Thesis for doctoral degree (Ph.D.)2008

Neural Stem Cell-Mediated EnhancedGene Therapy of Malignant Gliomas

Using a Novel Plant Thymidine Kinase

Zahidul Khan

Thesis for doctoral degree (Ph.D.) 2008

Zahidul Khan

Neu

ral Stem C

ell-Mediated E

nhan

ced Gen

e Therapy of M

alignan

t Gliom

as

From the Department of Clinical Neuroscience

Karolinska Institutet, Stockholm, Sweden

NEURAL STEM CELL-MEDIATED ENHANCED GENE THERAPY OF MALIGNANT GLIOMAS

USING A NOVEL PLANT THYMIDINE KINASE

Zahidul Khan

Stockholm 2008

All previously published papers were reproduced with permission from the publisher. The cover picture is from my own experiments showing a human malignant glioma xenograft. Published by Karolinska Institutet. © Zahidul Khan, 2008 ISBN 978-91-7409-147-2

2008

Gårdsvägen 4, 169 70 Solna

Printed by

At last, when I woke from my slumber and opened my eyes, I saw thee standing by me, flooding my sleep with thy smile. How I had feared that the path was long and wearisome, and the struggle to reach thee was hard!

From “The Journey” by Rabindranath Tagore (1861 – 1941),

A Bengali poet, novelist and composer, the first Asian Nobel laureate in literature (1913)

To my parents

ABSTRACT Malignant gliomas are the most common brain tumors in adults. Approximately, 190,000 new patients are diagnosed with malignant gliomas worldwide every year. Unfortunately, the prognosis of malignant gliomas still remains very poor, despite advances in management. One of the major complicating factors in managing malignant gliomas is their invasiveness, which leads to formation of multiple tumor satellites and thereby inevitable recurrence. Therefore, it is imperative to develop new treatment modalities which effectively address this issue. Suicide gene therapy strategies have been applied against gliomas, but their success have been limited due to inefficient gene transfer technologies, inappropriate enzymatic activities, limited prodrug penetration into the brain, or low bystander effects. The studies presented in this thesis identify a novel suicide gene which can circumvent gene and prodrug related limitations, utilize neural stem cells as gene delivery vehicle and use the histone deacetylase inhibitor 4-phenylbutyrate (4-PB) as modulators of the gene therapy.

First, using the herpes simplex virus 1 thymidine kinase/ ganciclovir (HSV-tk/GCV) gene therapy approach, we investigated 4-PB for its effect on HSV-tk gene expression and the bystander effect in glioma cell cultures. The bystander effect is to a great extent mediated by gap junction communication (GJC). Gap junctions are intercellular channels that permit small water soluble molecules to be exchanged between cells. Each cell contributes half of the channel composed of connexin proteins (Cxs). Cx43 is shown to be involved in facilitating the bystander effect. We found that 4-PB upregulated the Cx43 level, amplified the GJC, as well as potentiated the recombinant HSV-tk gene expression in rat glioma cells stably expressing HSV-tk. As a result, a dramatic enhancement of bystander effect by administration of the HSV-tk prodrug GCV together with 4-PB was found.

Neural stem cells show extensive tropism for brain tumors and could be exploited to carry and deliver suicide genes. We characterized the human embryonic midbrain stem cell line NGC-407 and found that these cells possessed migratory capacity toward intracranial glioblastoma xenografts. Interestingly, 4-PB was also found to enhance functional GJC between NGC-407 cells and human glioma cells. This finding was supported by upregulation, increased membranous deposition, as well as decreased phosphorylation of Cx43 in NGC-407 cells. 4-PB effect on the differentiation of NGC-407 cells is also needed to be considered, as epigenetic mechanisms are known to play a pivotal role in embryonic development. We found that 4-PB inhibited both astrocytic and neuronal differentiation of NGC-407 cells and preserved their immature phenotype. How the enhancement of GJC together with the differentiation inhibitory effect of 4-PB in NGC-407 cells influence the stem cell-mediated suicide gene therapy in vivo remains to be investigated.

Finally and most importantly, we present here, the tomato thymidine kinase 1 (ToTK1) from its identification and characterization to its in vitro and in vivo testing as a suicide gene. Deposited GenBank sequences were searched for human thymidine kinase 1 homologs in plants and ToTK1 was identified. ToTK1 was found specific for the nucleoside analog azidothymidine (AZT), known to successfully pass the blood-brain barrier. ToTK1 efficiently phosphorylated AZT and surprisingly also phosphorylated AZT monophosphate into AZT diphosphate, an additional rate limiting step for AZT activation. The kinetic analysis and in vitro testing in human glioma cells indicated that ToTK1/AZT combination is superior to HSV-tk/GCV approach as a suicide gene therapy strategy. When NGC-407 cells expressing ToTK1 were implanted into nude rat brain along with U87MG human glioblastoma cells, the glioblastoma xenograft growth was substantially attenuated upon exposure to AZT as indicated by magnetic resonance imaging and autopsy analysis. Significantly improved survival was also observed for the treated animals, showing that the ToTK1/AZT suicide gene prodrug combination with stem cell-mediated gene delivery represents a promising new treatment of malignant gliomas.

Key words: malignant glioma, suicide gene therapy, bystander effect, gap junction, neural stem cell, 4-phenylbutyrate, epigenetics, cell differentiation

LIST OF PUBLICATIONS This thesis is based on the following papers, which will be referred to in the text by their roman numerals (I-IV)

I. Ole Ammerpohl, Dorit Thormeyer, Zahidul Khan, Ioulia B. Appelskog, Zoran Gojkovic, Per M. Almqvist and Tomas J. Ekström. HDACi phenylbutyrate increases bystander killing of HSV-tk transfected glioma cells. Biochem Biophys Res Commun. 2004 Nov 5;324(1):8-14.

II. Zahidul Khan*, Monira Akhtar, Thomas Asklund, Bengt Juliusson, Per M. Almqvist and Tomas J. Ekström. HDAC inhibition amplifies gap junction communication in neural progenitors: potential for cell-mediated enzyme prodrug therapy. Exp Cell Res. 2007 Aug 1;313(13):2958-67.

III. Zahidul Khan*, Monira Akhtar and Tomas J. Ekström. HDAC inhibitor 4-phenylbutyrate preserves immature phenotype of human embryonic midbrain stem cells: implication of the involvement of DNA hypomethylation. Submitted manuscript

IV. Zahidul Khan, Wolfgang Knecht, Mette Willer, Elzbieta Rozpedowska, Peter Kristoffersen, Anders Ranegaard Clausen, Birgitte Munch-Petersen, Zoran Gojkovic, Jure Piškur and Tomas J. Ekström. Plant thymidine kinase 1 is an efficient suicide gene for brain tumor therapy. Submitted manuscript

* Corresponding author

RELATED PUBLICATIONS Shucheng Yin, Ada Girnita, Zahidul Khan, Sandra Andersson, Magnus Axelson, Monica Nister, Olle Larsson, Tomas J Ekström, Thomas Strömberg and Leonard Girnita. Targeting insulin-like growth factor-1 receptor (IGF-1R) by picropodophyllin (PPP) as a treatment option for glioblastoma. Submitted manuscript Markus Dehnhardt, Myroslav V. Zoriy, Zahidul Khan, Guido Reifenberger, Tomas J. Ekström, J. Sabine Becker, Karl Zilles and Andreas Bauer. Element distribution is altered in a zone surrounding human glioblastoma multiforme. J Trace Elem Med Biol. 2008;22(1):17-23. Thomas Asklund, Anna-Maria Marino, Christofer Juhlin, Anastasios Sofiadis, Zahidul Khan, Catharina Larsson, Roger Henriksson and Tomas J Ekström. Cooperative cytotoxicity by HDAC inhibitors and induction of Cx43 by RTKIs in GBM cells. Submitted manuscript

CONTENTS 1 Introduction................................................................................................ 1

1.1 Malignant gliomas ................................................................................ 1 1.1.1 Origin of gliomas ......................................................................................................1 1.1.2 Histopathological features ........................................................................................2 1.1.3 Molecular features ....................................................................................................2 1.1.4 Treatment hurdles .....................................................................................................2 1.1.5 New therapeutic strategies........................................................................................3

1.2 Gene therapy: suicide gene therapy .................................................... 4 1.2.1 Suicide genes ............................................................................................................4 1.2.2 Properties of the enzyme encoded by a suicide gene...............................................6 1.2.3 Prodrugs and the blood-brain barrier........................................................................6 1.2.4 Gene delivery systems ..............................................................................................7 1.2.5 Bystander effect ........................................................................................................9

1.3 Gap junction communication (GJC) ................................................... 11 1.3.1 Regulation of gap junction communication ...........................................................12

1.4 Neural stem cell (NSC) ...................................................................... 13 1.4.1 Markers for neural stem cells .................................................................................13 1.4.2 Neural stem cell differentiation ..............................................................................14 1.4.3 Potentials of neural stem cells ................................................................................14

1.5 Epigenetics......................................................................................... 15 1.5.1 Histone modifications: the acetylation...................................................................15 1.5.2 DNA methylation....................................................................................................16 1.5.3 Relationship between epigenetic mechanisms.......................................................17 1.5.4 HDAC inhibitors and suicide gene therapy ...........................................................18

2 Aims of the study..................................................................................... 20

3 Materials and methods............................................................................ 21

3.1 Materials and cell lines ....................................................................... 21 3.1.1 Drugs and chemicals...............................................................................................21 3.1.2 Malignant glioma cell lines ....................................................................................21 3.1.3 Neural stem cell lines..............................................................................................21

3.2 Methods.............................................................................................. 21 3.2.1 The neural stem cell line NGC-407 and its characterization (II) ..........................21 3.2.2 Identification and characterization of a new suicide gene (IV).............................22 3.2.3 Vector construction and establishment of stable cell clones .................................23 3.2.4 FACS analysis (I)....................................................................................................24 3.2.5 Fluorescent dye transfer test for functional GJC (I and II)....................................25 3.2.6 Cell proliferation and viability assays (I, II and IV) ..............................................25 3.2.7 LUMA analysis (III) ...............................................................................................25 3.2.8 Animal studies (IV) ................................................................................................26 3.2.9 Experimental magnetic resonance imaging (MRI) (IV)........................................27

4 Results and discussions ........................................................................ 28

4.1 4-PB enhances GJC-mediated bystander effect (I) ........................... 28

4.2 4-PB increases HSV-tk transgene expression (I) .............................. 28

4.3 NGC-407 cells possess NSC characteristics (II) ............................... 29

4.4 NGC-407 cells show tropism for malignant gliomas (IV) ................... 29

4.5 4-PB enhances GJC between NGC-407 and glioma cells (II)........... 30

4.6 4-PB preserves stemness of NGC-407 cells (III) ............................... 30

4.7 4-PB induces Global DNA hypomethylation in NGC-407 cells (III) ... 31

4.8 ToTK1 is a novel suicide gene (IV) .................................................... 32

4.9 NSC-mediated ToTK1/AZT therapy for malignant gliomas (IV) ........ 32

5 Conclusion and Future Perspectives.................................................... 34

6 Acknowledgements................................................................................. 36

7 References ............................................................................................... 38

LIST OF ABBREVIATIONS 4-PB 4- phenylbutyrate AGA 18 -glycirrhetinic acid AZT Azidothymidine, Zidovudine AZT-DP Azidothymidine diphosphate AZT-MP Azidothymidine monophosphate bFGF Basic fibroblast growth factor Cx43 Connexin 43 dN Deoxyribonucloside dNK Deoxyribonucleoside kinase dNMP Deoxyribonucloside monophosphate DNMT DNA methyltransferase DTT Dithiothreitol EGF Epidermal growth factor EGFR FACS

Epidermal growth factor receptor Fluorescence-activated cell sorting

GBM Glioblastoma multiforme GCV Ganciclovir GFAP Glial fibrillary acidic protein GFP Green fluorescent protein GJC Gap junction communication HDAC Histone deacetylase HSV-tk Herpes simplex virus 1 thymidine kinase LUMA Luminometric methylation assay MAPK Mitogen-activated protein kinases MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NPC Neural progenitor cell NSC Neural stem cells PBS Phosphate buffered saline PDGFR PKC

Platelet-derived growth factor receptor Protein kinase C

PTEN RFP

Phosphatase and tensin homolog Red fluorescent protein

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis TBST Tris-buffered saline tween-20 TDP Thymidine diphopshate Thd Thymidine TK1 Thymidine kinase 1 TMP Thymidine monophosphate ToTK1 Tomato thymidine kinase 1 TSA Trichostatin A WHO World Health Organization XTT Sodium 3´-[1-(phenylaminocarbonyl)- 3,4-tetrazolium]-bis (4-

methoxy-6-nitro) benzene sulfonic acid hydrate

1

1 INTRODUCTION 1.1 MALIGNANT GLIOMAS

The World Health Organization (WHO) classifies primary brain tumors into four prognostic grades, grade I to IV, based on their severity of malignancy (Louis, 2007). Grade III and grade IV gliomas are categorized as malignant gliomas which account for approximately 70% of all primary brain tumors in adult (Wen and Kesari, 2008). The most malignant form, glioblastoma multiforme (GBM), accounts for 60-70% of the malignant gliomas (Louis, 2007, Wen and Kesari, 2008). The incidence of malignant gliomas is approximately 5 cases per 100,000 people per annum where men are more affected than women (Wen and Kesari, 2008). Approximately 190,000 new patients were diagnosed with, and 142,000 patients died of, malignant gliomas worldwide in 2002 according to the WHO latest estimates in GLOBOCAN 2002 (Parkin et al., 2005). This represents a median survival of less than one year for the vast majority of malignant glioma patients. Virtually, none of the patients is cured of their illness. This failure in finding a curative treatment lies in the pathogenesis and histological features of malignant gliomas. 1.1.1 Origin of gliomas

Although, gliomas have long been considered to be originated by neoplastic transformation of glial cells (astrocytes or oligodendrocytes), neural stem cells (NSCs) have recently been proposed as the source of gliomas (Sanai et al., 2005). This NSC-hypothesis of glioma-origin stems from the findings that adult brain contains NSCs (Eriksson et al., 1998, Sanai et al., 2004) and the genes, for example epidermal growth factor receptor (EGFR), phosphatase and tensin homolog (PTEN), Hedgehog, nestin etc., which are involved in neurogenesis and gliogenesis, play pivotal roles in gliomagenesis (Sanai et al., 2005). In fact, a small population of cells within gliomas have been identified by their expression of the NSC-marker CD133 (Ignatova et al., 2002) and it was reported that only CD133 positive GBM cells form tumor in vivo (Singh et al., 2004). The presence of multiple cell types within the same tumor, such as oligoastrocytoma, and the similarity in migratory capacity and pattern between glioma cells and NSCs also support the NSC-theory of glioma-origin. This thesis, however, deals with NSCs from the perspective of glioma treatment by exploiting their migratory capacity. Therefore, NSCs and their properties will be further described from that perspective in the ‘Neural Stem Cell’ and subsequent sections.

2

1.1.2 Histopathological features

Malignant glioma cells extensively infiltrate into normal brain parenchyma through the dense network of neuronal and glial cell processes. They also disseminate via cerebrospinal fluid pathways and meninges giving rise to mulifocal gliomas (Claes et al., 2007). GBM can even be present as multicentric gliomas (Salvati et al., 2003). Despite this aggressive invasiveness, malignant gliomas do not metastasize outside the nervous system. Based on the glial phenotype, malignant gliomas can be histopathologically classified as astrocytic, oligodendroglial or oligoastrocytic. However, marked phenotypical heterogeneity within a malignant glioma is frequently observed. Microscopically, WHO grade III gliomas (e.g. anaplastic astrocytoma) show high cellularity, cellular and nuclear anaplasia, and mitoses. In addition to these features, WHO grade IV gliomas (e.g. GBM) show microvascular proliferation and/ or necrosis and sometimes, perinecrotic palisading (Claes et al., 2007). 1.1.3 Molecular features

Malignant transformation in gliomas results from the sequential accumulation of genetic aberrations and the alteration of growth-factor signaling pathways which are relatively well characterized. GBM which arises most commonly de novo (primary GBM) is characterized by amplification and mutations of EGFR, deletion of PTEN and p16 and loss of heterozygosity of chromosome 10q (Furnari et al., 2007). Some GBMs arise by malignant transformation of low-grade astrocytomas (secondary GBM). They are characterized by p53 mutations, PDGFR overexpression, abnormalities in the p16 and retinoblastoma (Rb) pathways, and loss of heterozygosity of chromosome 10q (Furnari et al., 2007). Despite their genetic differences, primary and secondary glioblastomas are histopathologically indistinguishable and respond similarly to conventional treatment. 1.1.4 Treatment hurdles

Malignant gliomas are conventionally treated by surgery, radiotherapy and chemotherapy. Maximal surgical resection is performed whenever possible, although complete excision is not possible as diffuse infiltration usually spreads beyond the radiologically evident boundaries of the tumor (Wen and Kesari, 2008). Radiotherapy is the mainstay treatment for malignant gliomas (Stupp et al., 2005). However, these tumors also become refractory to radiotherapy and again, the invasiveness through the brain parenchyma renders the invasive microfoci to escape localized radio-therapeutic assault. Thus, the survival benefit with radiotherapy is still counted in weeks and months. Chemotherapy has limited impact on the outcome of malignant gliomas, despite taking an increasingly important place in the treatment (Wen and Kesari, 2008). Major complicating factors in the drug treatment of malignant gliomas

3

include the presence of a variably disrupted blood-brain barrier (BBB) which complicates drug delivery to the tumor, an insufficient therapeutic index and a lack of specificity of chemotherapeutic agents for the tumor cells (Deeken and Loscher, 2007, Muldoon et al., 2007, Penas-Prado and Gilbert, 2007). Moreover, the intratumoral heterogeneity on the cytopathological, transcriptional and genomic levels, results in the emergence of drug-resistant cell subpopulations. As a result, malignant gliomas invariably recur with immediate fatality, in spite of the aggressive multimodality treatment. Therefore, it is imperative to develop new therapeutic strategies. 1.1.5 New therapeutic strategies

The improved understanding of the molecular pathogenesis of malignant gliomas has opened up possibilities of targeted molecular therapies tailored to the unique biology of an individual’s malignant glioma. Molecular therapies targeting receptor tyrosine kinase pathways (e.g. EGFR) (Mellinghoff et al., 2005) (Ren et al., 2007), rapamycin pathway (Galanis et al., 2005), histone deacetylase (HDAC) (Phuphanich et al., 2005), farnesyl transferase (Cloughesy et al., 2006), angiogenesis (Vredenburgh et al., 2007) etc. are under extensive investigation and a number of them are in clinical trials. To date, none of the targeted therapies has proven clinically effective for malignant gliomas (Salgaller and Liau, 2006, Lukas et al., 2007, Penas-Prado and Gilbert, 2007). This is presumably due to the failure of a single agent to address the complexity of signaling pathways and the poor penetration of drugs across the BBB(Deeken and Loscher, 2007, Muldoon et al., 2007). Therapies targeting multiple pathways in combination with the conventional treatment are under investigation (Doherty et al., 2006). Other investigational therapies include immunotherapy and gene therapy (Avgeropoulos and Batchelor, 1999, Lukas et al., 2007).

4

1.2 GENE THERAPY: SUICIDE GENE THERAPY

Gene therapy is a paradigm for modifying the cellular genome for therapeutic benefit. Although gene therapy was initially investigated for diseases known to be due to well characterized genetic mutations, most of the current clinical protocols concentrate on the treatment of cancer (Freeman and Zwiebel, 1993, Whartenby et al., 1995). Cancer gene therapy strategies are divided into four main areas: (1) transfer of genes to enhance immunogenicity; (2) transfer of tumor-suppressor genes; (3) transfer of altered drug-resistance genes; and (4) transfer of suicide genes (Yazawa et al., 2002). The suicide gene therapy paradigm is based on the transduction of a target cell with a gene coding for an enzyme. When the transduced cell subsequently is exposed to a prodrug, the first step which is usually the rate limiting step in activating the prodrug towards a cytotoxic compound is catalyzed by the introduced enzyme (Culver et al., 1992, Freeman et al., 1993). Suicide gene therapy is, therefore, also called enzyme-prodrug therapy. Success in a suicide gene therapy strategy relies on the following components:

A suicide gene encoding an enzyme with comprehensive catalytic activity A suitable prodrug with adequate access into the tumor Targeted gene delivery and sufficient transgene expression Sufficient bystander cell killing effect

1.2.1 Suicide genes

Both mammalian and non-mammalian genes encoding prodrug activating enzymes have been investigated as suicide genes in cancer gene therapy. The most notable non-mammalian enzymes are herpes simplex virus 1 thymidine kinase (HSV-tk) (Oldfield et al., 1993), bacterial cytosine deaminase (Huber et al., 1994) and nitroreductase (Palmer et al., 2004), insect Drosophila melanogaster deoxyribonucleoside kinases (Zheng et al., 2000, Knecht et al., 2007) and rabbit liver carboxylesterases (Danks et al., 1998) etc. Human enzymes e.g. deoxycytidine kinase (Hapke et al., 1996) and cytochrome P450 (Kan et al., 2002), which are either absent or markedly down regulated in cancer have also been investigated in suicide gene therapy strategies. No plant enzymes have so far been investigated for the purpose of suicide gene therapy. The therapeutic approach which uses deoxyribonucleoside kinases (dNKs) such as HSV-tk, exploits the differential proliferative behavior of cancer cells and host cells and causes the killing of highly proliferating cancer cells without affecting the host cells. Brain is unique in this perspective as the vast majority of brain cells are terminally differentiated and non-proliferative, whereas brain tumors are amongst the most fast growing tumors. Therefore, malignant gliomas have been the prime target of suicide gene therapy using dNKs. Proliferating cells replicate their DNA in each cell cycle. The building stones for DNA synthesis are deoxyribonucleotides which are

5

synthesized by de novo and salvage pathways. During salvage, deoxyribonuclosides (dN) are phosphorylated into the corresponding monophosphates, diphosphates and finally into triphosphates. The salvage pathway is also the route of activation of a number of anti-viral prodrugs e.g. ganciclovir (GCV) activated by HSV-tk, and other established anti-cancer nucleoside analog prodrugs (Sandrini and Piskur, 2005). The first step, the phosphorylation of dN or nucleoside analogs to dN monophosphates

Figure 1 Suicide gene therapy strategy: Deoxyribonucleoside kinase (dNK)/ nucleoside analog prodrug (NA) approach. Based on delivery methods, foreign dNK gene is either integrated into the cancer cell genome or expressed in the delivering cells (1); cancer cells are exposed to NA (2); dNK catalyzes NA to NA-monophosphate (NA-P) conversion, which is the first and the rate limiting activation step of NA(3); subsequently, NA-P is converted to NA-diphosphate (NA-P-P) (4a) and then to the active form NA-triphosphates (NA-P-P-P) (4b) by endogenous dNKs; the activated NA incorporates into the replicating DNA causing strand breaks and cell death (5). (dNMP), or the respective analogs, is catalyzed by dNKs and is often the rate-limiting step. Therefore, dNKs have been the prime candidates as suicide genes in cancer gene therapy in combination with nucleoside analog prodrugs. The monophosphate forms of nucleoside analog in suicide gene therapy system are subsequently converted by endogenous cellular kinases to triphosphate forms which then incorporate into the replicating DNA by competitively inhibiting incorporation of endogenous DNA precursor nucleotides into DNA (Figure 1). As a result, DNA synthesis is terminated leading to cell death.

5

synthesized by de novo and salvage pathways. During salvage, deoxyribonuclosides (dN) are phosphorylated into the corresponding monophosphates, diphosphates and finally into triphosphates. The salvage pathway is also the route of activation of a number of anti-viral prodrugs e.g. ganciclovir (GCV) activated by HSV-tk, and other established anti-cancer nucleoside analog prodrugs (Sandrini and Piskur, 2005). The first step, the phosphorylation of dN or nucleoside analogs to dN monophosphates

Figure 1 Suicide gene therapy strategy: Deoxyribonucleoside kinase (dNK)/ nucleoside analog prodrug (NA) approach. Based on delivery methods, foreign dNK gene is either integrated into the cancer cell genome or expressed in the delivering cells (1); cancer cells are exposed to NA (2); dNK catalyzes NA to NA-monophosphate (NA-P) conversion, which is the first and the rate limiting activation step of NA(3); subsequently, NA-P is converted to NA-diphosphate (NA-P-P) (4a) and then to the active form NA-triphosphates (NA-P-P-P) (4b) by endogenous dNKs; the activated NA incorporates into the replicating DNA causing strand breaks and cell death (5). (dNMP), or the respective analogs, is catalyzed by dNKs and is often the rate-limiting step. Therefore, dNKs have been the prime candidates as suicide genes in cancer gene therapy in combination with nucleoside analog prodrugs. The monophosphate forms of nucleoside analog in suicide gene therapy system are subsequently converted by endogenous cellular kinases to triphosphate forms which then incorporate into the replicating DNA by competitively inhibiting incorporation of endogenous DNA precursor nucleotides into DNA (Figure 1). As a result, DNA synthesis is terminated leading to cell death.

6

1.2.2 Properties of the enzyme encoded by a suicide gene

Activity of an enzyme is characterized by the kinetic parameter kcat/Km which is defined by Km and Vmax values set by the Michaelis-Menten equation:

where, V is the velocity of an enzymatic reaction. Vmax is the maximal reaction velocity that can be achieved by an enzyme under saturating substrate concentration [S]. The Michaelis-Menten constant Km is equal to the substrate concentration that gives half-saturation of the enzyme. It is thus equal to the substrate concentration at one-half the Vmax. kcat is the catalytic turnover number of the enzyme defining the maximum number of substrate molecules converted to product per unit of time. Therefore, the kcat/Km value indicates the catalytic efficiency of an enzyme. There are specific requirements for enzymes used in suicide gene therapy. An ideal suicide gene candidate would exhibit a highly different and increased selectivity, and a higher catalytic turnover, for the nucleoside analog compared to the endogenous enzymes (Knecht et al., 2007). Most preferably, the nucleoside analog should not be activated at all by the human dNKs. In other words, an effective enzyme-prodrug system should preferably have both low Km and high Vmax values for the prodrug, defining a favorable kcat/Km (Springer and Niculescu-Duvaz, 2000). Most of the existing enzyme-prodrug systems do not show these favorable kinetic properties. For example, HSV-tk has a 70-fold higher Km value for its prodrug ganciclovir (GCV) than for thymidine (Thd) (Mercer et al., 2002) which in vivo results in substrate out-competition of GCV by Thd (Springer and Niculescu-Duvaz, 2000, Knecht et al., 2007). In order to improve gene therapy efficiency, modifications of existing suicide genes such as mutations have been attempted, however, with limited success. Therefore, new suicide genes with better kinetic parameters of the encoding enzymes are needed. 1.2.3 Prodrugs and the blood-brain barrier

One of the major complicating factors of suicide gene therapy, specific for brain tumors is the presence of the blood-brain barrier (BBB) that restricts the passage of solutes (Basilion et al., 2000, Huynh et al., 2006). BBB is a membranic structure formed by the tight junctions between endothelial cells in central nervous system vessels. In addition to good pharmacological properties, systemically delivered prodrugs in suicide gene therapy systems of brain tumors need to sufficiently cross the BBB. Most prodrugs of the existing suicide gene therapy systems do not successfully pass the BBB. For example, GCV poorly penetrate into the brain and moreover, it becomes rapidly

7

eliminated from the brain by unclear mechanisms (Brewster et al., 1994). Enhanced delivery of drugs including the prodrugs into the brain is under extensive investigation. Some encouraging results have been observed in preclinical research (Brewster et al., 1994), though their clinical translation remains elusive. In this perspective, the antiretroviral agent zidovudine (azidothymidine, AZT), which is currently approved for human immunodeficiency virus (HIV) infection could be an ideal prodrug. The presence of azido group increases the lipophilic nature of AZT, allowing it to cross cell membranes easily by diffusion and thereby also to cross the BBB. AZT exhibits better pharmacological properties than many other nucleoside analogs e.g. GCV, with better solubility and better BBB penetration (Klecker et al., 1987, Blum et al., 1988). The cerebrospinal fluid-plasma ratio of AZT increases in a linear fashion with time after drug administration (Rolinski et al., 1997). Another advantageous feature of AZT in humans is that both steps, the phosphorylation to monophosphate and to diphosphate are poorly supported by endogenous cellular dNKs, and represent two rate limiting steps in AZT activation towards a cytotoxic compound (Furman et al., 1986, Lavie et al., 1997). Therefore, the differential properties of the human dNKs and an AZT-specific dNK will limit AZT activation and cell killing effect exclusively to targeted tumor cells and will spare normal cells. However, the need for an enzyme to phosphorylate both of those rate limiting steps requires identification of a suitable enzyme which can be used in suicide gene therapy in combination with AZT. To date none of the enzymes encoded by suicide genes has strongly indicated their specificity for AZT. 1.2.4 Gene delivery systems

Gene therapy has long suffered from inefficient gene transfer technologies. Suicide genes are traditionally delivered into malignant gliomas by either viral or non-viral vectors. Retrovirus, adenovirus, adeno-associated virus and herpes simplex virus 1 in their replication-defective or replication-competent and tumor-selective forms have been studied as gene delivery systems (Ma et al., 2002, Chiocca et al., 2003, Hampl et al., 2003, Rainov and Kramm, 2003, Wang et al., 2003). However, their success has been limited due to insertional mutagenesis, short-term transgene expression, low transgene capacity, neurovirulence, recombination with wild type, immunogenecity and above all, poor penetration into the glioma tissue (Lawler et al., 2006). Intratumoral grafting of engineered viral vector produing cells, usually fibroblasts has also been attempted. But these cells are not tumor tropic and therefore, unable to target disseminated tumor cells (Ram et al., 1997, Rainov and Kramm, 2003, Lawler et al., 2006). Non-viral gene delivery systems such as liposomes also face similar distribution challenges as viral systems. In addition, they have lower in vivo transfection efficiencies and shorter transgene expression times (Li et al., 2005) (Yoshida et al., 2004). Interestingly, NSCs have recently emerged as attractive delivery vehicle in

8

gene therapeutic strategies of brain diseases where diffuse and targeted gene delivery is critical for success (Benedetti et al., 2000, Ehtesham et al., 2002a, Ehtesham et al., 2002b, Brown et al., 2003, Yang et al., 2004, Li et al., 2005, Muller et al., 2006). NSCs have inherent migratory capacity toward intracranial pathology including stroke, neurodegenerative diseases and brain tumors, can simultaneously express transgenes and are less immunogenic (Muller et al., 2006) (Staflin et al., 2004, Lawler et al., 2006). When they are implanted intracranially contralateral to malignant gliomas or even intravenously they display robust tropism for the tumor (Aboody et al., 2000, Brown et al., 2003, Yip et al., 2003). These properties of NSCs can be exploited to deliver a therapeutic or antitumor gene into the brain tumor. Therefore, NSCs as delivery vehicle of suicide genes can eliminate a number of viral and non-viral vector-related limitations of suicide gene therapy. If the NSCs armed with a suicide gene that kills dividing cells, become problematic themselves in term of tumorogenicity, they will self-eliminate which is a built-in safety bonus. Although NSCs are less immunogenic, an immune reaction against NSCs intermixed with a tumor raises little concern because that serves only to enhance bystander tumor killing effect (Freeman et al., 1997). Several investigators have already

Figure 2 Tropism of neural stem cells for malignant gliomas. Schematic diagram showing neural stem cells injected intracranially contralateral to brain tumor are migrating towards the main and distant sites of the tumor through the corpus callosum. (From Müller F-J et al. (2006), Nature Review Neuroscience) demonstrated that NSCs can efficiently deliver tumor-toxic genes, for example HSV-tk, CD, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and cytokines IL-4 and IL-12, reduce tumor volume and increase survival (Benedetti et al., 2000, Ehtesham et al., 2002a, Ehtesham et al., 2002b, Li et al., 2005, Uhl et al.,

9

2005, Kim et al., 2006). Interestingly, NSCs have also been shown to possess inherent tumor killing property (Staflin et al., 2004). 1.2.5 Bystander effect

The bystander effect in a suicide therapy system can be defined as the cytotoxic effect on non-genetically modified tumor cells after prodrug administration when only a fraction of the tumor mass (in case of viral and non-viral gene deliveries) or only suicide gene carrying cells are genetically modified to express an activating enzyme (Takamiya et al., 1992, Freeman et al., 1993). While this effect is necessary for success in viral and non-viral based suicide gene therapy strategies, the NSC-based approach entirely relies on this effect. The dNK/nucleoside analog therapeutic system requires cell-to-cell contact between genetically modified and unmodified cells to display a bystander effect as the toxic compounds (activated nucleoside analog) are phosphorylated and are, therefore, not diffusible across cell membranes (Niculescu-Duvaz and Springer, 2005). The transfer of activated nucleoside analogs from cell to cell mainly requires the existence of gap junctional communications (GJCs) (Fick et al., 1995, Elshami et al., 1996, Mesnil et al., 1996). Other mechanisms of bystander effect could also be involved.

Figure 3 Bystander effect in stem cell-mediated suicide gene therapy. Neural stem cells (NSC) carrying a suicide gene (SG) migrate to glioblastoma multiforme (GBM) and establish gap junction communication (GJC) with the GBM cells. When they are exposed to nucleoside analog prodrug (NA), enzymes encoded by the suicide gene catalyze the conversion of the NA to NA monophosphate, the bottleneck step to generate active NA triphosphate (NA-TP). NA-TP generated by the stem cell is transferred to neighboring GBM cells through GJC and causes cell death by terminating DNA synthesis. The cells where GJC are not optimal (indicated by white arrows) do not get killed, while sufficient GJC (indicated by black arrows) results in cell death.

9

2005, Kim et al., 2006). Interestingly, NSCs have also been shown to possess inherent tumor killing property (Staflin et al., 2004). 1.2.5 Bystander effect

The bystander effect in a suicide therapy system can be defined as the cytotoxic effect on non-genetically modified tumor cells after prodrug administration when only a fraction of the tumor mass (in case of viral and non-viral gene deliveries) or only suicide gene carrying cells are genetically modified to express an activating enzyme (Takamiya et al., 1992, Freeman et al., 1993). While this effect is necessary for success in viral and non-viral based suicide gene therapy strategies, the NSC-based approach entirely relies on this effect. The dNK/nucleoside analog therapeutic system requires cell-to-cell contact between genetically modified and unmodified cells to display a bystander effect as the toxic compounds (activated nucleoside analog) are phosphorylated and are, therefore, not diffusible across cell membranes (Niculescu-Duvaz and Springer, 2005). The transfer of activated nucleoside analogs from cell to cell mainly requires the existence of gap junctional communications (GJCs) (Fick et al., 1995, Elshami et al., 1996, Mesnil et al., 1996). Other mechanisms of bystander effect could also be involved.

Figure 3 Bystander effect in stem cell-mediated suicide gene therapy. Neural stem cells (NSC) carrying a suicide gene (SG) migrate to glioblastoma multiforme (GBM) and establish gap junction communication (GJC) with the GBM cells. When they are exposed to nucleoside analog prodrug (NA), enzymes encoded by the suicide gene catalyze the conversion of the NA to NA monophosphate, the bottleneck step to generate active NA triphosphate (NA-TP). NA-TP generated by the stem cell is transferred to neighboring GBM cells through GJC and causes cell death by terminating DNA synthesis. The cells where GJC are not optimal (indicated by white arrows) do not get killed, while sufficient GJC (indicated by black arrows) results in cell death.

10

The mechanism of cell death in dNK/nucleoside analog gene therapy is apoptosis and tumor cells are known to phagocytose nearby apoptotic vesicles (Freeman et al., 1993, Samejima and Meruelo, 1995, Hamel et al., 1996) (Denning and Pitts, 1997). Therefore, release of apoptotic vesicles by dying tumor cells or gene delivering stem cells and then, phagocytosis of those vesicles by neighboring tumor cells could be another potential mechanism of bystander effect. Phagocytosis of material e.g., hydrolases or other lytic enzymes from dying cells to bystander cells has also been suggested as a mechanism for the bystander effect (Freeman and Zwiebel, 1993, Pope et al., 1997, Sturtz et al., 1997) (Denning and Pitts, 1997). Another effect called the “Good Samaritan effect” has also been described in suicide gene therapy. This effect refers to the observation that transfected cells can be protected from the active drug, presumably by lowering the concentration of the cytotoxic metabolites through GJC (Wygoda et al., 1997, Andrade-Rozental et al., 2000). This can be considered as beneficial, especially in NSC-mediated suicide gene therapy, because the transgenes will last longer, producing more toxic metabolites, thus enhancing the bystander effect. In viral vector-based suicide gene therapy, immune system also plays an important role in enhancing the bystander effect in vivo through the release of cytokines (Freeman et al., 1997). Therefore, the therapeutic effect is found be more pronounced in immunocompetent animals. However, immunogenicity is also one of the main disadvantages of viral vectors, as it causes loss of transgene expression and a systemic immune response is detrimental (Basilion et al., 2000).

11

1.3 GAP JUNCTION COMMUNICATION (GJC)

Gap junctions are cylindrical hydrophilic channels spanning the plasma membranes of two adjacent cells, with each cell contributing half the channel (hemichannel) termed connexon. Each connexon is a multimeric assembly of six proteins named connexins (Evans and Martin, 2002). 21 types of connexins (Cxs) have been so far characterized in humans (King and Lampe, 2005). Connexin 43 (Cx43) is the most widely expressed connexin and begins to assemble in the plasma membrane at very early stage of embryonic development (Willecke et al., 2002, Saez et al., 2003, Houghton, 2005). Gap junctions exist in all multi-cellular organisms and in almost all cell types in these organisms (Evans and Martin, 2002). They differ from other membrane channels since they exist between two cells, are relatively nonspecific, and the molecular movement through the channels occurs by passive diffusion. The gap junctional channels have an apparent selectivity based principally on molecular size, allowing the movement of molecules smaller than 1000 Da, such as ions, nucleotides, second messengers, amino acids, short peptides and other metabolites (Saez et al., 2003). Consequently, GJC is an important mechanism for maintaining cellular homeostasis, proliferation and differentiation. Gap junction channels typically remain open, and only close under specific circumstances. These include the pathophysiological state of the tissue or organism, the presence of external stimuli and the cell cycle. As the cells proceed through the cell cycle, the GJC decreases and becomes remarkably diminished at the mitotic phase (Lampe and Lau, 2004). In conditions with increased mitotic activity such as cancers, GJC is often found severely impaired (Mesnil et al., 2005). For example, malignant gliomas show markedly diminished GJC which is also correlated with a lack or very low expression of Cx43 (Huang et al., 1999). A variety of other diseases including some forms of neuropathy, hereditary deafness, cataracts, skin diseases and heart diseases also display deficient or improper GJC (Willecke et al., 2002).

Figure 4 Schematic diagram showing gap junction channels. A gap junction channel is formed by assembling two hemi-channels (connexons), each of which contributed by apposing cell membranes of two neighboring cells. Each connexon is composed of six connexin proteins.

12

1.3.1 Regulation of gap junction communication

The regulation of GJC is largely mediated by connexin synthesis and phosphorylation. Connexin synthesis determines the number of gap junctions, while phosphorylation regulates channel permeability and gating mechanism. Although there are some contrasting reports, it is generally accepted that phosphorylation of Cx43 is associated with closure of gap junction channels. Cx43 phosphorylation can also occur during their oligomerization into connexon prior to reaching the cell membrane and thus may affect the connexon trafficking and assembly with partner cell connexon (Lampe and Lau, 2000). It also plays a role in connexon disassembly and degradation processes. Cx43 is phosphorylated at different serine (Ser) residues in the C-terminal region by various kinases including mitogen-activated protein (MAP) kinases, protein kinase C (PKC), protein kinase A and casein kinase 1 (Lampe and Lau, 2000). It has been demonstrated that the level of GJC is predictive for the extent of the bystander effect in suicide gene therapy (Mesnil and Yamasaki, 2000, Asklund et al., 2003). Therefore, the efficacy of this therapy could be improved by enhancing the GJC which can be achieved either by inducing connexin expression with pharmacological agents or by transferring connexin gene. It has been shown that by transduction of a foreign Cx43 gene into malignant glioma cells increase bystander effect in HSV-tk/GCV suicide gene therapy approach (Marconi et al., 2000).

Figure 5 Regulation of gap junction communication. Schematic diagrams (A and B) and atomic force microscope (AFM) images (C and D) show an open gap junction channel when the channel forming connexin 43 is not phosphorylated (A and C) and a closed gap junction channel when the connexin 43 is phosphorylated (B and D). (AFM images from Nicholson et al (2003), Journal of Cell Science).

13

1.4 NEURAL STEM CELL (NSC)

A stem cell is a cell that can continuously self-renew and under right conditions, has the ability to differentiate into many different cell types that make up the organism. Self-renewal is defined as the ability to generate daughter cells identical to their mother. Differentiation is the structural and functional specialization of cells and tissues during development, which occurs by the gradual maturation of cells with specialized structures and functions from unspecialized precursors as a result of reprogramming of gene expression. Stem cells can be totipotent, pluripotent or multipotent. Totipotent stem cells are the cells which can develop into any cell found in the multi-cellular organism, such as the fertilized egg or early embryonic cell. Approximately four days after fertilization in humans, the totipotent cells start to specialize and form a cluster of cells known as a blastocyst. Inner cell mass of the blastocyst consists of pluripotent stem cells. They can give rise to all cell types in the human body except for extraembryonic tissue (e.g. placenta) and thus, unlike totipotent stem cells, can not develop into a complete organism. After further development, some of these cells are finally restricted to give rise to one or a few cell types, so they become multipotent stem cells. NSCs are multipotent stem cells which can give rise to neurons and glial cells (Gage, 2000). These cells must also migrate to specific locations in the neuraxis. NSCs can be obtained from embryonic (< 8 weeks gestational age) or fetal (> 8 weeks gestational age) neural tissue (Uchida et al., 2000). Certain adult brain regions, such as subventricular zone of lateral ventricle and hippocampus also contain NSCs and they can as well be harvested (Eriksson et al., 1998, Johansson et al., 1999). 1.4.1 Markers for neural stem cells NSCs are isolated by expression of their markers CD133 (prominin-1) or nestin. Nestin is a class VI intermediate filament protein and has been the most extensively used marker to identify NSCs within various areas of the developing nervous system and in cultured cells in vitro. Specifically, nestin is considered as a marker for dividing and migrating NSCs (Gilyarov, 2008). It is expressed both in neuronal and glial progenitors as well as in their common progenitors. Mature cells in the nervous system do not express nestin. However, it can be re-expressed in the brain cells under certain pathological conditions such as brain injuries and neoplastic transformation (Dahlstrand et al., 1992, Holmin et al., 1997). The transient expression of nestin in NSCs has been suggested to be a major step in the neural differentiation pathway (Lendahl et al., 1990), the exact role, however, is still undefined. As a cytoskeletal component, nestin has been suggested to take part in stabilizing cell structure and coordinating changes in intracellular dynamics (Herrmann and Aebi, 2000, Michalczyk and Ziman, 2005), which are presumably necessary for dividing and migrating NSCs. Other multipotent

14

stem cells such as hematopoietic stem cells and pancreatic islet stem cells have also been shown to express nestin (Lumelsky et al., 2001, Shih et al., 2001). The cell surface antigen CD133 is a 120 kDa membranous glycosylated protein and has been used to directly isolate NSCs. Due to its location in membranous protrusions, a functional role of plasma membrane organization has been attributed to CD133 (Corbeil et al., 2001), although the specific role is still unclear. However, for NSCs, it has been suggested that CD133 expression may sustain a stem cell phenotype and upon its repression, a cell becomes committed to differentiation (Marzesco et al., 2005). Its expression also correlates with the proliferative behavior of NSCs (Kania et al., 2005). CD133 is also used as a marker for hematopoietic stem cells and cancer stem cells (Mizrak et al., 2008). 1.4.2 Neural stem cell differentiation NSCs require growth factors such as basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF) for their proliferative expansion and simultaneously keeping the stem cell phenotype in vitro. They start differentiating along neuronal and glial lineages upon withdrawal of the growth factors (Cattaneo and McKay, 1990). Neuronal class III -tubulin (Tuj1) is used to characterize postmitotic, immature neurons and the class III intermediate filament protein glial fibrillary acidic protein (GFAP) is used to characterize the differentiated astrocytes. The differentiation processes are regulated by the dynamic interplay between extracellular cues including cytokine signaling and intracellular programmes such as epigenetic modification. There is increasing evidence that epigenetic mechanisms involving, for example, changes in DNA methylation and histone modification are closely associated with fate specification of NSCs. These epigenetic alterations could provide coordinated systems for regulating gene expression at each step of neural cell differentiation. 1.4.3 Potentials of neural stem cells

In addition to delivering suicide genes into brain tumors, NSCs can also be genetically engineered to express and deliver therapeutically important exogenous genes for many intractable neurodegenerative diseases. They have also great potential in neuroregenerative medicine. Exploiting their differentiation property, NSCs could be used to replace degenerating nervous tissue. These cells have also an inherent biology to spontaneously secrete therapeutically important molecules such as glial cell derived neurotrophic factor (GDNF), brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF) and various other neurotrophins which can protect and rescue sick neurons (Blesch and Tuszynski, 2001, Ourednik et al., 2002, Svendsen and Langston, 2004). NSCs are also attractive tools to explore mechanisms of brain development and carcinogenesis.

15

1.5 EPIGENETICS

Epigenetics is the study of genomic properties that influence phenotype without directly involving genotype (DNA sequence). The best example of epigenetic changes in eukaryotic biology is the process of cellular differentiation. Although differentiated cells and their embryonic precursors are genetically identical, the differentiated cells have radically different expression profiles and highly specialized functions. This suggests that epigenetic mechanisms must play a pivotal role in the differentiation process. Major epigenetic mechanisms include DNA methylation and histone modifications.

Figure 6 Epigenetic mechanisms (From Qiu et al (2006), Nature). 1.5.1 Histone modifications: the acetylation Histones are the main protein components of chromatin. Four pairs of histone proteins (mostly H2A, H2B, H3 and H4) form a nucleosome, the primary repeating unit of

15

1.5 EPIGENETICS

Epigenetics is the study of genomic properties that influence phenotype without directly involving genotype (DNA sequence). The best example of epigenetic changes in eukaryotic biology is the process of cellular differentiation. Although differentiated cells and their embryonic precursors are genetically identical, the differentiated cells have radically different expression profiles and highly specialized functions. This suggests that epigenetic mechanisms must play a pivotal role in the differentiation process. Major epigenetic mechanisms include DNA methylation and histone modifications.

Figure 6 Epigenetic mechanisms (From Qiu et al (2006), Nature). 1.5.1 Histone modifications: the acetylation Histones are the main protein components of chromatin. Four pairs of histone proteins (mostly H2A, H2B, H3 and H4) form a nucleosome, the primary repeating unit of

16

chromatin. DNA is wrapped around nucleosomes which are then folded through a series of successively higher order structures to eventually form a chromosome. This DNA packaging provides the basic physical structure for an added layer of regulatory control for gene expression. Nucleosomal structure can be altered by chromatin remodeling factors and histone modifications to allow regulated access to the DNA. Histones can be modified by acetylation, methylation, citrullination, phosphorylation, sumoylation, ubiquitination and ADP-ribosylation. Histone acetylation is the most abundant and reversible chromatin modifications and involves two classes of enzymes, histone acetyltransferases (HATs) and histone deacetylases (HDACs). HATs add acetyl groups at lysine residues within histone tails while HDACs remove them. The level of histone acetylation is the result of these converse activities of HATs and HDACs. The acetyl groups neutralize the positive charge of lysines and thereby their electrostatic interactions with the negatively charged DNA phosphate backbone. This results in relaxation of the chromatin and increased access of transcription factors to the DNA (Jenuwein and Allis, 2001). HDACs can act as transcription repressors, due to histone deacetylation, and consequently promote chromatin condensation. The HDAC activity is reported to be necessary for the progression of differentiation of embryonic stem cells (Lee et al., 2004) and hematopoietic stem cells (Koipally et al., 1999). When embryonic rodent hippocampal stem cells are differentiated into neurons and astrocytes, an upregulated HDAC activity is found (Ajamian et al., 2003). Accordingly, the HDAC inhibitor trichostatin A (TSA) inhibits the oligodendrocytic differentiation of rodent NSCs (Marin-Husstege et al., 2002, Shen et al., 2005) and promotes self-renewal of hematopoietic stem cells (Young et al., 2004). 1.5.2 DNA methylation

DNA methylation involves the addition of a methyl group to the number 5 carbon of the cytosine pyrimidine ring of DNA. In adult somatic tissues, DNA methylation typically occurs in a CpG dinucleotide context; non-CpG methylation is prevalent in embryonic stem cells. In humans, the process of DNA methylation is carried out by three enzymes, DNA methyltransferases 1, 3A and 3B (DNMT1, DNMT3A, DNMT3B). It is thought that DNMT3A and DNMT3B are the de novo methyltransferases that set up DNA methylation patterns early in development. DNMT1 is the proposed maintenance methyltransferase that is responsible for copying DNA methylation patterns to the daughter strands during DNA replication (Chen and Li, 2004). DNA methylation is known to regulate single gene expression in the differentiation process and in most cases expressed genes are demethylated at promoter regions (Roloff and Nuber, 2005). For example, astrocytic differentiation of forebrain

17

progenitors is activated following demethylation of a promoter region of the astrocytic gene GFAP (Takizawa et al., 2001). 1.5.3 Relationship between epigenetic mechanisms

DNA methylation may impact on the transcription of genes in two ways. First, the methylation of DNA may itself physically impede the binding of transcriptional proteins to the gene and secondly, and likely more importantly, methylated DNA may be bound by proteins known as methyl-CpG-binding domain proteins (MBDs). MBD proteins then recruit additional proteins to the locus, such as HDACs and other chromatin remodelling proteins that can modify histones, thereby forming compact, inactive chromatin termed silent chromatin (Fuks et al., 2000, Bachman et al., 2001). This link between DNA methylation and chromatin structure is very important. In particular, loss of methyl-CpG-binding protein 2 (MeCP2) has been implicated in Rett syndrome and methyl-CpG-binding domain protein 2 (MBD2) mediates the transcriptional silencing of hypermethylated genes in cancer (Nan et al., 1998). It is also observed that chromatin regions which contain hypoacetylated histones contain hypermethylated DNA (Jones and Wolffe, 1999) and chemical compounds such as TSA which inhibit HDAC activity induce DNA demethylation (Selker, 1998, Ou et al., 2007).

Figure 7 Relationship between epigenetic mechanisms (From Harrison’s Principles of Internal Medicine 16th Edition)

18

1.5.4 HDAC inhibitors and suicide gene therapy HDAC inhibitors selectively modulate gene expression by inducing histone acetylation and thereby altered chromatin structure (Johnstone, 2002). In addition to histones, HDACs have many other protein substrates such as DNA binding proteins, transcription factors, hormone receptors and cytoskeleton proteins, which are involved in regulation of gene expression. Inhibition of HDAC acitivity thus can cause accumulation of acetylated forms of these proteins and alter their function (Xu et al., 2007). HDAC inhibitors induce different phenotypes in various transformed cells which ranges from growth arrest, differentiation, senescence to different forms of cell death. In comparison, normal cells are relatively more resistant to HDAC inhibitors-induced cell death. Therefore, HDAC inhibitors have been widely touted as one of the most promising anti-cancer drugs and at least 16 of them including butyrates have already entered clinical studies in patients with hematologic and solid tumors including malignant gliomas (http://clinicaltrials.gov) (Gilbert et al., 2001, Baker et al., 2002, Phuphanich et al., 2005).

Figure 8 Enhancement of bystander effect by the HDAC inhibitor 4-Phenylbutyrate (4-PB) in stem cell-mediated suicide gene therapy. Neural stem cells (NSC) carrying a suicide gene (SG) causes glioblastoma multiforme (GBM) cell death upon exposure to nucleoside analog prodrug (NA) and generation of active NA (NA-TP) as described in Figure 3. The cell death does not occur if there is not enough gap junction communication (GJC) as the middle cartoon depicts (white arrows indicate suboptimal GJC). However, if the cells are treated with NA in combination with 4-PB, the GJC between the stem cells and GBM cells could be enhanced (single headed black arrows) resulting in increased tumor cell killing and thereby improved efficacy of the suicide gene therapy. As GBM cells are known to have weak GJC, its enhancement by 4-PB (double headed black arrows) and transfer of phosphorylated NA through them could also play an important role in suicide gene therapy.

19

Interestingly, the HDAC inhibitor 4-Phenylbutyrate (4-PB) has also been shown to increase GJC in malignant glioma cellls through upregulation of Cx43 expression (Asklund et al., 2004). 4-PB thus may be a promising adjuvant in suicide gene therapy of malignant gliomas to enhance gap junction-mediated bystander cell killing effect. 4-PB is a short chain fatty acid which can be administered orally and is in clinical use since more than 20 years for urea cycle disorders (Summar, 2001, Wilcken, 2004).

20

2 AIMS OF THE STUDY The overall aim of this thesis was to explore possibilities to improve a suicide gene therapy approach of malignant gliomas from different aspects including finding a suicide gene with better catalytic property for a suitable prodrug, enhancement of bystander effect and targeted gene delivery into the tumor. In doing so, we specifically aimed at

Investigating HDAC inhibitors, specifically 4-PB as a candidate modulator of suicide gene therapy with respect to gap junction-mediated bystander effect

Investigating a human neural stem cell line as delivery vehicle of suicide genes by examining its tropism for malignant gliomas and its interaction, specifically gap junction formation, with the glioma cells

Investigating 4-PB effects on the stem cell line, especially on its self-renewal, differentiation and hetero-cellular GJC with malignant glioma cells, and

Finally, identifying a better suicide gene specific for a suitable prodrug and investigating its efficacy in in vitro and in vivo malignant glioma models using the stem cell-mediated gene delivery strategy.

21

3 MATERIALS AND METHODS

Routine laboratory materials and methods such as culture of commercially available cell lines, immunohistochemistry, Western blot analysis and RT-PCR are described in each paper included in this thesis. Other specific materials and methods are described in this section. 3.1 MATERIALS AND CELL LINES 3.1.1 Drugs and chemicals Following are the main drugs and chemicals used: 4-PB (Triple Crown America, USA), TSA (Wako Chemicals GmbH, Germany), GCV (Cymevene, Roche AB), AZT (Sigma A2169), 18 -glycyrrhetinic acid (AGA) (Sigma) and Methyl- -cyclodextrin (Kleptose CRYSMEB) (Roquette). 3.1.2 Malignant glioma cell lines The rat glioma cell line C6 (I), human GBM cell lines U343MGa (II) and U87MG (II and IV) were used in various experiments in this thesis. They were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen) supplemented with 10% fetal bovine serum (FBS), 100 µg/ml streptomycin, and 100 µg/ml penicillin. 3.1.3 Neural stem cell lines The human embryonic/fetal neural stem cell lines NGC-407 (II, III and IV) and HNSC.100 (II) were used in this thesis. NGC-407 cell line established by NsGene A/S (Ballerup, Denmark) in collaboration with Prof. Olle Lindwall, Lund University, is described below. HNSC.100 cell line (provided by ZGene A/S, Hørsholm, Denmark) was obtained from human fetal diencephalon and telencephalon (gestational age 10–10.5 weeks) and is a well characterized neural stem cell line (Villa et al., 2000). 3.2 METHODS 3.2.1 The neural stem cell line NGC-407 and its characterization (II) Human embryonic tissue was derived from an elective first trimester abortion, and was recovered with the permission of the Ethics committee of Lund University (ethical permission no. LU 549-02). Cells had been isolated from the ventral mesencephalon of the 7-week-old embryo and immortalized using a retroviral vector

22

containing the v-myc oncogene. Immediately after dissection, the embryonic neural tissue was cut in a drop of cell dissociation solution (Sigma). The cut tissue was centrifuged at 1000 rpm for 5 min and resuspended in fresh culture medium. Four days after seeding, the cell cultures were transduced at an MOI of 1 with a retroviral vector construct consisting of the gag–v-myc fragment of v-myc (GenBank accession number AF033809) cloned as an EcoRI/DraI fragment between EcoRI and HpaI sites of pLXSN (Clontech) (GenBank accession number M28248). A gene conferring neomycin resistance present in the vector allowed selection of transduced cells by adding geneticin at a concentration of 800 g/ml. The immortalized NGC-407 cells were grown as an adherent monolayer in DMEM/F-12 (Invitrogen) supplemented with N2 supplement 1x (Invitrogen), 0.5% human serum albumin (Sigma), 0.6% glucose (Sigma), 5 mM HEPES (Invitrogen), B27 supplement (Invitrogen), 40 ng/ml basic FGF (bFGF) (R&D Systems) and 20 ng/ml epidermal growth factor (EGF) (R&D Systems). The medium was renewed every 2–3 days. The cells were differentiated by removing the growth factors from the culture medium 24 hours after seeding. Characterization of the cells was performed by western blot and immunohistochemical analysis of the NSC markers nestin and CD133 and the differentiated cell markers -tubulin III for neurons and GFAP for astrocytes.

3.2.2 Identification and characterization of a new suicide gene (IV) A novel suicide gene of plant origin, a tomato thymidine kinase 1 (ToTK1) was identified and characterized in collaboration with Prof. Jure Piškur, Lund Univeristy, the Technical University of Denmark and ZGene A/S (Hørsholm, Denmark). Identification Deposited GenBank sequences were searched for human thymidine kinase 1 (TK1) homologs from plants and several putative TK1 were obtained. In this way, a TK1 homolog from tomato, Lycopersicon esculentum (Expressed Sequence Tag ACCN BG129197, clone cTOF23J5) of 513 bp was found. Several sequences containing open reading frames (ORFs) encoding a kinase were sub-cloned from genomic DNA and cDNA libraries or plasmids obtained from the originator laboratories and inserted into the EcoRI/BamHI site of pGEX-2T (Amersham-Pharmacia). The corresponding original plasmid for ToTK1 was obtained from the Clemson University Genomic Institute and the insert complete sequence was determined giving an ORF of 234 amino acid residues (GenBank accession number AF514775). The plasmid was used as a template in the PCR-mediated sub-cloning into the EcoRI/BamHI site of pGEX-2T (Amersham-Pharmacia). The resulting expression plasmid was named P579. The TK-deficient E. coli strain KY895 (F-, tdk-1, ilv) was transformed with the plasmid P579. Growth of colonies was tested with logarithmic dilutions of various nucleoside analogs including AZT to determine the lethal dose (LD100) of the nucleoside analogs.

23

Recombinant expression and purification

ToTK1 was overexpressed in KY895 as fusion proteins with an N-terminal glutathione S-transferase (GST) fusion tag and purified by gluthatione-Sepharose affinity chromatography followed by thrombin cleavage. Briefly, transformed KY895 were grown to an OD600 nm of 0.5–0.6 in LB/ampicillin (100 g/ml) medium at 37°C and protein expression was induced by addition of 100 M isopropyl- -D-thiogalactopyranoside. The cells were further grown for 4 h at 25°C and subsequently harvested by centrifugation. The cell pellet was homogenized by sonication in the binding buffer A (20 mM NaPO4, pH 7.3; 150 mM NaCl; 10% glycerol and 0.1% Triton X-100) in the presence of a protease inhibitor cocktail (Complete, ethylenediaminetetraacetic acid free from Roche Dignostics), subjected to centrifugation at 10 000 g for 30 min, filtered and loaded onto a 1 ml column (glutathione-Sepharose from Amersham Pharmacia) pre-equilibrated in binding buffer A. After loading of the sample, the column was washed with 20 ml of binding buffer A. Subsequently, the column was washed with 2.5 ml of 10 mM ATP/MgCl2 in buffer A and incubated for 1 h at room temperature and then for 30 min at 4°C. The column was washed again with 5 ml of buffer A and 1 ml of thrombin (50 U/ml) solution was applied on the column. The column was gently shaken overnight at 4°C and the protein was eluted from the column in buffer A. Enzyme assay dN kinase activity was determined by initial velocity measurements based on four time samples by the DE-81 filter paper assay using tritium-labeled nucleoside substrates and liquid scintillation. The standard assay conditions were: 50 mM Tris-HCl, pH 8.0, 2.5 mM MgCl2, 10 mM dithiothreiotol, 0.5 mM CHAPS, 3 mg/ml bovine serum albumin, 2.5 mM ATP and the 3H-labeled acceptor substrate at different concentrations. The dNMP kinase activities were determined by initial velocity measurements based on four time samples with assay conditions described above. The reaction was terminated by thermal inactivation of the enzyme and the products of the monophosphate kinase reaction were analyzed by thin layer chromatography spotting 5 µl of the terminated reaction on polyethyleneimine-cellulose plates. The separation took place in 0.25 M LiCl as the developing solvent. The spots were identified under UV using a marker mix of the corresponding nucleotides (25 nmol of each), cut out and the radioactivity extracted with 0.2 M KCl/ 0.1 M HCl and determined by liquid scintillation. One unit (U) of kinase activity is defined as 1 µmol of the corresponding monophosphate or diphosphate product formed per minute. The kinetic data were evaluated by nonlinear regression analysis using the Michaelis-Menten equation v = Vmax x [S]/(Km + [S]). The equations were fitted to all available data in a global fit. 3.2.3 Vector construction and establishment of stable cell clones Vectors carrying EGFP/HSV-tk sense/antisense (I)

24

The cloning of pBudCE4.1 EGFP/HSV-tk sense and pBudCE4.1 EGFP/HSV-tk antisense carrying an EGFP (enhanced green fluorescent protein) gene driven by an EF-1 a promoter, and a HSV-tk gene in either sense or antisense orientation driven by a cytomegalovirus (CMV) promoter was performed as described by Asklund et al (Asklund et al., 2003). pDsRed2-C1 (red fluorescent protein, RFP) was used as provided (Clontech). Rat glioma C6 cells were transfected with either pBudCE4.1 EGFP/tk sense or pBudCE4.1 EGFP/tk antisense or pDsRed2-C1 using Lipofectamin Plus reagent (Invitrogen) according to the manufacturer’s instruction. The selection of positive clones was started after 60 h by either adding zeocin (Invitrogen; 400µg/ml) to pBudCE4.1 EGFP/tk sense and pBudCE4.1 EGFP/tk antisense transfected cells, or geneticin (Invitrogen) to a final concentration of 500µg/ml to pDsRed2-C1 transfected cells. Vectors carrying ToTK1 (IV) The cDNA of ToTK1 was cloned into a retrovirus vector based on the Moloney murine leukemia (MLV) virus to generate a replication-deficient recombinant retrovirus containing the ToTK1. DNA fragments were amplified with Pfu polymerase (Stratagene) using primers with designed flanking restriction enzyme sites. ToTK1 constructs were cut with BglII/XhoI and cloned into the BglII/XhoI site of the pLCXSN plasmid vector (Clontech) under the control of the CMV promoter. The plasmid was purified using the Qiagen plasmid kit (QIAGEN) and the DNA sequence of the constructed plasmid was verified by DNA sequence determination. pLCXSN alone and the vector containing HSV-tk (cloned into BamHI/XhoI site) were used as controls. U87MG cells were transduced with the retroviral vectors containing either ToTK1 or HSV-tk. NGC-407 cells were transfected with ToTK1 containing pCI plasmid (Promega) using FuGENE 6 Transfection Reagent (Roche) and following the manufacturer’s instructions. Positive NGC-407 cell clones were then selected with hygromycin. The recombinant cells were also tested for sensitivity to AZT. 3.2.4 FACS analysis (I) C6 cells carrying EGFP/HSV-tk were mixed with the C6 cells carrying only RFP and co-cultured. They were treated with GCV, 4-PB and AGA separately or in combination. To determine the number of EGFP positive, the cell cultures were subjected to FACS analysis. After one wash with PBS, cells were trypsinized, pelleted and adjusted to 1–5 x106 cells/ ml in PBS containing 10% FCS. Cell cytometry was performed using a FACScan cell analyzer (Becton–Dickinson Bioscience, San Jose, CA, USA). EGFP detection was conducted at 530/30 nm. Fluorescence >650 nm was measured for internal reference. WinMDI2.8 (http://facs.scripps.edu) was used for analyzing FACS data.

25

3.2.5 Fluorescent dye transfer test for functional GJC (I and II)

Background Goldberg et al (Goldberg et al., 1995) first described the fluorescent dye transfer method based on the fluorescent probes calcein and dioctadecyl tetramethylindocarbocyanine perchlorate (DiI) to study GJC. Calcein is a water-soluble acetomethylic (AM) ester which is hydrolyzed intracellularly by cellular esterases to generate green fluorescent calcein. Calcein can diffuse via gap junctions to neighboring cells as a marker for functional gap junction coupling. Cell coupling is defined as calcein diffusion from the labeled cells to at least one adjacent cell. On the other hand, DiI is a lipophilic tracer and thus membrane-bound and non-transferable from cell to cell. When cells are double labeled with calcein and DiI and mixed with unlabeled cells, the DiI stain makes it possible to distinguish primary calcein-labeled cells from gap junction-mediated calcein-labeled cells. Procedure Donor cells were incubated for 20 min at 37 °C with 10 M DiI (Molecular Probes, V-22885) and 5 M calcein-AM (Molecular Probes, C-3099) in serum free medium. These double labeled cells were rinsed with PBS, trypsinized, rinsed again with PBS and added to a subconfluent monolayer of unlabeled recipient cells at a ratio of 1:50 (labeled: unlabeled). Calcein dye transfer was documented using a fluorescent microscope and analyzing acquired images. The gap junction blocker AGA was used to validate the gap junction-mediated calcein transfer. 3.2.6 Cell proliferation and viability assays (I, II and IV) Cells were seeded in equal number in 3-4 replicates. The viability was monitored by MTT (Sigma) or XTT (Roche) colorimetric assays which are based on the measurement of tetrazolium salt MTT or XTT reduction by the viable cells. The cells were incubated with MTT dissolved in serum-free culture medium (0.5 mg/ml) for 30 min at 37 °C. The reduced MTT product formazan dye was solubilized by adding isopropanol. Where XTT assay was used, the defrosted XTT labeling mixture was directly added to the cell culture following manufacturer’s instruction. After incubation, plates were read at a microplate reader (570 nm with 650 nm reference for MTT and 450 nm with 690 nm reference for XTT). AGA was used to investigate the bystander effect (I and IV). The IC50 value was calculated as the mean value of these experiments using SigmaPlot® and the formula A = Amax/(1+(I/IC)) (IV). 3.2.7 LUMA analysis (III)

26

Genomic DNA of proliferating and differentiating NGC-407 cells cultured with or without 4-PB was extracted using the GeneElute Mammalian Genomic DNA Miniprep kit (Sigma). DNA was quantified using NanoDrop ND-1000 spectrophotometer. Global DNA methylation was assessed by LUMA as described in detail elswhere (Karimi et al., 2006). Briefly, genomic DNA (200–500 ng) was separately cleaved with HpaII + EcoRI and MspI + EcoRI and the degree of cleavage was subsequently quantified by pyrosequencing using PSQ96TM MA system and software (Biotage AB). The calculated ratio HpaII/MspI (ranging from 0 to1) represents the degree of global DNA methylation – the more HpaII/MspI ratio, the less methylation, and vice versa. 3.2.8 Animal studies (IV) Nude rat model of human glioblastoma multiforme for studying the efficacy of ToTK1/AZT therapy All animal experiments were conducted in accordance with the ethical guidelines from Stockholm local ethics committee. Male nude rats ((Hsd:RH-rnu), Harlan, Germany), aged 8-9 weeks with an average weight of 150-200 g were anesthetized by isoflurane inhalation and on a stereo tactic platform under microscopic guidance, a burr hole of <1 mm diameter was made in the skull, 2 mm right lateral and 3 mm posterior to bregma. A total of 3x105 NGC-407-ToTK1 and U87MG cells mixed at the indicated ratios in a 4µl volume were slowly injected 5 mm deep from the skull surface using a 5 µl volume Hamilton syringe. Starting 24 hours after implantation, the rats were treated intraperitoneally with AZT 400 mg/kg of body weight per day in 2 divided doses for upto 21 days. The control groups received equal volume of vehicle (15 % Methyl- -cyclodextrin). After sacrificing the animals, brains were quickly collected and immediately frozen in 2-methyl butane (Sigma) cooled by dry ice. The brains were then transferred into a -750C freezer until they were sectioned by a cryostat onto SuperFrost Plus glass slides (Menzel-Gläser) at a 14 µm thickness. In vivo investigation of NGC-407 cell tropism for malignant gliomas (IV) The NGC-407 cells used in this experiment were transfected with pCI-EGFP plasmid expressing green fluorescent protein (GFP) using the Nucleofector™ (Amaxa Biosystems, Germany) following the provided protocol. The cells were selected for their resistance to hygromycin. As described in the previous section, a burr hole in the skull of nude rats was created at 2 mm right lateral to bregma. 1.5x105 U87MG cells in 3 l volume were injected through the burr hole, 3.5 mm deep from the skull surface. The tumors were allowed to grow for 7 days. Then, 5x105 NGC-407-GFP cells were injected just contralateral to the U87MG cells implantation. 14 days after NGC-407 cells implantation, the animals were sacrificed and collected brains were sectioned and analyzed by immunohistochemistry to trace the NGC-407 cells by their GFP expression.

27

3.2.9 Experimental magnetic resonance imaging (MRI) (IV) MRI experiments were performed using a 4.7 T horizontal bore animal MR system operating on a Paravision (version 3.0.2) software platform (Bruker Biospec, Germany). Animals were anesthetized with, and maintained on isofluorane, and body temperature was maintained with a heated air stream. Respiratory activity was monitored continuously throughout the experiments. Multislice T2-weighted images of animal brains were acquired in the coronal, sagittal and axial planes (TR 2,500 ms, TE 95.6 ms, slice thickness 1.5 mm or 0.5 mm, FOV 3.5 cm, 5-16 slices depending on tumor size).

28

4 RESULTS AND DISCUSSIONS Malignant gliomas are amongst the most devastating and intractable of all cancers. In spite of exhausting all available therapeutic options including surgical resection, radiotherapy and chemotherapy, the median survival time after diagnosis is only 12 months for the most common and aggressive glioma GBM (Wen and Kesari, 2008). The success of suicide gene therapy against them has been limited due to inefficient gene transfer technologies, insufficient expression of the transgene, limited prodrug penetration into the brain, inappropriate enzymatic activities or low bystander killing effects. In this thesis, all these components of suicide gene therapy have been addressed by in vitro and in vivo experimentations with a goal to improve this promising therapeutic approach for malignant gliomas. 4.1 4-PB ENHANCES GJC-MEDIATED BYSTANDER EFFECT (I) First, stable C6 rat glioma cells expressing either RFP or EGFP/HSV-tk sense or antisense orientation were established in order to investigate bystander tumor cell killing effect by MTT assay and validate it by FACS analysis. Co-culture of RFP carrying cells and EGFP/HSV-tk sense or antisense carrying cells were treated with GCV in the presence or absence of 4-PB. Co-cultures consisting of HSV-tk antisense cells served as HSV-tk negative control. We found that 4-PB dramatically increased GCV-induced cell killing of the culture where a portion of cells expressed the HSV-tk sense. When the gap junction blocker AGA was added to the culture, the 4-PB enhancement was substantially reduced. FACS analysis showed that AGA did not prevent death of the HSV-tk positive cells. This suggests that the 4-PB-mediated enhanced cell killing is due to the death of bystander HSV-tk negative (RFP positive) cells which should be gap junction dependent as bystander effect is known to be mediated to a large extent by GJC (Fick et al., 1995). 4-PB was found to increase the gap junction protein Cx43 expression, which supports the role of GJC in the 4-PB enhancement of the bystander effect. Cx43 is also known as a tumor suppressor independent of its function of gap junction formation (Rose et al., 1993, Moorby and Patel, 2001, Zhang et al., 2003). Therefore, upregulation of Cx43 by 4-PB can be beneficial from two aspects, enhancement of bystander effect as well as suppression of tumor growth. 4.2 4-PB INCREASES HSV-TK TRANSGENE EXPRESSION (I) FACS analysis also revealed that death of the cells carrying HSV-tk sense was extensive in presence of 4-PB along with GCV compared with the cell death by GCV treatment alone. To explore the mechanism behind this increased cell death, we performed RT-PCR to measure HSV-tk expression in those cells. Interestingly, the

29

CMV-promoter driven HSV-tk-expression was substantially increased by 4-PB, while expression of endogenous glucose 6-phosphate dehydrogenase was unchanged. This corresponds with previous reports that HDAC inhibitors can activate silenced transgenes (Rosenqvist et al., 2002). The elevated HSV-tk expression might have led to an enhanced effect of the ganciclovir treatment by an increased GCV phosphorylation rate. Thus, the enhanced bystander killing effect caused by the combined treatment with 4-PB and GCV could be inhibited by AGA, but it could not prevent the gap junction independent death of the cells carrying HSV-tk sense. 4.3 NGC-407 CELLS POSSESS NSC CHARACTERISTICS (II) By definition, NSCs should have unlimited self-renewal capacity and differentiate to both neurons and glial cells. As the human embryonic midbrain cell line NGC-407 had been immortalized for uniformity and in vitro expansion, their self-renewal capacity is predictable. We aimed to investigate multipotentiality of these cells in vitro. NSCs are grown and maintained in a medium which is supplemented with mitogens to promote their proliferation (Gage, 2000). In such a proliferating condition, we found that all NGC-407 cells expressed the NSC-marker nestin abundantly and uniformly, while a very few number of cells expressed astrocytic marker GFAP and neuronal marker -tubulin III. NSCs are known to start differentiating upon mitogen withdrawal from the culture media (Cattaneo and McKay, 1990). When bFGF and EGF was removed from the media, majority of NGC-407 cells differentiated to astrocytes and a smaller portion of the cells generated neurons. Importantly, immunocytochemical staining showed that there is minimal overlap between staining of the respective differentiated cell markers GFAP and -tubulin III, which is consistent with lineage specific differentiation. A recent report on this cell line by other investigators verifies this observation (Roybon et al., 2008). 4.4 NGC-407 CELLS SHOW TROPISM FOR MALIGNANT GLIOMAS (IV) One of the important issues in NSC-mediated therapy for brain tumors is selecting an appropriate NSC type through investigation of the efficiency of cell engraftment, survival, transgene expression and migration. Several studies have demonstrated the ability of NSCs to migrate towards malignant gliomas. However, those studies mostly used NSCs and glioma cells, both of murine origin (Aboody et al., 2000, Glass et al., 2005, Shah et al., 2005, Staflin et al., 2007). NGC-407 cells are of human embryonic origin and had been obtained, propagated and maintained following GLP guidelines which renders them to be possibly utilized in both preclinical and clinical studies. Therefore, we examined these cells for their in vivo survival, transgene expression as well as migratory capacity towards human glioblastoma xenografts. NGC-407 cells carrying a GFP reporter gene were implanted into the nude rat brain, contralateral to the hemisphere bearing a glioblastoma xenograft. NGC-407 cells were traced by their

30

GFP expression at different time points after the stem cell implantation. We found that NGC-407/GFP cells extensively migrated towards the GBM xenografts in the opposite hemisphere through corpus callosum. They infiltrated the tumor bed and were found intermingling with the tumor cells inside the xenograft 2 weeks after their implantation. This also demonstrates that NGC-407 cells can survive in vivo and can efficiently express the foreign gene GFP. 4.5 4-PB ENHANCES GJC BETWEEN NGC-407 AND GLIOMA CELLS (II) NSC-mediated suicide gene therapy has been applied in animal malignant glioma models. Many in vitro and in vivo reports demonstrated that NSCs carrying a suicide gene can efficiently eliminate glioma cells (Benedetti et al., 2000, Ehtesham et al., 2002a, Ehtesham et al., 2002b, Brown et al., 2003) (Yang et al., 2004, Li et al., 2005). However, the heterocellular GJC between NSCs and the glioma cells and the extent of bystander effect on which this therapeutic approach entirely depends, have not been properly addressed. By fluorescent dye transfer method, we investigated functional GJC between NGC-407 and human glioblastoma cell lines in vitro. We found that both proliferating and differentiating NGC-407 cells were capable to form functional GJC with human glioblastoma cells, although the extent of dye tranfer from the differentiating cells was relatively limited, which corresponds with downregulated Cx43 expression in these cells. After observing the homocellular GJC modulatory effect of 4-PB in glioma cells (I), we were interested to examine the 4-PB effect in this heterocellular GJC. 4-PB upregulated C×43 expression along with increased membranous deposition in both proliferating and differentiating NGC-407 cells. Simultaneously, Ser 279/282 phosphorylation of C×43 known to be catalyzed by MAP kinase was inhibited by 4-PB in both these cell types. As a result, we observed an amplified homocellular GJC among NGC-407 cells, as well as a substantial increase of heterocellular GJC between both proliferating and differentiating NGC-407 cells and human glioblastoma cells. It is important to see that 4-PB enhances GJC in both types of the NGC-407 cells. It is thus likely that 4-PB will maintain an optimum level of the heterocellular GJC irrespective of the differentiation status of NSCs in vivo. Therefore, 4-PB could be an attractive adjuvant drug for enhancement of bystander effect in NSC-mediated suicide gene therapy of malignant gliomas.

4.6 4-PB PRESERVES STEMNESS OF NGC-407 CELLS (III) 4-PB effects on NSC differentiation also needs to be accounted for in utilization of 4-PB in NSC-mediated suicide gene therapy. We treated NGC-407 cells with 4-PB in both their proliferating and differentiating conditions and examined their differentiation status by stem cell markers nestin and CD133, and differentiated cell markers GFAP and -tubulin III. We found that 4-PB inhibited both astrocytic and neuronal

31

differentiation of NGC-407 cells indicated by downregulation of the respective markers and suppression of morphological characteristics of mature cells such as formation of a delicate network. This corresponds with the reports that HDAC activity is essential for the progression of differentiation of embryonic stem cells including NSCs and HDAC inhibitors such as TSA and sodium butyrate inhibits glial differentiation (Marin-Husstege et al., 2002, Ajamian et al., 2003, Lee et al., 2004, Shen et al., 2005). However, it has also been reported that HDAC inhibitor valproic acid stimulates neuronal differentiation of adult NSCs (Hsieh et al., 2004). These contrasting effects of HDAC inhibitors on neural stem cell differentiation has been suggested as being cell type specific (Roloff and Nuber, 2005, Meshorer, 2007). Importantly, we have also found that 4-PB preserved the stem cell phenotype of the NGC-407 cells. Nestin became entirely degraded 5 days after induction of differentiation. However, some of the differentiating cells retained the integrity of nestin upon exposure to 4-PB. CD133 was also found upregulated in these 4-PB cells. As Cx43 became downregulated in differentiated NGC-407 cells, resulting in their reduced ability to form GJC with glioma cells (II), this differentiation inhibitory and stemness preserving effect of 4-PB may prove useful for NGC-407-mediated suicide gene therapy. 4.7 4-PB INDUCES GLOBAL DNA HYPOMETHYLATION IN NGC-407

CELLS (III) As epigenetic mechanisms are integrated with each other (Ng and Bird, 1999, Fuks et al., 2000, Bachman et al., 2001) and play a pivotal role during embryonic development and NSC differentiation (Li, 2002) (Hsieh and Gage, 2004, Roloff and Nuber, 2005, Kondo, 2006, Meshorer, 2007), we investigated the global DNA methylation levels in NGC-407 cells upon differentiation and after 4-PB treatment. We found that NGC-407 cells become globally DNA hypermethylated 5 days after induction of differentiation, which is consistent with the report that adult brain contains one of the most highly methylated DNA in the body (Ehrlich et al., 1982). The hypermethylation corresponds with upregulation and nuclear localization of the de novo DNA methyltransferase DNMT3B which is known to play relatively more important role during early development (Okano et al., 1999). When the cells were treated with 4-PB, the global DNA methylation level was decreased. This correlates with the observed differentiation inhibitory effect of 4-PB, as global DNA hypomethylation has been shown to inhibit differentiation of embryonic stem cells, (Jackson et al., 2004). The hypomethylation may be explained by the demethylating effect of 4-PB as a HDAC inhibitor as reported previously for TSA (Chen and Pikaard, 1997, Selker, 1998, Ou et al., 2007). However, we also found some corresponding changes in the expression and localization of DNMTs. DNMT3B was located more in the cytoplasm than in the nuclei, and the DNMT1 protein level was decreased, after 4-PB treatment.

32

4.8 TOTK1 IS A NOVEL SUICIDE GENE (IV) By kinetic analysis, ToTK1 was found to efficiently phosphorylate AZT, known to successfully pass the BBB (Klecker et al., 1987, Rolinski et al., 1997). Importantly, in contrast to other nucleoside kinases/substrates, ToTK1 also phosphorylated AZT-monophosphates to AZT-diphosphates, the advantageous second rate limiting step in AZT activation, which human endogenous cellular kinases are unable to catalyze (Lavie et al., 1997). The Vmax and Km parameters of ToTK1 for AZT and Thd were very similar. In contrast, most of the enzymes encoded by suicide genes in cancer gene therapy have higher Km value for their respective nucleoside analog prodrugs than for the endogenous precursors for DNA synthesis (Mercer et al., 2002), which results in substrate out-competition of the prodrugs in vivo. ToTK1 had also a favorable catalytic efficiency (kcat/Km) for AZT defined by its low Km and high Vmax values, which most of the frequently used suicide gene systems such as HSV-tk/GCV lack (Springer and Niculescu-Duvaz, 2000). Altogether, the kinetic parameters indicate that ToTK1 is catalytically superior to other suicide genes including HSV-tk. We then compared the efficiency of ToTK1/AZT suicide gene therapy approach with HSV-tk/GCV system in an in vitro GBM model. ToTK1 dramatically increased the sensitivity of human GBM cells to AZT, while HSV-tk increased the sensitivity to GCV modestly. IC50 for AZT was 7.4 mM in parental U87MG cells compared to 16 µM in U87MG-ToTK1 cells, showing approximately 500-fold increase in AZT sensitivity when the cells were transduced with ToTK1. On the other hand, GCV sensitivity was increased only about 60-fold when the cells were carrying HSV-tk. IC50 for GCV was 1.6 mM in parental U87MG cells and 28.7 µM in U87MG-HSV-tk cells. These results show that ToTK1 and AZT combination is more efficient than the combination of HSV-tk and GCV as a suicide gene therapy approach. 4.9 NSC-MEDIATED TOTK1/AZT THERAPY FOR MALIGNANT GLIOMAS

(IV) NGC-407 carrying ToTK1 were mixed with parental U87MG cells at 1:1, 1:3 and 1:30 ratios. Then they were co-implanted intracranially into nude rats. Another group of animals received only parental U87MG cells. The rats were treated with AZT or equal volume of vehicle intraperitoneally for up to 21 days starting 1 day after the implantation. The AZT-treated rats which received cells at 1:1 and 1:3 ratios, survived on average 6-7 days longer than the vehicle treated rats. The improvement of survival was statistically significant (P < 0.05). Tumor growth was followed by MRI on day 12-13 and 22-23 post-implantation and tumor volume was measured using the images. Comparison of tumor size between the first and second MRI follow-ups in the vehicle treated animals showed that after initial establishment, the tumor grew very fast. Upon exposure to AZT, tumor size were 25% and 50% of the respective vehicle treated

33

tumors at 1:1 and 1:3 ratios of NGC-407-ToTK1 and U87MG cells. There was no difference in tumor size where only U87MG cells were injected, which indicates that AZT does not have an effect in absence of ToTK1. At the 1:30 ratio, exposure to AZT resulted in an apparent reduced tumor size and increased survival by 3 days, although this was not statistically significant. It is important to note here that the reduced tumor size was observed in immunocompromised animals which were used to prevent rejection of human cell grafts. Immune response is known to improve the efficacy of suicide gene therapy by enhancing bystander effect (Freeman et al., 1997). Therefore, the observed effect may be even greater in immunocompetent animals or in a clinical setting. However, considering the rapid growth of tumor, surgically debulked tumors will presumably require several injections of ToTK1 expressing stem cells in a clinical setting, a procedure which is technically not realistic in animal studies. Although this experimental set-up does not closely mimic the clinical therapeutic approach where a glioblastoma patient gets treatment for an existing tumor, it does provide evidence in favor of the effectiveness of ToTK1 in vivo.

34

5 CONCLUSION AND FUTURE PERSPECTIVES Different suicide gene therapies have been applied in animal models of malignant gliomas. A number of clinical trials have also been conducted, albeit with very limited success (Oldfield et al., 1993, Ram et al., 1997, Palu et al., 1999, Trask et al., 2000). Some hurdles, concerning specifically gene delivery system, transgene expression, enzyme activity, brain penetration and activation of prodrug and bystander effect, must be overcome before suicide gene therapy will become a clinically efficient treatment of malignant gliomas. The papers included in this thesis address all these issues. Using the herpes simplex virus thymidine kinase/ganciclovir (HSV-tk/GCV) suicide gene therapy system, we found that the histone deacetylase (HDAC) inhibitor 4-phenylbutyrate (4-PB) dramatically enhanced the gap junction-mediated bystander glioma cell killing effect and also activated the recombinant HSV-tk gene expression. As neural stem cells (NSCs) has emerged as a potential gene delivery vehicle, we characterized the human embryonic midbrain stem cell line NGC-407 and showed that these cells can be used to carry and deliver a suicide gene into malignant gliomas. NGC-407 cells were able to express a foreign gene over a long period of time in vivo, possessed extensive migratory capacity for xenografted human malignant gliomas and had the ability to form gap junction communicaiton (GJC) with the glioma cells. Interestingly, 4-PB enhanced this heterocellular GJC between the NGC-407 and glioma cells as it did for the homocellular GJC among glioma cells. 4-PB also inhibited differentiation of NGC-407 cells and preserved their stem cell phenotype. These findings are important for a NSC-mediated suicide gene therapy paradigm as we also found that differentiated NGC-407 cells had less ability to form GJC with glioma cells. Taken together, 4-PB can be used to improve NSC-mediated suicide gene therapy through enhancing the bystander effect. Finally, we identified and thoroughly characterized a novel suicide gene of plant origin, a tomato thymidine kinase 1 (ToTK1) which is specific for the prodrug zidovudine (AZT) known to successfully pass the blood-brain-barrier. One of the unique properties of this kinase is its ability to catalyze the AZT monophosphate to diphosphate reaction in addition to monophosphate conversion. Both these steps are rate limiting in AZT activation. In an in vitro malignant glioma model, we found ToTK1/AZT combination superior to HSV-tk/GCV approach as a suicide gene therapy system. In order to examine in vivo efficacy, we delivered ToTK1 into human glioblastoma xenografts by recombinant NGC-407 cells. After exposure to AZT, the

35

xenografts were substantially smaller. Significantly improved survival was also observed for the treated animals. Various approaches are being investigated to improve suicide gene therapy strategy.

These include mutations of suicide genes, use of multiple genes, new prodrugs for old enzymes, new formulations of prodrugs, development of efficient vectors, stem cell delivery of suicide genes and radiosensitization of tumor cells (Benedetti et al., 2000, Grignet-Debrus et al., 2000, Miura et al., 2002, Desaknai et al., 2003, Wiewrodt et al., 2003, Niculescu-Duvaz and Springer, 2005). From this perspective, the ToTK1/AZT combination along with stem cell-mediated gene delivery represents a promising new treatment for malignant gliomas. How 4-PB modulates this NGC-407 cell-mediated ToTK1/AZT therapy of malignant gliomas as an adjunct remains to be further investigated. As malignant gliomas are usually rapidly fatal, a survival benefit of only a few months with new therapeutic regimens is considered clinically meaningful in most clinical trials on this cancer (Stupp et al., 2005). The superiority over the clinically attempted HSV-tk/GCV system, the observed growth inhibitory effect, and significant survival benefit in animal models justify aiming for clinical trials of ToTK1/AZT suicide gene therapy on malignant glioma patients, particularly with the immediately fatal recurrent disease. This therapeutic strategy along with the stem cell-mediated gene delivery may also be applied for other kinds of solid tumors.

36

6 ACKNOWLEDGEMENTS Numerous people have in many ways helped me on the journey of making this thesis. I would like to thank in particular: Professor Tomas J. Ekstöm, my main supervisor. It is difficult to overstate my gratitude to you. When I first came to KI, you opened your arms to me and you have kept them open all these years. You guided, supported and encouraged me in every step of this thesis work and at the same time, you gave me enough freedom to build up my scientific confidence. I thank you for your friendliness and always having time for a question or a discussion. I am also grateful for giving me so many travel opportunities. Per M. Almqvist, my co-supervisor for initiating my PhD project and for your guidance and support all these years. Zoran Gojkovic, President and CEO of ZGene A/S, Denmark, for your continuous encouragement and support. You always inspired me to give my best. I also thank you for entertaining the visits to Copenhagen (many times), Athens, St. Louis and Seattle. Professor Jure Piškur at Lund University for your enormous support in preparing the final manuscript (on the tomato kinase). Your generosity is unparalleled. Börje Bjelke who first invited me at KI and gave me opportunity to learn animal surgical techniques and experimental MRI. ZGene A/S (Hørsholm, Denmark) for the financial support during this final year of my study. All the people at ZGene A/S - Peter Kristoffersen, Bo Skaaning Jensen, Eva Lin and others for your contribution. I would like to specifically thank Mette Willer and Annette Glidov for coming to Stockholm many times to help me performing animal studies. Former members of Tomas Ekström group: Ole Ammerpohl for introducing me in the cell culture work, Ioulia Appelskog for teaching me western blot analysis, Henrik von Horn for your warm friendship, Irina Svechnikova, Sabine Dörhöfer, Claes Holmgren and Sofia Johansson - it was great working with you. Current members of Tomas Ekström group: Mohsen Karimi, Tzvetomira Philipova, Anna-Maria Marino, Ghada Hagali and all the guest students for all the good time we shared in the lab. Thomas Asklund, one of my best friends in Stockholm and a colleague who always tries to understand me. Thanks for all those Irish coffee. That Irish restaurant is now probably my third most visited place in Stockholm after my home and lab! I am also very thankful for your guidance in understanding the Swedish culture. There is always something to learn from you! All the co-authors in my publications, especially Wolfgang Knecht and Birgitte Munch-Petersen for your contribution in the Paper IV.

37

Kicki Edwardsson, Anna-Lena, Leo and others at the animal facility (AKM) for your support and taking care. My gratitude to all the rats who have donated their lives to science. Britt Meijer at neurosurgery lab for your help during the early days in this PhD journey. Professor Sture Liljequist for providing me the cryostat mictotome for sectioning. Stefan Brené and Rouslan Sitnikov for your help in performing the MRI experiments. Colleagues at CMM 1st floor: Hannah Lindström and Daniel Johansson of former Mats Persson group for making long hours at the cell culture room enjoyable, Joya Marinova of former Georgy Bakalkin group for your help in analyzing western blot images. All the members of Gunner Norstedt, Catharina Larsson, Lars Terenius, Kristina Broliden and Clas Johansson groups for making CMM:01 exciting and enjoyable place. Yin-Choy Chuan, thanks for inviting me to play badminton with you. I enjoyed. Colleagues from other floors at CMM for creating a good research environment. Maral from 3rd floor for your help in immunohistochemistry, Gunther Weber for your help with the CCD camera. My writing room colleagues Anna Lindblom and Anna Ökvist for creating an excellent condition to write this thesis. Delphi Post for administrative help and Katarina Zannin for your patience in handling our mess. Daniel Uvehag for your help with my computer. My medical school friend Biplob (Stockholm) for your encouragement before I moved to Sweden, Alam, Hakim, Ajay and Sharif for that nice trip to St. Martin Island during my last visit to Bangladesh. Bipu, Arif and Rashed vai, I miss you all. My high school friends Nazim and Biplob (New York), I miss you too. Forgive me for not being able to keep myself in touch with you. Doctors of Bangladeshi origin in Stockholm and Uppsala for arranging fabulous spring and summer parties every year and serving home-made delicious food. My best friend and my little uncle Kamal Pasha for your hospitality and everything you did during our visit to Los Angeles last year. That was the best time I had abroad. My brothers and sisters Shahed, Saiful, Rumana and Tania for all your support and love. My parents who are everything to me. You give me strength and hope to keep believing in myself. Without your love and blessing I wouldn't be where I am today. Abba, you provided everything I would like and before I wished for. But due to this thesis work, I could not even be present beside you when you needed me most during your life-threatening illness. You know, only my body was away! Amma, only you can dissect my tears! Monira for your love. Thanks for accompanying me during this PhD journey and facing enormous difficulties with courage. Without your patience and determination, life would be more difficult. I wish all the best with your own PhD journey.

38

7 REFERENCES Aboody, K. S., Brown, A., Rainov, N. G., Bower, K. A., Liu, S., Yang, W., Small, J. E.,

Herrlinger, U., Ourednik, V., Black, P. M., Breakefield, X. O. and Snyder, E. Y., (2000). Neural stem cells display extensive tropism for pathology in adult brain: evidence from intracranial gliomas. Proc Natl Acad Sci USA. 97, 12846-12851.

Ajamian, F., Suuronen, T., Salminen, A. and Reeben, M., (2003). Upregulation of class II histone deacetylases mRNA during neural differentiation of cultured rat hippocampal progenitor cells. Neurosci Lett. 346, 57-60.

Andrade-Rozental, A. F., Rozental, R., Hopperstad, M. G., Wu, J. K., Vrionis, F. D. and Spray, D. C., (2000). Gap junctions: the "kiss of death" and the "kiss of life". Brain Res Brain Res Rev. 32, 308-315.

Asklund, T., Appelskog, I. B., Ammerpohl, O., Ekström, T. J. and Almqvist, P. M., (2004). Histone deacetylase inhibitor 4-phenylbutyrate modulates glial fibrillary acidic protein and connexin 43 expression, and enhances gap-junction communication, in human glioblastoma cells. Eur J Cancer. 40, 1073-1081.

Asklund, T., Appelskog, I. B., Ammerpohl, O., Langmoen, I. A., Dilber, M. S., Aints, A., Ekström, T. J. and Almqvist, P. M., (2003). Gap junction-mediated bystander effect in primary cultures of human malignant gliomas with recombinant expression of the HSVtk gene. Exp Cell Res. 284, 185-195.

Avgeropoulos, N. G. and Batchelor, T. T., (1999). New treatment strategies for malignant gliomas. Oncologist. 4, 209-224.

Bachman, K. E., Rountree, M. R. and Baylin, S. B., (2001). Dnmt3a and Dnmt3b are transcriptional repressors that exhibit unique localization properties to heterochromatin. J Biol Chem. 276, 32282-32287.

Baker, M. J., Brem, S., Daniels, S., Sherman, B. and Phuphanich, S., (2002). Complete response of a recurrent, multicentric malignant glioma in a patient treated with phenylbutyrate. J Neurooncol. 59, 239-242.

Basilion, J. P., Ichikawa, T. and Chiocca, E. A., (2000). Gene therapy of brain tumors: problems presented by physiological barriers. Neurosurg Focus. 8.

Benedetti, S., Pirola, B., Pollo, B., Magrassi, L., Bruzzone, M. G., Rigamonti, D., Galli, R., Selleri, S., Di Meco, F., De Fraja, C., Vescovi, A., Cattaneo, E. and Finocchiaro, G., (2000). Gene therapy of experimental brain tumors using neural progenitor cells. Nat Med. 6, 447-450.

Blesch, A. and Tuszynski, M. H., (2001). GDNF gene delivery to injured adult CNS motor neurons promotes axonal growth, expression of the trophic neuropeptide CGRP, and cellular protection. J Comp Neurol. 436, 399-410.

Blum, M. R., Liao, S. H., Good, S. S. and de Miranda, P., (1988). Pharmacokinetics and bioavailability of zidovudine in humans. Am J Med. 85, 189-194.

Brewster, M. E., Raghavan, K., Pop, E. and Bodor, N., (1994). Enhanced delivery of ganciclovir to the brain through the use of redox targeting. Antimicrob Agents Chemother. 38, 817-823.

Brown, A. B., Yang, W., Schmidt, N. O., Carroll, R., Leishear, K. K., Rainov, N. G., Black, P. M., Breakefield, X. O. and Aboody, K. S., (2003). Intravascular delivery of neural stem

39

cell lines to target intracranial and extracranial tumors of neural and non-neural origin. Hum Gene Ther. 14, 1777-1785.

Cattaneo, E. and McKay, R., (1990). Proliferation and differentiation of neuronal stem cells regulated by nerve growth factor. Nature. 347, 762-765.

Chen, T. and Li, E., (2004). Structure and function of eukaryotic DNA methyltransferases. Curr Top Dev Biol. 60, 55-89.

Chen, Z. J. and Pikaard, C. S., (1997). Epigenetic silencing of RNA polymerase I transcription: a role for DNA methylation and histone modification in nucleolar dominance. Genes Dev. 11, 2124-2136.

Chiocca, E. A., Aghi, M. and Fulci, G., (2003). Viral therapy for glioblastoma. Cancer J. 9, 167-179.

Claes, A., Idema, A. J. and Wesseling, P., (2007). Diffuse glioma growth: a guerilla war. Acta Neuropathol. 114, 443-458.

Cloughesy, T. F., Wen, P. Y., Robins, H. I., Chang, S. M., Groves, M. D., Fink, K. L., Junck, L., Schiff, D., Abrey, L., Gilbert, M. R., Lieberman, F., Kuhn, J., DeAngelis, L. M., Mehta, M., Raizer, J. J., Yung, W. K., Aldape, K., Wright, J., Lamborn, K. R. and Prados, M. D., (2006). Phase II trial of tipifarnib in patients with recurrent malignant glioma either receiving or not receiving enzyme-inducing antiepileptic drugs: a North American Brain Tumor Consortium Study. J Clin Oncol. 24, 3651-3656.

Corbeil, D., Roper, K., Fargeas, C. A., Joester, A. and Huttner, W. B., (2001). Prominin: a story of cholesterol, plasma membrane protrusions and human pathology. Traffic. 2, 82-91.

Culver, K. W., Ram, Z., Wallbridge, S., Ishii, H., Oldfield, E. H. and Blaese, R. M., (1992). In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science. 256, 1550-1552.

Dahlstrand, J., Collins, V. P. and Lendahl, U., (1992). Expression of the class VI intermediate filament nestin in human central nervous system tumors. Cancer Res. 52, 5334-5341.

Danks, M. K., Morton, C. L., Pawlik, C. A. and Potter, P. M., (1998). Overexpression of a rabbit liver carboxylesterase sensitizes human tumor cells to CPT-11. Cancer Res. 58, 20-22.

Deeken, J. F. and Loscher, W., (2007). The blood-brain barrier and cancer: transporters, treatment, and Trojan horses. Clin Cancer Res. 13, 1663-1674.

Denning, C. and Pitts, J. D., (1997). Bystander effects of different enzyme-prodrug systems for cancer gene therapy depend on different pathways for intercellular transfer of toxic metabolites, a factor that will govern clinical choice of appropriate regimes. Hum Gene Ther. 8, 1825-1835.

Desaknai, S., Lumniczky, K., Esik, O., Hamada, H. and Safrany, G., (2003). Local tumour irradiation enhances the anti-tumour effect of a double-suicide gene therapy system in a murine glioma model. J Gene Med. 5, 377-385.

Doherty, L., Gigas, D. C., Kesari, S., Drappatz, J., Kim, R., Zimmerman, J., Ostrowsky, L. and Wen, P. Y., (2006). Pilot study of the combination of EGFR and mTOR inhibitors in recurrent malignant gliomas. Neurology. 67, 156-158.

Ehrlich, M., Gama-Sosa, M. A., Huang, L. H., Midgett, R. M., Kuo, K. C., McCune, R. A. and Gehrke, C., (1982). Amount and distribution of 5-methylcytosine in human DNA from different types of tissues of cells. Nucleic Acids Res. 10, 2709-2721.

40

Ehtesham, M., Kabos, P., Gutierrez, M. A., Chung, N. H., Griffith, T. S., Black, K. L. and Yu, J. S., (2002a). Induction of glioblastoma apoptosis using neural stem cell-mediated delivery of tumor necrosis factor-related apoptosis-inducing ligand. Cancer Res. 62, 7170-7174.

Ehtesham, M., Kabos, P., Kabosova, A., Neuman, T., Black, K. L. and Yu, J. S., (2002b). The use of interleukin 12-secreting neural stem cells for the treatment of intracranial glioma. Cancer Res. 62, 5657-5663.

Elshami, A. A., Saavedra, A., Zhang, H., Kucharczuk, J. C., Spray, D. C., Fishman, G. I., Amin, K. M., Kaiser, L. R. and Albelda, S. M., (1996). Gap junctions play a role in the 'bystander effect' of the herpes simplex virus thymidine kinase/ganciclovir system in vitro. Gene Ther. 3, 85-92.

Eriksson, P. S., Perfilieva, E., Bjork-Eriksson, T., Alborn, A. M., Nordborg, C., Peterson, D. A. and Gage, F. H., (1998). Neurogenesis in the adult human hippocampus. Nat Med. 4, 1313-1317.

Evans, W. H. and Martin, P. E., (2002). Gap junctions: structure and function (Review). Mol Membr Biol. 19, 121-136.

Fick, J., Barker, F. G., 2nd, Dazin, P., Westphale, E. M., Beyer, E. C. and Israel, M. A., (1995). The extent of heterocellular communication mediated by gap junctions is predictive of bystander tumor cytotoxicity in vitro. Proc Natl Acad Sci U S A. 92, 11071-11075.

Freeman, S. M., Abboud, C. N., Whartenby, K. A., Packman, C. H., Koeplin, D. S., Moolten, F. L. and Abraham, G. N., (1993). The "bystander effect": tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res. 53, 5274-5283.

Freeman, S. M., Ramesh, R. and Marrogi, A. J., (1997). Immune system in suicide-gene therapy. Lancet. 349, 2-3.

Freeman, S. M. and Zwiebel, J. A., (1993). Gene therapy of cancer. Cancer Invest. 11, 676-688.

Fuks, F., Burgers, W. A., Brehm, A., Hughes-Davies, L. and Kouzarides, T., (2000). DNA methyltransferase Dnmt1 associates with histone deacetylase activity. Nat Genet. 24, 88-91.

Furman, P. A., Fyfe, J. A., St Clair, M. H., Weinhold, K., Rideout, J. L., Freeman, G. A., Lehrman, S. N., Bolognesi, D. P., Broder, S., Mitsuya, H. and et al., (1986). Phosphorylation of 3'-azido-3'-deoxythymidine and selective interaction of the 5'-triphosphate with human immunodeficiency virus reverse transcriptase. Proc Natl Acad Sci U S A. 83, 8333-8337.

Furnari, F. B., Fenton, T., Bachoo, R. M., Mukasa, A., Stommel, J. M., Stegh, A., Hahn, W. C., Ligon, K. L., Louis, D. N., Brennan, C., Chin, L., DePinho, R. A. and Cavenee, W. K., (2007). Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev. 21, 2683-2710.

Gage, F. H., (2000). Mammalian neural stem cells. Science. 287, 1433-1438.

Galanis, E., Buckner, J. C., Maurer, M. J., Kreisberg, J. I., Ballman, K., Boni, J., Peralba, J. M., Jenkins, R. B., Dakhil, S. R., Morton, R. F., Jaeckle, K. A., Scheithauer, B. W., Dancey, J., Hidalgo, M. and Walsh, D. J., (2005). Phase II trial of temsirolimus (CCI-779) in recurrent glioblastoma multiforme: a North Central Cancer Treatment Group Study. J Clin Oncol. 23, 5294-5304.

Gilbert, J., Baker, S. D., Bowling, M. K., Grochow, L., Figg, W. D., Zabelina, Y., Donehower, R. C. and Carducci, M. A., (2001). A phase I dose escalation and bioavailability study

41

of oral sodium phenylbutyrate in patients with refractory solid tumor malignancies. Clin Cancer Res. 7, 2292-2300.

Gilyarov, A. V., (2008). Nestin in central nervous system cells. Neurosci Behav Physiol. 38, 165-169.

Glass, R., Synowitz, M., Kronenberg, G., Walzlein, J. H., Markovic, D. S., Wang, L. P., Gast, D., Kiwit, J., Kempermann, G. and Kettenmann, H., (2005). Glioblastoma-induced attraction of endogenous neural precursor cells is associated with improved survival. J Neurosci. 25, 2637-2646.

Goldberg, G. S., Bechberger, J. F. and Naus, C. C., (1995). A pre-loading method of evaluating gap junctional communication by fluorescent dye transfer. Biotechniques. 18, 490-497.

Grignet-Debrus, C., Cool, V., Baudson, N., Degreve, B., Balzarini, J., De Leval, L., Debrus, S., Velu, T. and Calberg-Bacq, C. M., (2000). Comparative in vitro and in vivo cytotoxic activity of (E)-5-(2-bromovinyl)-2'-deoxyuridine (BVDU) and its arabinosyl derivative, (E)-5-(2-bromovinyl)-1-beta-D-arabinofuranosyluracil (BVaraU), against tumor cells expressing either the Varicella zoster or the Herpes simplex virus thymidine kinase. Cancer Gene Ther. 7, 215-223.

Hamel, W., Magnelli, L., Chiarugi, V. P. and Israel, M. A., (1996). Herpes simplex virus thymidine kinase/ganciclovir-mediated apoptotic death of bystander cells. Cancer Res. 56, 2697-2702.

Hampl, J. A., Camp, S. M., Mydlarz, W. K., Hampl, M., Ichikawa, T., Chiocca, E. A., Louis, D. N., Sena-Esteves, M. and Breakefield, X. O., (2003). Potentiated gene delivery to tumors using herpes simplex virus/Epstein-Barr virus/RV tribrid amplicon vectors. Hum Gene Ther. 14, 611-626.

Hapke, D. M., Stegmann, A. P. and Mitchell, B. S., (1996). Retroviral transfer of deoxycytidine kinase into tumor cell lines enhances nucleoside toxicity. Cancer Res. 56, 2343-2347.

Herrmann, H. and Aebi, U., (2000). Intermediate filaments and their associates: multi-talented structural elements specifying cytoarchitecture and cytodynamics. Curr Opin Cell Biol. 12, 79-90.

Holmin, S., Almqvist, P., Lendahl, U. and Mathiesen, T., (1997). Adult nestin-expressing subependymal cells differentiate to astrocytes in response to brain injury. Eur J Neurosci. 9, 65-75.

Houghton, F. D., (2005). Role of gap junctions during early embryo development. Reproduction. 129, 129-135.

Hsieh, J. and Gage, F. H., (2004). Epigenetic control of neural stem cell fate. Curr Opin Genet Dev. 14, 461-469.

Hsieh, J., Nakashima, K., Kuwabara, T., Mejia, E. and Gage, F. H., (2004). Histone deacetylase inhibition-mediated neuronal differentiation of multipotent adult neural progenitor cells. Proc Natl Acad Sci U S A. 101, 16659-16664.

Huang, R. P., Hossain, M. Z., Sehgal, A. and Boynton, A. L., (1999). Reduced connexin43 expression in high-grade human brain glioma cells. J Surg Oncol. 70, 21-24.

Huber, B. E., Austin, E. A., Richards, C. A., Davis, S. T. and Good, S. S., (1994). Metabolism of 5-fluorocytosine to 5-fluorouracil in human colorectal tumor cells transduced with the cytosine deaminase gene: significant antitumor effects when only a small percentage of tumor cells express cytosine deaminase. Proc Natl Acad Sci U S A. 91, 8302-8306.

42

Huynh, G. H., Deen, D. F. and Szoka, F. C., Jr., (2006). Barriers to carrier mediated drug and gene delivery to brain tumors. J Control Release. 110, 236-259.

Ignatova, T. N., Kukekov, V. G., Laywell, E. D., Suslov, O. N., Vrionis, F. D. and Steindler, D. A., (2002). Human cortical glial tumors contain neural stem-like cells expressing astroglial and neuronal markers in vitro. Glia. 39, 193-206.

Jackson, M., Krassowska, A., Gilbert, N., Chevassut, T., Forrester, L., Ansell, J. and Ramsahoye, B., (2004). Severe global DNA hypomethylation blocks differentiation and induces histone hyperacetylation in embryonic stem cells. Mol Cell Biol. 24, 8862-8871.

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

Johansson, C. B., Momma, S., Clarke, D. L., Risling, M., Lendahl, U. and Frisen, J., (1999). Identification of a neural stem cell in the adult mammalian central nervous system. Cell. 96, 25-34.

Johnstone, R. W., (2002). Histone-deacetylase inhibitors: novel drugs for the treatment of cancer. Nat Rev Drug Discov. 1, 287-299.

Jones, P. L. and Wolffe, A. P., (1999). Relationships between chromatin organization and DNA methylation in determining gene expression. Semin Cancer Biol. 9, 339-347.

Kan, O., Kingsman, S. and Naylor, S., (2002). Cytochrome P450-based cancer gene therapy: current status. Expert Opin Biol Ther. 2, 857-868.

Kania, G., Corbeil, D., Fuchs, J., Tarasov, K. V., Blyszczuk, P., Huttner, W. B., Boheler, K. R. and Wobus, A. M., (2005). Somatic stem cell marker prominin-1/CD133 is expressed in embryonic stem cell-derived progenitors. Stem Cells. 23, 791-804.

Karimi, M., Johansson, S., Stach, D., Corcoran, M., Grander, D., Schalling, M., Bakalkin, G., Lyko, F., Larsson, C. and Ekstrom, T. J., (2006). LUMA (LUminometric Methylation Assay)--a high throughput method to the analysis of genomic DNA methylation. Exp Cell Res. 312, 1989-1995.

Kim, S. K., Kim, S. U., Park, I. H., Bang, J. H., Aboody, K. S., Wang, K. C., Cho, B. K., Kim, M., Menon, L. G., Black, P. M. and Carroll, R. S., (2006). Human neural stem cells target experimental intracranial medulloblastoma and deliver a therapeutic gene leading to tumor regression. Clin Cancer Res. 12, 5550-5556.

King, T. J. and Lampe, P. D., (2005). Temporal regulation of connexin phosphorylation in embryonic and adult tissues. Biochim Biophys Acta. 1719, 24-35.

Klecker, R. W., Jr., Collins, J. M., Yarchoan, R., Thomas, R., Jenkins, J. F., Broder, S. and Myers, C. E., (1987). Plasma and cerebrospinal fluid pharmacokinetics of 3'-azido-3'-deoxythymidine: a novel pyrimidine analog with potential application for the treatment of patients with AIDS and related diseases. Clin Pharmacol Ther. 41, 407-412.

Knecht, W., Rozpedowska, E., Le Breton, C., Willer, M., Gojkovic, Z., Sandrini, M. P., Joergensen, T., Hasholt, L., Munch-Petersen, B. and Piskur, J., (2007). Drosophila deoxyribonucleoside kinase mutants with enhanced ability to phosphorylate purine analogs. Gene Ther. 14, 1278-1286.

Koipally, J., Renold, A., Kim, J. and Georgopoulos, K., (1999). Repression by Ikaros and Aiolos is mediated through histone deacetylase complexes. Embo J. 18, 3090-3100.

Kondo, T., (2006). Epigenetic alchemy for cell fate conversion. Curr Opin Genet Dev. 16, 502-507.

43

Lampe, P. D. and Lau, A. F., (2000). Regulation of gap junctions by phosphorylation of connexins. Arch Biochem Biophys. 384, 205-215.

Lampe, P. D. and Lau, A. F., (2004). The effects of connexin phosphorylation on gap junctional communication. Int J Biochem Cell Biol. 36, 1171-1186.

Lavie, A., Schlichting, I., Vetter, I. R., Konrad, M., Reinstein, J. and Goody, R. S., (1997). The bottleneck in AZT activation. Nat Med. 3, 922-924.

Lawler, S. E., Peruzzi, P. P. and Chiocca, E. A., (2006). Genetic strategies for brain tumor therapy. Cancer Gene Ther. 13, 225-233.

Lee, J. H., Hart, S. R. and Skalnik, D. G., (2004). Histone deacetylase activity is required for embryonic stem cell differentiation. Genesis. 38, 32-38.

Lendahl, U., Zimmerman, L. B. and McKay, R. D., (1990). CNS stem cells express a new class of intermediate filament protein. Cell. 60, 585-595.

Li, E., (2002). Chromatin modification and epigenetic reprogramming in mammalian development. Nat Rev Genet. 3, 662-673.

Li, S., Tokuyama, T., Yamamoto, J., Koide, M., Yokota, N. and Namba, H., (2005). Bystander effect-mediated gene therapy of gliomas using genetically engineered neural stem cells. Cancer Gene Ther. 12, 600-607.

Louis, D. N., Ohgaki, H., Wiestler, O.D., Cavenee, W.K., (2007). WHO classification of tumours of the central nervous system. International Agency for Research on Cancer Press, Lyon.

Lukas, R. V., Boire, A. and Nicholas, M. K., (2007). Emerging therapies for malignant glioma. Expert Rev Anticancer Ther. 7, S29-36.

Lumelsky, N., Blondel, O., Laeng, P., Velasco, I., Ravin, R. and McKay, R., (2001). Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science. 292, 1389-1394.

Ma, H. I., Lin, S. Z., Chiang, Y. H., Li, J., Chen, S. L., Tsao, Y. P. and Xiao, X., (2002). Intratumoral gene therapy of malignant brain tumor in a rat model with angiostatin delivered by adeno-associated viral (AAV) vector. Gene Ther. 9, 2-11.

Marconi, P., Tamura, M., Moriuchi, S., Krisky, D. M., Niranjan, A., Goins, W. F., Cohen, J. B. and Glorioso, J. C., (2000). Connexin 43-enhanced suicide gene therapy using herpesviral vectors. Mol Ther. 1, 71-81.

Marin-Husstege, M., Muggironi, M., Liu, A. and Casaccia-Bonnefil, P., (2002). Histone deacetylase activity is necessary for oligodendrocyte lineage progression. J Neurosci. 22, 10333-10345.

Marzesco, A. M., Janich, P., Wilsch-Brauninger, M., Dubreuil, V., Langenfeld, K., Corbeil, D. and Huttner, W. B., (2005). Release of extracellular membrane particles carrying the stem cell marker prominin-1 (CD133) from neural progenitors and other epithelial cells. J Cell Sci. 118, 2849-2858.

Mellinghoff, I. K., Wang, M. Y., Vivanco, I., Haas-Kogan, D. A., Zhu, S., Dia, E. Q., Lu, K. V., Yoshimoto, K., Huang, J. H., Chute, D. J., Riggs, B. L., Horvath, S., Liau, L. M., Cavenee, W. K., Rao, P. N., Beroukhim, R., Peck, T. C., Lee, J. C., Sellers, W. R., Stokoe, D., Prados, M., Cloughesy, T. F., Sawyers, C. L. and Mischel, P. S., (2005). Molecular determinants of the response of glioblastomas to EGFR kinase inhibitors. N Engl J Med. 353, 2012-2024.

44

Mercer, K. E., Ahn, C. E., Coke, A., Compadre, C. M. and Drake, R. R., (2002). Mutation of herpesvirus thymidine kinase to generate ganciclovir-specific kinases for use in cancer gene therapies. Protein Eng. 15, 903-911.

Meshorer, E., (2007). Chromatin in embryonic stem cell neuronal differentiation. Histol Histopathol. 22, 311-319.

Mesnil, M., Crespin, S., Avanzo, J. L. and Zaidan-Dagli, M. L., (2005). Defective gap junctional intercellular communication in the carcinogenic process. Biochim Biophys Acta. 1719, 125-145.

Mesnil, M., Piccoli, C., Tiraby, G., Willecke, K. and Yamasaki, H., (1996). Bystander killing of cancer cells by herpes simplex virus thymidine kinase gene is mediated by connexins. Proc Natl Acad Sci U S A. 93, 1831-1835.

Mesnil, M. and Yamasaki, H., (2000). Bystander effect in herpes simplex virus-thymidine kinase/ganciclovir cancer gene therapy: role of gap-junctional intercellular communication. Cancer Res. 60, 3989-3999.

Michalczyk, K. and Ziman, M., (2005). Nestin structure and predicted function in cellular cytoskeletal organisation. Histol Histopathol. 20, 665-671.

Miura, F., Moriuchi, S., Maeda, M., Sano, A., Maruno, M., Tsanaclis, A. M., Marino, R., Jr., Glorioso, J. C. and Yoshimine, T., (2002). Sustained release of low-dose ganciclovir from a silicone formulation prolonged the survival of rats with gliosarcomas under herpes simplex virus thymidine kinase suicide gene therapy. Gene Ther. 9, 1653-1658.

Mizrak, D., Brittan, M. and Alison, M. R., (2008). CD133: molecule of the moment. J Pathol. 214, 3-9.

Moorby, C. and Patel, M., (2001). Dual functions for connexins: Cx43 regulates growth independently of gap junction formation. Exp Cell Res. 271, 238-248.

Muldoon, L. L., Soussain, C., Jahnke, K., Johanson, C., Siegal, T., Smith, Q. R., Hall, W. A., Hynynen, K., Senter, P. D., Peereboom, D. M. and Neuwelt, E. A., (2007). Chemotherapy delivery issues in central nervous system malignancy: a reality check. J Clin Oncol. 25, 2295-2305.

Muller, F. J., Snyder, E. Y. and Loring, J. F., (2006). Gene therapy: can neural stem cells deliver? Nat Rev Neurosci. 7, 75-84.

Nan, X., Ng, H. H., Johnson, C. A., Laherty, C. D., Turner, B. M., Eisenman, R. N. and Bird, A., (1998). Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature. 393, 386-389.

Ng, H. H. and Bird, A., (1999). DNA methylation and chromatin modification. Curr Opin Genet Dev. 9, 158-163.

Niculescu-Duvaz, I. and Springer, C. J., (2005). Introduction to the background, principles, and state of the art in suicide gene therapy. Mol Biotechnol. 30, 71-88.

Okano, M., Bell, D. W., Haber, D. A. and Li, E., (1999). DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell. 99, 247-257.

Oldfield, E. H., Ram, Z., Culver, K. W., Blaese, R. M., DeVroom, H. L. and Anderson, W. F., (1993). Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum Gene Ther. 4, 39-69.

45

Ou, J. N., Torrisani, J., Unterberger, A., Provencal, N., Shikimi, K., Karimi, M., Ekstrom, T. J. and Szyf, M., (2007). Histone deacetylase inhibitor Trichostatin A induces global and gene-specific DNA demethylation in human cancer cell lines. Biochem Pharmacol. 73, 1297-1307.

Ourednik, J., Ourednik, V., Lynch, W. P., Schachner, M. and Snyder, E. Y., (2002). Neural stem cells display an inherent mechanism for rescuing dysfunctional neurons. Nat Biotechnol. 20, 1103-1110.

Palmer, D. H., Mautner, V., Mirza, D., Oliff, S., Gerritsen, W., van der Sijp, J. R., Hubscher, S., Reynolds, G., Bonney, S., Rajaratnam, R., Hull, D., Horne, M., Ellis, J., Mountain, A., Hill, S., Harris, P. A., Searle, P. F., Young, L. S., James, N. D. and Kerr, D. J., (2004). Virus-directed enzyme prodrug therapy: intratumoral administration of a replication-deficient adenovirus encoding nitroreductase to patients with resectable liver cancer. J Clin Oncol. 22, 1546-1552.

Palu, G., Cavaggioni, A., Calvi, P., Franchin, E., Pizzato, M., Boschetto, R., Parolin, C., Chilosi, M., Ferrini, S., Zanusso, A. and Colombo, F., (1999). Gene therapy of glioblastoma multiforme via combined expression of suicide and cytokine genes: a pilot study in humans. Gene Ther. 6, 330-337.

Parkin, D. M., Bray, F., Ferlay, J. and Pisani, P., (2005). Global cancer statistics, 2002. CA Cancer J Clin. 55, 74-108.

Penas-Prado, M. and Gilbert, M. R., (2007). Molecularly targeted therapies for malignant gliomas: advances and challenges. Expert Rev Anticancer Ther. 7, 641-661.

Phuphanich, S., Baker, S. D., Grossman, S. A., Carson, K. A., Gilbert, M. R., Fisher, J. D. and Carducci, M. A., (2005). Oral sodium phenylbutyrate in patients with recurrent malignant gliomas: a dose escalation and pharmacologic study. Neuro-oncol. 7, 177-182.

Pope, I. M., Poston, G. J. and Kinsella, A. R., (1997). The role of the bystander effect in suicide gene therapy. Eur J Cancer. 33, 1005-1016.

Rainov, N. G. and Kramm, C. M., (2003). Recombinant retrovirus vectors for treatment of malignant brain tumors. Int Rev Neurobiol. 55, 185-203.

Ram, Z., Culver, K. W., Oshiro, E. M., Viola, J. J., DeVroom, H. L., Otto, E., Long, Z., Chiang, Y., McGarrity, G. J., Muul, L. M., Katz, D., Blaese, R. M. and Oldfield, E. H., (1997). Therapy of malignant brain tumors by intratumoral implantation of retroviral vector-producing cells. Nat Med. 3, 1354-1361.

Ren, H., Yang, B. F. and Rainov, N. G., (2007). Receptor tyrosine kinases as therapeutic targets in malignant glioma. Rev Recent Clin Trials. 2, 87-101.

Rolinski, B., Bogner, J. R., Sadri, I., Wintergerst, U. and Goebel, F. D., (1997). Absorption and elimination kinetics of zidovudine in the cerebrospinal fluid in HIV-1-infected patients. J Acquir Immune Defic Syndr Hum Retrovirol. 15, 192-197.

Roloff, T. C. and Nuber, U. A., (2005). Chromatin, epigenetics and stem cells. Eur J Cell Biol. 84, 123-135.

Rose, B., Mehta, P. P. and Loewenstein, W. R., (1993). Gap-junction protein gene suppresses tumorigenicity. Carcinogenesis. 14, 1073-1075.

Rosenqvist, N., Hard Af Segerstad, C., Samuelsson, C., Johansen, J. and Lundberg, C., (2002). Activation of silenced transgene expression in neural precursor cell lines by inhibitors of histone deacetylation. J Gene Med. 4, 248-257.

46

Roybon, L., Hjalt, T., Christophersen, N. S., Li, J. Y. and Brundin, P., (2008). Effects on differentiation of embryonic ventral midbrain progenitors by Lmx1a, Msx1, Ngn2, and Pitx3. J Neurosci. 28, 3644-3656.

Saez, J. C., Berthoud, V. M., Branes, M. C., Martinez, A. D. and Beyer, E. C., (2003). Plasma membrane channels formed by connexins: their regulation and functions. Physiol Rev. 83, 1359-1400.

Salgaller, M. L. and Liau, L. M., (2006). Current status of clinical trials for glioblastoma. Rev Recent Clin Trials. 1, 265-281.

Salvati, M., Caroli, E., Orlando, E. R., Frati, A., Artizzu, S. and Ferrante, L., (2003). Multicentric glioma: our experience in 25 patients and critical review of the literature. Neurosurg Rev. 26, 275-279.

Samejima, Y. and Meruelo, D., (1995). 'Bystander killing' induces apoptosis and is inhibited by forskolin. Gene Ther. 2, 50-58.

Sanai, N., Alvarez-Buylla, A. and Berger, M. S., (2005). Neural stem cells and the origin of gliomas. N Engl J Med. 353, 811-822.

Sanai, N., Tramontin, A. D., Quinones-Hinojosa, A., Barbaro, N. M., Gupta, N., Kunwar, S., Lawton, M. T., McDermott, M. W., Parsa, A. T., Manuel-Garcia Verdugo, J., Berger, M. S. and Alvarez-Buylla, A., (2004). Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration. Nature. 427, 740-744.

Sandrini, M. P. and Piskur, J., (2005). Deoxyribonucleoside kinases: two enzyme families catalyze the same reaction. Trends Biochem Sci. 30, 225-228.

Selker, E. U., (1998). Trichostatin A causes selective loss of DNA methylation in Neurospora. Proc Natl Acad Sci U S A. 95, 9430-9435.

Shah, K., Bureau, E., Kim, D. E., Yang, K., Tang, Y., Weissleder, R. and Breakefield, X. O., (2005). Glioma therapy and real-time imaging of neural precursor cell migration and tumor regression. Ann Neurol. 57, 34-41.

Shen, S., Li, J. and Casaccia-Bonnefil, P., (2005). Histone modifications affect timing of oligodendrocyte progenitor differentiation in the developing rat brain. J Cell Biol. 169, 577-589.

Shih, C. C., Weng, Y., Mamelak, A., LeBon, T., Hu, M. C. and Forman, S. J., (2001). Identification of a candidate human neurohematopoietic stem-cell population. Blood. 98, 2412-2422.

Singh, S. K., Hawkins, C., Clarke, I. D., Squire, J. A., Bayani, J., Hide, T., Henkelman, R. M., Cusimano, M. D. and Dirks, P. B., (2004). Identification of human brain tumour initiating cells. Nature. 432, 396-401.

Springer, C. J. and Niculescu-Duvaz, I., (2000). Prodrug-activating systems in suicide gene therapy. J Clin Invest. 105, 1161-1167.

Staflin, K., Honeth, G., Kalliomaki, S., Kjellman, C., Edvardsen, K. and Lindvall, M., (2004). Neural progenitor cell lines inhibit rat tumor growth in vivo. Cancer Res. 64, 5347-5354.

Staflin, K., Lindvall, M., Zuchner, T. and Lundberg, C., (2007). Instructive cross-talk between neural progenitor cells and gliomas. J Neurosci Res. 85, 2147-2159.

Stupp, R., Mason, W. P., van den Bent, M. J., Weller, M., Fisher, B., Taphoorn, M. J., Belanger, K., Brandes, A. A., Marosi, C., Bogdahn, U., Curschmann, J., Janzer, R. C.,

47

Ludwin, S. K., Gorlia, T., Allgeier, A., Lacombe, D., Cairncross, J. G., Eisenhauer, E. and Mirimanoff, R. O., (2005). Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 352, 987-996.

Sturtz, F. G., Waddell, K., Shulok, J., Chen, X., Caruso, M., Sanson, M., Snodgrass, H. R. and Platika, D., (1997). Variable efficiency of the thymidine kinase/ganciclovir system in human glioblastoma cell lines: implications for gene therapy. Hum Gene Ther. 8, 1945-1953.

Summar, M., (2001). Current strategies for the management of neonatal urea cycle disorders. J Pediatr. 138, S30-39.

Svendsen, C. N. and Langston, J. W., (2004). Stem cells for Parkinson disease and ALS: replacement or protection? Nat Med. 10, 224-225.

Takamiya, Y., Short, M. P., Ezzeddine, Z. D., Moolten, F. L., Breakefield, X. O. and Martuza, R. L., (1992). Gene therapy of malignant brain tumors: a rat glioma line bearing the herpes simplex virus type 1-thymidine kinase gene and wild type retrovirus kills other tumor cells. J Neurosci Res. 33, 493-503.

Takizawa, T., Nakashima, K., Namihira, M., Ochiai, W., Uemura, A., Yanagisawa, M., Fujita, N., Nakao, M. and Taga, T., (2001). DNA methylation is a critical cell-intrinsic determinant of astrocyte differentiation in the fetal brain. Dev Cell. 1, 749-758.

Trask, T. W., Trask, R. P., Aguilar-Cordova, E., Shine, H. D., Wyde, P. R., Goodman, J. C., Hamilton, W. J., Rojas-Martinez, A., Chen, S. H., Woo, S. L. and Grossman, R. G., (2000). Phase I study of adenoviral delivery of the HSV-tk gene and ganciclovir administration in patients with current malignant brain tumors. Mol Ther. 1, 195-203.

Uchida, N., Buck, D. W., He, D., Reitsma, M. J., Masek, M., Phan, T. V., Tsukamoto, A. S., Gage, F. H. and Weissman, I. L., (2000). Direct isolation of human central nervous system stem cells. Proc Natl Acad Sci U S A. 97, 14720-14725.

Uhl, M., Weiler, M., Wick, W., Jacobs, A. H., Weller, M. and Herrlinger, U., (2005). Migratory neural stem cells for improved thymidine kinase-based gene therapy of malignant gliomas. Biochem Biophys Res Commun. 328, 125-129.

Wang, W. J., Tai, C. K., Kasahara, N. and Chen, T. C., (2003). Highly efficient and tumor-restricted gene transfer to malignant gliomas by replication-competent retroviral vectors. Hum Gene Ther. 14, 117-127.

Wen, P. Y. and Kesari, S., (2008). Malignant gliomas in adults. N Engl J Med. 359, 492-507.

Whartenby, K. A., Abboud, C. N., Marrogi, A. J., Ramesh, R. and Freeman, S. M., (1995). The biology of cancer gene therapy. Lab Invest. 72, 131-145.

Wiewrodt, R., Amin, K., Kiefer, M., Jovanovic, V. P., Kapoor, V., Force, S., Chang, M., Lanuti, M., Black, M. E., Kaiser, L. R. and Albelda, S. M., (2003). Adenovirus-mediated gene transfer of enhanced Herpes simplex virus thymidine kinase mutants improves prodrug-mediated tumor cell killing. Cancer Gene Ther. 10, 353-364.

Wilcken, B., (2004). Problems in the management of urea cycle disorders. Mol Genet Metab. 81 Suppl 1, S86-91.

Villa, A., Snyder, E. Y., Vescovi, A. and Martinez-Serrano, A., (2000). Establishment and properties of a growth factor-dependent, perpetual neural stem cell line from the human CNS. Exp Neurol. 161, 67-84.

48

Willecke, K., Eiberger, J., Degen, J., Eckardt, D., Romualdi, A., Guldenagel, M., Deutsch, U. and Sohl, G., (2002). Structural and functional diversity of connexin genes in the mouse and human genome. Biol Chem. 383, 725-737.

Vredenburgh, J. J., Desjardins, A., Herndon, J. E., 2nd, Dowell, J. M., Reardon, D. A., Quinn, J. A., Rich, J. N., Sathornsumetee, S., Gururangan, S., Wagner, M., Bigner, D. D., Friedman, A. H. and Friedman, H. S., (2007). Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res. 13, 1253-1259.

Wygoda, M. R., Wilson, M. R., Davis, M. A., Trosko, J. E., Rehemtulla, A. and Lawrence, T. S., (1997). Protection of herpes simplex virus thymidine kinase-transduced cells from ganciclovir-mediated cytotoxicity by bystander cells: the Good Samaritan effect. Cancer Res. 57, 1699-1703.

Xu, W. S., Parmigiani, R. B. and Marks, P. A., (2007). Histone deacetylase inhibitors: molecular mechanisms of action. Oncogene. 26, 5541-5552.

Yang, S. Y., Liu, H. and Zhang, J. N., (2004). Gene therapy of rat malignant gliomas using neural stem cells expressing IL-12. DNA Cell Biol. 23, 381-389.

Yazawa, K., Fisher, W. E. and Brunicardi, F. C., (2002). Current progress in suicide gene therapy for cancer. World J Surg. 26, 783-789.

Yip, S., Aboody, K. S., Burns, M., Imitola, J., Boockvar, J. A., Allport, J., Park, K. I., Teng, Y. D., Lachyankar, M., McIntosh, T., O'Rourke, D. M., Khoury, S., Weissleder, R., Black, P. M., Weiss, W. and Snyder, E. Y., (2003). Neural stem cell biology may be well suited for improving brain tumor therapies. Cancer J. 9, 189-204.

Yoshida, J., Mizuno, M., Fujii, M., Kajita, Y., Nakahara, N., Hatano, M., Saito, R., Nobayashi, M. and Wakabayashi, T., (2004). Human gene therapy for malignant gliomas (glioblastoma multiforme and anaplastic astrocytoma) by in vivo transduction with human interferon beta gene using cationic liposomes. Hum Gene Ther. 15, 77-86.

Young, J. C., Wu, S., Hansteen, G., Du, C., Sambucetti, L., Remiszewski, S., O'Farrell, A. M., Hill, B., Lavau, C. and Murray, L. J., (2004). Inhibitors of histone deacetylases promote hematopoietic stem cell self-renewal. Cytotherapy. 6, 328-336.

Zhang, Y. W., Kaneda, M. and Morita, I., (2003). The gap junction-independent tumor-suppressing effect of connexin 43. J Biol Chem. 278, 44852-44856.

Zheng, X., Johansson, M. and Karlsson, A., (2000). Retroviral transduction of cancer cell lines with the gene encoding Drosophila melanogaster multisubstrate deoxyribonucleoside kinase. J Biol Chem. 275, 39125-39129.