by HpaII/EcoRI and HpaII/MspI (0.54±0.07 and 0.26±0.003, respectively) than patients with MDS. It has been shown that during disease development, the promoter region becomes hypermethylated while the non-promoter region becomes hypomethylated (Feinberg and Tycko 2004). Hence, a considerable methylation status change in both promoter and non-promoter region could end up giving the impression of no change in the total methylation level quantitatively. Moreover, no significant difference in global methylation, measured by HpaII/EcoRI and HpaII/MspI, were found between low-risk and high-risk MDS, as defined by IPSS score (p=0.22 and 0.53, respectively) (Paper III).

Figure 9. The effects of azacytidine on cell proliferation (DNA synthesis) obtained after either a single or fractionated dose of azacytidine. Cells were exposed to a single dose for 24 hours or fractionated total dose over 48 (A) or 72 (B) hours; followed by incubation for 48 hours in drug-free medium.

48

Gene specific methylation was assessed for p15INK4B, which is the most studied gene in MDS, and suggested to be a relevant gene for MDS progression. p15INK4B is usually unmethylated in healthy controls. In the present cohort, consisting mainly of patients with low and intermediate-risk MDS we found p15INK4B

methylation in 33% of patients with MDS, more frequent in high-risk than in low-risk MDS (p=0.003) (Paper III), which is in line with previously published data (Tien et al. 2001).

Figure 10. Comparison of global methylation, HpaII/EcoRI (A) and HpaII/MspI (B), between patients and age-matched healthy controls. Significant changes in hypomethylation between patients and age-matched healthy controls were evident (p=0.01 and 0.009, respectively for HpaII/EcoRI and HpaII/MspI).

Figure 11. Global methylation in MDS and healthy controls. MDS patients were divided into two groups depending on p15INK4B methylation status and compared with age-matched healthy controls. A) HpaII/EcoRI ratios: A significant difference was detected between p15INK4B unmethylated and methylated cases (p=0.02) and between unmethylated p15INK4B cases and controls (p=0.003) B) HpaII/MspI ratios: A significant difference was detected between p15INK4B

unmethylated and methylated cases (p=0.03) and between unmethylated p15INK4B

cases and controls (p=0.003).

49

In addition, our result showed a significant difference in global methylation between p15INK4B methylated and unmethylated patients, both for HpaII/EcoRI (p=0.02) and HpaII/MspI (p=0.03) (Figure 11) (Paper III). Hence we demonstrated for the first time, differences in global methylation status between normal individuals and low-risk MDS, and between low- and high-risk MDS, as defined by p15INK4B

methylation status. We studied global methylation over time in 8 patients. Four had stable disease and stable methylation levels. Interestingly, only 2/4 patients with disease progression showed an increase in methylation. Hence, variations in methylation status might occur during development and progression of MDS but whether this is essential for progression of MDS, or whether the different levels of methylation merely reflects different variants of the disease should be further investigated.

5.2.2 Methylation status in relation to outcome of treatment We studied the role of gene-specific methylation in relation to the outcome of

induction chemotherapy (Figure 12). Interestingly, p15INK4B methylation alone was not associated with complete remission (CR) rate (p=0.25) (Paper IV). However, patients with methylation of p15INK4B plus one other gene had a significantly lower CR rate than other patients, and none of the patients with promoter methylation of p15INK4B, E-cadherin (CDH) and Hypermethylated in Cancer 1 (HIC) genes achieved CR. Among the three genes, hypermethylation of CDH showed the strongest association with CR rate (p=0.008). The role of CDH methylation was retained also in the multivariate analysis (Paper IV). The finding of hypermethylation as a potential mechanism behind resistance to conventional chemotherapeutic agents is a first-in-class finding that may shed light over the poor outcome of patients with high-risk

Figure 12. Percentage of CR in patients in comparison with methylation of P15, CDH and HIC genes.

50

MDS, and also the high incidence of relapse even after intensive regimens such as stem cell transplantation. Since our study is relatively small, 52 patients, the observation has first to be confirmed in an independent material. If confirmed, this may lead to profound changes in the concept for how demethylating and conventional chemotherapeutic agents should be combined.

5.2.3 Hypomethylating effects of 5-azacytidine In the present studies, we showed that 5-azacytidine induces both global and

gene-specific hypomethylation (Paper I and II).

5.2.3.1 Global DNA methylation 5-azacytidine induced global DNA hypomethylation (paper II) in myeloid

(P39 and HL60) and T-cell (Jurkat) lines. After 48 hours, 0.5 M 5-azacytidine induced significant Global DNA methylation for both the parameters (HpaII/MspIand HpaII/EcoRI) (Figure 13).

5.2.3.2 Gene specific methylation 5-azacytidine induced gene specific methylation of CDH1, ER and HIC gene

in the P39 myeloid cell line (Figure 14) (Paper I). We tested concentrations between 0.5, 1.0 and 2.0 M for during 24, 48 and 72 hours. The hypomethylating effect was more prominent with longer incubation, which probably is explained by the fact that azacytidine mainly acts on cells in S-phase. The strongest hypomethylating effect was

Figure 13. 5-azacytidine (Aza) induced global DNA hypomethylation and apoptosis. (A) Aza induced time-dependant effect on HpaII/MspI ratio indicating global DNA hypomethylation pattern in P39 cells. Significant hypomethylation was observed with 0.5 M after 24 hours (p=0.03) and with 1.0 M Aza after 48 hours (p=0.006). (B) Global DNA hypomethylation of treated and FACS-sorted apoptotic and non-apoptotic cells in comparison to untreated and unsorted cells in P39, HL60 and Jurkat cells. Cells were treated with 1 M Aza for 24 hours.

51

observed after incubation with 1 M for 72 hours. Interestingly, hypomethylation was less pronounced after incubation with concentrations higher than 1 M.

5.3 Apoptosis Imbalance in apoptotic processes plays a significant role in MDS. As

mentioned earlier, increased apoptotic rate in low-risk MDS and decreased apoptotic rate during disease progression is still a poorly understood phenomenon. The high incidence of drug resistance to conventional chemotherapy is a key finding of MDS, as well as the poor ability of marrow stem cells to regenerate after cytotoxic injury. Previous papers have shown 5-azacytidine can induce apoptosis as well as hypomethylation (Silverman et al. 2002; Ueno et al. 2002; Khan et al. 2006), but no studies have investigated the relation between these two phenomena. Although the efficacy of 5-azacytidine is suggested to be related to its hypomethylating properties, the link between the hypomethylating and apoptotic properties of 5-azacytidine has not been clarified. Such knowledge is essential for developing better schedules for azacytidine, as well as combination therapies including azacytidine, and was therefore a target for our studies.

5.3.1 Morphological assay of apoptosis 5-azacytidine induced apoptosis in a dose-dependent manner (0.5-2.0 M),

measured by morphology in P39 and HL60 cells (Figure 15) (Paper I& II); and also

Figure 14. 5-azacytidine-induced gene specific methylation of CDH1, ER and HIC gene, measured by denaturing gradient gel electrophoresis (DGGE).

52

by active caspase-3 and loss of m (Paper II). 5-azacytidine concentration of 0.1 M failed to induce significant apoptosis.

5.3.1.1 Bcl2- family members and cytochrome c

Bcl-2 family proteins are important regulators of the apoptosis process. Imbalance of Bcl-2 proteins towards a pro-apoptotic state leads to apoptosis, and vice versa. 5-azacytidine induced activation of pro-apoptotic proteins, such as Bid and Bax in P39 cells (Paper II). Additionally, inactivation of anti-apoptotic proteins, such as Bcl-2, was also evident (Paper II). In addition to inducing a pro-apoptotic imbalance of Bcl-2 family proteins, 5-azacytidine also induced release of cytochrome c from mitochondria (Figure 16).

Figure 15. 5-azacytidine-induced morphological apoptotic changes in P39 cells ater 24 hours. Significant apoptosis was observed with 0.5 and 1.0 M Aza (p<0.001) (Paper II).

Figure 16. 5-azacytidine-induced cytochrome c release, which was related to Bid cleavage. Activation of Bid and Bax, inactivation cleabage of Bcl-2 and relase of cytochrome c was dose-dependent.

53

5.3.1.2 Mitochondrial involvement Loss of m is considered to be one of the key regulators of apoptosis and a

well established indicator of mitochondria-related apoptosis. 5-azacytidine induced loss of m in a dose-dependent manner (Figure 17).

5.3.1.3 Induction of DNA damage Caspase-2 activation has been shown to link to DNA damage and to act

upstream of mitochondria (Robertson et al. 2002). Incubation of P39 cells with 5-azacytidine led to dose-dependent activation and processing of caspase-2 (Figure 18) (Paper II). However, inhibition of caspase-2 by Z-VDVAD-fmk failed to block loss of m and apoptosis-related morphological changes (Paper II).

PARP is another enzyme which can act as a sensor for DNA damage and become activated (Bouchard et al. 2003). Inhibition of PARP by nicotinamide (NA) and 3-aminobenzamide (3AB) only partially blocked loss of m. For example, NA and 3AB caused 32 and 34% reduction of loss of m, respectively, when P39 cells

Figure 17. (A) 5-azacytidine induced mitochondrial alterations in P39, HL60 and Jurkat cells (Paper II). (B) Significant loss of m was achieved after 12 hours of incubation with 0.5 M 5-azacytidine in P39 cells (p=0.013 and 0.002 with 0.5 and 1.0 M; respectively; Paper II). However, no significant increase in apoptosis was achieved with 0.1 M of 5-azacytidine at any time point (12, 24 and 48 hours).

54

were incubated with 1 M of 5-azacytidine for 24 hours. In addition, PARP inhibitors also failed to block 5-azacytidine-induced caspase-3 or apoptotic morphology (Paper II, data not shown). Interestingly, PARP inhibitors also partially suppressed activation (38% reduction) and processing of caspase-2 (Paper II). Hence, these findings suggested that 5-azacytidine induced involvement of PARP in mitochondrial alteration, and working upstream of caspase-2.

Figure 18. 5-azacytidine-induced DNA damage evidenced by activation of caspase-2.

Figure 19. (A) 5-azacytidine-induced dose-dependent enzymatic activation of caspase-3.( B) Role of caspase inhibitors (10 M) in Aza-induced loss of m(M1) measured by flowcytometry using the fluorescent dye tetramethylrhodamine ethyl ester (TMRE).

55

5.3.1.4 Caspase activation As with caspase-2, 5-azacytidine also induced dose-dependent activation of

caspase-3. (Figure 19A)(Paper II). However, caspase-8, which is an indicator of receptor-mediated apoptosis, was not involved in 5-azacytidine-induced apoptosis (Paper II). Interestingly, when several caspase inhibitors, such as Z-VDVAD-fmk (caspase-2), ZDEVD-fmk (caspase-3-like), Z-IETD-fmk (caspase-8), Z-LEHD-fmk (caspase-9) and Z-VAD-fmk (pan-caspase inhibitor), were used at concentrations from 10-40 M, it neither blocked 5-azacytidine-induced apoptosis-related mitochondrial alterations (Figure 19B) nor morphological changes (Paper II).

5.3.1.5 Cytosolic acidification Intracellular acidification in the non-apoptotic fraction of cells has been

reported to link with apoptosis induced by cytotoxic drugs. In analogy, 5-azacytidine induced intra-cellular acidification in non-apoptotic fraction of P39 cells (Paper II). A decrease of 3.2, 9.7 and 11.3 %, measured by mean florescence unit (MFI), were observed after 24 hours with 0.5, 1.0 and 2.0 M azacytidine, respectively (Figure 20A). Monensin stabilized the 5-azacytidine-induced intracellular pH, an increase of 8.2% MFI when treated with combination of monensin and 1 M 5-azacytidine compared to only 1 M 5-azacytidine (Figure 20B). Nonetheless, in contrast to what have been reported for several other cytotoxic drugs (Ahmad et al. 2004; Kanno et al.2004), this stabilization of intracellular pH could neither prevent the early loss of m

nor the late apoptosis-related morphological changes (Paper II).

Figure 20. (A) 5-azacytidine-induced cytosolic acidification. (B) Stabilization of 5-azacytidine-induced cytosolic acidification by monensin.

56

5.4 Overview of 5-azacytidine-induced hypomethylation and apoptosis

5-azacytidine induced both hypomethylation and apoptosis in doses that were not very toxic and therefore could be related to clinically relevant doses. Interestingly, initiation of apoptosis preceded detectable changes in methylation pattern by 12 hours, but developed in parallel between 24 and 48 hours. However, the initial absence of hypomethylation might be explained by a low initial and specific hypomethylation not detected by LUMA, leading to re-expression of single crucial genes, which might be significant to induce some of the apoptotic pathways in the leukaemic cells. Nonetheless, 5-azacytidine-induced decrease in global methylation was only evident in the apoptotic but still viable cells, whereas, the non-apoptotic cells retained a methylation pattern similar to untreated cells. Our data also indicate that azacytidine affects cells in culture according to their cell cycle status, as sorted non-apoptotic cells could be induced to apoptosis after a second exposure of azacytidine (Paper II).

Figure 21. Proposed mechanism of 5-azacytidine (Aza) induced hypomethylation and apoptosis. DNA hypomethylation causes apoptosis, independently or by DNA damage-mediated caspase-2 activation and/or of Bcl-2 family proteins and mitochondrial involvement. Inhibition of PARP, caspase-2, cytosolic acidification, as well as caspases, separately did not block Aza-induced apoptosis. Hence, Aza induces apoptosis in multiple and separately regulated pathways.

57

Conventionally, apoptotic changes in various cellular compartments are regarded as mechanistically linked events in a single "program", where activation of caspases and proteolysis of intracellular substrates represent a final common pathway to cell death. Our work shows that apoptotic pathways may be regulated independently. Thus, cells were dying by apoptosis despite prevention of mitochondrial deterioration by nicotinamide or 3-aminobenzamide, or inhibition of caspases, or stabilization of intracellular pH (Figure 21).

Re-expression of epigenetically silenced gene has been reported to induce apoptosis (Hopkins-Donaldson et al. 2000; Levine et al. 2003; Reu et al. 2006). In addition, global methylation has been shown to be associated with chromosomal instability (Ehrlich 2002; Deng et al. 2006), and thereby lead to a DNA damage mediated apoptosis. Hence, these could be hypothesized as a link between 5-azacytidine-induced hypomethylation and apoptosis.

5.5 Efficacy of 5-azacytidine in MDS when other drugs fail The exact mechanism for the effectiveness of 5-azacytidine is yet to be

discovered. However, in the light of our findings we propose the following hypotheses:

1. 5-azacytidine acts via multiple but separately regulated apoptotic pathways, as well as hypomethylation. If one of the pathways is blocked, the apoptotic mechanism can still prevail.

2. Hypermethylation of certain genes may be a mechanism behind resistance to conventional chemotherapeutic agents. Therefore 5-azacytidine could re-express the genes and thereby overcome drug resistance of cancer cells.

58

6. CONCLUSIONSBased on these studies, it can be concluded that:

5-azacytidine induced dose-dependent hypomethylation as well as a complex pattern of apoptotic mechanisms, regulated by multiple and separate pathways. Furthermore, apoptosis and hypomethylation occurred in parallel.

Optimal cellular effects of 5-azacytidine can be achieved by shorter exposure times. Translated to shorter treatment duration in patients, this may eventually lead to superior and more cost-effective clinical treatment schedules, and improved quality of life for the patients.

Patients with MDS showed a lower overall degree of global methylation than normal healthy individuals. Global methylation status did not correlate with conventional risk factors in MDS, but was more pronounced in patients with severe erythroid failure.

Global and gene-specific DNA methylation patterns in MDS correlated with each other in the sense that more advanced MDS, characterized by p15INK4B

methylation showed higher global DNA methylation than those with unmethylated p15INK4B.

Patients with stable disease maintained their global DNA methylation level over time, while disease progression was associated with an increase in global DNA methylation in half of the cases. Hence, increased methylation is not a prerequisite for disease progression.

P15INK4B methylation plus methylation of CDH or HIC gene was more frequent in patients with conventional high-risk features, such as adverse karyotype and high percentage of marrow blasts. DNA hypermethylation of multiple surrogate genes was associated with the percentage of blasts and marrow CD34 expression, but not with the degree of inefficient haematopoiesis.

We demonstrated a significant negative impact of promoter methylation status on the outcome of conventional induction chemotherapy in high-risk MDS and AML following MDS. This finding may have significant impact on the clinical management of patients with high-risk myeloid disease.

59

7. FUTURE PERSPECTIVES For future MDS research within the field of methylation, it is fundamental to

obtain improved understanding of the genetic and epigenetic mechanisms underlying the development of MDS. The molecular and cellular mechanisms of MDS resulting in aberrant proliferation, differentiation, dysplasia and apoptosis, need to be understood in order to improve treatment for patients with MDS. We need improved, quicker and uncomplicated methods for detection of global and gene-specific DNA methylation for assessment of prognosis as well as for optimal treatment decisions. The rate of scientific progress in the experimental studies of methylation and other aspects of epigenetics is currently quite rapid and we can expect new results, which might include surprises as well as controversies. Our studies show that the combination of conventional chemotherapy and hypomethylating agents is a very relevant topic for future studies.

60

8. ACKNOWLEDGEMENTS I have looked forward to writing this section; not only because it means that I

have almost finished writing my kappa (!), but also because I would like to express my sincere gratitude to the numerous people that have helped me along the path. This long journey would not have been enjoyable and memorable without their presence. It was a fantastic experience working with you all.

Dr. Eva Hellström-Lindberg, my supervisor, for giving me the opportunity to be a PhD student under your supervision. I am grateful for your excellent guidance and encouragement. You really amazed me with your all time positive attitude. I appreciate and cherish the constant moral support that I received from you. Thank you for teaching me the art of scientific paper writing and presentation.

Professor Tomas J Ekström, my co-supervisor, for always having time to talk through my problems whenever I call you, even when I know that you are ‘never free’! I really admired your sense of humour and easy-going attitude. I am grateful for all the scientific discussions we had over the time and enjoyed working in your lab.

Professor Moustapha Hassan, my previous co-supervisor, for being there more as a friend than a co-supervisor. I appreciate for all the ‘short’ meetings I had with you during my training. I really enjoyed our time to time discussion about ‘life’ and ‘evolution’ outside the boundary of research. Dr. Zuzana Hassan, for sharing your research knowledge with me. I cherish the nice barbeque parties at your place every summer.

Professor Boris Zhivotovsky, our collaborator in IMM, thank you for sharing your scientific knowledge in the field of apoptosis and allowing me to work in your lab. Dr. Vladimir Gogvadze, our collaborator in IMM, thank you for being such a wonderful person. I valued the scientific discussions we had and also for your all time positive attitude. It was a privilege to know and work with you. Pinelopi, Ulrika, Salvador, Nabil, Helin, Magnus, Margaretha and Roshan, Apoptosis group in IMM, for all the help and support during my work in your lab.

Dr. Anni Aggerholm, our collaborator in Denmark, for teaching and sharing your vast knowledge regarding the bisulfite-DGGE technique. I really appreciated your warm personality and friendly attitude. Professor Peter Hokland and Karin Brændstrup, our collaborators in Århus University, Denmark, for your continuous support and productive collaboration.

Professor Jan Palmblad and Professor Jan Bolinder, present and past Heads of the Department of Medicine, respectively; Professor Per Ljungman, Head of Haematology clinics for providing excellent working environment.

Berit Lecomte and Elenor Nyman, for their welcoming personality and all the help related to administrative and economical work during my PhD period; and Klas Karlsson and Sören Lehmann for providing a good working environment.

61

MDS group, KI: Lalla (Ann-Mari) Forsblom, a very special person, for being there more as a friend than a colleague and for taking care of me during the whole PhD period. I relished our ‘small’ discussions about art, music, book, cooking, computer, ‘o.s.v’. I appreciated your warm personality and friendly attitude, and trying to solve all the problems related to work and life. I thank you for being incredibly kind to me. Jan Schmidt-Mende and Ramin Tehranchi, for your friendship, social life, table-tennis, innerbandy, badminton, ski and other activities during my PhD period. Monica Jansson, for your friendly attitude. I have disturbed you at so many occasions during my PhD period, but surprisingly have not found you troubled any of those times. Thank you for being so kind to me. Maryam Nikpour, Michael Grövdal and Martin Jädersten for creating a nice working environment in the lab. I thank you for your friendship.

Associate Professor Alf Grandien and Dr. Kari Hogstrand, from CIM, for helping me with the PCR problem. Åsa-Lena Dackland, from Pathology dept., for helping me to sort cells in the pathology department and working long hours.

Christina Nilsson, Ann Wallblom, Ann Svensson, Sofia Bengtzén, Kristina Friberg, Birgitta Åhnsén, Emma Emanuelsson, Kerstin Jönsson, Ulrika Felldin, Alexandra Treschow, Kyriakos Konstantinidi, Hayrettin Guven, Evren Alici, Christian Unger, Hernán Concha-Quesada and Johanna Forsman, from Haematology lab, for creating an excellent working environment in the lab. I thank you for sharing your skills and knowledge to help me with my problems. Special thanks to Birgitta Stellan and Mari Gilljan, for all time cheerful approach, kind words and encouragement.

Mohsen Karimi, for teaching me the exciting new method, LUMA. Sofia Johansson, for teaching me the software to analyse LUMA results.

Professor Curt Peterson, Dr. Malin Lindqvist and Professor Peter Söderkvist, from Linköping University, for introducing me in the field of research. Md. Delwar Hossain, my teacher from Ideal High School, who taught me to have dreams and most importantly the capacity to pursue it.

Marina Vita, Parvaneh Afsharian and Gayane Avetisyan, my friends in the haematology lab, for your friendship, enjoyable coffee-breaks and all time support. Life would not been so much fun during this period without your presence. Song Hairong, my ‘crazy’ Chinese friend, for all your funny stories, activities and friendship; Lusine Aghajanova for your friendship. Behnam Sadeghi and Solaiman for your friendship and discussions in the coffee-breaks.

Riyadh, Lipi vabi and Tajrian, for your long-term friendship and affection. Thank you for being there whenever we needed you. Akter vai, for your friendship. I shall never forget the beautiful memories we had together during my stay in Linköping. Mukut, Ayaz, Shumon and Titu, my very good friends, for your unconditional friendship and the nice few months we spend together in Stockholm.

62

Asad vai and Chumki vabi, for your friendship, support and kindness. Lipi, for your ‘paglamies’, Amzad and Nahid, Monon and Loren, Rony and Tanjila, for sharing all the fun we had while living in Röntgenvägen 1. Arup and Shimul, for becoming our close friends within a short time and for your warm personalities. Mahbub vai and Ela apa, for being our ‘boro vai’ in Sweden and for your guidance. Jewel vai and Nishi vabi; Disha and Polash; for your friendship. Sohel (choto), for your computer related helps and movies. Maroof, Raj and Porag for your friendship. I shall cherish the wonderful gatherings and lovely memories we had together. To all my friends from CMC, for taking care of me and being there during ‘the ALMOST end of life!’.

Zaki Ahmed Khan, my father, for having trust in every step in my life. Qazi Hosne Ara, my mother, for your prayers. Nasrin Khan, for you enthusiastic and all time positive attitude. Shagor vaiya, Zami, Rimi and Abid, my brothers and sister, for your invaluable and endless support and love throughout my life. Khan Ali Imam and Anisa Imam for your unconditional love and blessings. Nana (Monowar Rahman Khan), Nanu, Ruma, Shuma and Nipa for their love and incredible support throughout my life.

Tania, my wife, for your love and generous support. Thank you for having patience especially during the ‘hectic’ days of PhD training.

Rohana, our daughter, for simply just being there and bringing new meaning to all the things in my life.

These works were supported by the grants from Swedish Cancer Society, Stockholm Cancer Society, Nordic Cancer Union, Swedish Children Cancer Society and Swedish Research Council.

63

9. REFERENCES Acehan D,Jiang X, et al. (2002). "Three-dimensional structure of the apoptosome:

implications for assembly, procaspase-9 binding, and activation." Mol Cell9(2): 423-32.

Aggerholm A,Holm MS, et al. (2006). "Promoter hypermethylation of p15INK4B, HIC1, CDH1, and ER is frequent in myelodysplastic syndrome and predicts poor prognosis in early-stage patients." Eur J Haematol 76(1): 23-32.

Ahmad KA,Clement MV, et al. (2004). "Resveratrol inhibits drug-induced apoptosis in human leukemia cells by creating an intracellular milieu nonpermissive for death execution." Cancer Res 64(4): 1452-9.

Antonsson B,Montessuit S, et al. (2000). "Bax oligomerization is required for channel-forming activity in liposomes and to trigger cytochrome c release from mitochondria." Biochem J 345 Pt 2: 271-8.

Appelbaum FR and Anderson J (1998). "Allogeneic bone marrow transplantation for myelodysplastic syndrome: outcomes analysis according to IPSS score." Leukemia 12 Suppl 1: S25-9.

Aul C,Bowen DT, et al. (1998a). "Pathogenesis, etiology and epidemiology of myelodysplastic syndromes." Haematologica 83(1): 71-86.

Aul C,Germing U, et al. (1998b). "Increasing incidence of myelodysplastic syndromes: real or fictitious?" Leuk Res 22(1): 93-100.

Aul C,Giagounidis A, et al. (2001). "Epidemiological features of myelodysplastic syndromes: results from regional cancer surveys and hospital-based statistics." Int J Hematol 73(4): 405-10.

Bachman KE,Rountree MR, et al. (2001). "Dnmt3a and Dnmt3b are transcriptional repressors that exhibit unique localization properties to heterochromatin." J Biol Chem 276(34): 32282-7.

Bain BJ (1996). "The bone marrow aspirate of healthy subjects." Br J Haematol94(1): 206-9.

Baylin SB (2005). "DNA methylation and gene silencing in cancer." Nat Clin Pract Oncol 2 Suppl 1: S4-11.

Baylin SB and Ohm JE (2006). "Epigenetic gene silencing in cancer - a mechanism for early oncogenic pathway addiction?" Nat Rev Cancer 6(2): 107-16.

Bellacosa A,Cicchillitti L, et al. (1999). "MED1, a novel human methyl-CpG-binding endonuclease, interacts with DNA mismatch repair protein MLH1." Proc Natl Acad Sci U S A 96(7): 3969-74.

Bennett JM,Catovsky D, et al. (1976). "Proposals for the classification of the acute leukaemias. French-American-British (FAB) co-operative group." Br J Haematol 33(4): 451-8.

Bennett JM,Catovsky D, et al. (1982). "Proposals for the classification of the myelodysplastic syndromes." Br J Haematol 51(2): 189-99.

Beran M (2000). "Intensive chemotherapy for patients with high-risk myelodysplastic syndrome." Int J Hematol 72(2): 139-50.

Bestor TH (2000). ''The DNA methyltransferases of mammals.'' Hum Mol Genet9(16): 2395-402.

Bird A (2002). "DNA methylation patterns and epigenetic memory." Genes Dev16(1): 6-21.

Bird A (2007). "Perceptions of epigenetics." Nature 447(7143): 396-8. Bird AP and AP Wolffe (1999). ''Methylation-induced repression--belts, braces, and

chromatin.'' Cell 99(5): 451-4.

64

Bjorkman SE (1956). "Chronic refractory anemia with sideroblastic bone marrow; a study of four cases." Blood 11(3): 250-9.

Block M,Jacobson LO, et al. (1953). "Preleukemic acute human leukemia." J Am Med Assoc 152(11): 1018-28.

Bouchard VJ,Rouleau M, et al. (2003). "PARP-1, a determinant of cell survival in response to DNA damage." Exp Hematol 31(6): 446-54.

Bowen D,Culligan D, et al. (2003). "Guidelines for the diagnosis and therapy of adult myelodysplastic syndromes." Br J Haematol 120(2): 187-200.

Broliden PA,Dahl IM, et al. (2006). "Antithymocyte globulin and cyclosporine A as combination therapy for low-risk non-sideroblastic myelodysplastic syndromes." Haematologica 91(5): 667-70.

Carlile GW,Smith DH, et al. (2004). "Caspase-3 has a nonapoptotic function in erythroid maturation." Blood 103(11): 4310-6.

Carthew RW (2006). "Molecular biology. A new RNA dimension to genome control." Science 313(5785): 305-6.

Catalano L,Selleri C, et al. (2000). "Prolonged response to cyclosporin-A in hypoplastic refractory anemia and correlation with in vitro studies." Haematologica 85(2): 133-8.

Cazzola M,Invernizzi R, et al. (2003). "Mitochondrial ferritin expression in erythroid cells from patients with sideroblastic anemia." Blood 101(5): 1996-2000.

Chan GW,Foss FM, et al. (2003). "Reduced-intensity transplantation for patients with myelodysplastic syndrome achieves durable remission with less graft-versus-host disease." Biol Blood Marrow Transplant 9(12): 753-9.

Cheson BD,Greenberg PL, et al. (2006). "Clinical application and proposal for modification of the International Working Group (IWG) response criteria in myelodysplasia." Blood 108(2): 419-25.

Childs AC,Phaneuf SL, et al. (2002). "Doxorubicin treatment in vivo causes cytochrome C release and cardiomyocyte apoptosis, as well as increased mitochondrial efficiency, superoxide dismutase activity, and Bcl-2:Bax ratio." Cancer Res 62(16): 4592-8.

Chiu YL and Rana TM (2002). "RNAi in human cells: basic structural and functional features of small interfering RNA." Mol Cell 10(3): 549-61.

Chou JJ,Li H, et al. (1999). "Solution structure of BID, an intracellular amplifier of apoptotic signaling." Cell 96(5): 615-24.

Claessens YE,Bouscary D, et al. (2002). "In vitro proliferation and differentiation of erythroid progenitors from patients with myelodysplastic syndromes: evidence for Fas-dependent apoptosis." Blood 99(5): 1594-601.

Claessens YE,Park S, et al. (2005). "Rescue of early-stage myelodysplastic syndrome-deriving erythroid precursors by the ectopic expression of a dominant-negative form of FADD." Blood 105(10): 4035-42.

Clement MV,Ponton A, et al. (1998). "Apoptosis induced by hydrogen peroxide is mediated by decreased superoxide anion concentration and reduction of intracellular milieu." FEBS Lett 440(1-2): 13-8.

Cohen GM (1997). "Caspases: the executioners of apoptosis." Biochem J 326 ( Pt 1):1-16.

Cohen MB and Glazer RI (1985). "Cytotoxicity and the inhibition of ribosomal RNA processing in human colon carcinoma cells." Mol Pharmacol 27(2): 308-13.

Craven SE,French D, et al. (2005). "Loss of Hspa9b in zebrafish recapitulates the ineffective hematopoiesis of the myelodysplastic syndrome." Blood 105(9): 3528-34.

Crompton M (2000). "Mitochondrial intermembrane junctional complexes and their role in cell death." J Physiol 529 Pt 1: 11-21.

65

Cross SH,Meehan RR, et al. (1997). "A component of the transcriptional repressor MeCP1 shares a motif with DNA methyltransferase and HRX proteins." Nat Genet 16(3): 256-9.

Csankovszki G,Nagy A, et al. (2001). "Synergism of Xist RNA, DNA methylation, and histone hypoacetylation in maintaining X chromosome inactivation." J Cell Biol 153(4): 773-84.

Daskalakis M,Nguyen TT, et al. (2002). "Demethylation of a hypermethylated P15/INK4B gene in patients with myelodysplastic syndrome by 5-Aza-2'-deoxycytidine (decitabine) treatment." Blood 100(8): 2957-64.

De Botton S,Sabri S, et al. (2002). "Platelet formation is the consequence of caspase activation within megakaryocytes." Blood 100(4): 1310-7.

de Murcia JM,Niedergang C, et al. (1997). "Requirement of poly(ADP-ribose) polymerase in recovery from DNA damage in mice and in cells." Proc Natl Acad Sci U S A 94(14): 7303-7.

De Witte T,Van Biezen A, et al. (1997). "Autologous bone marrow transplantation for patients with myelodysplastic syndrome (MDS) or acute myeloid leukemia following MDS. Chronic and Acute Leukemia Working Parties of the European Group for Blood and Marrow Transplantation." Blood 90(10): 3853-7.

Debatin KM,Poncet D, et al. (2002). "Chemotherapy: targeting the mitochondrial cell death pathway." Oncogene 21(57): 8786-803.

Deng G,Nguyen A, et al. (2006). "Regional hypermethylation and global hypomethylation are associated with altered chromatin conformation and histone acetylation in colorectal cancer." Int J Cancer 118(12): 2999-3005.

Doench JG,Petersen CP, et al. (2003). "siRNAs can function as miRNAs." Genes Dev 17(4): 438-42.

Earnshaw WC (1995). "Nuclear changes in apoptosis." Curr Opin Cell Biol 7(3): 337-43.

Egger G,Liang G, et al. (2004). "Epigenetics in human disease and prospects for epigenetic therapy." Nature 429(6990): 457-63.

Ehrlich M (2002). "DNA methylation in cancer: too much, but also too little." Oncogene 21(35): 5400-13.

Eskes R,Antonsson B, et al. (1998). "Bax-induced cytochrome C release from mitochondria is independent of the permeability transition pore but highly dependent on Mg2+ ions." J Cell Biol 143(1): 217-24.

Eskes R,Desagher S, et al. (2000). "Bid induces the oligomerization and insertion of Bax into the outer mitochondrial membrane." Mol Cell Biol 20(3): 929-35.

Esteller M,Corn PG, et al. (2001). "A gene hypermethylation profile of human cancer." Cancer Res 61(8): 3225-9.

Estey E,Thall P, et al. (1997). "Effect of diagnosis (refractory anemia with excess blasts, refractory anemia with excess blasts in transformation, or acute myeloid leukemia [AML]) on outcome of AML-type chemotherapy." Blood90(8): 2969-77.

Fadeel B,Orrenius S, et al. (2000). "The most unkindest cut of all: on the multiple roles of mammalian caspases." Leukemia 14(8): 1514-25.

Fadeel B,Zhivotovsky B, et al. (1999). "All along the watchtower: on the regulation of apoptosis regulators." Faseb J 13(13): 1647-57.

Feinberg AP and Tycko B (2004). "The history of cancer epigenetics." Nat Rev Cancer 4(2): 143-53.

Feinberg AP and Vogelstein B (1983). "Hypomethylation distinguishes genes of some human cancers from their normal counterparts." Nature 301(5895): 89-92.

66

Fenaux P (2005). "Response assessments in low-risk and high-risk myelodysplastic syndromes (MDS)." Semin Oncol 32(4 Suppl 5): S11-5.

Fenaux P,Mufti GJ, et al. (2007). "Azacitidine (AZA) Treatment Prolongs Overall Survival (OS) in Higher-Risk MDS Patients Compared with Conventional Care Regimens (CCR): Results of the AZA-001 Phase III Study." ASH annual meeting abstracts 110(11): Abstract #817.

Feng Q,Cao R, et al. (2002). "Identification and functional characterization of the p66/p68 components of the MeCP1 complex." Mol Cell Biol 22(2): 536-46.

Feng Q and Zhang Y (2001). "The MeCP1 complex represses transcription through preferential binding, remodeling, and deacetylating methylated nucleosomes." Genes Dev 15(7): 827-32.

Fernando P,Kelly JF, et al. (2002). "Caspase 3 activity is required for skeletal muscle differentiation." Proc Natl Acad Sci U S A 99(17): 11025-30.

Fleury C,Mignotte B, et al. (2002). "Mitochondrial reactive oxygen species in cell death signaling." Biochimie 84(2-3): 131-41.

Fontenay-Roupie M,Bouscary D, et al. (1999). "Ineffective erythropoiesis in myelodysplastic syndromes: correlation with Fas expression but not with lack of erythropoietin receptor signal transduction." Br J Haematol 106(2): 464-73.

Fraga MF,Ballestar E, et al. (2005). "Epigenetic differences arise during the lifetime of monozygotic twins." Proc Natl Acad Sci U S A 102(30): 10604-9.

Fuks F,Hurd PJ, et al. (2003). "The DNA methyltransferases associate with HP1 and the SUV39H1 histone methyltransferase." Nucleic Acids Res 31(9): 2305-12.

Gattermann N (2000). "From sideroblastic anemia to the role of mitochondrial DNA mutations in myelodysplastic syndromes." Leuk Res 24(2): 141-51.

Germing U,Strupp C, et al. (2007). "Epidemiologic features in myelodysplastic syndromes." Leukemia Research 31(Supplement 1): S1.

Gersuk GM,Beckham C, et al. (1998). "A role for tumour necrosis factor-alpha, Fas and Fas-Ligand in marrow failure associated with myelodysplastic syndrome." Br J Haematol 103(1): 176-88.

Glasspool RM,Teodoridis JM, et al. (2006). "Epigenetics as a mechanism driving polygenic clinical drug resistance." Br J Cancer 94(8): 1087-92.

Gogvadze V,Robertson JD, et al. (2001). "Cytochrome c release occurs via Ca2+-dependent and Ca2+-independent mechanisms that are regulated by Bax." J Biol Chem 276(22): 19066-71.

Goldstein JC,Waterhouse NJ, et al. (2000). "The coordinate release of cytochrome c during apoptosis is rapid, complete and kinetically invariant." Nat Cell Biol2(3): 156-62.

Goll MG,Kirpekar F, et al. (2006). "Methylation of tRNAAsp by the DNA methyltransferase homolog Dnmt2." Science 311(5759): 395-8.

Green DR and Reed JC (1998). "Mitochondria and apoptosis." Science 281(5381):1309-12.

Greenberg P,Cox C, et al. (1997). "International scoring system for evaluating prognosis in myelodysplastic syndromes." Blood 89(6): 2079-88.

Griffiths GJ,Dubrez L, et al. (1999). "Cell damage-induced conformational changes of the pro-apoptotic protein Bak in vivo precede the onset of apoptosis." J Cell Biol 144(5): 903-14.

Hamilton-Paterson JL (1949). "Pre-leukaemic anaemia." Acta Haematol 2(5): 309-16. Hanahan D and Weinberg RA (2000). "The hallmarks of cancer." Cell 100(1): 57-70. Hellstrom-Lindberg E (1995). "Efficacy of erythropoietin in the myelodysplastic

syndromes: a meta-analysis of 205 patients from 17 studies." Br J Haematol89(1): 67-71.

67

Hellstrom-Lindberg E,Ahlgren T, et al. (1998). "Treatment of anemia in myelodysplastic syndromes with granulocyte colony-stimulating factor plus erythropoietin: results from a randomized phase II study and long-term follow-up of 71 patients." Blood 92(1): 68-75.

Hellstrom-Lindberg E,Gulbrandsen N, et al. (2003). "A validated decision model for treating the anaemia of myelodysplastic syndromes with erythropoietin + granulocyte colony-stimulating factor: significant effects on quality of life." Br J Haematol 120(6): 1037-46.

Hellstrom-Lindberg E,Robert KH, et al. (1992). "A predictive model for the clinical response to low dose ara-C: a study of 102 patients with myelodysplastic syndromes or acute leukaemia." Br J Haematol 81(4): 503-11.

Hellstrom-Lindberg E,Schmidt-Mende J, et al. (2001). "Apoptosis in refractory anaemia with ringed sideroblasts is initiated at the stem cell level and associated with increased activation of caspases." Br J Haematol 112(3): 714-26.

Hendrich B,Abbott C, et al. (1999). "Genomic structure and chromosomal mapping of the murine and human Mbd1, Mbd2, Mbd3, and Mbd4 genes." Mamm Genome 10(9): 906-12.

Hendrich B and Bird A (1998). "Identification and characterization of a family of mammalian methyl-CpG binding proteins." Mol Cell Biol 18(11): 6538-47.

Hengartner MO (2000). "The biochemistry of apoptosis." Nature 407(6805): 770-6. Herman JG (1999). "Hypermethylation of tumor suppressor genes in cancer." Semin

Cancer Biol 9(5): 359-67. Herman JG and Baylin SB (2003). "Gene silencing in cancer in association with

promoter hypermethylation." N Engl J Med 349(21): 2042-54. Herman JG,Civin CI, et al. (1997). "Distinct patterns of inactivation of p15INK4B

and p16INK4A characterize the major types of hematological malignancies." Cancer Res 57(5): 837-41.

Herman JG,Latif F, et al. (1994). "Silencing of the VHL tumor-suppressor gene by DNA methylation in renal carcinoma." Proc Natl Acad Sci U S A 91(21): 9700-4.

Herman JG,Merlo A, et al. (1995). "Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers." Cancer Res 55(20): 4525-30.

Hirai H (2003). "Molecular mechanisms of myelodysplastic syndrome." Jpn J Clin Oncol 33(4): 153-60.

Hirai H,Okada M, et al. (1988). "Relationship between an activated N-ras oncogene and chromosomal abnormality during leukemic progression from myelodysplastic syndrome." Blood 71(1): 256-8.

Hirpara JL,Clement MV, et al. (2001). "Intracellular acidification triggered by mitochondrial-derived hydrogen peroxide is an effector mechanism for drug-induced apoptosis in tumor cells." J Biol Chem 276(1): 514-21.

Hofmann WK and Koeffler HP (2005). "Myelodysplastic syndrome." Annu Rev Med56: 1-16.

Hofmann WK,Lubbert M, et al. (2004). "Myelodysplastic syndromes." Hematol J5(1): 1-8.

Hopkins-Donaldson S,Bodmer JL, et al. (2000). "Loss of caspase-8 expression in highly malignant human neuroblastoma cells correlates with resistance to tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis." Cancer Res 60(16): 4315-9.

68

Horiike S,Yokota S, et al. (1997). "Tandem duplications of the FLT3 receptor gene are associated with leukemic transformation of myelodysplasia." Leukemia11(9): 1442-6.

Howe RB,Porwit-MacDonald A, et al. (2004). "The WHO classification of MDS does make a difference." Blood 103(9): 3265-70.

Hutvagner G and Zamore PD (2002a). "A microRNA in a multiple-turnover RNAi enzyme complex." Science 297(5589): 2056-60.

Hutvagner G and Zamore PD (2002b). "RNAi: nature abhors a double-strand." Curr Opin Genet Dev 12(2): 225-32.

Ishizaki Y,Jacobson MD, et al. (1998). "A role for caspases in lens fiber differentiation." J Cell Biol 140(1): 153-8.

Jaattela M (2004). "Multiple cell death pathways as regulators of tumour initiation and progression." Oncogene 23(16): 2746-56.

Jacobson MD,Weil M, et al. (1997). "Programmed cell death in animal development." Cell 88(3): 347-54.

Jadersten M,Montgomery SM, et al. (2005). "Long-term outcome of treatment of anemia in MDS with erythropoietin and G-CSF." Blood 106(3): 803-11.

Jaenisch R and Bird A (2003). "Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals." Nat Genet 33 Suppl:245-54.

Janowski BA,Younger ST, et al. (2007). "Activating gene expression in mammalian cells with promoter-targeted duplex RNAs." Nat Chem Biol 3(3): 166-73.

Jones PA and Baylin SB (2002). "The fundamental role of epigenetic events in cancer." Nat Rev Genet 3(6): 415-28.

Jones PA and PW Laird (1999). ''Cancer epigenetics comes of age.'' Nat Genet 21(2): 163-7.

Jones PA and Takai D (2001). "The role of DNA methylation in mammalian epigenetics." Science 293(5532): 1068-70.

Kam PC and Ferch NI (2000). "Apoptosis: mechanisms and clinical implications." Anaesthesia 55(11): 1081-93.

Kaminskas E,Farrell A, et al. (2005). "Approval summary: azacitidine for treatment of myelodysplastic syndrome subtypes." Clin Cancer Res 11(10): 3604-8.

Kaneda M,Okano M, et al. (2004). "Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting." Nature429(6994): 900-3.

Kanno S,Shouji A, et al. (2004). "Effects of naringin on cytosine arabinoside (Ara-C)-induced cytotoxicity and apoptosis in P388 cells." Life Sci 75(3): 353-65.

Kantarjian H,Issa JP, et al. (2006). "Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study." Cancer106(8): 1794-803.

Kantarjian H,Oki Y, et al. (2007). "Results of a randomized study of 3 schedules of low-dose decitabine in higher-risk myelodysplastic syndrome and chronic myelomonocytic leukemia." Blood 109(1): 52-7.

Karimi M,Johansson S, et al. (2006). "LUMA (LUminometric Methylation Assay)--a high throughput method to the analysis of genomic DNA methylation." Exp Cell Res 312(11): 1989-95.

Karon M,Sieger L, et al. (1973). "5-Azacytidine: a new active agent for the treatment of acute leukemia." Blood 42(3): 359-65.

Kennedy NJ,Kataoka T, et al. (1999). "Caspase activation is required for T cell proliferation." J Exp Med 190(12): 1891-6.

Khan R,Aggerholm A, et al. (2006). "A pharmacodynamic study of 5-azacytidine in the P39 cell line." Exp Hematol 34(1): 35-43.

69

Kim GD,Ni J, et al. (2002). "Co-operation and communication between the human maintenance and de novo DNA (cytosine-5) methyltransferases." Embo J21(15): 4183-95.

Kim R (2005). "Recent advances in understanding the cell death pathways activated by anticancer therapy." Cancer 103(8): 1551-60.

Kiziltepe T,Hideshima T, et al. (2007). "5-Azacytidine, a DNA methyltransferase inhibitor, induces ATR-mediated DNA double-strand break responses, apoptosis, and synergistic cytotoxicity with doxorubicin and bortezomib against multiple myeloma cells." Mol Cancer Ther 6(6): 1718-27.

Klose RJ and Bird AP (2006). "Genomic DNA methylation: the mark and its mediators." Trends Biochem Sci 31(2): 89-97.

Kornblith AB,Herndon JE, 2nd, et al. (2002). "Impact of azacytidine on the quality of life of patients with myelodysplastic syndrome treated in a randomized phase III trial: a Cancer and Leukemia Group B study." J Clin Oncol 20(10): 2441-52.

Kouzarides T (2002). "Histone methylation in transcriptional control." Curr Opin Genet Dev 12(2): 198-209.

Krajewski S,Tanaka S, et al. (1993). "Investigation of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes." Cancer Res 53(19): 4701-14.

Krammer PH (1999). "CD95(APO-1/Fas)-mediated apoptosis: live and let die." Adv Immunol 71: 163-210.

Krammer PH (2000). "CD95's deadly mission in the immune system." Nature407(6805): 789-95.

Kroemer G,Zamzami N, et al. (1997). "Mitochondrial control of apoptosis." Immunol Today 18(1): 44-51.

Kurata M,Hasegawa M, et al. (2006). "Expression dynamics of drug resistance genes, multidrug resistance 1 (MDR1) and lung resistance protein (LRP) during the evolution of overt leukemia in myelodysplastic syndromes." Exp Mol Pathol81(3): 249-54.

Kuwabara T,Hsieh J, et al. (2004). "A small modulatory dsRNA specifies the fate of adult neural stem cells." Cell 116(6): 779-93.

Lassus P,Opitz-Araya X, et al. (2002). "Requirement for caspase-2 in stress-induced apoptosis before mitochondrial permeabilization." Science 297(5585): 1352-4.

Lee Y,Ahn C, et al. (2003). "The nuclear RNase III Drosha initiates microRNA processing." Nature 425(6956): 415-9.

Lei H,Oh SP, et al. (1996). "De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells." Development 122(10): 3195-205.

Leist M and Jaattela M (2001). "Four deaths and a funeral: from caspases to alternative mechanisms." Nat Rev Mol Cell Biol 2(8): 589-98.

Leone G,Teofili L, et al. (2002). "DNA methylation and demethylating drugs in myelodysplastic syndromes and secondary leukemias." Haematologica87(12): 1324-41.

Leonhardt H,Page AW, et al. (1992). "A targeting sequence directs DNA methyltransferase to sites of DNA replication in mammalian nuclei." Cell71(5): 865-73.

Lepelley P,Soenen V, et al. (1994). "Expression of the multidrug resistance P-glycoprotein and its relationship to hematological characteristics and response to treatment in myelodysplastic syndromes." Leukemia 8(6): 998-1004.

Levine JJ,Stimson-Crider KM, et al. (2003). "Effects of methylation on expression of TMS1/ASC in human breast cancer cells." Oncogene 22(22): 3475-88.

70

Lewis JD,Meehan RR, et al. (1992). "Purification, sequence, and cellular localization of a novel chromosomal protein that binds to methylated DNA." Cell 69(6): 905-14.

Li E,Bestor TH, et al. (1992). "Targeted mutation of the DNA methyltransferase gene results in embryonic lethality." Cell 69(6): 915-26.

Li LC,Okino ST, et al. (2006). "Small dsRNAs induce transcriptional activation in human cells." Proc Natl Acad Sci U S A 103(46): 17337-42.

Li LH,Olin EJ, et al. (1970). "Cytotoxicity and mode of action of 5-azacytidine on L1210 leukemia." Cancer Res 30(11): 2760-9.

Linman JW and Bagby C, Jr. (1976). "The preleukemic syndrome: clinical and laboratory features, natural course, and management." Nouv Rev Fr Hematol Blood Cells 17(1-2): 11-31.

List A,Beran M, et al. (2003). "Opportunities for Trisenox (arsenic trioxide) in the treatment of myelodysplastic syndromes." Leukemia 17(8): 1499-507.

List A,Dewald G, et al. (2006a). "Lenalidomide in the myelodysplastic syndrome with chromosome 5q deletion." N Engl J Med 355(14): 1456-65.

List AF,Baker AF, et al. (2006b). "Lenalidomide: targeted anemia therapy for myelodysplastic syndromes." Cancer Control 13 Suppl: 4-11.

Lithgow T,van Driel R, et al. (1994). "The protein product of the oncogene bcl-2 is a component of the nuclear envelope, the endoplasmic reticulum, and the outer mitochondrial membrane." Cell Growth Differ 5(4): 411-7.

Liu Q,Rand TA, et al. (2003). "R2D2, a bridge between the initiation and effector steps of the Drosophila RNAi pathway." Science 301(5641): 1921-5.

Liu X,Kim CN, et al. (1996). "Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c." Cell 86(1): 147-57.

Llave C,Xie Z, et al. (2002). "Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA." Science 297(5589): 2053-6.

Lockshin RA and Zakeri Z (2004). "Apoptosis, autophagy, and more." Int J Biochem Cell Biol 36(12): 2405-19.

Lu LJ and Randerath K (1979). "Effects of 5-azacytidine on transfer RNA methyltransferases." Cancer Res 39(3): 940-9.

Luger S and Sacks N (2002). "Bone marrow transplantation for myelodysplastic syndrome--who? when? and which?" Bone Marrow Transplant 30(4): 199-206.

Luzzatto A (1907). "Sull anemia grave megaloblastica senza reporto ematologico correspondente (anemia pseudoaplastica)." Riv Veneta Med Venezia 47: 193-212.

Malcovati L,Germing U, et al. (2007). "Time-dependent prognostic scoring system for predicting survival and leukemic evolution in myelodysplastic syndromes." J Clin Oncol 25(23): 3503-10.

Mansat-de Mas V,Bezombes C, et al. (1999). "Implication of radical oxygen species in ceramide generation, c-Jun N-terminal kinase activation and apoptosis induced by daunorubicin." Mol Pharmacol 56(5): 867-74.

Martin C and Zhang Y (2005). "The diverse functions of histone lysine methylation." Nat Rev Mol Cell Biol 6(11): 838-49.

Martin SJ (1993). "Apoptosis: suicide, execution or murder?" Trends Cell Biol 3(5): 141-4.

Mayer B and Oberbauer R (2003). "Mitochondrial regulation of apoptosis." News Physiol Sci 18: 89-94.

Medema RH (2004). "Optimizing RNA interference for application in mammalian cells." Biochem J 380(Pt 3): 593-603.

71

Meehan RR,Lewis JD, et al. (1989). "Identification of a mammalian protein that binds specifically to DNA containing methylated CpGs." Cell 58(3): 499-507.

Merlo A, JG Herman, L Mao, DJ Lee, E Gabrielson, PC Burger, SB Baylin and D Sidransky (1995). ''5' CpG island methylation is associated with transcriptional silencing of the tumour suppressor p16/CDKN2/MTS1 in human cancers.'' Nat Med 1(7): 686-92.

Mignotte B and JL Vayssiere (1998). ''Mitochondria and apoptosis.'' Eur J Biochem252(1): 1-15.

Miller KB, K Kim, FS Morrison, JN Winter, JM Bennett, RS Neiman, DR Head, PA Cassileth and MJ O'Connell (1992). ''The evaluation of low-dose cytarabine in the treatment of myelodysplastic syndromes: a phase-III intergroup study.'' Ann Hematol 65(4): 162-8.

Minn AJ, P Velez, SL Schendel, H Liang, SW Muchmore, SW Fesik, M Fill and CB Thompson (1997). ''Bcl-x(L) forms an ion channel in synthetic lipid membranes.'' Nature 385(6614): 353-7.

Miwa M and Masutani M (2007). "PolyADP-ribosylation and cancer." Cancer Sci.Molldrem JJ,Caples M, et al. (1997). "Antithymocyte globulin for patients with

myelodysplastic syndrome." Br J Haematol 99(3): 699-705. Molldrem JJ,Leifer E, et al. (2002). "Antithymocyte globulin for treatment of the

bone marrow failure associated with myelodysplastic syndromes." Ann Intern Med 137(3): 156-63.

Morales-Ramirez P,Vallarino-Kelly T, et al. (2007). "Mechanisms of DNA breaks induction in vivo by 5-azacytidine: paths of micronucleus induction by azaC." J Appl Toxicol.

Morel P, M Hebbar, JL Lai, A Duhamel, C Preudhomme, E Wattel, F Bauters and P Fenaux (1993). ''Cytogenetic analysis has strong independent prognostic value in de novo myelodysplastic syndromes and can be incorporated in a new scoring system: a report on 408 cases.'' Leukemia 7(9): 1315-23.

Muchmore SW,Sattler M, et al. (1996). "X-ray and NMR structure of human Bcl-xL, an inhibitor of programmed cell death." Nature 381(6580): 335-41.

Mufti GJ (2004). "Pathobiology, classification, and diagnosis of myelodysplastic syndrome." Best Pract Res Clin Haematol 17(4): 543-57.

Mundle SD (2001). "Apoptosis in MDS: a new perspective." Cancer Treat Res 108:101-9.

Mundle SD,Ali A, et al. (1999a). "Evidence for involvement of tumor necrosis factor-alpha in apoptotic death of bone marrow cells in myelodysplastic syndromes." Am J Hematol 60(1): 36-47.

Mundle SD,Reza S, et al. (1999b). "Correlation of tumor necrosis factor alpha (TNF alpha) with high Caspase 3-like activity in myelodysplastic syndromes." Cancer Lett 140(1-2): 201-7.

Munoz-Gamez JA,Martin-Oliva D, et al. (2005). "PARP inhibition sensitizes p53-deficient breast cancer cells to doxorubicin-induced apoptosis." Biochem J386(Pt 1): 119-25.

Murakami T,Li X, et al. (1995). "Induction of apoptosis by 5-azacytidine: drug concentration-dependent differences in cell cycle specificity." Cancer Res55(14): 3093-8.

Napoli C,Lemieux C, et al. (1990). "Introduction of a Chimeric Chalcone Synthase Gene into Petunia Results in Reversible Co-Suppression of Homologous Genes in trans." Plant Cell 2(4): 279-289.

Negrin RS, DH Haeuber, A Nagler, Y Kobayashi, J Sklar, T Donlon, M Vincent and PL Greenberg (1990). ''Maintenance treatment of patients with

72

myelodysplastic syndromes using recombinant human granulocyte colony-stimulating factor.'' Blood 76(1): 36-43.

Negrin RS, R Stein, K Doherty, J Cornwell, J Vardiman, S Krantz and PL Greenberg (1996). ''Maintenance treatment of the anemia of myelodysplastic syndromes with recombinant human granulocyte colony-stimulating factor and erythropoietin: evidence for in vivo synergy.'' Blood 87(10): 4076-81.

Ng HH,Zhang Y, et al. (1999). "MBD2 is a transcriptional repressor belonging to the MeCP1 histone deacetylase complex." Nat Genet 23(1): 58-61.

Okano M,Bell DW, et al. (1999). "DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development." Cell 99(3): 247-57.

Okano M,Xie S, et al. (1998a). "Cloning and characterization of a family of novel mammalian DNA (cytosine-5) methyltransferases." Nat Genet 19(3): 219-20.

Okano M,Xie S, et al. (1998b). "Dnmt2 is not required for de novo and maintenance methylation of viral DNA in embryonic stem cells." Nucleic Acids Res26(11): 2536-40.

Parker JE, KL Fishlock, A Mijovic, B Czepulkowski, A Pagliuca and GJ Mufti (1998). '''Low-risk' myelodysplastic syndrome is associated with excessive apoptosis and an increased ratio of pro- versus anti-apoptotic bcl-2-related proteins.'' Br J Haematol 103(4): 1075-82.

Parker JE and GJ Mufti (2000). ''Excessive apoptosis in low risk myelodysplastic syndromes (MDS).'' Leuk Lymphoma 40(1-2): 1-24.

Parker JE and GJ Mufti (2004). ''The myelodysplastic syndromes: a matter of life or death.'' Acta Haematol 111(1-2): 78-99.

Pervaiz S and Clement MV (2002). "A permissive apoptotic environment: function of a decrease in intracellular superoxide anion and cytosolic acidification." Biochem Biophys Res Commun 290(4): 1145-50.

Poulain S,Lepelley P, et al. (2000). "Expression of the multidrug resistance-associated protein in myelodysplastic syndromes." Br J Haematol 110(3): 591-8.

Prokhortchouk A,Hendrich B, et al. (2001). "The p120 catenin partner Kaiso is a DNA methylation-dependent transcriptional repressor." Genes Dev 15(13): 1613-8.

Quesnel B,Guillerm G, et al. (1998). "Methylation of the p15(INK4b) gene in myelodysplastic syndromes is frequent and acquired during disease progression." Blood 91(8): 2985-90.

Quillet-Mary A,Jaffrezou JP, et al. (1997). "Implication of mitochondrial hydrogen peroxide generation in ceramide-induced apoptosis." J Biol Chem 272(34): 21388-95.

Rajapaksa R,Ginzton N, et al. (1996). "Altered oncoprotein expression and apoptosis in myelodysplastic syndrome marrow cells." Blood 88(11): 4275-87.

Raza A (2000). "Anti-TNF therapies in rheumatoid arthritis, Crohn's disease, sepsis, and myelodysplastic syndromes." Microsc Res Tech 50(3): 229-35.

Raza A,Reeves JA, et al. (2007). "Phase II study of lenalidomide in transfusion-dependent, low- and intermediate-1-risk myelodysplastic syndromes with karyotypes other than deletion 5q." Blood.

Reik W (2007). "Stability and flexibility of epigenetic gene regulation in mammalian development." Nature 447(7143): 425-32.

Reu FJ, SI Bae, L Cherkassky, DW Leaman, D Lindner, N Beaulieu, AR MacLeod and EC Borden (2006). ''Overcoming resistance to interferon-induced apoptosis of renal carcinoma and melanoma cells by DNA demethylation.'' J Clin Oncol 24(23): 3771-9.

73

Rhee I,Bachman KE, et al. (2002). "DNMT1 and DNMT3b cooperate to silence genes in human cancer cells." Nature 416(6880): 552-6.

Rhee I,Jair KW, et al. (2000). "CpG methylation is maintained in human cancer cells lacking DNMT1." Nature 404(6781): 1003-7.

Rhoades CP and Barker WH (1938). "Refractory anemia: an analysis of one hundred cases." JAMA 110: 794-96.

Ribera JM,Cervantes F, et al. (1987). "A multivariate analysis of prognostic factors in chronic myelomonocytic leukaemia according to the FAB criteria." Br J Haematol 65(3): 307-11.

Ribrag V,Suzan F, et al. (2003). "Phase II trial of CPT-11 in myelodysplastic syndromes with excess of marrow blasts." Leukemia 17(2): 319-22.

Rice L (2006). "Drug evaluation: AMG-531 for the treatment of thrombocytopenias." Curr Opin Investig Drugs 7(9): 834-41.

Riedl SJ and Salvesen GS (2007). "The apoptosome: signalling platform of cell death." Nat Rev Mol Cell Biol 8(5): 405-13.

Robertson JD,Enoksson M, et al. (2002). "Caspase-2 acts upstream of mitochondria to promote cytochrome c release during etoposide-induced apoptosis." J Biol Chem 277(33): 29803-9.

Robertson KD,Ait-Si-Ali S, et al. (2000). "DNMT1 forms a complex with Rb, E2F1 and HDAC1 and represses transcription from E2F-responsive promoters." Nat Genet 25(3): 338-42.

Rosato RR,Almenara JA, et al. (2003). "The histone deacetylase inhibitor MS-275 promotes differentiation or apoptosis in human leukemia cells through a process regulated by generation of reactive oxygen species and induction of p21CIP1/WAF1 1." Cancer Res 63(13): 3637-45.

Rosenfeld C and List A (2000). "A hypothesis for the pathogenesis of myelodysplastic syndromes: implications for new therapies." Leukemia 14(1):2-8.

Rountree MR, KE Bachman and SB Baylin (2000). ''DNMT1 binds HDAC2 and a new co-repressor, DMAP1, to form a complex at replication foci.'' Nat Genet25(3): 269-77.

Runde V, T de Witte, R Arnold, A Gratwohl, J Hermans, A van Biezen, D Niederwieser, M Labopin, MP Walter-Noel, A Bacigalupo, N Jacobsen, P Ljungman, E Carreras, HJ Kolb, C Aul and J Apperley (1998). ''Bone marrow transplantation from HLA-identical siblings as first-line treatment in patients with myelodysplastic syndromes: early transplantation is associated with improved outcome. Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation.'' Bone Marrow Transplant 21(3): 255-61.

Saarni MI and Linman JW (1973). "Preleukemia. The hematologic syndrome preceding acute leukemia." Am J Med 55(1): 38-48.

Saito M and Ishikawa F (2002). "The mCpG-binding domain of human MBD3 does not bind to mCpG but interacts with NuRD/Mi2 components HDAC1 and MTA2." J Biol Chem 277(38): 35434-9.

Sansom OJ,Maddison K, et al. (2007). "Mechanisms of disease: methyl-binding domain proteins as potential therapeutic targets in cancer." Nat Clin Pract Oncol 4(5): 305-15.

Sansom OJ,Zabkiewicz J, et al. (2003). "MBD4 deficiency reduces the apoptotic response to DNA-damaging agents in the murine small intestine." Oncogene22(46): 7130-6.

Sanz GF, MA Sanz, T Vallespi, MC Canizo, M Torrabadella, S Garcia, D Irriguible and JF San Miguel (1989). ''Two regression models and a scoring system for

74

predicting survival and planning treatment in myelodysplastic syndromes: a multivariate analysis of prognostic factors in 370 patients.'' Blood 74(1): 395-408.

Sarin A,Wu ML, et al. (1996). "Different interleukin-1 beta converting enzyme (ICE) family protease requirements for the apoptotic death of T lymphocytes triggered by diverse stimuli." J Exp Med 184(6): 2445-50.

Schendel SL,Xie Z, et al. (1997). "Channel formation by antiapoptotic protein Bcl-2." Proc Natl Acad Sci U S A 94(10): 5113-8.

Schmidt-Mende J, V Gogvadze, E Hellstrom-Lindberg and B Zhivotovsky (2006). ''Early mitochondrial alterations in ATRA-induced cell death.'' Cell Death Differ 13(1): 119-28.

Schmitz I,Kirchhoff S, et al. (2000). "Regulation of death receptor-mediated apoptosis pathways." Int J Biochem Cell Biol 32(11-12): 1123-36.

Schreiber V,Dantzer F, et al. (2006). "Poly(ADP-ribose): novel functions for an old molecule." Nat Rev Mol Cell Biol 7(7): 517-28.

Shen L,Kantarjian H, et al. (2005). "CpG Island Methylation Is a Poor Prognostic Factors in Myelodysplastic Syndrome Patients and Is Reversed by Decitabine Therapy-Results of a Phase III Randomized Study." ASH Annual Meeting Abstracts 106(11): Abstract #790.

Shimizu S,Narita M, et al. (1999). "Bcl-2 family proteins regulate the release of apoptogenic cytochrome c by the mitochondrial channel VDAC." Nature399(6735): 483-7.

Silverman LR, EP Demakos, BL Peterson, AB Kornblith, JC Holland, R Odchimar-Reissig, RM Stone, D Nelson, BL Powell, CM DeCastro, J Ellerton, RA Larson, CA Schiffer and JF Holland (2002). ''Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B.'' J Clin Oncol 20(10): 2429-40.

Silverman LR,Holland JF, et al. (1993). "Effects of treatment with 5-azacytidine on the in vivo and in vitro hematopoiesis in patients with myelodysplastic syndromes." Leukemia 7 Suppl 1: 21-9.

Simizu S,Umezawa K, et al. (1998). "Induction of hydrogen peroxide production and Bax expression by caspase-3(-like) proteases in tyrosine kinase inhibitor-induced apoptosis in human small cell lung carcinoma cells." Exp Cell Res238(1): 197-203.

Sloand EM,Kim S, et al. (2002). "Fas-mediated apoptosis is important in regulating cell replication and death in trisomy 8 hematopoietic cells but not in cells with other cytogenetic abnormalities." Blood 100(13): 4427-32.

Snyder RD and Lachmann PJ (1989). "Differential effects of 5-azacytidine and 5-azadeoxycytidine on cytotoxicity, DNA-strand breaking and repair of X-ray-induced DNA damage in HeLa cells." Mutat Res 226(3): 185-90.

Span LF,Vierwinden G, et al. (2005). "Programmed cell death is an intrinsic feature of MDS progenitors, predominantly found in the cluster-forming cells." Exp Hematol 33(4): 435-42.

Srinivasula SM,Ahmad M, et al. (1998). "Autoactivation of procaspase-9 by Apaf-1-mediated oligomerization." Mol Cell 1(7): 949-57.

Steuber CP,Krischer J, et al. (1996). "Therapy of refractory or recurrent childhood acute myeloid leukemia using amsacrine and etoposide with or without azacitidine: a Pediatric Oncology Group randomized phase II study." J Clin Oncol 14(5): 1521-5.

Strahl BD and Allis CD (2000). "The language of covalent histone modifications." Nature 403(6765): 41-5.

75

Struhl K (1998). "Histone acetylation and transcriptional regulatory mechanisms." Genes Dev 12(5): 599-606.

Sugimoto K,Hirano N, et al. (1993). "Mutations of the p53 gene in myelodysplastic syndrome (MDS) and MDS-derived leukemia." Blood 81(11): 3022-6.

Tabara H,Yigit E, et al. (2002). "The dsRNA binding protein RDE-4 interacts with RDE-1, DCR-1, and a DExH-box helicase to direct RNAi in C. elegans." Cell109(7): 861-71.

Tan YJ,Beerheide W, et al. (1999). "Biophysical characterization of the oligomeric state of Bax and its complex formation with Bcl-XL." Biochem Biophys Res Commun 255(2): 334-9.

Taussig DC,Davies AJ, et al. (2003). "Durable remissions of myelodysplastic syndrome and acute myeloid leukemia after reduced-intensity allografting." J Clin Oncol 21(16): 3060-5.

Tehranchi R, B Fadeel, AM Forsblom, B Christensson, J Samuelsson, B Zhivotovsky and E Hellstrom-Lindberg (2003). ''Granulocyte colony-stimulating factor inhibits spontaneous cytochrome c release and mitochondria-dependent apoptosis of myelodysplastic syndrome hematopoietic progenitors.'' Blood101(3): 1080-6.

Tehranchi R, R Invernizzi, A Grandien, B Zhivotovsky, B Fadeel, AM Forsblom, E Travaglino, J Samuelsson, R Hast, L Nilsson, M Cazzola, R Wibom and E Hellstrom-Lindberg (2005). ''Aberrant mitochondrial iron distribution and maturation arrest characterize early erythroid precursors in low-risk myelodysplastic syndromes.'' Blood 106(1): 247-53.

Tien HF,Tang JH, et al. (2001). "Methylation of the p15(INK4B) gene in myelodysplastic syndrome: it can be detected early at diagnosis or during disease progression and is highly associated with leukaemic transformation." Br J Haematol 112(1): 148-54.

Turner BM (2002). "Cellular memory and the histone code." Cell 111(3): 285-91. Uchida T,Kinoshita T, et al. (1997). "Hypermethylation of the p15INK4B gene in

myelodysplastic syndromes." Blood 90(4): 1403-9. Ueno M,Katayama K, et al. (2002). "Mechanisms of 5-azacytidine (5AzC)-induced

toxicity in the rat foetal brain." Int J Exp Pathol 83(3): 139-50. van Loo G,Saelens X, et al. (2002). "The role of mitochondrial factors in apoptosis: a

Russian roulette with more than one bullet." Cell Death Differ 9(10): 1031-42. Vardiman JW, NL Harris and RD Brunning (2002). ''The World Health Organization

(WHO) classification of the myeloid neoplasms.'' Blood 100(7): 2292-302. Vey N (2004). ''Arsenic trioxide for the treatment of myelodysplastic syndromes.''

Expert Opin Pharmacother 5(3): 613-21. Vidal DO,Lopes LF, et al. (2007). "Drug resistance and methylation in

myelodysplastic syndrome." Curr Pharm Biotechnol 8(2): 77-81. Villunger A,Michalak EM, et al. (2003). "p53- and drug-induced apoptotic responses

mediated by BH3-only proteins puma and noxa." Science 302(5647): 1036-8. Wang X (2001). "The expanding role of mitochondria in apoptosis." Genes Dev

15(22): 2922-33. Wang XM,Wang X, et al. (1998). "Effects of 5-azacytidine and butyrate on

differentiation and apoptosis of hepatic cancer cell lines." Ann Surg 227(6): 922-31.

Watanabe S,Ichimura T, et al. (2003). "Methylated DNA-binding domain 1 and methylpurine-DNA glycosylase link transcriptional repression and DNA repair in chromatin." Proc Natl Acad Sci U S A 100(22): 12859-64.

Wattel E,De Botton S, et al. (1997). "Long-term follow-up of de novo myelodysplastic syndromes treated with intensive chemotherapy: incidence of

76

long-term survivors and outcome of partial responders." Br J Haematol 98(4): 983-91.

Wattel E,Solary E, et al. (1998). "Quinine improves the results of intensive chemotherapy in myelodysplastic syndromes expressing P glycoprotein: results of a randomized study." Br J Haematol 102(4): 1015-24.

Wattel E,Solary E, et al. (1999). "A prospective study of autologous bone marrow or peripheral blood stem cell transplantation after intensive chemotherapy in myelodysplastic syndromes. Groupe Francais des Myelodysplasies. Group Ouest-Est d'etude des Leucemies aigues myeloides." Leukemia 13(4): 524-9.

Weaver IC,Cervoni N, et al. (2004). "Epigenetic programming by maternal behavior." Nat Neurosci 7(8): 847-54.

Weil M, MC Raff and VM Braga (1999). ''Caspase activation in the terminal differentiation of human epidermal keratinocytes.'' Curr Biol 9(7): 361-4.

Wolter KG,Hsu YT, et al. (1997). "Movement of Bax from the cytosol to mitochondria during apoptosis." J Cell Biol 139(5): 1281-92.

Wong AH,Gottesman, II, et al. (2005). "Phenotypic differences in genetically identical organisms: the epigenetic perspective." Hum Mol Genet 14 Spec No 1: R11-8.

Yoder JA,Soman NS, et al. (1997). "DNA (cytosine-5)-methyltransferases in mouse cells and tissues. Studies with a mechanism-based probe." J Mol Biol 270(3): 385-95.

Yokota S, H Kiyoi, M Nakao, T Iwai, S Misawa, T Okuda, Y Sonoda, T Abe, K Kahsima, Y Matsuo and T Naoe (1997). ''Internal tandem duplication of the FLT3 gene is preferentially seen in acute myeloid leukemia and myelodysplastic syndrome among various hematological malignancies. A study on a large series of patients and cell lines.'' Leukemia 11(10): 1605-9.

Yoshida Y, T Nakahata, A Shibata, M Takahashi, Y Moriyama, K Kaku, T Masaoka, T Kaneko and S Miwa (1995). ''Effects of long-term treatment with recombinant human granulocyte-macrophage colony-stimulating factor in patients with myelodysplastic syndrome.'' Leuk Lymphoma 18(5-6): 457-63.

Zang DY,Goodwin RG, et al. (2001). "Expression of tumor necrosis factor-related apoptosis-inducing ligand, Apo2L, and its receptors in myelodysplastic syndrome: effects on in vitro hemopoiesis." Blood 98(10): 3058-65.

Zermati Y,Garrido C, et al. (2001). "Caspase activation is required for terminal erythroid differentiation." J Exp Med 193(2): 247-54.

Zeschnigk M,Lich C, et al. (1997). "A single-tube PCR test for the diagnosis of Angelman and Prader-Willi syndrome based on allelic methylation differences at the SNRPN locus." Eur J Hum Genet 5(2): 94-8.

Zhivotovsky B and S Orrenius (2005). ''Caspase-2 function in response to DNA damage.'' Biochem Biophys Res Commun 331(3): 859-67.

Zou H,Henzel WJ, et al. (1997). "Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3." Cell90(3): 405-13.

77

10. APPENDICES