ELUCIDATING THE ROLE OF AUTOPHAGY IN ZEBRAFISH MODELS OF LIVER CANCER

ELUCIDATING THE ROLE OF AUTOPHAGY IN ZEBRAFISH MODELS OF LIVER CANCER SIM HUEY FEN, TINA (B.Sc.(Hons.)), NUS NATIONAL UNIVERSITY OF SINGAPORE 2011 ELUCIDATING THE ROLE OF AUTOPHAGY IN ZEBRAFISH MODELS OF LIVER CANCER SIM HUEY FEN, TINA (B.Sc.(Hons.)), NUS A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE DEPARTMENT OF BIOLOGICAL SCIENCES NATIONAL UNIVERSITY OF SINGAPORE 2011 Acknowledgements Acknowledgements I would like to take this opportunity to express my greatest gratitude to my honorific supervisors, Professor Gong Zhiyuan (Department of Biological Sciences, Faculty of Science, NUS), and Associate Professor Shen Han Ming (Department of Epidemiology and Public Health, Yong Loo Lin School of Medicine, NUS) for taking me under their wing, as well as for their patience, and invaluable advice and encouragement offered throughout the entirety of my research. I would also like to express my gratitude to fellow laboratory mates from Prof Gong’s laboratory, both past and present, namely, Weiling, Grace, Li Zhen, Zhou Li, Cai Xia, Anh Tuan, Lili, Sahar, Xiao Qian, Hongyan, Xiaoyan, Shen Yuan, Ti Weng, Yin Ao, Myintzu, Yan Tie, Huiqing and Zhengyuan for their advice and encouragement. Special thanks to Jianzhou for the engaging brainstorming sessions and assistance rendered during my research; and Lora for making my stay in the lab a very memorable one. I also appreciate the favours from the members of A/P Shen’s laboratory, especially Zhou Jing, even though I am not a full member of the laboratory. I would also like to take this opportunity to thank staff from the fish facility in the department for their efforts in maintaining quality fish stocks as well as general assistance rendered. In addition, I would like to take this opportunity to show my appreciation to the National University of Singapore for providing me with the graduate research scholarship during these few years. Last but not least, I would like to express my love and gratitude to my dearest family for their love and unwavering support while I pursue my dreams and passion in research. i Table of contents Table of contents Acknowledgements i Table of Contents ii-iv Summary v-vi List of Tables vii List of Figures viii-x List of Common Abbreviations xi-xii Introduction 1. The use of zebrafish as a model organism for the study of cancer 1.1 The use of zebrafish as a model organism in vertebrate developmental biology studies 1.2 The use of zebrafish as a model organism in cancer studies 1.3 Generation of inducible Xmrk and cmyc driven liver cancer models 1.3.1 The use of inducible promoter systems to control transgene expression 1.3.2 The use of transposable elements to generate transgenic zebrafish 1.3.3 Xmrk and cmyc driven liver cancer models 2. An introduction to autophagy and its involvement in cancer 2.1 Autophagy as a basal and inducible cellular process 2.2 Molecular machinery of autophagy 2.3 Regulation of autophagy 2.4 The use of EGFP-LC3 and mRFP-EGFP-LC3 to study autophagy in vivo 2.5 Autophagy in cancer 3. Rationale of current study 3.1 Establishing Tg(fabp10:egfp-lc3) and Tg(fabp10:mrfp-egfp-lc3) transgenic zebrafish for studying autophagy 3.2 Establishing the role of autophagy in liver cancer 3.3 Objectives of study Materials and Methods 1. Zebrafish care and maintenance and tissue collection 1.1 Zebrafish maintenance 1.2 Zebrafish spawning and embryo collection 1.3 Microinjection into one-cell stage embryos 1 2 2 4 7 7 8 9 11 11 13 18 23 25 29 29 30 31 33 34 34 34 35 ii Table of contents 1.4 Larval screening and photo-documentation 1.5 Liver tissue collection from adult zebrafish 2. RNA applications 2.1 Total RNA isolation from embryos and larvae 2.2 RNA quantification 2.3 Ac mRNA preparation 3. DNA applications 3.1 Total DNA isolation 3.2 cDNA synthesis 3.3 Construction of plasmid DNA 3.3.1 Restriction endonuclease digestion of DNA 3.3.2 Recovery of DNA fragments from DNA gel 3.3.3 Ligation 3.3.4 Competent cell preparation 3.3.5 Transformation and re-transformation reactions 3.3.6 Colony screening 3.3.7 Bacterial cell culture and plasmid amplification 3.3.8 Plasmid isolation and purification 3.4 Polymerase chain reaction 3.4.1 Standard PCR 3.4.2 2-step reverse transcription polymerase chain reaction (2-step RT-PCR) 3.5 DNA vectors 3.5.1 pLF3.0-EGFP 3.5.2 pEGFP-LC3 3.5.3 pmRFP-EGFP-LC3 3.5.4 pMDS6 3.5.5 pAc-SP6 4. Protein applications 4.1 Total protein isolation and extraction 4.2 Protein quantification 4.3 Protein detection 4.3.1 SDS-PAGE gel casting 4.3.2 Running SDS-PAGE 4.3.3 Protein transfer from PAGE gel to Polyvinylidene fluoride (PVDF) membrane 4.3.4 Immunoblotting 4.3.5 Protein band detection 4.3.6 Calculating band intensity from immunoblots 5. Starvation and chemical treatments 5.1 Starvation treatments 5.2 Chemical treatments 5.3 Doxycycline induction of Xmrk and cmyc transgene expression 36 36 37 37 38 38 39 39 40 41 41 43 43 44 45 46 46 46 47 48 50 52 52 53 54 55 56 57 57 57 58 58 59 60 60 61 61 62 62 63 64 iii Table of contents 6. External photo-documentation of adult treated transgenic fish 7. Cryosectioning 7.1Cryostat sections of larvae 7.2 Cryostat sections of adult tissues 7.3 Hoechst staining 8. Confocal imaging 8.1 Live confocal imaging 8.2 Confocal imaging of sections 9. Bioinformatics and sequence analyses 10. Analyses of liver transcriptome data by deep sequencing 64 65 65 66 66 66 66 67 67 68 Results 1. Conserved autophagic machinery in zebrafish 1.1 Atg gene expression in zebrafish 1.2 Conservation of Atg proteins in zebrafish 1.3 Expression of autophagic genes and proteins in zebrafish 1.4 Conserved autophagic machinery in zebrafish 2. Establishment and characterisation of Tg(fabp10:egfp-lc3) transgenic line 2.1 Establishing Tg(fabp10:egfp-lc3) transgenic line 2.2 Characterisation of Tg(fabp10:egfp-lc3) transgenic line 3. Establishment and characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line 3.1 Establishing Tg(fabp10:mrfp-egfp-lc3) transgenic line 3.2 Characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line 4. Establishing the role of autophagy in liver cancer 4.1 Preliminary data implicating autophagy inhibition in liver cancer 4.2 Establishing double transgenic lines with liver-specific oncogene overexpression and EGFP-LC3 expression to study the role of autophagy in liver cancer 4.2.1 Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) 4.2.2 Tg(fabp10:TA; TRE:cmyc; krt4:RFP) 69 70 70 76 77 81 85 85 93 99 99 104 108 110 112 115 116 Discussion 1. Conserved autophagic machinery in zebrafish 2. Establishment and characterisation of Tg(fabp10:egfp-lc3) transgenic line 3. Establishment and characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line 4. Establishing the role of autophagy in liver cancer 5. Future work and conclusions 124 125 130 References 143 134 135 140 iv Summary Summary Autophagy is an evolutionarily conserved cellular process that involves autophagosomal sequestration of cytoplasm, long-lived proteins and organelles, and subsequent lysosomal degradation, in response to cellular stress such as starvation. It is also involved in the degradation of superfluous organelles, proteins and protein aggregates. In addition to its main role in regulating basal cellular homeostasis, it has also been implicated in a number of diseases. Presently, there are evidences implicating autophagy induction and inhibition during carcinogenesis; however the exact role of autophagy in carcinogenesis remains to be elucidated. With the zebrafish gaining popularity as a model organism in the study of human diseases including cancers, we aim to investigate the role of autophagy in liver cancer using our established zebrafish liver cancer models. As little study is so far conducted on autophagy in zebrafish, we first characterized autophagy in zebrafish, and found the autophagic machinery and function are well-conserved between human and zebrafish. We also established and characterized a transgenic line, Tg(fabp10:egfp-lc3), with constitutively liver-specific egfp-lc3 expression, in order to visualize the autophagic flux in the liver. We found that this transgenic line is able to produce EGFP-LC3 puncta under starvation and everolimus treatment, both processes inducing autophagy. Preliminary studies using our liver cancer transgenic zebrafish with inducible expression of cmyc or Xmrk oncogene in the liver suggested that autophagy may be inhibited during liver carcinogenesis. To further study this phenomenon, we crossed Tg(fabp10:egfp-lc3) with these oncogene transgenic lines in order to visualize the autophagic flux during tumour initiation, progression and regression in vivo, and we obtained evidences suggesting autophagy inhibition during liver carcinogenesis. We further observed that autophagy may have been de-repressed during tumour v Summary regression, although more studies need to be performed to validate our observations. Finally, we also developed a transgenic line with constitutively liver-specific mrfpegfp-lc3 expression. This transgenic line was developed based on the concept of lysosomal quenching of GFP fluorescence, with the mRFP-EGFP-LC3 reporter labelling autophagosomes yellow and autolysosomes red, thus allowing autophagic flux to be measured by comparing the ratio between yellow and red puncta. This will be the first attempt in generating a transgenic organism expressing this novel reporter. We hope that this second transgenic line will be useful and complement the first transgenic line in elucidating autophagic flux during liver carcinogenesis as well as other liver diseases. vi List of tables List of tables Table 1 Restriction enzymes used for sub-cloning 42 Table 2 Primers used in PCR 49 Table 3 Primers used in 2-step RT-PCR 51 Table 4 atg gene expression in the zebrafish 72 Table 5 Comparison of zebrafish and human Atg proteins 73 vii List of figures List of figures Figure 1 Schematic illustration of the autophagy pathway and its core machinery in mammalian cells. 17 Figure 2 Regulation of mammalian autophagy. 22 Figure 3 Autophagy in tumourigenesis and anti-cancer therapy. 28 Figure 4 Construct map of pLF3.0-EGFP. 52 Figure 5 Construct map of pEGFP-LC3. 53 Figure 6 Construct map of pmRFP-EGFP-LC3. 54 Figure 7 Construct map of pMDS6. 55 Figure 8 Construct map of pAC-SP6. 56 Figure 9 Conservation of LC3 protein in zebrafish. 74 Figure 10 Conservation of Atg5 protein in zebrafish. 75-76 Figure 11 Temporal expression patterns of lc3 and atg5 from 1 dpf to 5 dpf zebrafish larvae. 79 Figure 12 LC3-I and LC3-II protein levels from 1 dpf to 5 dpf zebrafish larvae. 80 Figure 13 Autophagy induction by rapamycin. 83 Figure 14 Autophagy induction is necessary for larvae to survive starvation. 84 Figure 15 The pDS-FABP10-EGFP-LC3 plasmid used generation of Tg(fabp10:egfp-lc3) transgenic line. the 86 Figure 16 Sequence identity exists between rat and zebrafish LC3 proteins. 87 Figure 17 Liver-specific EGFP-LC3 expression after microinjection of pDS-FABP10-EGFP-LC3. 89 Figure 18 EGFP-LC3 expression in the progeny of the five founders. 90 in viii List of figures Figure 19 EGFP-LC3 positive larvae in Line 1 harbours germline transmission of the fabp10:egfp-lc3 expression cassette. 91-92 Figure 20 Autophagy induction via starvation. Figure 21 EGFP-LC3 puncta formation in the presence of everolimus and chloroquine. Figure 22 LC3-I and LC3-II after chemical exposure for 48 hours. Figure 23 The pDS-FABP10-mRFP-EGFP-LC3 plasmid used in the generation of Tg(fabp10:mrfp-egfp-lc3) transgenic line. 100 Figure 24 Liver-specific mRFP-EGFP-LC3 expression after microinjecting the pDS-FABP10-mRFP-EGFP-LC3 plasmid into one-cell stage embryos. 101 Figure 25 Transgenic larvae harbour germline transmission of the fabp10:mrfp-egfp-lc3 expression cassette. 102-103 Figure 26 Starvation induces autophagy. Figure 27 Red and yellow puncta formation in the presence of everolimus and chloroquine. 106-107 Figure 28 Over-expression of cmyc and Xmrk led to liver tumour formation in stable transgenic zebrafish induced using 60 µg/ml doxycycline. 109 Figure 29 mTOR activation may cause autophagy inhibition in fish with Xmrk over-expression. 111 Figure 30 Schematic showing the genotypes of progeny resulting from crossing Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) with Tg(fabp10:egfp-lc3) and doxycycline exposure and control groups. 113 Figure 31 Schedule for doxycycline exposure and sampling time points. 114 Figure 32 Tumour progression manifested as increasingly severe gross and cellular morphology in transgenic fish over-expressing Xmrk. 118-119 95 96-97 98 105 ix List of figures Figure 33 Immunoblot analyses of Xmrk over-expressing fish showing increased free EGFP. 120 Figure 34 Tumour progression in fish over-expressing cmyc, manifesting as liver overgrowth and disruption to cell morphology. 121-122 Figure 35 Immunoblot analyses if cmyc over-expressing fish showing increased free EGFP and reduced p62 levels during tumour regression. 123 x List of common abbreviations List of common abbreviations 4EBP1 Ac/Ds AMP AMPK APS Atg ATP BSA cDNA cmyc CTP CQ DAPk DEPC DMSO DNA dNTP DTT Dox dpf EDTA EGFP EGFR eIF2α ENU ER ERK EVE FABP Gly GTP HA HCC HRP kb kD krt4 LB LC3 MAPK MCS eukaryotic initiation factor 4E binding protein 1 Activator/Dissociation adenosine monophosphate 5’-AMP-activated protein kinase ammonium persulphate autophagy related adenosine triphosphate bovine serum albumin DNA complementary to RNA cellular homologue of v-myc oncogene cytidine triphosphate chloroquine death-associated protein kinase diethyl pyrocarbonate dimethyl sulfoxide deoxyribonucleic acid deoxyribonucleotide dithiothreitol doxycycline days post fertilisation ethylenediaminetetraacetic acid enhanced green fluorescent protein epidermal growth factor receptor eukaryotic initiation factor 2 α ethylnitrosourea endoplasmic reticulum extracellular signal-regulated kinase everolimus fatty acid binding protein glycine guanosine triphosphate hepatocelluar adenoma hepatocellular carcinoma horse radish peroxidase kilobase kilo Dalton keratin 4 promoter Luria Bertani light chain 3 mitogen activated kinase multiple cloning site xi List of common abbreviations mpf mRFP mRNA mTORC1 NCBI NTC NTP p-S6 PBS PCR PE PFA PI3K PTEN PVDF RAP RNA rpm RT-PCR S6K SDS-PAGE T-PER TBS TEMED tetR TGF-α TILLING TOR TPM TRE TSC TTP Ulk1 UTP VEGF VHL Xmrk ZFIN months post fertilisation monomeric red fluorescent protein messenger RNA mammalian TOR complex 1 National Centre for Biotechnology Information no template control nucleoside triphosphate phosphorylated S6 phosphate buffered saline polymerase chain reaction phosphotidylethanolamine paraformaldehyde phosphatidylinositol 3-kinase phosphatase and tensin homologue polyvinylidene fluoride rapamycin ribonucleic acid revolutions per minute reverse-transcriptase PCR p70 ribosomal S6 kinase sodium dodecyl sulphate polyacrylamide gel electrophoresis tissue protein extraction reagent tris buffered saline tetramethylethylenediamine tetracycline repressor transforming growth factor α targeting-induced local lesions in genomes target of rapamycin transcript per million tetracycline responsive element tuberous sclerosis complex thymidine triphosphate Uncoordinated 51-like kinase uridine 5’-triphosphate vascular endothelial growth factor von Hippel-Lindau Xiphophorus melanoma receptor kinase gene zebrafish information network xii Introduction Introduction 1 Introduction 1. The use of zebrafish as a model organism for the study of cancer 1.1 The use of zebrafish as a model organism in vertebrate developmental biology studies The zebrafish (Danio rerio) is an excellent model organism for the study of vertebrate developmental biology and genetics due to its superior physical attributes when compared with rodent models. External fertilisation and development as well as optically transparent embryos allow early developmental processes to be studied easily under a dissecting microscope. In addition, traits such as high fecundity with the ability of a mating pair producing hundreds of embryos per week and short generation times of approximately three months further make possible the use of zebrafish for genetics analyses. Moreover, its ease of breeding and small size facilitate animal husbandry, and its low maintenance costs make zebrafish an ideal organism to complement the rodents in developmental biology and genetics studies (Dooley and Zon, 2000; Wixon, 2000). In addition to its physical attributes, the popularity of zebrafish as an experimental organism has also led to the development of various experimental methods and techniques. Whole-mount in situ hybridisation and the use of anti-sense morpholino knockdown in transparent embryos facilitate the study of developmental processes in vivo. Forward genetic screens employing ethylnitrosourea (ENU)induced point mutations allow the zebrafish to be used for screening developmental mutants and identifying genes involved in early development for further analyses (Driever et al., 1996; Haffter et al., 1996). Although other model organisms such as Drosophila and Caenorhabditis elegans are also amendable to large scale screens, the absence of vertebrate-specific organs such as kidney and neural crest cells make them inferior to the zebrafish in the study of specific developmental issues involving these 2 Introduction organs. The zebrafish is also amendable to reverse genetics via an approach known as targeting-induced local lesions in genomes, or TILLING, which is adapted from plant genomics studies (McCallum et al., 2000). DNA from progeny derived from ENUmutagenised parents is sequenced to look for mutations that are present in a gene-ofinterest. Progeny identified to carry the desired mutations are then propagated to maintain the line to study the function of the gene-of-interest. Furthermore, fluorescent proteins can be expressed in specific organs in order to study their development over time by placing the gene encoding fluorescent proteins under the control of organ-specific promoters (Higashijima et al., 1997; Long et al., 1997). This is especially useful in early development when the larvae are largely transparent which facilitates the observation of the organ under study. This method will also allow specific genes-of-interest to be expressed under the control of a specific promoter to study their function. Moreover, efforts have been made to map zebrafish genes onto chromosomes by both genetic and physical mapping (Geisler et al., 1999; Hukriede et al., 1999; Postlethwait et al., 1998). With advances in sequencing technologies, the zebrafish genome sequence is also made available as well for positional cloning. In spite of the large evolutionary distance between zebrafish and human, the syntenic relationship between their genomes further allow the use of the zebrafish in the identification of gene functions of unknown genes in both zebrafish and humans (Barbazuk et al., 2000). In all, these attributes of the zebrafish has positioned it to overcome the shortcomings of mouse models as well as made it relevant in developmental biology and genetics studies. 3 Introduction 1.2 The use of zebrafish as a model organism in cancer studies Fish has been used to study cancer since the early 1900s (Stern and Zon, 2003). Early studies involving Xiphophorus, or swordtail fish, ascertained the existence of an oncogene which is the causative agent of malignant melanoma in the species (Wittbrodt et al., 1989). The oncogene, now known as the Xiphophorus melanoma receptor kinase gene or Xmrk, is homologous to the human Epidermal Growth Factor (EGF) receptor, and is found to be masked by its tumour-suppressor at a separate locus and studies have shown that it plays an important role in elucidating the functional roles of receptor tyrosine kinases in tumourigenesis (Wittbrodt et al., 1989; Wittbrodt et al., 1992). Over the years, other fishes such as the medaka (Oryzias latipes) and zebrafish are also used in carcinogenicity testing (Law, 2001). The zebrafish has been most appropriately used as a model organism for the study of cancer (Amatruda et al., 2002; Feitsma and Cuppen, 2008; Stoletov and Klemke, 2008). First used as a chemical carcinogenesis model, the zebrafish developed mainly hepatic neoplasms after exposure to the carcinogen diethylnitrosamine (Stanton, 1965). More recent studies further showed that the zebrafish is capable of developing other types of neoplasms depending on the types of chemical carcinogens used and developmental stage at which they were exposed to (Spitsbergen et al., 2000a, b). Furthermore, it has been shown that the tumours that developed from such exposure were very similar to human tumours at a histological level, lending support to the use of zebrafish as a model to study chemically-induced cancer (Amatruda et al., 2002; Spitsbergen et al., 2000a). Apart from histological similarities between zebrafish and human tumours, comparisons between human and zebrafish liver cancer gene signatures revealed significant conservation between the two, suggesting the conservation of mechanisms that are involved in liver 4 Introduction carcinogenesis between the two (Lam and Gong, 2006; Lam et al., 2006). The ease of housing and processing zebrafish in large numbers for such studies enables these studies to generate data with greater statistical powers than mammalian cancer studies involving mice and rats (Stern and Zon, 2003). Despite the low cost and technical simplicity of chemical carcinogenesis, the low incidence of tumour occurrence coupled with late tumour onset and heterogeneity and variability of tumour genetic background and location make chemical carcinogenesis a less popular means of studying cancer in zebrafish. As the zebrafish is amendable to both forward and reverse genetics in the study of vertebrate developmental biology and genetics, these techniques can also be adapted to study cancer, and discover and analyse existing and novel oncogenes and tumour suppressors in zebrafish. Forward genetic screens were carried out to look for mutants displaying phenotypes related to cancer, such as proliferation defects which identified mutants with loss-of-function in bmyb and separase (Shepard et al., 2007; Shepard et al., 2005). Tp53 mutants were also isolated in reverse genetic screens utilising ENU-induced mutagenesis and were found to be involved in the development of malignant peripheral nerve sheath tumours (Berghmans et al., 2005). Xenotransplantation of mammalian cancer cells from various sources into zebrafish of different developmental stages were also performed successfully with excellent results. Such studies were useful in delineating the dynamics of microtumour formation, angiogenesis and vasculature pruning, as well as cellular invasion and metastasis in vivo (Feitsma and Cuppen, 2008; Stoletov and Klemke, 2008). Stoletov et al. studied how expression of RhoC and VEGF by human cancer cells aid in their metastasis, with RhoC expression inducing an amoeboid-type of invasion characterised by the formation of a rounded cell morphology and extensive 5 Introduction and dynamic membrane blebbing and protrusions, and VEGF expression inducing vessel permeabilisation and vessel remodelling to aid the intravasation of cancer cells into the remodelled blood vessels (Stoletov et al., 2007). The technique of generating transgenic zebrafish with the ability to control the expression of various genes in the study of gene function can also be extended to the study of oncogenes and tumour suppressors. Through the use of tissue-specific promoters, one can control the expression of oncogenes or mutant tumour suppressors in a tissue-specific manner. Transgenic zebrafish expressing gene fusions of these genes-of-interest with fluorescent proteins can be generated in order to select stable transgenic progeny, and more importantly, to facilitate in vivo monitoring and imaging of cells expressing these gene fusions (Stoletov and Klemke, 2008). Langenau et al. made use of this strategy to over-express human cmyc under the control of the zebrafish rag2 promoter, and obtained transgenic zebrafish overexpressing cmyc in T-cells with the concomitant development of T-cell acute lymphoblastic leukaemia (Langenau et al., 2003). Other transgenic zebrafish overexpressing activated BRAF, zebrafish bcl2 and activated human kRASG12D among others were also developed and they provided ample opportunities for studying the mechanisms of oncogene-driven carcinogenesis and the cross-talk among different signalling pathways (Amatruda and Patton, 2008; Langenau et al., 2005b; Langenau et al., 2007; Patton et al., 2005). Moreover, these models are amendable to pharmacological testing and novel drug discovery as the zebrafish embryos and larvae are permeable to many watersoluble compounds and they can be arrayed in a manner that facilitates highthroughput screening. Moreover, screens for compounds and chemicals that can exert their effects in zebrafish larvae are more powerful compared to screens using cell 6 Introduction lines as they are able to suppress the disease phenotype in a whole organism in vivo and they do not harbour any target bias (Stern and Zon, 2003). Compounds that are active in suppressing cancer phenotypes can be identified for further testing, as evidenced by screens for anti-angiogenic compounds (Wang et al., 2010; Zhang et al., 2009). 1.3 Generation of inducible Xmrk and cmyc driven liver cancer models 1.3.1 The use of inducible promoter systems to control transgene expression Langenau et al. generated the first transgenic zebrafish expressing human cmyc under the control of a zebrafish tissue-specific mitfa promoter which made for an excellent model to study the mechanism of cmyc-driven T-cell acute lymphoblastic leukaemia (Langenau et al., 2003). However, the authors found that the resultant transgenic zebrafish showed rapid onset of cancer that displayed efficient invasion and metastasis to other tissues and organs, leading to early lethality and requiring in vitro fertilisation for line maintenance. To overcome this problem, the same authors made use of the Cre/lox and heat shock promoter systems to control the expression of the cmyc oncogene (Feng et al., 2007; Langenau et al., 2005a). Initially the Cre/lox system displayed low disease penetrance when the conditional transgenic progeny carrying germline insertion of the construct rag2-loxP-dsRED2-loxP-EGFP-mMyc were injected with the Cre mRNA; with the complementation of the heat shock promoter controlling Cre expression, a higher rate of recombination events took place after heat shock which resulted in more progeny expressing the rag2-EGFP-mMyc cassette and thus, developing T-cell acute lymphoblastic leukaemia. Thus the spatiotemporal expression of transgenes can be controlled in conditional transgenic lines and they also allow the maintenance of the line by either in-crossing or out-crossing. 7 Introduction To follow tumour initiation, progression and even regression in the zebrafish, one must be able to control the switching on or off of oncogene expression (Feitsma and Cuppen, 2008). The tetracycline-responsive systems, which include the Tet-on and Tet-off systems, can exert such control over transgene expression. The tetracycline-responsive system was first established and demonstrated in Hela cells to control transgene expression (Gossen and Bujard, 1992). In the early versions of the system, the tetracycline repressor (tetR) from the tetracycline-resistant operon in Escherichia coli was fused to the activating domain of the virion protein 16 from the herpes simplex virus to form a tetracycline-controlled transactivator. tetR then activates transcription from a minimal promoter generated by the combination of human cytomegalovirus promoter and tet operator sequences. The isolation of a mutant tetR, known as rtTA, that requires the presence of doxycycline (Dox) in order to bind to and activate transcription of the minimal promoter and further refinements to the promoter system increase its efficiency and reduce its background activities and further gave birth to the Tet-on and Tet-off systems (Urlinger et al., 2000). This system will allow us to control not only the expression of the gene, but also the degree of expression by controlling Dox concentration. 1.3.2 The use of transposable elements to generate transgenic zebrafish The standard method of generating transgenic zebrafish is via the injection of DNA constructs containing gene expression cassettes consisting of a promoter, a transgene and polyA signals into one-cell stage embryos in the hope that the construct will be taken up and integrated into the genome of the injected embryos for germline transmission. This method has been the sole method in early transgenic fish studies (Higashijima et al., 1997; Long et al., 1997). However, the rate of transgenesis 8 Introduction employing this method is often less than satisfactory, with germline transmission rates of less than 10%. Hence there is a need to increase transgenesis rates to facilitate transgenic zebrafish generation. The reported use of the Tol2 transposition system from medaka and Sleeping Beauty transposition system from salmonid provided a new impetus in generating transgenic zebrafish with significantly higher transgenesis and gene expression rates (Davidson et al., 2003; Kawakami et al., 2000). Transposable elements can be found in genomes of most vertebrates and are composed of transposon DNA and a recombinase-coding DNA sequence that is involved in transposon trans-activation (Ivics et al., 2004). Recently, the Activator (Ac)/Dissociation (Ds) transposition system first isolated in maize by McClintock was found to work in zebrafish with comparable transgenesis and gene expression rates (Emelyanov et al., 2006; McClintock, 1950). This study also showed that the Ac/Ds transposable element is versatile enough to be active in zebrafish and human cells, eliminating the requirement for host cell-specific factors for transposition to occur. Furthermore, our laboratory recently reported the successful use of the Ac/Ds transposable element in the generation of transgenic zebrafish expressing zebrafish kRASV12 under the control of the liver-specific fabp10 promoter in modelling oncogene-driven liver cancer (Her et al., 2003; Nguyen et al., 2011). 1.3.3 Xmrk and cmyc driven liver cancer models Our laboratory has also successfully made use of the Tet-on system to generate transgenic zebrafish with inducible, liver-specific expression of oncogenes to model liver cancer. In our zebrafish, a liver-specific fabp10 promoter was used to drive the constitutive expression of the rtTA transactivator, which will activate transcription of the oncogene found downstream of the minimal promoter found in a 9 Introduction separate construct in the presence of Dox. We used two different oncogenes to model liver cancer, namely the Xiphophorus melanoma receptor kinase gene or Xmrk, an Epidermal Growth Factor receptor (EGFR) homologue, and mouse cmyc (Huang et al., 2011). Xmrk was first isolated in Xiphophorus, which was the main causative agent for melanoma development in the fish species. Xmrk was found to harbour activating mutations in the extracellular domain, explaining its tumourigenicity in vitro and in vivo (Winnemoeller et al., 2005). The suitability of using Xmrk to drive liver carcinogenesis was further supported by studies implicating dysregulated EGFR signalling in human hepatocellular carcinoma or HCC (Berasain et al., 2009). It is noteworthy to point out that the EGFR pathway is chronically stimulated during inflammation in the liver following injury and that tumourigenesis is favoured in the backdrop of inflammation. Further studies implicated the over-expression of EGFR and its ligands in human HCC (Avila et al., 2006; Berasain et al., 2007; Breuhahn et al., 2006; Castillo et al., 2006). Thus it would be interesting to uncover the role chronic Xmrk over-expression plays in liver carcinogenesis. We also generated Dox-inducible liver-specific over-expression of mouse cmyc. Initially isolated from chicken DNA, cmyc is the cellular homologue of the vmyc oncogene found in avian myelocytomatosis retrovirus MC29 (Vennstrom et al., 1982). The oncogenic potential of cmyc was first demonstrated by its involvement in the progression of human Burkitt’s lymphoma, due to a translocation event between chromosome 8 and one of the three chromosomes containing genes encoding immunoglobulins (Spencer and Groudine, 1991). cmyc encodes a transcription factor that heterodimerises with its partner protein MAX via interaction with its carboxyterminal basic-helix-loop-helix-zipper domain. The heterodimer binds to E-boxcontaining DNA and facilitates transcription and gene expression (Blackwood and 10 Introduction Eisenman, 1991; Eilers and Eisenman, 2008; Pelengaris et al., 2002). cmyc expression is positively correlated with cell growth and proliferation by promoting cell-cycle progression; it also inhibits cellular terminal differentiation of most cell types as well as sensitises cells to apoptosis (Amati, 2001; Amati et al., 1998; Dang, 1999; Eilers, 1999). In addition, cmyc expression is also involved in cellular reprogramming to form induced pluripotent stem cells (Eilers and Eisenman, 2008). In light of these, it is of interest to uncover links that connect dysregulated cmyc expression and carcinogenesis. Early studies using mouse models provided evidence linking hepatocyte-specific cmyc over-expression with chronic hepatic proliferation and increased incidence of cancer, and it potentiates transforming growth factor (TGF)-α in the development of hepatocellular carcinoma (Calvisi and Thorgeirsson, 2005). However, the role of dysregulated cmyc expression in carcinogenesis is still elusive, although several hypotheses have been proposed (Dang et al., 2005; Eilers and Eisenman, 2008; Pelengaris et al., 2002). These two models that we have generated will provide us with ample material to study the roles of oncogene-driven liver carcinogenesis, and to study the intricate cross-talk between various signalling pathways. They also represent good models for chemical and drug screens that target both EGFR and cmyc signalling pathways in a whole organism. 2. An introduction to autophagy and its involvement in cancer 2.1 Autophagy as a basal and inducible cellular process Autophagy is a ubiquitous cellular process that involves the delivery of cellular cytoplasmic components for degradation in the lysosome. Early studies led to the identification of at least three forms of autophagy in mammalian cells: chaperone- 11 Introduction mediated autophagy, microautophagy and macroautophagy. They differ with respect to the method of sequestering and delivering cytoplasmic components to the lysosome for degradation and their physiological functions (Levine and Kroemer, 2008). In microautophagy, cytoplasmic components are taken up by protrusion, septation or invagination of the vacuolar membrane in the yeast or the lysosomal membrane in eukaryotes; whereas chaperone-mediated autophagy requires chaperones to facilitate the translocation of unfolded, soluble proteins across the lysosomal membrane (Wang and Klionsky, 2003; Yang and Klionsky, 2009). In macroautophagy, doublemembrane vesicles called autophagosomes sequester cytoplasmic components, which then fuse with lysosomes with the concomitant delivery of the inner single-membrane vesicles for lysosomal degradation and recycling of resultant nutrients (Yang and Klionsky, 2009). We will only focus on macroautophagy (referred to as autophagy hereafter) in this dissertation. Autophagy is an evolutionarily conserved cellular process involved in the degradation of a cell’s cytoplasm, long-lived proteins and organelles in response to cellular stressors such as starvation. Autophagy is also involved in the degradation of damaged or superfluous organelles and proteins and protein aggregates, in addition to its role in regulating cellular homeostasis at a basal level (Klionsky, 2007; Mizushima et al., 2008). The ability of autophagy to carry out large-scale degradation of cellular components requires it to be tightly controlled, as unregulated autophagy can have catastrophic consequences. Due to its far-reaching effects in regulating cellular turnover and homeostasis, autophagy deregulation is found to be implicated in various diseases such as neurodegeneration, myopathies, infection and immunity, as well as cancer (Levine and Kroemer, 2008; Mizushima et al., 2008). 12 Introduction Autophagy was first identified in mammalian cells by Christine de Duve in the 1950s as part of a lysosomal degradative mechanism (Klionsky, 2007). Subsequent genetic screens for yeast mutants affecting protein turnover led to the identification of autophagy-related (Atg) genes and their protein products that are crucial for the autophagy process (Yang and Klionsky, 2009). To date, more than 30 Atg genes have been identified in yeast, and more than a dozen orthologues in higher organisms have also been identified (Kourtis and Tavernarakis, 2009; Rubinsztein et al., 2007). 2.2 Molecular machinery of autophagy Autophagy is a multi-step process. Autophagy induction leads to the formation of a phagophore that undergoes nucleation and elongation to form a doublemembrane vesicle called the autophagosome which sequesters cytoplasmic components in the process. The autophagosome subsequently fuses with endosomes forming amphisomes, or with lysosomes forming autolysosomes where the sequestered materials are then released into the lysosome to be degraded by lysosomal enzymes and the resultant amino acids and other nutrients are subsequently reused for use by the cell (Klionsky, 2007; Levine and Kroemer, 2008). This section serves to briefly summarise the current knowledge of the molecular mechanisms of autophagy. Induction of autophagy requires the activity of Atg1, a serine/theronine kinase, which is found in a complex with Atg13 and Atg17 (Kabeya et al., 2005; Kamada et al., 2000; Matsuura et al., 1997). Studies have demonstrated that the formation of this protein kinase complex is positively correlated with an increase in autophagic activity. Uncoordinated 51-like kinases (Ulk) 1 and 2 are mammalian homologues of Atg1, while FIP200 has been recently identified to form a complex with Ulk1 and was proposed to be an Atg17 homologue (Chan et al., 2007; Hara and Mizushima, 2009; 13 Introduction Hara et al., 2008; Yan et al., 1998). Furthermore, studies by Jung et al. provided evidence of the existence of a functional mammalian homologue of Atg13, proving that the induction machinery is well conserved in higher vertebrates (Jung et al., 2009). Vesicle nucleation requires the activity of a protein complex comprising a class III phosphatidylinositol 3-kinase (PI3K) Vps34, Vps15, Atg6 and Atg14 (Levine and Kroemer, 2008). In yeast, Vps34 activity is regulated by its partner Vps15, and Atg14 serves to specify the complex for autophagy (Obara et al., 2006; Stack et al., 1995). Vps34 activity is crucial for autophagy and it is thought that the production of PI(3)P by Vps34 serves to recruit downstream effectors to the complex for nucleation events (Yang and Klionsky, 2009). Similar to yeast, mammalian class III PI3K, hVps34 forms a complex with p150, a homologue of Vps15, as well as Beclin1, the mammalian homologue of Atg6, and the complex is required for autophagy, proving that autophagy is a highly conserved process from yeast to mammals (Liang et al., 1999; Panaretou et al., 1997; Volinia et al., 1995). Expansion of the double-membrane structure requires the activity of two ubiquitin-like protein conjugation systems, namely, the Atg12 and Atg8 systems (Yang and Klionsky, 2009). The Atg12 system is found to be essential for preautophagosomal membrane formation, while Atg8 conjugation is required for formation of complete autophagosomes by mediating membrane tethering and hemifusion (Kabeya et al., 2000; Mizushima et al., 2001; Nakatogawa et al., 2007). Atg12 is first activated by the E1-like enzyme, Atg7, which is then transferred to the E2-like enzyme, Atg10, before being conjugated to Atg5 (Mizushima et al., 1998a). Atg12-Atg5 complex then interacts with Atg16 to form a Atg12-Atg5-Atg16 multimeric complex facilitated by homo-oligomerisation of Atg16 that is essential for 14 Introduction autophagy (Kuma et al., 2002). Similarly, Atg8 is first cleaved by Atg4, exposing a terminal glycine, it is then activated by Atg7, before being transferred to another E2like enzyme Atg3 for conjugation to a membrane lipid, phosphotidylethanolamine (PE) (Ichimura et al., 2004; Ichimura et al., 2000). Similar to other Atg proteins, Atg12 and Atg8 systems remain highly conserved and are functional in mammalian autophagy and their mammalian counterparts have been identified and characterised such as the mammalian counterpart of Atg16, Atg16L (Mizushima et al., 2003; Mizushima et al., 1998b; Mizushima et al., 2002). Similarly, Atg8 has several mammalian homologues, namely MAP1LC3 or LC3, GATE16, GABARAP, and Atg8L, and they are processed similarly to the yeast Atg8 with LC3 being the most abundant in autophagosomal membranes and therefore used as a marker to monitor autophagic activity (Hemelaar et al., 2003; Kabeya et al., 2000; Kabeya et al., 2004; Tanida et al., 2006). During autophagosomal membrane expansion, both Atg12-Atg5Atg16L and LC3-PE conjugates can be found to decorate the autophagosomal membrane, with the Atg12-Atg5-Atg16L mostly localised on the outer membrane and released from the membrane shortly before or after autophagosome formation while LC3-PE was found on both inner and outer membranes (Kabeya et al., 2000; Kirisako et al., 1999; Mizushima et al., 2003; Mizushima et al., 2001). LC3-PE found on the outer leaflet of the autophagosomal membrane is released from the autophagosomal membrane by Atg4 cleavage, while those found on the inner leaflet remains in the autophagosome and is delivered together with the sequestered materials to the lysosome for degradation. The reversible process of Atg8-PE conjugation, where Atg4 can also liberate Atg8 from PE, allows Atg8 to be recycled and used in another conjugation reaction, allowing efficient autophagy to occur (Kirisako et al., 2000). It is also shown that the Atg12-Atg5-Atg16L complex is required for efficient LC3 15 Introduction lipidation and for specifying the site of LC3 lipidation in mammalian autophagy (Fujita et al., 2008). Other Atg proteins not mentioned above play other roles that are equally important to autophagy. An integral membrane protein Atg9 is thought to play the role of trafficking membrane to the forming autophagosome (He et al., 2006; Noda et al., 2000). Atg9 movement to the forming autophagosome requires Atg23 and Atg27, which function in a heterotrimeric complex, while retrieval of Atg9 from the forming autophagosome requires the combined action of Atg2 and Atg18 in addition to the Atg1-Atg13-Atg17 complex (Legakis et al., 2007; Reggiori et al., 2004; Yen et al., 2007). Atg11 and Atg17 may act as scaffold proteins in the recruitment of Atg proteins (Yang and Klionsky, 2009). Atg15 is thought to exert lipase activity due to the presence of a lipase active site motif (Teter et al., 2001). Yeast Atg22 was found to aid in the efflux of amino acids resulting from autophagic activity (Yang et al., 2006). 16 Introduction Figure 1 Schematic illustration of the autophagy pathway and its core machinery in mammalian cells. Autophagy is a multi-step process with the involvement of several Atg proteins at each step. Figure reproduced from Levine and Kroemer (Levine and Kroemer, 2008). 17 Introduction 2.3 Regulation of autophagy The Target of Rapamycin (TOR) kinase is the key regulator of autophagy in eukaryotes. Apart from regulating autophagy, it also regulates various cellular processes involved in cell growth such as transcription, translation, cell size and cytoskeletal organisation (Schmelzle and Hall, 2000). Mammalian TOR (mTOR) signalling is crucial in the integration of various cellular signals and cell stressors (Corradetti and Guan, 2006). mTOR can be incorporated into two TOR complexes, mTORC1 and mTORC2, where they phosphorylate different cellular targets and possess different functions and different sensitivities to rapamycin (Jacinto et al., 2004; Loewith et al., 2002). mTORC1 is rapamycin-sensitive and acts as a potent inhibitor of autophagy, where it inhibits autophagy in the presence of growth factors and nutrients (Levine and Kroemer, 2008). Growth factors stimulate receptor tyrosine kinases, where they activate mTORC1 via PI3K-Akt signalling which in turn, inhibit autophagy (Esclatine et al., 2009; Levine and Kroemer, 2008; Lum et al., 2005). mTOR senses changes in cellular energy via the 5’-AMP-activated protein kinase (AMPK). Low energy status in the cell is represented by a high AMP to ATP ratio, and activates AMPK where it inhibits mTOR signalling to activate autophagy (Corradetti et al., 2004; Meijer and Codogno, 2011). mTOR inhibits autophagy by phosphorylating Atg13 in the Atg1-Atg13-Atg17 complex; hyper-phosphorylated Atg13 has reduced affinity for Atg1 and Atg17, leading to reduced Atg1-Atg13-Atg17 complex formation and concomitantly, reduced autophagy induction (Klionsky, 2005). Other regulatory molecules that exert influence on autophagy also include the eukaryotic initiation factor 2α (eIF2α) which responds to stressors like nutrient deprivation, and endoplasmic reticulum (ER) stress, mitogen-activated (MAP) kinases, extracellular signal-regulated kinases (ERK1/2), death-associated protein kinase 18 Introduction (DAPk), the heterotrimeric G protein Gi3, tumour suppressor p53 and others (Esclatine et al., 2009). Pharmacological agents that target mTOR as well as the phosphotidylinositol 3-kinases (PI3K) involved in autophagy regulation can be used to either induce or inhibit the autophagic process. As mentioned above, rapamycin and its analogues such as everolimus can be used to induce autophagy by relieving mTOR-induced inhibition (Rubinsztein et al., 2007). 3-Methyladenine, a potent PI3K inhibitor, can also be used to inhibit autophagy by inhibiting the activity of class III PI3K (Blommaart et al., 1997; Petiot et al., 2000; Seglen and Gordon, 1982). Lysosomal proton pump inhibitors such as bafilomycin A1 and lysosomotropic alkalines such as chloroquine can also be used to inhibit autophagy effectively (Levine and Kroemer, 2008). Apoptosis is a programmed cell death process discovered during insect metamorphosis and during embryogenesis and development of both vertebrates and invertebrates (Glucksmann, 1965; Lockshin and Williams, 1965a, b, c; Saunders, 1966). It is characterised by caspase activation, poly (ADP-ribose) polymerase I (PARP1) proteolysis and DNA fragmentation cuminating in cell shrinkage, chromatin condensation and apoptotic body formation (Giansanti et al., 2011). It operates via two pathways, extrinsic and intrinsic, with the extrinsic pathway activated via receptor binding of molecules belonging to the Tumour Necrosis Factor (TNF) family, and the intrinsic pathway activated by stimuli converging and acting on the mitochondria. Both pathways culminate in the activation of caspases, which serves to degrade cellular components at the final step of apoptosis, following which the degraded components are engulfed by phagocytes to prevent an inflammatory response at the end of the process. 19 Introduction The intimate relationship between apoptosis and autophagy was discovered when Boya et al. reported that mammalian cells were observed to undergo apoptosis when they were devoid of nutrients and when autophagy was inhibited through the use of chemical inhibitors or genetic silencing by siRNAs (Boya et al., 2005). Boya et al. further observed that cells with autophagic vesicles were able to recover when cultured under optimal conditions following nutrient deprivation, although cells with disrupted mitochondrial transmembrane potential were still committed to die when cultured under the same conditions, strongly suggesting a pro-survival role of autophagy under cellular stress. However, Notte et al. noted that in the event that the stress is too severe or the duration of the stress is too long, autophagy may also participate in cell death on observations that numerous autophagic vesicles were found in dying cells (Notte et al., 2011). More studies need to be performed to determine if autophagy does indeed participate actively in cell death either on its own, or together with other cell death mechanisms, or if these observations merely a consequence of a failed attempt to preserve cell viability in times of stress (Levine and Yuan, 2005). Further studies revealed that Bcl-2 family members were found to regulate both pathways: anti-apoptotic Bcl-2 and Bcl-XL were found to bind to and inhibit Beclin1 through the Beclin1 BH3 domain possibly to inhibit autophagy (Maiuri et al., 2007; Zhou et al., 2011). The involvement of Atg5 in both autophagy and apoptosis further supports the intricate cross-talk between the two processes: tumour cells with atg5 over-expression were found to be more susceptible to apoptotic stimuli, while silencing of atg5 results in partial resistance to chemotherapy (Yousefi et al., 2006). This intricate relationship between autophagy and apoptosis is further convoluted by findings where autophagy abrogation re-sensitised tumour cells to apoptogenic stimuli 20 Introduction as well as enforced autophagy leading to cell demise in apoptosis-resistant cells either through cooperation with other cell death mechanisms or massive autophagy-induced cell death (Chen et al., 2010; Scarlatti et al., 2009; Shen and Codogno, 2011; Xie et al., 2011; Zhivotovsky and Orrenius, 2010). Thus it would be interesting to delve more deeply into the relationship of autophagy and apoptosis and to make use of this newly-acquired knowledge in designing new cancer therapeutics. 21 Introduction Figure 2 Regulation of mammalian autophagy. The intricate relationship between various signalling pathways and autophagy regulation is depicted in the figure above. Figure reproduced from Yang and Klionsky (Yang and Klionsky, 2010). 22 Introduction 2.4 The use of EGFP-LC3 and mRFP-EGFP-LC3 to study autophagy in vivo The past decade has seen an exponential increase in the research on autophagy, especially its implication in diseases hence there is a need to establish techniques to study autophagy both in vitro and in vivo. The gold standard of analysing autophagic activity requires the use of electron microscopy to identify autophagosomes and autolysosomes (Mizushima et al., 2010). However such a method is both time consuming and requires expertise in the correct identification of autophagic vesicles. Thus there is a need to establish a more reliable and accessible method of analysing autophagy status. The finding that LC3, the mammalian homologue of Atg8, undergoes PE lipidation for its localisation to autophagosomes provides an opportunity to detect autophagic activity. LC3 can be found in mammalian cells in two forms, LC3-I and the PE-conjugated LC3-II. LC3-I is found in the cytosol while LC3-II is usually found to be associated with autophagosomal membranes and its formation is positively correlated with autophagosome formation and thus autophagic activity (Kabeya et al., 2000). Thus immunoblotting of LC3-I and LC3-II provide an easy read-out of autophagic activity in both in vitro and in vivo experimental conditions (Klionsky et al., 2008; Mizushima et al., 2010). The introduction of protein fusions to fluorescent proteins further allow autophagic analyses to be easily performed using fluorescent microscopes in real time. The use of EGFP-LC3 fusion protein was first established by Kabeya et al. where they reported the association of LC3-II with autophagosomal membranes forming discrete EGFP puncta. They further established that LC3-I was freely available in the cytosol forming a diffuse cytoplasmic distribution (Kabeya et al., 2000). This study establishes the use of EGFP-LC3 to analyse autophagic activity, where an upregulation of autophagy will see a concomitant increase in GFP puncta representing 23 Introduction autophagosomes. The intracellular localisation of EGFP-LC3 thus becomes a complementary method for the analyses of autophagic flux in addition to immunoblotting for LC3 (Kabeya et al., 2000). Since then, EGFP-LC3 is used to study autophagy dynamics in a variety of in vitro conditions using various cell lines as well as in the generation of transgenic mice to study autophagic dynamics in a variety of conditions such as starvation (Mizushima et al., 2004). As autophagosome formation is an intermediate step in a highly dynamic cellular process, EGFP-LC3 puncta observed at any specific time point usually reflects the delicate balance between the rate of autophagosome formation and the rate at which they are converted to autolysosomes (Mizushima et al., 2010). As such, it would be more appropriate to measure autophagic flux to determine cellular autophagic activity. Autophagic flux refers to the dynamic process of autophagosome formation, delivery of engulfed cellular substrates to the lysosome and subsequent degradation of the substrates within the lysosome. One method of measuring autophagic flux is to quantify EGFP-LC3 puncta numbers present in cells, as increased autophagic activity often leads to increased autophagosome formation and therefore, EGFP-LC3 puncta formation. However, a major pitfall exists as increased puncta numbers may also be due to reduced autophagic degradation that leads to an accumulation of autophagosomes. Therefore the number of EGFP-LC3 puncta that can be quantified does not correlate to cellular autophagic activity per se and several other assays are required in addition to observing EGFP-LC3 puncta formation to determine autophagic flux. Another useful assay to measure autophagic flux was developed based on the concept of lysosomal quenching of GFP fluorescence. As mentioned, autophagosomes fuse with lysosomes as part of its maturation process to degrade 24 Introduction engulfed cellular substrates. However, the low pH achieved through autophagosomelysosome fusions quench GFP fluorescence, making it difficult to trace EGFP-LC3 delivery to lysosomes; this also hampers efforts to demonstrate the co-localisation of EGFP-LC3 puncta with lysosomes (Bampton et al., 2005; Kabeya et al., 2000). On the other hand, RFP exhibits stable fluorescence in acidic compartments making it useful in labelling acidic autophagic vesicles. A novel reporter consisting of monomeric RFP (mRFP) and EGFP fused to LC3 was developed to study autophagosome-lysosome fusion as well as to provide a means of measuring autophagic flux (Kimura et al., 2007). Using this novel construct, autophagosomes are labelled yellow (due to presence of both GFP and RFP) and autolysosomes are labelled red and autophagic flux can be measured simply by observing the ratio of both yellow and red signals. Increases in autophagic flux are indicated by increases in both yellow and red puncta while blockages in fusion and maturation of autophagosomes into autolysosomes are indicated by increases in yellow puncta without concomitant increases in red puncta, making it easier to study autophagic flux in real time. We can accurately study autophagic activity with the use of the mRFPEGFP-LC3 reporter together with other assays. 2.5 Autophagy in cancer Dysregulated autophagy is implicated in various diseases such as neurodegeneration, myopathies, infection and immunity, as well as cancer (Levine and Kroemer, 2008; Mizushima et al., 2008). Among them, the role of autophagy in cancer still remains elusive. The first evidence linking autophagy as a tumour suppression mechanism came from the discovery that Atg6/Beclin1 is a haploinsufficient tumour suppressor (Karantza-Wadsworth et al., 2007; Liang et al., 25 Introduction 1999; Qu et al., 2003; Yue et al., 2003). Other studies implicating Atg4C, Atg5 and Atg7 deficiencies to increased rates of tumourigenesis as well as the findings that Beclin1 cofactors act as tumour suppressors further demonstrated the tumour suppressive function of autophagy (Brech et al., 2009; Marino et al., 2007; Mathew et al., 2007b; Takamura et al., 2011). Further studies revealed that most oncogenes like class I PI3K and Akt inhibit autophagy while tumour suppressors such as phosphatase and tensin homologue (PTEN) and tuberous sclerosis complex 1 and 2 (TSC1 and TSC2) induce autophagy (Maiuri et al., 2009). Moreover, autophagy may also be required to kill tumour cells efficiently in certain circumstances as shown by Giaccia and colleagues where they demonstrated that addition of a compound STF-62247 induced autophagy and vacuolisation in von Hippel-Lindau (VHL)-deficient renal cell carcinoma cells (Turcotte et al., 2008). On the other hand, autophagy induction may represent a means for tumour cells to survive various cellular insults such as hypoxic conditions, metabolic stress and chemotherapeutic stress (Chen and Debnath, 2010; Degenhardt et al., 2006). Although autophagy may have tumour suppressive functions, autophagy is still required for the survival of established tumour cells in response to various stresses. As a result of their higher proliferation rates, tumour cells have higher demands for nutrients and oxygen and they increase their autophagic activity in order to survive this metabolic stress (Degenhardt et al., 2006). Similarly, tumour cells found in the interior of poorly vascularised tumours survive hypoxic conditions by up-regulating their autophagic activity, in turn protecting them from apoptosis and necrosis. Furthermore, autophagy may confer pro-survival benefits in metastasising tumour cells by protecting them from anoikis (Fung et al., 2008). Autophagy also serves to help tumour cells survive radio- or chemo-therapeutic stress. A study by Apel et al. 26 Introduction provided evidence that siRNA-induced inhibition of autophagy re-sensitised cancer cells to radiation-induced cell death, suggesting that cancer cells co-opt autophagy as a cell survival mechanism, which is in line with its function of promoting cell survival during periods of stress (Apel et al., 2008). Thus autophagy is likened to a doubleedged sword, as autophagy has both tumour suppressive and pro-tumourigenic properties (Mathew et al., 2007a; White and DiPaola, 2009). There is a consensus that autophagy as a tumour suppressive mechanism exerts its effects via damage mitigation and limiting metabolic stress and genomic instability (KarantzaWadsworth et al., 2007; Mathew et al., 2007b). Tumourigenesis is favoured in an inflammatory environment, and autophagy serves to limit inflammation and therefore exerts its tumour suppressive function (Degenhardt et al., 2006). However, autophagy is also co-opted as a survival mechanism by established tumour cells facing hypoxic conditions and metabolic stress, enabling long-term cell survival against a backdrop of apoptotic defects (Degenhardt et al., 2006; Lum et al., 2005). Autophagy may also help tumour cells establish dormancy in the face of severe stress, and the ability to regenerate and form viable tumour cells when favourable environmental conditions return (White and DiPaola, 2009). Lastly, autophagy may also be up-regulated in tumour cells facing therapeutic stress, as demonstrated by Apel et al. (Apel et al., 2008). Thus it is hypothesised that autophagy be up-regulated during early carcinogenesis in order for it to exert its anti-cancer function, while inhibited in extant tumours to re-sensitise them to stress and chemo- and radio-therapy. As such, more work needs to be done to establish this paradigm as well as to determine the efficacy of such an approach. 27 Introduction Figure 3 Autophagy in tumourigenesis and anti-cancer therapy. Autophagy is likened to a double-edged sword possessing both tumour suppressive and protumourigenic properties. Autophagy can either promote tumour cell survival or kills tumour cells as a result of anti-cancer therapy; hence it is a novel target in anti-cancer therapy. Figure reproduced from Maiuri et al. (Maiuri et al., 2009). 28 Introduction 3. Rationale of current study 3.1 Establishing Tg(fabp10:egfp-lc3) and Tg(fabp10:mrfp-egfp-lc3) transgenic zebrafish for studying autophagy The study of autophagy in a model organism has been made possible by the use of EGFP-LC3 fusion protein and this was demonstrated by Mizushima et al. when they generated the first transgenic mouse with ubiquitous and constitutive expression of EGFP-LC3 (Mizushima et al., 2004). Therefore, it is envisioned that the establishment of a transgenic zebrafish expressing egfp-lc3 will be useful in the study of autophagy in this popular vertebrate model, where little study has been conducted. A zebrafish transgenic line with liver-specific egfp-lc3 expression will be generated in this study in order to study autophagy in the liver. This makes sense as autophagy was first studied in rat livers following glucagon injection and that autophagy plays an important role in nutrient and energy metabolism and homeostasis in the liver (Yin et al., 2008). In addition, reduced autophagic activity was implicated in liver inflammation and carcinogenesis involving alpha-1-antitrypsin deficiency, suggesting that functional autophagy is required for proper protein aggregate disposal (Perlmutter, 2009). To generate a transgenic line with constitutively liver-specific egfp-lc3 expression, the liver-specific fabp10 promoter will be used to restrict egfp-lc3 expression to the liver tissue. In order to increase transgenesis and germline transmission rates, the Ac/Ds transposition system will be used as well. This transgenic line will be useful in studying autophagy in liver development as well as in various liver injuries. It will be useful for studying the role of autophagy in liver cancer using our liver cancer models, and will also be useful for chemical and drug screens that target the autophagy pathway. 29 Introduction As a proof of concept, we will also generate a transgenic line with constitutively liver-specific mrfp-egfp-lc3 expression. This will be the first attempt in generating a transgenic organism expressing this novel reporter. We hope to employ this transgenic line in liver cancer study as well as other studies involving liver development or liver diseases in the zebrafish. Similar methodology used in the generation of the EGFP-LC3 transgenic line will be applied here. 3.2 Establishing the role of autophagy in liver cancer Studies have also demonstrated both pro-tumourigenic and anti-tumourigenic functions of autophagy in liver cancer. Mice heterozygous for Beclin1 have a higher propensity to develop spontaneous hepatocellular carcinoma (HCC), suggesting an anti-tumourigenic function (Qu et al., 2003). Ding et al. demonstrated in their study that HCC cell lines have lowered levels of Atg gene expression as well as autophagic activity compared to normal hepatic cell lines; they also found reduced levels of Beclin1 mRNA and protein levels in HCC tissue samples compared to adjacent nontumourigenic tissues from the same patient. They also demonstrated that lowered autohagic activity in the malignant cell lines and HCC tissues was especially prominent when anti-apoptotic proteins such as Bcl-XL was over-expressed, further suggesting that autophagy defects synergised with apoptotic defects in promoting liver carcinogenesis, suggesting that antophagy has an anti-tumourigenic function (Ding et al., 2008). In contrast, Longo et al. demonstrated that autophagy functions as a survival mechanism in liver cancer cells and protected them from anthocyanininduced apoptosis, thus suggesting a pro-tumorigenic role of autophagy (Longo et al., 2008). As can be seen, the role of autophagy in liver cancer is currently not well established, although it is proposed that autophagy mainly exerts a tumour 30 Introduction suppressive function during early stages of carcinogenesis while it is co-opted by extant tumour cells as a means to survive metabolic and chemo- or radio-therapeutic stress. Thus we would like to examine this phenomenon using our zebrafish liver cancer models. In order to do this, we will examine the autophagic flux during liver tumour initiation, progression and regression using our zebrafish liver tumour models. We hope to derive more detailed information on the involvement of autophagy in liver cancer, and hope that this information will help in designing combinatorial chemotherapy for liver cancer patients using autophagy inducers or inhibitors at the appropriate stages of treatment together with established chemotherapeutic drugs for liver cancer. 3.3 Objectives of the study We would like to propose the following objectives for this study: 1. Validation of conservation of autophagic machinery in the zebafish 2. Establishment and characterisation of a zebrafish transgenic line with constitutively liver-specific egfp-lc3 3. Establishment and characterisation of a zebrafish transgenic line with constitutively liver-specific mrfp-egfp-lc3 expression 4. Analysing autophagic flux in zebrafish transgenic lines with inducible liver tumour zebrafish models by crossing egfp-lc3 and oncogene-expressing transgenic lines We hope that by achieving these objectives we will be able to prove our hypothesis that autophagy serves an anti-cancer function in the early stages of cancer and it may be co-opted by cancer cells as a cell survival mechanism at later stages. This will greatly impact the field and provide information for combinatorial drug 31 Introduction therapy using existing drugs and autophagy inducers or inhibitors at appropriate stages of cancer to treat liver cancer patients. 32 Materials and Methods Materials and Methods 33 Materials and Methods 1. Zebrafish care and maintenance and tissue collection 1.1 Zebrafish maintenance Wild type and transgenic adult zebrafish were maintained in the National University of Singapore Aquarium at Block S2 Level 2 according to the method as described in the Zebrafish Book available online at the Zebrafish Information Network (ZFIN) (http://zfin.org/zf_info/zfbook/zfbk.html) (Westerfield, 2000) and in compliance with the Institutional Animal Care and Use Committee (IACUC) guidelines. Fish were kept in the aquarium with a photoperiod of 14 h light, 10 h dark. Fish were fed brine shrimp (World Aquafeeds, USA) twice a day, with additional flakes being fed in the late afternoon during the spawning period. 1.2 Zebrafish spawning and embryo collection To collect embryos, breeding adult fish were separated into pairs with each pair being separated using a divider in a tank the night before. Spawning was induced by the removal of the divider and allowing the fish to mate on the following morning. Following that, the breeding pair was removed and the embryos that were found on the bottom of the tank were drained using a sieve and placed in a petri dish containing egg water (final concentration of 60 µg/ml sea salt dissolved in RO water) and methylene blue (final concentration of 0.0002 %), and staged accordingly (Kimmel et al., 1995). The required developmental stages were presented as hours postfertilisation (dpf). Embryos were placed in an incubator with a constant temperature of 28.5 °C to facilitate staging. To rear embryos, embryos were kept in the incubator with daily changes of egg water and removal of dead embryos and larvae until they were approximately 4 dpf, before they were transferred into larger tanks containing egg water only. The 34 Materials and Methods larvae were transferred into tanks in the aquatic system when they were approximately 21 dpf. 1.3 Microinjection into one-cell stage embryos Needles were prepared from glass capillaries using the P-97 Flaming/Brown Micropipette puller (Sutter Instrument, USA) using optimised heat and pull time conditions. Microinjection was carried out using the FemtoJet microinjector (Eppendorf, Germany) using optimised pressure and microinjection time settings. Embryos were placed on a glass slide with excess egg water drained off using KimWipes tissues (Kimberly-Clark Co., USA) before microinjecting under a dissecting microscope. Microinjected embryos were then returned to the petri dishes to be reared. To generate the Tg(fabp10:egfp-lc3) transgenic line, 50 ng/µl of the plasmid pDS-FABP10-EGFP-LC3 was mixed with 500 ng/µl of Ac mRNA, and the mixture was microinjected into one-cell stage embryos. A total volume of 1 nl was injected into each embryo. A total of 25 pg of plasmid and 250 pg of Ac mRNA were microinjected into each embryo. To generate the Tg(fabp10:mrfp-egfp-lc3) transgenic line, 50 ng/µl of the plasmid pDS-FABP10-mRFP-EGFP-LC3 was mixed with 50 ng/µl Ac mRNA, and the mixture was micro-injected into one-cell stage embryos. A total volume of 1 nl was injected into each embryo. A total of 25 pg of plasmid and 25 pg of Ac mRNA were micro-injected into each embryo. 35 Materials and Methods 1.4 Larval screening and photo-documentation Transgenic larvae were screened based on fluorescence expression, embryos were reared until approximately 3 dpf before being screened using a fluorescence microscope (Carl Zeiss AG, Germany). Larvae expressing desired fluorescence were siphoned with a plastic pipette and reared in a separate petri dish containing egg water. Larvae were photo-documented using the Carl Zeiss Axiovert 200M microscope (Carl Zeiss, Germany), using the 2.5X, 5X and 10X objectives after anaesthetised using 0.1 % phenoxy-ethanol (Sigma Aldrich, USA). GFP signals were detected using the blue filter while RFP signals were detected using the green filter. Images were taken using the attached Axiocam HRC camera mounted on the microscope and accompanying software. Images were further processed using Adobe Photoshop software (Adobe, USA). 1.5 Liver tissue collection from adult zebrafish Adult transgenic fish were sacrificed to collect their liver tissue for histology and protein analyses. Briefly, each fish was first anaesthetised in ice water, after which an incision was made from the anus across the belly along the midline, up across the pectoral fins and finally along the trunk ending at the anus, with the skin removed to expose the liver tissue. Liver tissue was then removed from the trunk using a pair of sterile forceps and placed in sterilised microcentrifuge tubes. Tumourigenic livers from 3 transgenic fish were pooled together in a single tube, while up to 10 non-tumourigenic livers were pooled together in 1 tube. All tubes were snapped frozen using liquid nitrogen and subsequently stored at -80 °C. 36 Materials and Methods 2. RNA applications 2.1 Total RNA isolation from embryos and larvae Embryos and larvae at different stages of development were collected in sterilised microcentrifuge tubes and snap frozen using liquid nitrogen after draining off excess egg water. Total RNA was then extracted using TRIZOL reagent (Invitrogen, USA) according to manufacturer’s instructions with some modifications. All centrifugation steps were performed at 4 °C. 300 µl of TRIZOL reagent was first added to each sample and homogenised using a motorised pestle before adding the remaining 700 µl of the reagent. Each sample was then incubated at room temperature for 10 minutes to allow nucleoproteins to dissociate; this was followed by the addition of 200 µl of chloroform. The mixture was shaken vigorously by hand for 20 seconds before incubating at room temperature for an additional 5 minutes. Samples were centrifuged at 12,000 g for 15 minutes, after which the aqueous layer was carefully transferred to a separate sterilised tube. 500 µl isopropanol was added to each tube to precipitate RNA following which the tubes were inverted 10 times by hand and incubated at room temperature for 20 minutes. The mixtures were vortexed for 5 seconds before undergoing centrifugation at 12,000 g for 10 minutes. Excess isopropanol was removed and 1 ml of 75 % cold ethanol was added to wash the resulting pellet. The pellets were centrifuged again at 7,500 g for 5 minutes before draining the ethanol. The pellets were left to air-dry for approximately 5 minutes before the addition of diethyl pyrocarbonate (DEPC)-treated water to fully dissolve the pellet. Aqueous RNA samples were heated at 55 °C for 10 minutes. 37 Materials and Methods 2.2 RNA quantification The aqueous RNA was quantified using Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientific, USA). The RNA quality was verified by running the samples in 1 % agarose gel between 80 to 100 V for approximately 20 minutes, upon which the gel was photo-documented using the Gel Doc XR Gel Documentation System (Bio-Rad, USA). RNA samples were then aliquot into separate tubes and stored at -80 °C. 2.3 Ac mRNA preparation The pAc-SP6 plasmid was first linearised using BamHI, and gel purified to yield linearised DNA. The linearised DNA was used to generate capped mRNA in vitro using the mMessage mMachine SP6 kit (Ambion, USA). Briefly, approximately 2 µg of linearised template was mixed with 2 µl of 10X Reaction Buffer (containing unspecified concentrations of salts, buffer, DTT and other ingredients), 10 µl of 2X NTP/CAP (10 mM ATP; 10 mM CTP; 10 mM UTP; 2 mM GTP; 8 mM cap analogue), 2 µl of SP6 enzyme mix (buffered 50 % glycerol containing RNA polymerase, RNase inhibitor, and other components) and nuclease-free water to give a 20 µl reaction volume. The mixture was thoroughly mixed and incubated at 30 °C for 2 hours, following which 1 µl of TURBO DNase (2 units/µl) was added to the mixture and incubated at 37 °C for an additional 15 minutes. The resultant capped RNA was precipitated by lithium chloride. Briefly, 1 µl 0.5 M EDTA pH 8, 2.5 µl 4 M lithium chloride and 75 µl 100 % cold ethanol was added to the mixture and left to precipitate on ice for approximately 30 minutes, following which the mixture was centrifuged at 4 °C at 14,000 rpm for 20 minutes. The supernatant was carefully removed and the pellet was washed using 500 µl of 75 % cold ethanol and vortexed. 38 Materials and Methods The resultant mixture was then centrifuged at 4 °C at 14,000 rpm for 10 minutes. The ethanol was decanted, and the pellet was air-dried for approximately 5 minutes before being dissolved in DEPC-treated water. The aqueous capped mRNA was then quantified using the Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientific, USA), and the quality was verified by running a 1 % agarose gel between 80 to 100 V for approximately 20 minutes. The gel was later photo-documented using the Gel Doc XR Gel Documentation System (Bio-Rad, USA) to ensure only a single band of size between 1 kb and 1.5 kb was seen. mRNA was then aliquot into separate tubes and stored at -80 °C. 3. DNA applications 3.1 Total DNA isolation Larvae at different stages of development were collected and DNA extraction buffer (10 mM Tris-HCl pH8.2; 10 mM EDTA; 200 mM NaCl; 0.5 % SDS), proteinase K and RNase were added according to established protocol. Larvae were homogenized and incubated at 56 °C for approximately 2 hours, with vortexing every 30 minutes during incubation. An equal volume of phenol-chloroform was added and mixed and centrifuged at 13,000 rpm for 10 minutes. The supernatant was then transferred to a separate sterilized microcentrifuge tube and 0.1 volume of 3 M sodium acetate (pH 5.2) and 2 volumes of 100 % ethanol were added, mixed and incubated at room temperature for 15 minutes, following which the mixture was centrifuged at 13,000 rpm for 5 minutes to remove ethanol. The resulting pellet was rinsed with 200 µl of 70 % ethanol and centrifuged at 13,000 rpm for 5 minutes, following which the excess ethanol was drained off. The pellet was air-dried for 39 Materials and Methods approximately 5 minutes and TE buffer was added to fully dissolve the pellet. The aqueous DNA samples were aliquot into separate tubes and stored at -20 °C. For genotyping Polymerase Chain Reaction (PCR), individual embryos or larvae were placed in sterilised microcentrifuge tubes, and covered with 20 µl of extraction buffer consisting of 1X PCR Buffer from the GoTaq Flexi DNA Polymerase kit (Promega, USA) and proteinase K (1:500 ratio). The mixture was subsequently incubated at 60 °C for at least 2 hours, and vortexed after 1 hour of incubation. The mixture was quickly centrifuged to pellet the mixture and heated at 95 °C for 10 minutes to eliminate proteinase K activity. The mixture was subsequently stored at -20 °C. 1 µl of the mixture was used to run 1 reaction of genotyping PCR. 3.2 cDNA synthesis cDNA synthesis was carried out using isolated total RNA using the SuperScript II Reverse Transcriptase kit (Invitrogen, USA). Briefly, total RNA was pre-treated with DNase to remove contaminating DNA. 20 µl of aqueous total RNA containing 2 µg of RNA, and 2 µl of 0.5 µg/µl oligo-dT and 10 mM dNTPs were mixed in a sterilised microcentrifuge tube and incubated at 65 °C for 5 minutes. 8 µl of 5X First Strand Buffer (250 mM Tris-HCl, pH 8.3; 375 mM KCl; 15 mM MgCl2) were added together with 4 µl of 0.1 M DTT and 2 µl of RNaseOUT (40 units/µl) into the mixture and incubated at 42 °C for 2 minutes, following which 2 µl of SuperScript II Reverse Transcriptase was added. The entire reaction mixture was incubated at 42 °C for 50 minutes, and then at 70 °C for 15 minutes to inactivate reverse transcriptase activity. The resultant cDNA was diluted 10 times using autoclaved 40 Materials and Methods MilliQ water and quantified using the Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientific, USA). 3.3 Construction of plasmid DNA 3.3.1 Restriction endonuclease digestion of DNA Restriction endonuclease digestion of plasmid DNA was performed to release desired fragments from a particular plasmid for sub-cloning purposes, or to linearise a plasmid in order to insert a DNA fragment into the said plasmid. All restriction enzymes used were purchased from New England Biolabs, and digestion reactions were performed at 37 °C for 2 hours or according to manufacturer’s recommendations. Typically 10 µg of plasmid were used in restriction digestion reactions. All digestion reactions were performed in a total volume of either 60 µl or 100 µl, with the volume of restriction enzymes not exceeding 10 % of total reaction volume to prevent star activity. To stop the reaction, the mixture was immediately placed on ice, following which they were mixed with 6X Loading Dye (Fermentas, USA) and run on a 1 % agarose gel to separate the fragments. DNA electrophoresis was run at a constant voltage of between 80 to 100 V for 20 to 45 minutes, depending on the estimated size of fragment and linearised plasmid. Table 1 shows the restriction enzymes used for subcloning in this study: 41 Materials and Methods Table 1. Restriction enzymes used for sub-cloning Plasmid used Released fragment pLF-3.0 EGFP EGFP pEGFP-LC3 EGFP-LC3 pmRFP-EGFP-LC3 mRFP-EGFP-LC3 pMDS6 Nil, linearise the plasmid for sub-cloning Restriction enzymes used MluI, NheI MluI, NheI MluI, NheI MluI, NotI 42 Materials and Methods 3.3.2 Recovery of DNA fragments from agarose gel Gel purification is one of the most effective methods to recover and purify DNA fragments that were released or digested by restriction enzymes. Gel purification reactions in this study were performed using the QIAquick Gel Extraction Kit (Qiagen, USA) according to manufacturer’s instructions. Briefly, gel slices containing the DNA band of interest were cut from the agarose gel, weighed, and 3 volumes of Buffer QG (containing guanidinium thiocyanate) were added to 1 volume of gel. The resultant mixture was incubated at 50 °C for 10 minutes with occasional vortexing to completely melt the gel. This was following by the addition of 10 µl of 3 M sodium acetate, pH 5.0. For fragments larger than 4 kb, 1 gel volume of 100 % isopropanol was added to the mixture. The entire mixture was then loaded into the QIAquick column and centrifuged at 14,000 rpm for 1 minute. To completely remove traces of agarose gel, an additional 500 µl of Buffer QG was added and centrifuged using the same conditions. The column was washed using 750 µl of Buffer PE and centrifuged. The column was centrifuged for an additional minute after decanting to remove residual Buffer PE. 30 µl of autoclaved MilliQ water was added to the column and the column was left to incubate at room temperature for approximately 2 minutes before centrifuging at 14,000 rpm for 1 minute to elute the DNA in the column. The DNA was quantified and stored at -20 °C. 3.3.3 Ligation Ligation reactions between DNA fragments and vector DNA were typically carried out in a 10 µl volume reaction consisting of 1 µl of 10X ligation buffer (0.3 M Tris-HCl, pH 7.8; 0.1 M MgCl2; 0.1 M DDT and 5 mM ATP) (NEB, USA), 1 µl T4 DNA ligase (NEB, USA) and variable volumes of vector DNA, insert DNA and 43 Materials and Methods autoclaved MilliQ water. Vector DNA and insert DNA were added according to a molar ratio of approximately 1:3 or 1:4. The ligation reaction was mixed well and left to incubate at 4 °C overnight. Occasionally, 1:10 molar ratios were attempted if reactions using molar ratios of 1:3 or 1:4 did not yield successful ligation. 3.3.4 Competent cell preparation Competent bacterial cells are usually used for transformation and retransformation reactions. To prepare competent cells, One Shot MAX Efficiency DH5α-T1 competent cells (Invitrogen, USA) or any other types of bacterial cells were first streaked onto blank Luria Bertani (LB) (Invitrogen, USA) agar plates and incubated at 37 °C overnight. One colony was picked from the plate and inoculated into 25 ml of LB media and incubated at 37 °C with constant agitation at 250 rpm. 2.5 ml of the culture was re-innoculated into another 25 ml of LB media and incubated at 37 °C overnight with constant agitation. The next morning, 0.5 ml of the overnight culture was inoculated into 50 ml of LB media and incubated at 37 °C with constant agitation for 2 to 3 hours, until the OD600 reached between 0.5 and 0.6. The culture was then chilled on ice for 15 minutes after transferring the culture into 50 ml Falcon tubes. The competent cells were pelleted by centrifuging at 2,500 rpm at 4 °C for 15 minutes, following which the supernatant was decanted and the tubes were inverted and air-dried to completely remove the supernatant. Cells were suspended in 17 ml of CCMB solution (80 mM CaCl2.2H2O; 20 mM MnCl2.4H2O; 10 mM MgCl2.6H20, 10 mM CH3COOK, 100 ml glycerol), following which the solution was chilled on ice for 20 minutes. The solution was further centrifuged at 3,000 rpm at 4 °C for 15 minutes, following which the supernatant was decanted. The cells were subsequently resuspended in 4 ml of CCMB, and incubated on ice for 30 minutes, after which they 44 Materials and Methods were aliquot into sterilised pre-chilled microcentrifuge tubes in aliquots of 100 µl. The tubes were snap frozen using liquid nitrogen and stored at -80 °C. 3.3.5 Transformation and re-transformation reactions Transformation and re-transformation reactions can be prepared using the freshly made competent cells. For a transformation reaction, a tube of competent cells was first thawed on ice before adding 5 µl of ligation product. The mixture was then incubated on ice for 20 minutes before being subjected to 2 minutes of heat shock at 37 °C. The mixture was then incubated on ice for an additional 5 minutes, after which 1 ml of LB broth was added to the mixture. The entire mixture was then incubated at 37 °C for 1 hour with constant agitation at 250 rpm. The mixture was centrifuged at 4,000 rpm for 4 minutes following incubation, and excess LB solution was decanted, leaving approximately 100 µl of LB solution in the tube. The competent cells were suspended using the 100 µl of LB solution, following which the competent cell mixture was spread on a LB agar plate containing appropriate antibiotics (30 µg/ml Kanamycin or 100 µg/ml Ampicillin) using glass beads. The LB agar plate was incubated at 37 °C overnight. Re-transformation reactions were performed in order to amplify a given plasmid for other experiments. For a re-transformation reaction, a tube of competent cells was first thawed on ice before adding 0.1 µl of plasmid DNA. The mixture was then incubated on ice for 5 minutes, subjected to 2 minutes of heat shock at 37 °C, and then further incubated on ice for an additional 5 minutes. Subsequently, the mixture was spread on a LB agar plate containing appropriate antibiotics (30 µg/ml Kanamycin or 100 µg/ml Ampicillin) using glass beads. The LB agar plate was incubated at 37 °C overnight. 45 Materials and Methods 3.3.6 Colony screening To verify successful ligation and transformation reactions, colony screening PCR was performed. Colonies were first marked in numerical order and picked using white sterile pipette tips and inoculated into PCR tubes containing 10 µl of antibioticcontaining LB broth. The PCR tubes were further incubated at 37 °C for 30 minutes prior to colony screening PCR using various primers, of which one primer was designed to anneal to the vector DNA while its complementary primer was designed to anneal to the insert DNA. Such a design allows us to determine the presence of insert as well as the orientation of insertion from a single PCR reaction. PCR reactions were performed according to section 3.4.1. 3.3.7 Bacterial cell culture and plasmid amplification After selecting for colonies containing the desired inserts in the correct orientations, 1 µl of the culture from the PCR tubes were inoculated into tubes not containing more than 5 ml of antibiotic-containing LB broth. The cultures were incubated at 37 °C overnight with constant agitation at 250 rpm. 3.3.8 Plasmid isolation and purification Following an overnight incubation, the plasmid was isolated and purified using the Wizard Plus SV Miniprep DNA Purification System (Promega, USA). Overnight culture was centrifuged at 4,000 rpm at 4 °C for 5 minutes, and the supernatant decanted. The pellet was re-suspended in 250 µl of Cell Re-suspension solution (100 mg/ml RNase A; 10 mM EDTA; 25 mM Tris-HCl, pH 7.5), following which 250 µl of Cell Lysis solution (0.02 M NaOH; 1 % SDS) was added to the bacterial suspension. The tubes were inverted 6 times by hand to mix the bacterial 46 Materials and Methods suspension and incubated at room temperature for 2 minutes. 10 µl of Alkaline Protease solution was subsequently added and the suspension was mixed by inverting the tube 5 times by hand, following which the tube was further incubated at room temperature for 5 minutes. 350 µl of Neutralisation solution (4.09 M guanidine hydrochloride; 0.759 M CH3COOK; 2.12 M glacial CH3COOH, pH 4.2) was then added and the tubes were inverted 5 times by hand. The mixture was centrifuged at 14,000 rpm for 10 minutes to pellet the bacterial cell debris, and the supernatant was transferred to a fresh spin column provided by the kit. The spin column containing the supernatant was further centrifuged at 14,000 rpm for 1 minute, following which the flow-through was discarded and 750 µl of Column Wash Buffer (162.8 mM CH3COOK; 22.6 mM Tris-HCl; 0.109 mM EDTA, pH 8.0; 75 % ethanol) was added to wash the plasmid and the spin column was further centrifuged at 14,000 rpm for 1 minute. After discarding the flow-through, an additional 250 µl of Column Wash Buffer was added and the spin column was centrifuged for a second wash step. The spin column was then transferred to a new sterile microcentrifuge tube to elute the plasmid DNA. 50 µl of autoclaved MilliQ water was added to the spin column and the spin column was incubated at room temperature for 5 minutes before being centrifuged at 14,000 rpm for 1 minute to elute the plasmid DNA. The plasmid DNA was then quantified and 1 µl of plasmid DNA was subjected to gel electrophoresis using a 1 % agarose gel to verify and compare migration speed with its predecessors. 3.4 Polymerase chain reaction Polymerase chain reactions are useful to verify as well as amplify DNA fragments in a single reaction. PCR reactions have many applications and variations 47 Materials and Methods in a molecular biology laboratory and we highlight some of the variations that were utilised in this study in this section. 3.4.1 Standard PCR Standard PCR reactions were performed using GoTaq Flexi DNA Polymerase kit (Promega, USA) with a total reaction volume of 10 µl consisting of 2 µl of 5X Green GoTaq Flexi Buffer, 1.2 µl of 25 mM Magnesium Chloride solution, 1 µl of dNTP mixture (2 mM ATP; 2 mM GTP; 2 mM CTP; 2 mM TTP), 0.5 µl of 10 µM sense primer, 0.5 µl of 10 µM anti-sense primer, 0.2 µl of Taq polymerase (5 units/µl) and 0.5 µl of template DNA, with the addition of autoclaved MilliQ water adjusted accordingly to obtain a total reaction volume of 10 µl. For colony screening and genotyping PCR, 1 µl of bacterial culture was usually added. The PCR programme was typically set up with the following cycling parameters: an initial denaturation step at 95 °C for 5 minutes, followed by varying number of cycles comprising a three-step cycling of denaturation at 94 °C for 30 seconds, annealing at 60 °C for 30 seconds and extension at 72 °C for 1 minute. The entire PCR programme ends typically with a final extension at 72 °C for 10 minutes. The annealing temperature as well as the extension times can both be adjusted for optimised PCR reactions depending on the primers used as well as the lengths of amplicon. Typically, the annealing temperature will be set to 2 °C lower than the lower temperature of a pair of primers, while the extension time is estimated based on the formula of 1 min/kb of amplicon. Table 2 shows the primers used for genotyping and other variants of standard PCR. 48 Materials and Methods Table 2. Primers used in PCR Gene name Primer Direction name 3'LFABPF Fragment Forward corresponding to egfp 5'LC3R Fragment Reverse corresponding to egfp Fragment 3’LFABPF Forward corresponding to mrfp-egfplc3 Fragment LC3R Reverse corresponding to mrfp-egfplc3 β-actin B-actin2F Forward β-actin B-actin2R Reverse Sequence 5'GTTCAAACAGCAGCAGGTCATTG 3' 5'GGTCTTCTCGGACGGTCTAGATCT 3' 5'GTTCAAACAGCAGCAGGTCATTG 3' 5'AAGAGCAAGGATGTTCGGGC3' 5'GAGAGAGGCTACAGCTTCAC3' 5'ACTCCTGCTTGCTAATCCAC3' 49 Materials and Methods 3.4.3 2-step reverse transcription polymerase chain reaction (2-step RT-PCR) The 2-step RT-PCR serves to amplify DNA fragments using mRNA templates. It usually starts with cDNA synthesis, as described in section 3.2, before the cDNA templates are used in a PCR reaction as described in section 3.4.1. Table 3 shows the primers used in 2-step RT-PCR. 50 Materials and Methods Table 3. Primers used in 2-step RT-PCR Gene Primer Direction Sequence name name lc3 LC3F Forward 5'TTGTAAAGACGGACTTCACAGGACG3' lc3 LC3R Reverse 5'AAGAGCAAGGATGTTCGGGC3' 5'GGTGTTGTTACCAGCTCCAAA3' atg5 Atg5F Forward 5'TGACCATCACTCCCAATAAGG3' atg5 Atg5R Reverse β-actin B-actin2F Forward 5'GAGAGAGGCTACAGCTTCAC3' β-actin B-actin2R Reverse 5'ACTCCTGCTTGCTAATCCAC3' 51 Materials and Methods 3.5 DNA Vectors 3.5.1 pLF3.0-EGFP This plasmid is a kind gift from Her et al. (Her et al., 2003). The plasmid consists of the liver-type fatty acid binding protein (FABP10) promoter attached to egfp. This construct allows egfp expression in the liver and the promoter was used to control egfp-lc3 expression in the liver. Figure 4 Construct map of pLF3.0-EGFP. 52 Materials and Methods 3.5.2 pEGFP-LC3 This plasmid is a kind gift from Kabeya et al. (Kabeya et al., 2000). The plasmid contains a fusion protein between rat lc3 and egfp. The egfp-lc3 fragment was used to generate the Tg(fabp10:egfp-1c3) transgenic line. CMV promoter EGFP pEGFP-LC3 5452 bp Kan / Neo rLC3 poly A signals Figure 5 Construct map of pEGFP-LC3. 53 Materials and Methods 3.5.3 pmRFP-EGFP-LC3 This plasmid is a kind gift from Kimura et al. (Kimura et al., 2007). The plasmid contains a fusion protein between rat lc3, egfp and mrfp. The mrfp-egfp-lc3 fragment was used to generate the Tg(fabp10:mrfp-egfp-1c3) transgenic line. Figure 6 Construct map of pmRFP-EGFP-LC3. 54 Materials and Methods 3.5.4 pMDS6 This plasmid is a kind gift from Emelyanov et al. (Emelyanov et al., 2006). This plasmid consists of a multiple cloning site (MCS) flanked by a pair of Ds elements to facilitate transposition in the presence of Ac mRNA. The fabp10:egfp-lc3 expression cassette was sub-cloned into the MCS and the whole construct and Ac mRNA were co-injected into one-cell stage embryos to generate Tg(fabp10:egfp-lc3). Similarly, the fabp10:mrfp-egfp-lc3 expression cassette was subcloned into the MCS and the whole construct and Ac mRNA were co-injected into one-cell stage embryos to generate Tg(fabp10:mrfp-egfp-lc3). Amp 5'DS pMDS6 3666 bp MCS 3'DS Figure 7 Construct map of pMDS6. 55 Materials and Methods 3.5.5 pAc-SP6 This plasmid was also a kind gift from Emelyanov et al. (Emelyanov et al., 2006). This plasmid was used to generate Ac mRNA via in vitro transcription as described in section 2.3. Figure 8 Construct map of pAc-SP6. 56 Materials and Methods 4. Protein applications 4.1 Total protein isolation and extraction To extract protein from larvae, larvae must first be de-chorionated using a pair of forceps followed by gentle washing by swirling larvae in a petri dish containing egg water to remove chorion membranes. Next, 50 larvae were placed in a sterilised microcentrifuge tube and 100 µl of Tissue Protein Extraction Reagent (T-PER) (proprietary detergent in 25 mM bicine; 150 mM sodium chloride, pH 7.6) (Thermo Scientific, USA) and 1 µl of protease inhibitor (Roche Applied Science, Germany) were added. The whole mixture was homogenised completely, following which the mixture was centrifuged at 14,000 rpm at 4 °C for 10 minutes. The supernatant was then transferred to a new tube and aliquots were used for protein quantification. 4.2 Protein quantification The Protein Assay Dye Reagent Concentrate (Bio-rad, USA) was used to quantify proteins using the microassay procedure using a microtiter plate, according to manufacturer’s instructions. Briefly, five dilutions of a protein standard, usually Bovine Serum Albumin (BSA) were prepared in autoclaved MilliQ water and 160 µl of each standard was pipeted into separate microtiter plate wells. For protein samples, 2 µl of each sample was pipeted into each well containing 158 µl of autoclaved MilliQ water, making up a total of 160 µl. This is followed by pipeting 40 µl of Dye Concentrate to each well, where they were mixed by gentle pipeting to avoid bubble formation. The entire microtiter plate was then incubated at room temperature for 5 minutes before the absorbance at 595 nm was measured using the xMark Microplate Spectrophotometer (Bio-rad, USA). Protein samples were stored at -80 °C. 57 Materials and Methods Following protein quantification, protein samples were aliquot into separate microcentrifuge tubes, and 6X loading dye (300 mM Tris-HCl, pH 6.8; 12 % (w/v) SDS; 0.6 % (w/v) bromophenol blue; 60 % (v/v) glycerol; 3 % beta-mecaptoethanol) was added to each sample. The samples were then mixed by gentle pipeting before being subjected to denaturation at 100 °C for 5 minutes. Denatured samples were stored at -20 °C if not used to run SDS-Polyacrylamide Gel Electrophoresis (SDSPAGE) immediately after denaturation. 4.3 Protein Detection 4.3.1 SDS-PAGE gel casting The SDS-PAGE gel must first be cast prior to running SDS-PAGE. A typical SDS-PAGE gel consists of a dis-continuous gel made up of a stacking gel used to compress protein samples into a thin line across the gel, and a resolving gel used to separate proteins according to molecular sizes. Usually 12 % resolving gels are cast and used in a typical SDS-PAGE experiment. Occasionally, 15 % resolving gels are used to resolve proteins with small molecular sizes while 10 % gels are used to resolve proteins with large molecular sizes. Gels were cast according to standard protocols from the manufacturers (Bio-rad, USA). Firstly, 1 mm spacer plates and short glass plates were wiped clean with autoclaved MilliQ water and KimWipes (Kimberly-Clark Co., USA), after which they were used to form a glass cassette sandwich using the casting frame. Varying amounts of autoclaved MilliQ water and 40 % Acrylamide/Bis solution, 37.5:1 mixture (Bio-rad, USA), 2.5 ml 1.5 M Tris-HCl, pH 8.8, 0.1 ml 10 % SDS, 0.1 ml 10 % ammonium persulphate (APS) (Bio-rad, USA), and 0.004 ml Tetramethylethylenediamine (TEMED) (Thermo Scientific, USA) were added in the stated order to obtain 10 ml of resolving gel solution. The gel solution 58 Materials and Methods was quickly mixed well by inverting the tube 10 times, following which the solution was pipeted into the glass cassette sandwich until a gap of approximately 1.5 cm between the top of the gel solution and top of the short plate was left. A layer of autoclaved MilliQ water was used to overlay the gel. The resolving gel mixture was then left to polymerise for 30 to 45 minutes. To prepare a 4 ml stacking gel, 2.9 ml of autoclaved MilliQ water was combined with 0.5 ml 40 % Acrylamide/Bis solution, 37.5:1 mixture (Bio-rad, USA), 0.5 ml 1.0 M Tris-HCl pH 6.8, 0.04 ml 10 % SDS, 0.04 ml 10 % APS, and 0.004 ml TEMED in the stated order and mixed. Next, the gel mixture was poured over the hardened resolving gel to the top of the short plate and the comb was placed between the glass cassette sandwich and left to polymerise for 20 minutes. 4.3.2 Running SDS-PAGE After both the resolving gel and stacking gel were polymerised, the entire gel sandwich was placed in the clamping frame. Usually 2 gels were cast and run at the same time; however, a buffer dam will be used when running only 1 gel. Once the 2 gel sandwiches were placed in perfect alignment in the clamping frame with the short plates facing inwards, the arms of the clamping frame will be slid into place to secure both gels. The entire set up was then placed in the Mini-Protean Tetra tank (Bio-rad, USA) and the entire tank was filled with 1X running buffer (19 mM Tris; 144 mM glycine; 0.075 % SDS; pH 8.3), after which samples were loaded with a pipette. Samples were allowed to settle at the bottom of their respective wells before placing the lid on the tank and connecting the lid’s electrical leads into the power supply (Biorad, USA). A constant voltage of 100 V was applied for approximately 1.5 hours, depending on the size of the proteins of interest and resolving gel percentage. 59 Materials and Methods 4.3.3 Protein transfer from PAGE gel to Polyvinylidene fluoride (PVDF) membrane Following SDS-PAGE, the gels were removed from the tank and their gel cassettes and the gels were prepared for transfer to a PVDF membrane. Briefly, the fibre pads and filter papers pre-soaked in 1X transfer buffer (25 mM Tris; 192 mM glycine; 0.01 % SDS; 20 % (v/v) methanol) were used to make a sandwich. The cassette was first placed with the grey side down in a tray filled with 1X transfer buffer and the sandwich was assembled in the following order: fibre pad, filter paper, SDS-PAGE gel, PVDF membrane, filter paper and fibre pad. The PVDF membrane was activated prior to adding to the sandwich by soaking in 100 % methanol for 5 seconds. A 15 ml Falcon tube was used to roll out air bubbles during each step in the sandwich assembly. The sandwich was then clamped in place by the cassette and placed in the trans-blot tank and filled to the brim with 1X transfer buffer. The entire tank was then run at a constant voltage of 100 V for 60 minutes to 90 minutes at 4 °C. 4.3.4 Immunoblotting All antibodies were purchased from Sigma Aldrich (secondary antibodies, anti-β-actin), MBL International (anti-LC3), Abnova (p62) and Covance (GFP). After transfer, the PVDF membrane was removed from the sandwich and placed in 10 ml of 5 % BSA / 1X TBST (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 0.1 % Tween 20) solution per 2 PVDF membranes for blocking at room temperature for 2 hours, following which the membrane was cut into various strips for incubation in primary antibody-containing BSA solution at 4 °C overnight with gentle agitation. Primary antibodies were usually diluted in 1: 1000 ratio in 5 % BSA/TBST solution, with the exception of β-actin, which was diluted in a 1:2000 ratio. The following day, the 60 Materials and Methods primary antibodies were drained out of the container and the membrane strips were washed with 15 ml of 1X TBST solution 3 times for 10 minutes each, following which horse radish peroxidise (HRP)-conjugated secondary antibodies were added in a 1:2000 dilution and incubated at room temperature for 2 hours. After the 2-hour incubation, the membranes were washed with 15 ml of 1X TBST solution 3 times for 10 minutes each. 4.3.5 Protein band detection Subsequently, the membranes were then incubated in a 2 ml Working solution containing equal parts of the Stable Peroxide Solution and Luminol/Enhancer Solution from the SuperSignal West Pico Chemiluminescent Substrate (Pierce, USA) for 1 minute. After removing the membranes from the Working solution excess solution was removed by tapping the edge of the membrane on an absorbent tissue before placing it in a film cassette, with the protein side facing up. A transparency slide cut into the shape of the cassette was then placed on top of the membrane and after ensuring that the transparency slide is clean a piece of X-ray film (Pierce, USA) was gently placed on top of the membrane and the film was exposed for 60 seconds. The film was then developed using appropriate developing solution and fixer using the film developer machine (Kodak, USA). Various exposure times were attempted to ensure sufficient exposure. A maximum of 10 minutes was used to expose a film. 4.3.6 Calculating band intensity from immunoblots The films were scanned and the band intensities were calculated using the Image J software (NIH, USA). Briefly, the bands were selected using the Rectangle Selection tool and the profile plot of each lane was calculated by using the Plot Lanes 61 Materials and Methods function. The Straight Line tool was selected and a straight line was drawn across each peak to enclose each peak. Each peak was then selected and the Label Peaks function was used to label each peak with its size being expressed as a percentage of the total size of all the measured peaks. The percentage intensity of each peak was then compared to the control peak to obtain each peak’s relative intensity after each peak was normalised using respective peaks’ loading control. This data was then plotted as a bar chart. 5. Starvation and chemical treatments 5.1 Starvation treatments Chloroquine was purchased from Sigma Aldrich and dissolved in autoclaved MilliQ water to give a stock solution 100 mM. For the set of experiments studying the effects of chloroquine in starvation, 50 larvae were placed in a 9 cm petri dish with 25 ml of egg water. Larvae were not fed from 3 dpf to 8 dpf in order to exhaust the nutrient supplies found in the yolk. Larvae were further starved from 8 dpf onwards until all larvae in an experimental group died or upon reaching experimental end point, whichever is earlier. Egg water was changed every other day during the entire period from 3 dpf until the end of the experimental duration. Separate pipettes were used to siphon off water in order to prevent cross-contamination from the fed groups as well as from groups exposed to chloroquine. Larvae in the fed group were fed chicken egg yolk suspension from 8 dpf onwards, where an entire egg yolk was ground and 50 ml of autoclaved water was added to give a suspension. Usually 100 µl of egg yolk suspension was fed to the larvae in the fed group daily. All larvae were reared at 28.5 °C in an incubator. 62 Materials and Methods For the set of experiments studying intracellular EGFP-LC3 and mRFPEGFP-LC3 distribution during starvation, 15 larvae were placed in one well containing 15 ml of egg water, using a 6-well plate. Egg water was changed every other day and 20 µl of egg yolk suspension was given to the fed control group daily. Larvae were starved until 12 dpf before processing for confocal microscopy imaging. 5.2 Chemical treatments Rapamycin and everolimus were purchased from Sigma Aldrich. Each of these was dissolved in 100 % DMSO to obtain stock solutions of 10 mM before being used in actual chemical exposures. All stock solutions were kept in the dark by wrapping the tubes with aluminium foil and stored at -20 °C. 2 dpf wild type larvae and 3 dpf transgenic larvae were used in these experiments. 3 dpf transgenic larvae were used so as to visualise intracellular GFP-LC3 distribution. Briefly, 15 larvae were placed in each well in a 6-well plate and 10 ml of egg water was added to each well. DMSO was used as a vehicle control and 100 % DMSO was added to yield a final concentration of 0.01 %, while rapamycin, everolimus and chloroquine stock solutions were added to yield a final concentration of 1 µM rapamycin or 1 µM everolimus and 50 µM chloroquine. The 6-well plates were further wrapped using aluminium foil to prevent light exposure and were incubated in a separate incubator at 28.5 °C for 48 hours. Fresh egg water and fresh chemicals were added after 24 hours incubation. Larvae were rinsed with clean egg water twice before being fixed for confocal microscopy imaging or being processed for immunoblotting. All liquid wastes were pooled, stored and disposed of according to established chemical waste disposal protocols in NUS. 63 Materials and Methods 5.3 Doxycycline induction of Xmrk and cmyc transgene expression Doxycycline was purchased from Sigma Aldrich and the stock solution was prepared by weighing and dissolving doxycycline powder in 40 ml of autoclaved distilled water in 50 ml Falcon tubes. The tubes were wrapped in aluminium foil to prevent light exposure and stored at -20 °C until use. 4 mpf adult transgenic and non-transgenic fish were screened, sorted and placed in tanks containing 5 litres of egg water, with each tank containing 40 fish. All tanks were placed in a dark room, and fish were fed twice daily with brine shrimp. Doxycycline stock solution was added to tanks containing transgenic fish requiring doxycycline-induced oncogene expression to yield a final concentration of 60 µg/ml doxycycline. Water in the tanks was changed and fresh doxycycline was added every other day until the end of the treatment period. During the 3-week recovery period, water in the tanks was changed every other day, but doxycycline was not added into any tank. The tanks were kept in the same room and under dark conditions. 6. External photo-documentation of adult treated transgenic fish Adult transgenic and non-transgenic fish were first anaesthetised in ice water and placed in a plastic weigh boat marked with a 1 cm scale bar. To show the gross internal morphology, the skin covering the belly was removed using a pair of scissors and forceps before replacing the fish back in the plastic weigh boat. Photographs were taken with a digital single-lens reflex camera, and images were processed using the Adobe Photoshop software (Adobe, USA), with the cropped images retaining their aspect ratio as the original images. 64 Materials and Methods 7. Cryosectioning 7.1 Cryostat sections of larvae Larvae were fixed using 4 % paraformaldehyde in PBS (PFA/PBS) at 4 °C overnight with gentle shaking in the dark by wrapping the tubes with aluminium foil, following which the larvae were washed 3 times using PBST (3.2 mM Na2HPO4; 0.5 mM KH2PO4; 1.3 mM KCl; 135 mM NaCl; 0.05 % Tween 20, pH 7.4) for 10 minutes each. The fixed larvae were then transferred into molten 2 % bactoagar/5 % sucrose in a cap detached from a microcentrifuge tube at 50 °C. Needles were used to orientate the larvae with their heads facing upwards before the agar solidified. The block was further trimmed using a razor after solidifying, and was transferred into 30 % sucrose solution to be incubated overnight at 4 °C. The block was subsequently placed on a frozen layer of Tissue Freezing Medium (Jung, Germany) covering the flat surface of the tissue holder. It was subsequently covered with a layer of Medium before freezing it using liquid nitrogen. The entire assembly was then placed inside the Leica CM 1900 cryo-microtome (Leica, USA) to equilibrate to -30 °C for 2 hours. 12 µm thick sections were made and placed on Superfrost Plus slides (Fisher, USA) before being dried on a hot plate at 42 °C for about 1 to 2 hours. Light was excluded by making a cover using aluminium foil and placing the cover over the slides. Following this, slides were washed twice using PBST for 5 minutes each. Slides were either processed for Hoechst staining or sealed immediately using 60 µl FluorSave Reagent (Merck, USA) and covered using 22 X 60 mm cover slips. The slides were left to dry at 4 °C overnight in slide holders. 65 Materials and Methods 7.2 Cryostat sections of adult tissues Adult fish were sacrificed and fixed in 4 % PFA/PBS in the morning, and the adult liver tissue and surrounding gut was removed from the fish and further fixed in 4 % PFA/PBS overnight with gentle shaking without exposing to light. The livers were washed 3 times using PBST for 5 minutes each, before being placed on a layer of Tissue Freezing Medium that covered the flat surface of the tissue holder. A layer of Tissue Freezing Medium was layered on top of it and the entire assembly was subsequently frozen using liquid nitrogen before being placed in the microtome to equilibrate for 2 hours. Downstream procedures were identical to cyro-sectioning of larvae. 7.3 Hoechst staining Washed slides were immediately stained using Hoechst staining reagent. Each slide was incubated with sufficient amount of Hoechst stain so that sections were in contact with the staining solution, and left to incubate at room temperature for 20 minutes in the dark. Subsequently, the slides were tilted to remove the staining solution and washed 4 times using PBST, for 10 minutes each. 60 µl of FluorSave Reagent was then added to each slide and each slide was covered using 22 X 60 mm cover slips. The slides were left to dry at 4 °C overnight in slide holders. 8. Confocal imaging 8.1 Live confocal imaging Live larvae were first anaesthetised using 0.1 % phenoxy-ethanol (Sigma Aldrich, USA) before being transferred to a glass bottom dish (MatTek Co., USA). 1 drop of molten 1.5 % low melting agarose gel was placed on top of each larva and the 66 Materials and Methods larva was subsequently positioned using a pair of needles before the agarose solidified. Larvae were positioned such that they were lying with their left sides on the bottom of the glass bottom dish. Larvae were subsequently imaged for EGFP-LC3 expression using the Zeiss LSM510 Meta scanning laser confocal microscope (Carl Zeiss Inc., Germany), using an excitation wavelength of 488 nm and 505-530 band-pass filters with the Zeiss EC Plan-Neofluar 40X objectives with 2.5 times digital zoom. Similarly, an excitation wavelength of 543 nm and a 590-620 band-pass filter were used to image mRFP expression. Images were collected and processed using Zeiss’s LSM Image Browser software (Carl Zeiss, Germany). 8.2 Confocal imaging of sections Sections were imaged using the Carl Zeiss LSM510 Meta scanning laser confocal microscope. Slides were placed with the cover slip-attached side facing downwards towards the objectives. Images were taken in a similar manner as described in section 8.1. For Hoechst staining, images were taken using an excitation wavelength of 405 nm. Scanning for Hoechst staining was performed after scanning for EGFP and mRFP signals. Images were all collected and processed using Zeiss’s LSM Image Browser software as mentioned. 9. Bioinformatics and sequence analyses The National Centre for Biotechnology Information (NCBI) database was searched for Atg proteins sequences from mouse, rat, human, African clawed frog, chicken, Caenorhabditis elegans, yeast and zebrafish. The blast2seq programme freely available from the NCBI website 67 Materials and Methods (http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastp&BLAST_PROGRAMS= blastp&PAGE_TYPE=BlastSearch&SHOW_DEFAULTS=on&BLAST_SPEC=blast 2seq&LINK_LOC=blasttab&LAST_PAGE=blastn&BLAST_INIT=blast2seq) was used to align human and zebrafish Atg proteins to produce Table 5. Next, the ClustalW2 multiple sequence alignment programme (http://www.ebi.ac.uk/Tools/msa/clustalw2/) was used to generate multiple sequence alignments of LC3 and Atg5, as well as phylogenetic trees, for sequence analyses. 10. Analysis of liver transcriptomic data by deep sequencing Liver transcriptomic data by deep sequencing was available in the lab and normalized data were used for analysis in this study. Only normal male liver transcriptomic data were used. Atg genes were searched and the average transcripts per million (TPM) were calculated and presented. Results are presented in Table 4. 68 Results Results 69 Results 1. Conserved autophagy machinery in zebrafish 1.1 Atg gene expression in zebrafish We first investigated atg gene expression in the zebrafish liver using available liver transcriptomic data generated by deep sequencing. We found that the zebrafish expresses numerous atg genes in the liver as shown in Table 4. Among them, atg13 has the highest average transcript per million (TPM) reads of 191.8964, while atg2 has the lowest average TPM reads of 0.2825. We did not find any transcripts corresponding to atg7 and atg12. 1.2 Conservation of Atg proteins in zebrafish To examine if the zebrafish genome encodes proteins required for the autophagic process, we first performed a search for zebrafish Atg proteins in the National Centre for Biotechnology Information (NCBI) database. As seen in Table 5, the zebrafish indeed has Atg proteins that are crucial for autophagy, including a few proteins such as Atg7 and Atg12 that are predicted from existing genomic information available in NCBI. Moreover, the human and zebrafish Atg protein orthologues displayed remarkable homology when they were aligned using the blast2seq programme found in the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastp&BLAST_PROGRAMS= blastp&PAGE_TYPE=BlastSearch&SHOW_DEFAULTS=on&BLAST_SPEC=blast 2seq&LINK_LOC=blasttab&LAST_PAGE=blastn&BLAST_INIT=blast2seq), with identity scores ranging from 50 % to 93 %. As we are planning to generate a transgenic line expressing liver-specific EGFP-LC3 and mRFP-EGFP-LC3, we decided to look at the LC3 protein in more detail. Further amino acid alignments of LC3 from Mus musculus (mouse), Rattus 70 Results norvegicus (rat), Homo sapiens (human), Xenopus laevis (African clawed frog), Gallus gallus (chicken), Caenorhabditis elegans, and Saccharomyces cerevisiae (yeast) with that of zebrafish showed high conservation of amino acid residues across the entire LC3 protein (Figure 9A). Figure 9B also showed the evolutionary relationship of LC3 from these organisms, suggesting that zebrafish LC3 is almost as conserved as LC3 from mammalian species despite being in a different branch. Atg5 plays an important role in the regulation of autophagy, thus we are also interested to find out if zebrafish Atg5 is similarly conserved. Amino acid alignment of Atg5 revealed that zebrafish Atg5 is highly similar to mammalian Atg5, as shown by the highly conserved amino acid sequence as well as from phylogenetic tree analyses (Figure 10). It can be concluded from the above evidences that zebrafish Atg proteins are conserved and that they may play similar functions as those found in mammals. 71 Results Table 4 atg gene expression in the zebrafish atg gene Refseq sequence ulk1a NM_001130631 atg2b XM_001340472 atg3 NM_200022 atg4b NM_001089352 atg4c NM_001002103 atg4d XM_001333057 atg5 NM_205618 beclin1 (atg6) NM_200872 map1lc31a (lc3) NM_214739 atg9a NM_001083031 atg10 NM_001037124 atg13 NM_200433 atg14 NM_001024812 Average TPM 0.43 0.28 13.83 16.23 0.77 0.29 5.65 4.57 87.11 0.50 12.29 191.90 2.71 72 Results Table 5 Comparison of zebrafish and human Atg proteins. Name Refseq sequence (Human) Refseq sequence (Zebrafish) Alignment scores Identities Positives E-value Atg1 (unc 51-like-kinase 1, Ulk1) NP_003556 NP_001124103 50 % 62 % 0 Atg3 NP_071933 NP_956316 82 % 90 % 2.00E-159 Atg4B NP_037457 NP_001082821 73 % 83 % 3.00E-177 Atg5 Atg6 (Beclin1) NP_004840 NP_003757 NP_991181 NP_957166 81 % 78 % 90 % 87 % 1.00E-138 0 Atg7 NP_006386 XP_002667343 (obsolete) 72 % 80 % 2.00E-138 Atg8 (LC3B) Atg9 NP_073729 NP_076990 NP_955898 NP_001076500 93 % 70 % 96 % 80 % 3.00E-59 0 Atg10 NP_113670 NP_001032201 50 % 50 % 1.00E-61 Atg12 NP_004698 XP_699602 72 % 78 % 2.00E-48 Atg13 NP_001136145 NP_956727 74 % 83 % 0 Atg14 NP_055739 NP_001019983 67 % 85 % 0 Atg16 Vps34 (PI3K Class III) NP_110430 NP_002638 NP_001017854 NP_001017550 68 % 84 % 83 % 91 % 5.00E-126 0 73 Results A B Figure 9 Conservation of LC3 protein in zebrafish. (A) Amino acid sequences from various species were compared using ClustalW2 software, with zebrafish LC3 showing evolutionarily conserved amino acid sequence. (B) A phylogram generated using the same ClustalW2 software, showing inferred evolutionary relationships among LC3 from various species. It also showed that zebrafish LC3 is closely related to mammalian LC3. 74 Results A 75 Results B Figure 10 Conservation of Atg5 protein in zebrafish. (A) Amino acid sequences from various species were compared using ClustalW2 software, with zebrafish Atg5 showing evolutionarily conserved amino acid sequence. (B) A phylogram generated using the same ClustalW2 software, showing inferred evolutionary relationships among LC3 from various species. It also showed that zebrafish Atg5 is closely related to mammalian LC3, and may probably be ancestral to rat, C.elegans and yeast LC3. 76 Results 1.3 Expression of autophagic genes and proteins in zebrafish In order to investigate if the autophagic genes are expressed during zebrafish embryogenesis, reverse-transcription polymerase chain reaction (RT-PCR) of lc3 and atg5 using RNA isolated from wild-type 1-cell stage embryos through 96 hours post fertilisation (hpf) larvae was performed to study their temporal expression patterns. We detected both lc3 and atg5 transcripts from 1-cell stage embryos onwards, suggesting maternal deposition of both transcripts (Figure 11). Activation of zygotic expression of genes beyond 4 hpf might help maintain the expression level of both lc3 and atg5 mRNAs during early embryonic development. Such expression patterns suggested that both lc3 and atg5 may play important roles during early embryonic development, especially the first four hours after fertilisation, during the cleavage stage of embryonic development. We then investigated if expression of lc3 corresponded to protein expression as well as phosphatidylethanolanime (PE) conjugation of LC3 by immunoblotting for LC3 proteins using anti-rat LC3 antibodies. Cytoplasmic LC3, or LC3-I, exists as a 16 kD species, while the PE-conjugated LC3-II has a size of 14 kD due to the increased migratory speed conferred by the PE moiety. Immunoblotting using proteins extracted from wild-type whole larvae from 1 dpf to 5 dpf was shown in Figure 12. Results revealed that LC3-I was expressed from 1 dpf onwards and that PE conjugation only occurred after 1 dpf, as evidenced by the presence of the 14 kD LC3-II band in 2 dpf samples. This result further proved that zebrafish LC3 undergoes similar posttranslational modification as its mammalian counterparts. Results further showed increases in LC3-II levels from 2 dpf to 4 dpf, suggesting that overall autophagic flux may be increased during this developmental period. However, there was a great reduction of LC3-I and LC3-II levels at 5 dpf indicating increased rates of 77 Results degradation in the lysosome from 5 dpf. We further observed that different batches of larvae kept under similar conditions produced varying levels of LC3-I and LC3-II throughout the first five days of development, suggesting that autophagy may be tightly regulated by very small changes in the environment. 78 Results A B Figure 11 Temporal expression patterns of lc3 and atg5. (A) Temporal expression of lc3 transcripts revealed maternal deposition of lc3, suggesting the importance of LC3 protein expression or autophagy function during early development. (B) Temporal expression of atg5 transcripts revealed that, similar to lc3, maternal deposition of atg5 occurred, highly suggesting an unknown role of Atg5 protein function or autophagy function during early embryonic development. β-actin served as loading control. 79 Results Figure 12 LC3-I and LC3-II protein levels from 1 dpf to 5 dpf zebrafish larvae. LC3-I conversion to LC3-II occurred from 2 dpf, as faint LC3-II levels can be observed from the immunoblot. LC3-II levels increased from 2 dpf to 4 dpf, suggesting induction of autophagy during this developmental period. Greatly reduced LC3-I and LC3-II levels were observed at 5 dpf, suggesting increased degradation. Results are representative of at least three independent experiments. β-actin served as loading control. 80 Results 1.4 Conserved autophagic machinery in zebrafish Having establishing that the zebrafish possesses atg genes and expresses Atg proteins and that zebrafish LC3 undergoes similar post-translational modification like its mammalian counterparts, we then assessed the functional conservation of the zebrafish autophagic machinery and its amenability to chemical manipulation. In mammalian and yeast cells, autophagy induction or increased autophagic flux is manifested as an increase in LC3-II levels although a blockage at the final stage of autophagy - lysosomal degradation - can also bring about an increase in LC3-II levels. An increase in LC3-I with concomitant reduction in LC3-II levels are often seen when autophagy inhibition is effected at the induction stage. As such, autophagy inducers and inhibitors can be used to evaluate the functionality of the autophagic machinery in the zebrafish by manipulating autophagy chemically. Rapamycin exerts its effect by inhibiting the mTORC1 complex, relieving its inhibitory effect on the ULK1-FIP200Atg17 complex necessary for autophagy induction, thereby enhancing autophagosome formation and LC3-II levels. However, this increase in LC3-II levels may not be significant due to continuous lysosomal degradation of LC3-II in the autolysosome. Thus the lysosomotropic alkaline chloroquine is also used simultaneously to discern the changes in LC3-II levels as it effectively interferes with the final stage of lysosomal degradation of sequestered materials. As chloroquine interferes with the degradation of sequestered materials in the autolysosomes, it can also be used as an autophagy inhibitor. To assess autophagy induction, we exposed wild-type 2 dpf larvae to 1 µM rapamycin for 8 hours. Another group of larvae were pre-treated with 50 µM chloroquine for 1 hour before being exposed to rapamycin for 7 hours. In addition, larvae were also exposed to 100% dimethyl sulfoxide (DMSO) vehicle in egg water 81 Results as vehicle controls. At the end of the treatment, samples with 1 hour pre-treatment with chloroquine followed by 7 hours of rapamycin exposure saw a more significant increase in LC3-II levels than the samples that were exposed to rapamycin only (Figure 13). This thus proves that the zebrafish autophagic machinery is conserved and is also amendable to chemical treatment like its mammalian counterparts. Autophagy is up-regulated by cells in response to cellular stress such as starvation in order to survive the stress period. Accordingly, cells fail to survive the stressor if they fail to induce autophagy as a means of cell survival. To evaluate this survival function of autophagy, 8 dpf larvae were continuously exposed to 50 µM chloroquine under both fed and starved conditions as shown in the schematic in Figure 14, and their survival rates were noted. Continuous exposure to chloroquine led to early deaths regardless whether the larvae were fed or starved, while larvae not exposed to chloroquine survived better, with starved larvae surviving slightly better than those that were fed. This may be correlated to the amount of debris present in the water that feeding may have caused, with the fed larvae exposed to feed that may have decayed in the duration of the treatment, leading to an accumulation of toxic waste products and/or excess carbon dioxide in the water that caused the death of the fed larvae. The early deaths of the larvae exposed to chloroquine may be due to an inhibition of protein turnover that is essential for the larvae to survive starvation. The above results led us to conclude that the autophagic machinery is highly conserved and obviously functional in the zebrafish, and that it functions very similarly to the mammalian and yeast autophagic machinery in ensuring cell survival in the face of cellular and environmental stressors. These observations provide biological basis for us to generate transgenic zebrafish expressing EGFP-LC3 and mRFP-EGFP-LC3 fusion proteins to study autophagy in vivo. 82 Results A B LC3-II 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 1.59 1.00 0.79 Water DMSO 0.89 LC3-II Rap Rap & CQ Figure 13 Autophagy induction by rapamycin. (A) Larvae treated with rapamycin showed an increase in LC3-II levels, indicating increased autophagic flux, while larvae treated with both rapamycin and chloroquine showed a significant increase in LC3-II. Results were representative of at least three independent experiments. β-actin served as loading control. (B) Quantification of LC3-II levels relative to β-actin loading control from three independent experiments. DMSO: Dimethyl sulfoxide, Rap: 1 µM rapamycin, CQ: 50 µM chloroquine. 83 Results A B Number of embryos survived Survival rate of larvae exposed to chloroquine 50 45 40 35 30 25 20 15 10 5 0 Feed, - CQ Starve, - CQ Feed, + CQ Starve, +CQ 0 2 4 6 8 10 12 14 16 Days post fertilisation (dpf) Figure 14 Autophagy induction is necessary for larvae to survive starvation. (A) Schematic showing the experimental set-up used to investigate the relevance of autophagy during starvation. (B) Survival rates of larvae exposed to 50 µM chloroquine were greatly reduced regardless of whether they were fed or starved. Fed and starved larvae in the absence of chloroquine survived better, although some deaths still occurred. +CQ: addition of 50 µM chloroquine, -CQ: no chloroquine. 84 Results 2. Establishment and characterisation of Tg(fabp10:egfp-lc3) transgenic line 2.1 Establishing Tg(fabp10:egfp-lc3) transgenic line As mentioned previously in the Introduction, autophagy is first studied in the livers of animal models, and that the study of autophagy has not been extended to zebrafish until recently by He et al. (He et al., 2009), therefore it would be of interest to us to study autophagy in zebrafish livers. Furthermore, our lab has generated zebrafish models of liver cancer by transgenic expression of oncogenes, thus it would be interesting to study the role of autophagy in liver cancer. To analyse the role of autophagy in hepato-carcinogenesis we set out to generate a zebrafish transgenic line with constitutive liver-specific egfp-lc3 expression. To generate the zebrafish transgenic line the egfp-lc3 expression cassette created by Kabeya et al. (Kabeya et al., 2000) was placed downstream under the control of a zebrafish liver-specific fabp10 promoter isolated by Her et al. (Her et al., 2003). We incorporated the Ac/Ds transposon system to facilitate transgenesis, generating the plasmid construct pDS-FABP10-EGFP-LC3 as shown in Figure 15. We further compared LC3 protein sequences from rat and zebrafish to ensure high homology between the two as the expression cassette utilises lc3 isolated from rat. Sequence alignment in Figure 16 revealed high sequence identity (93 %) between the two, especially at Gly120 where pre-LC3 protein is cleaved by Atg4B forming LC3-I, before PE conjugation can take place. This thus ensures that exogenous EGFP-LC3 will also be recognised by zebrafish Atg4B and be processed in a similar manner as endogenous zebrafish LC3. 85 Results 5'DS poly A signals rLC3 Amp E GFP pDS-FABP10-EGFP-LC3 8379 bp 3'DS FABP 10 Figure 15 The pDS-FABP10-EGFP-LC3 plasmid used in the generation of Tg(fabp10:egfp-lc3) transgenic line. The egfp-lc3 fragment was first subcloned into the plasmid construct containing the fabp10 promoter, before subcloning the fabp10:egfp-lc3 fragment into the plasmid construct containing Ds elements required for transposition. 86 Results Figure 16 Sequence identity exists between rat and zebrafish LC3 proteins. ClustalW2 was used to study the amino acid sequence of zebrafish and rat LC3. Glycine at position 120 is boxed in red. 87 Results Microinjection of the pDS-FABP10-EGFP-LC3 plasmid into one-cell stage embryos was performed to evaluate the ability of the plasmid in driving EGFP-LC3 expression in the liver. Faint EGFP-LC3 expression in the liver was first observed at 3 dpf after microinjection, with increasingly stronger EGFP-LC3 signals as the larvae developed in size. Following this evaluation, the pDS-FABP10-EGFP-LC3 plasmid was co-injected with Ac mRNA into one-cell stage embryos. As can be seen in Figure 17, injected larvae expressed EGFP-LC3 in the liver from 3 dpf and the EGFP-LC3 expression persisted until at least 6 dpf, allowing EGFP-LC3-expressing larvae to be selected and bred to maturity for further screening. Screening for founder fish was performed when the EGFP-LC3-positive larvae were bred to maturity at around four months post fertilisation (mpf). Each founder adult fish was crossed with its corresponding wild-type partner and their F1 progeny was collected and screened for EGFP-LC3 expression. In total, five out of 11 fish screened produced progeny with EGFP-LC3 expression. Out of the five founders, only one founder (Line 1) was used for subsequent characterisation as the other four showed ectopic EGFP-LC3 expression in the gut (Figure 18). Both EGFP-LC3 positive and negative larvae in Line 1 were processed for genotyping PCR. PCR results showed germline transmission of the fabp10-egfp-lc3 expression cassette in larvae with EGFP-LC3 expression in the liver, while their non-transgenic siblings did not harbour the expression cassette in their genome (Figure 19). Progeny from Line 1 also expressed increasingly higher levels of EGFP-LC3 in the liver as evidenced from their development from 3 dpf to 8 dpf. Immunoblotting for GFP in the liver protein lysates from transgenic adult livers showed the expression of EGFP-LC3 fusion protein as well as free EGFP while those from non-transgenic siblings did not (Figure 19). 88 Results Figure 17 Liver-specific EGFP-LC3 expression after microinjection of pDSFABP10-EGFP-LC3. (A-C) Wild type (5 dpf) un-injected zebrafish larvae did not express EGFP-LC3 in the liver. (D-F) Wild type zebrafish larva (6 dpf) with liverspecific expression after co-injecting pDS-FABP10-EGFP-LC3 and Ac mRNA. 89 Results Figure 18 EGFP-LC3 expression in the progeny of the five founders. Five different founders were identified producing progeny with EGFP-LC3 expression. The founders were named according to the sequence in which they were crossed with wild type fish. (A-D) Line 1 founder producing progeny with liver-specific EGFPLC3 expression, and was chosen for further characterisation. (E-T) Four other founders produced progeny with ectopic EGFP-LC3 expression. 90 Results A B 91 Results C Figure 19 EGFP-LC3 positive larvae in Line 1 harbours germline transmission of the fabp10:egfp-lc3 expression cassette. (A) Liver-specific EGFP-LC3 expression increased in intensity from 3 dpf to 8 dpf. (B) Genotyping PCR revealed germline transmission of the fabp10-egfp-lc3 expression cassette in transgenic progeny with EGFP-LC3 expression, while the non-transgenic siblings without EGFP-LC3 expression did not harbour the expression cassette in their genome. β-actin served as loading control. NTC: no template control. (C) Immunoblotting for EGFP-LC3 using anti-GFP antibodies revealed the presence of a band for both EGFP-LC3 and free EGFP in protein lysates from transgenic adult liver. No such bands were detected in protein lysates obtained from non-transgenic adult liver. β-actin served as loading control. 92 Results 2.2 Characterisation of Tg(fabp10:egfp-lc3) transgenic line We further characterised the F1 progeny from Line 1 for their ability to visualise the autophagic flux in vivo. Firstly, the EGFP-LC3 positive larvae were subjected to starvation from 3 dpf onwards until 12 dpf when they have exhausted their yolk supply in the yolk sac, while the control group were fed throughout the entire period. Live imaging using confocal microscopy was performed to visualise changes in the intracellular distribution of EGFP-LC3 in the liver. As expected, punctate structures of EGFP-LC3 were observed in the livers of larvae subjected to starvation, while the fed larvae showed diffused cytoplasmic distribution of EGFPLC3 (Figure 20). This result indicated that there was increased cellular turnover during starvation and that the autophagic machinery was conserved in the zebrafish. This result further proved that the autophagic flux can be studied in the zebrafish in vivo using confocal microscopy. We further investigated if the transgenic progeny were amendable to chemical manipulation of autophagy using everolimus, an analogue of rapamycin, and chloroquine. Larvae were exposed to everolimus, chloroquine and a combination of everolimus and chloroquine for 48 hours and imaged using the confocal microscope as shown in Figure 21. Control larvae displayed a diffused cytoplasmic distribution of EGFP-LC3, while small EGFP-LC3 puncta were observed in larvae exposed to everolimus, and to the combination of everolimus and chloroquine, as a result of an induction in autophagy. However, there was no observable EGFP-LC3 puncta in larvae exposed to chloroquine alone. This is unexpected as chloroquine exposure should have increased the number of observable EGFP-LC3 puncta due to disrupted lysosomal degradation of EGFP-LC3. However, immunoblotting for LC3 revealed that chemical exposure indeed caused changes in the levels of LC3-I and LC3-II, 93 Results similar to published data, as shown in Figure 22. The reason behind this is currently unknown. We also observed another phenomenon in this set of experiments. The EGFPLC3 puncta observed in larvae exposed to chemical inducers and inhibitors were smaller than those observed when larvae were subjected to starvation. This may be caused by different stimuli and therefore activation of subtly different signalling pathways although they may converge on to the mTOR pathway to control autophagy induction. Further studies are required to study this phenomenon. 94 Results Figure 20 Autophagy induction via starvation. (A) Control larvae that did not undergo starvation showed diffused cytoplasmic EGFP-LC3 distribution in hepatocytes. (B) Hepatocytes from starved larvae produced EGFP-LC3 puncta under starvation conditions, some of the autophagosomes are indicated by the white arrows. Images were taken at 1,000X magnification. 95 Results 96 Results Figure 21 EGFP-LC3 puncta formation in the presence of everolimus and chloroquine. 3 dpf larvae were exposed to 1 µM of everolimus, 50 µM of chloroquine or to a combination of both as indicated in the figure for 48 hours. 100% DMSO was dissolved in egg water to a final concentration of 0.01% and acted as vehicle control. (A-D) Hepatocytes showed diffused cytoplasmic EGFP-LC3 distribution in water and in vehicle control. (E-F) Small EGFP-LC3 puncta were observed in hepatocytes of larvae exposed to 1 µM everolimus. (G-H) No significant EGFP-LC3 puncta were observed in hepatocytes of larvae exposed to 50 µM chloroquine. (I-J) Larvae exposed to both 1 µM everolimus and 50 µM chloroquine had EGFP-LC3 puncta in their hepatocytes. EGFP-LC3 puncta were indicated with white arrows. Cell nuclei were indicated by blue Hoechst staining. EVE: 1 µM everolimus, CQ: 50 µM chloroquine. Images were taken at 1,000X magnification. 97 Results A B LC3-II 11.34 12 10 8 6 LC3-II 4 2 1.00 0.93 Water DMSO 1.38 1.80 0 EVE CQ EVE+CQ Figure 22 LC3-I and LC3-II after chemical exposure for 48 hours. (A) A small increase in LC3-II amounts were observed after larvae were exposed to everolimus, while a small increase in LC3-I was observed in larvae exposed to chloroquine. Larvae exposed to both had markedly increased LC3-I and LC3-II levels, indicative of autophagy induction. β-actin served as loading control. (B) Quantification of LC3II levels relative to β-actin loading control from three independent experiments. DMSO: Dimethyl sulfoxide, EVE: 1 µM everolimus, CQ: 50 µM chloroquine. 98 Results 3. Establishment and characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line 3.1 Establishing Tg(fabp10:mrfp-egfp-lc3) transgenic line As described earlier on the use of the mRFP-EGFP-LC3 fusion protein to detect and measure autophagic flux, we also attempted to generate a transgenic line with constitutively liver-specific mRFP-EGFP-LC3 expression. To generate this transgenic line, we sub-cloned the mrfp-egfp-lc3 expression cassette created by Kimura et al. (Kimura et al., 2007) downstream of the fabp10 promoter before subcloning the fabp10:mrfp-egfp-lc3 fragment into the pMDS6 plasmid construct to generate the pDS-FABP10-mRFP-EGFP-LC3 plasmid construct as shown in Figure 23. The plasmid was microinjected into one-cell stage embryos to evaluate the ability of the plasmid in driving mrfp-egfp-lc3 expression in the liver. As expected, both EGFP and mRFP signals were observed in the livers of 3 dpf microinjected larvae. The plasmid was also co-injected with Ac mRNA into one-cell stage embryos and larvae with both EGFP and mRFP expression as shown in Figure 24 were reared to maturity at around 4 mpf before screening for founder fish. Two founder adult fish were identified to produce progeny with liver-specific EGFP and mRFP fluorescence. Genotyping PCR using genomic DNA isolated from the progeny with EGFP and mRFP fluorescence showed germline transmission of the fabp10-mrfp-egfp-lc3 expression cassette, while their non-transgenic siblings did not (Figure 25). 99 Results 5'DS poly A signals rLC3 Amp EGFP pDS-FABP10-mRFP-EGFP-LC3 mRFP 9063 bp 3'DS FABP -10 Figure 23 The pDS-FABP10-mRFP-EGFP-LC3 plasmid used in the generation of Tg(fabp10:mrfp-egfp-lc3) transgenic line. The mrfp-egfp-lc3 fragment was first subcloned into the plasmid construct containing the fabp10 promoter, before subcloning the fabp10:mrfp-egfp-lc3 fragment into the plasmid construct containing Ds elements required for transposition. 100 Results Figure 24 Liver-specific mRFP-EGFP-LC3 expression after microinjecting the pDS-FABP10-mRFP-EGFP-LC3 plasmid into one-cell stage embryos. (A-C) Wild type (5 dpf) zebrafish larvae did not express mRFP-EGFP-LC3 in the liver. (D-F) Wild type zebrafish larva (4 dpf) with liver-specific mRFP-EGFP-LC3 expression after co-injecting the plasmid with Ac mRNA. 101 Results 102 Results Figure 25 Transgenic larvae harbours germline transmission of the fabp10:mrfpegfp-lc3 expression cassette. Genotyping PCR revealed germline transmission of the fabp10:mrfp-egfp-lc3 expression cassette in transgenic progeny with both EGFP and mRFP expression, while the non-transgenic siblings without both EGFP and RFP expression did not. Identical results were obtained from the two transgenic lines isolated. β-actin served as loading control. 103 Results 3.2 Characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line We next proceeded to characterise the F1 progeny of Line 2 for their ability to visualise the autophagic flux. Similar to the Tg(fabp10:egfp-lc3) transgenic line, the two-colour transgenic progeny were also subjected to starvation from 3 dpf to 12 dpf, while the control larvae were fed throughout the entire period. Live imaging using a confocal microscope revealed the formation of obvious red puncta corresponding to autolysosomes in starved larvae, although we did not observe any significant increases in yellow puncta corresponding to autophagosomes in the starved larvae (Figure 26). As this is an initial experiment, more studies need to be performed to confirm this observation. We also investigated if we can utilise the transgenic progeny to visualise the autophagic flux under chemical treatment using everolimus and chloroquine. Similarly, transgenic larvae were treated with everolimus, chloroquine and a combination of everolimus and chloroquine for 48 hours and imaged using a confocal microscope. Larvae treated with everolimus alone displayed increased numbers of red puncta, although we did not observe a corresponding increase in yellow puncta in the same larva (Figure 27). Larvae treated with chloroquine alone displayed more yellow puncta corresponding to autophagosome formation without any concomitant increase in red puncta indicative of blocked maturation of autophagosomes into autolysosomes, while larvae treated with a combination of both everolimus and chloroquine had increases in both red and yellow puncta, indicative of increased autophagic flux. As this is also an initial experiment to test the validity of our transgenic line, more studies are required to validate our results. 104 Results Figure 26 Starvation induces autophagy. (A) Control larvae that did not undergo starvation did not show obvious red and yellow puncta formation in the hepatocytes. (B) Hepatocytes of starved larvae produced red puncta corresponding to autolysosomes while there was no significant increase in the numbers of yellow puncta corresponding to autophagosomes. 105 Results 106 Results Figure 27 Red and yellow puncta formation in the presence of everolimus and chloroquine. 3 dpf transgenic larvae were treated with 1 µM everolimus, 50 µM chloroquine or a combination of both as indicated in the figure for 48 hours. Control larvae were placed in egg water for 48 hours. (A-B) No obvious red and yellow puncta were observed in control larvae. (C-D) Red puncta were observed without significant yellow puncta formation when larvae were treated with everolimus. (E-F) Slightly more yellow puncta were observed in larvae treated with 50 µM chloroquine. (G-H) Significantly increased red and yellow puncta were observed in larvae treated with both 1 µM everolimus and 50 µM chloroquine. Cell nuclei were indicated by blue Hoechst staining. EVE: 1 µM everolimus, CQ: 50 µM chloroquine. Images were taken at 400X magnification. 107 Results 4. Establishing the role of autophagy in liver cancer With the establishment and characterisation of the Tg(fabp10:egfp-lc3) transgenic line, we further investigated if we could extend the use of this transgenic line to study the involvement of autophagy in liver cancer using our liver cancer transgenic models (Huang et al., 2011; Li et al., 2011; Nguyen et al., 2011). Currently we have developed two liver cancer models based on oncogene over-expression in the liver. The Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) and the Tg(fabp10:TA; TRE:cmyc, krt4:RFP) transgenic lines were generated to study liver cancer, with the former line developing hepatocellular carcinoma (HCC) and the latter line developing hepatocellular adenoma (HA), as shown in Figure 28 (Huang et al., 2011). Moreover, the use of the Tet-on inducible system further allows us to control the temporal expression of the oncogene, allowing us to study tumour initiation, progression and regression. This would also facilitate our study of the autophagy status and role in liver tumourigenesis caused by oncogene over-expression. 108 Results Figure 28 Over-expression of cmyc and Xmrk led to liver tumour formation in stable transgenic zebrafish induced using 60 µg/ml doxycycline. Fish were treated with 60 µg/ml doxycycline from 21 dpf onwards for 30 days. (A, D, G) External gross morphology and histo-pathological analysis of a non-transgenic fish used as a control. (B, E, H) Fish over-expressing cmyc developed an enlarged liver and developing hepatocellular adenoma. (C, F, J) Fish over-expressing Xmrk developed a discoloured and enlarged liver. Histo-pathological analysis revealed the fish developing hepatocellular carcinoma. White dotted lines demarcate liver tissue from surrounding tissue in the fish. HA: hepatocellular adenoma, HCC: hepatocellular carcinoma. Figure adapted from Huang et al. and courtesy of Dr Zhen Li (Huang et al., 2011). 109 Results 4.1 Preliminary data implicating autophagy inhibition in liver cancer Activation of autophagy often entails the suppression of the mTOR pathway with a concomitant reduction in the phosphorylation of targets downstream of the mTOR complex 1 (mTORC1) such as p70 ribosomal S6 kinase (S6K), and eukaryotic initiation factor 4E binding protein 1 (4EBP1) (Corradetti and Guan, 2006). This indirectly leads to reduced amounts of phosphorylated S6 ribosomal protein (p-S6), a target of phospho-S6K. This relationship between mTORC1 activation and increased amounts of p-S6 can be exploited as readout for mTOR activation. Immunoblotting of lysates of both 4- and 8-month old Xmrk over-expressing zebrafish liver saw an increase in p-S6, reflecting mTOR activation, while the untreated transgenic and nontransgenic sibling fish did not (Figure 29). An increase in LC3-I levels were also observed in fish with elevated levels of p-S6, suggesting that autophagy may be blocked due to activation of the mTOR signalling pathway. However, we were unable to determine the amounts of LC3-II levels, thus it would be premature for us to conclude that autophagy blockage was caused by mTOR activation. This observation entails further investigation. 110 Results Figure 29 mTOR activation may cause autophagy inhibition in fish with Xmrk over-expression. Both 4 months and 8 months old transgenic fish treated with doxycycline (Dox) showed elevated p-S6 levels, indicative of mTOR activation. LC3I levels were also elevated in these Dox-induced transgenic fish compared to wild type treated or transgenic non-treated controls, suggesting autophagy inhibition. LC3II levels remained to be determined. Fish liver lysates were used for immunoblotting. β-actin served as loading control. 111 Results 4.2 Establishing double transgenic lines with liver-specific oncogene overexpression and EGFP-LC3 expression to study the role of autophagy in liver cancer In order to study the autophagic flux in the zebrafish liver cancer models in vivo, we crossed the Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) and Tg(fabp10:TA; TRE:cmyc; krt4:RFP) transgenic lines with Tg(fabp10:egfp-lc3) to produce double transgenic fish, with Figure 30 detailing the progeny genotypes that resulted from a cross involving Tg(fabp10:TA; TRE:Xmrk; krt4:GFP). Fish exposed to doxycycline (treatment group) and their respective controls were grouped as in Figure 30. Progeny resulting from crossing Tg(fabp10:TA; TRE:cmyc; krt4:RFP) and Tg(fabp10:egfp-lc3) were treated in an identical manner as the above. The double transgenic fish were induced to over-express Xmrk and cmyc respectively in order to develop liver cancer upon doxycycline exposure when they were approximately 4 mpf. Fish were sampled every two weeks to follow tumour initiation and progression, and doxycycline was withdrawn after six and a half continuous weeks of doxycycline exposure to effect tumour regression. Figure 31 showed the schedule for both doxycycline exposure as well as sampling time points for the double transgenic fish. External gross morphology, and internal gross morphology were taken note of and photodocumented, while liver tissue were taken from fish for processing for immunoblotting and confocal microscopy. 112 Results Figure 30 Schematic showing the genotypes of progeny resulting from crossing Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) with Tg(fabp10:egfp-lc3), and doxycycline exposure and control groups. Fish expressing Xmrk and egfp-lc3 (L+X+) as well as egfp-lc3 only (L+X-) were selected and grouped as shown in the schematic. Each group were further divided into two sub-groups, one of which was treated with 60 µg/ml doxycycline (denoted by red dotted circles) while the other group were not, therefore four different treatment groups will result. L: EGFP-LC3, X: Xmrk, D: doxycycline. 113 Results Adult double transgenic and control fish doxycycline treatment 1st sampling at 2.5 weeks Continue treatment 2nd sampling at 4.5 weeks Continue treatment 3rd sampling at 6.5 weeks Recovery (no doxycycline added from here on) 4th sampling at 7.5 weeks Continue recovery 5th sampling at 9.5 weeks Figure 31 Schedule for doxycycline exposure and sampling time points. Doxycycline treatment was continuous for six and a half weeks to allow tumour formation and development, following which doxycycline was withdrawn and tumours were allowed to regress. Sampling took place approximately every two weeks, with the exception of the fourth sampling time point, where it took place one week after doxycycline withdrawal. 114 Results 4.2.1 Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) Treated double transgenic fish over-expressing Xmrk developed liver tumours as expected within six and a half weeks of doxycycline exposure, with the tumours regressing following three weeks of doxycycline withdrawal (Figure 32). We observed the double transgenic fish developing increasingly larger bellies due to enlargement of the liver tissue as time progressed, with noticeable liver tissue discolouration. Upon dissection, the liver tissue had a firmer consistency, unlike the controls that had liver tissue with a softer consistency. These observations corresponded to the severity of the liver tumours developed, with firmer consistency positively correlated with severity of the tumours developed. The size of the liver tissue returned to near normal when regression was induced with doxycycline withdrawal, with the return of near normal liver consistency and colouration when compared to the controls. Confocal imaging of double transgenic fish livers overexpressing Xmrk also revealed altered cell morphology with increasing severity as the tumours developed. They also displayed diffused intracellular cytoplasmic EGFPLC3 and no obvious EGFP-LC3 puncta were observed. We further observed cell morphology returning back to near normal as tumours regressed, with the reappearance of EGFP-LC3 puncta three weeks after tumours were induced to regress (Figure 32). We also examined the protein levels of LC3-I, LC3-II, p62 as well as EGFPLC3 and free EGFP in these fish (Figure 33). p62 levels can be used to estimate the autophagy status as it interacts with LC3 via a LC3 recognition sequence (LRS) and is a substrate for autophagic degradation and accumulation of p62 is correlated with defective autophagy (Komatsu and Ichimura, 2010). We found that L+X+D+ fish had increased amounts of EGFP-LC3 in their lysates throughout the entire experimental 115 Results duration, despite the failure to detect significant EGFP-LC3 in the other controls. We hypothesised that this may be due to a blockage in autophagy, causing EGFP-LC3 to accumulate in L+X+D+ fish. However, the levels of free EGFP present did not concur with our hypothesis as significant amounts of free EGFP could still be detected in L+X+D+ fish during tumour induction as well as during regression. We did not detect any changes in p62 levels, thus we could not corroborate p62 levels with autophagy status. We were also unable to determine LC3-I and LC3-II levels accurately to arrive at any conclusion. We could only hypothesise that autophagy may be inhibited during liver tumour formation and development, as such, hypothesising that autophagy may serve as an anti-cancer mechanism in Xmrk-induced liver tumour formation. 4.3.2 Tg(fabp10:TA; TRE:cmyc; krt4:RFP) As with the Xmrk over-expressing double transgenic fish, the cmyc overexpressing double transgenic fish developed tumours that were visible due to their severely enlarged livers which manifested as severely enlarged bellies (Figure 34). However, we observed that the tumours that developed maintained their consistencies as well as colouration when compared to the controls. Liver overgrowth was observed in every fish that we sampled at the first three time points. Confocal imaging did not reveal any differences in distribution of intracellular EGFP-LC3, thus it was difficult for us to ascertain if there was any changes in the autophagic flux. We also performed protein analyses on the fish livers. As shown in Figure 35, double transgenic (L+M+D+) liver protein samples showed increased free EGFP levels in samples during tumour regression, suggesting that there was increased lysosomal degradation of EGFP-LC3, though the controls also had increased EGFP levels in the fourth and fifth time points that corresponded to the tumour regression 116 Results phase. The double transgenic protein samples also showed reduced p62 levels in the fourth and fifth time points, highly suggesting increased lysosomal degradation or increased autophagic flux. We also detected slight reduction in EGFP-LC3 levels during tumour regression in L+M+D+ fish, suggesting that EGFP-LC3 was probably degraded. This observation also corroborated with our observation and hypothesis that autophagy may be de-repressed during tumour regression. Immunoblotting for LC3-I and LC3-II did not provide any substantial evidence in supporting the notion that autophagy was inhibited during tumour progression and they remain to be investigated. Although the changes in protein levels in L+M+D+ recapitulated those in L+M-D- controls, those from the two other control groups (L+M+D- ad L+M-D+) were rather different, suggesting that different genotypes as well as addition of doxycycline may have some effect on the fish and on the autophagic process, though this phenomenon also remains to be investigated. 117 Results 118 Results Figure 32 Tumour progression manifested as increasingly severe gross and cellular morphology in transgenic fish over-expressing Xmrk. Fish were grouped according to sampling time point. Liver overgrowth was monitored as treatment time lapsed, with significant overgrowth by the third time point. Liver discolouration was observed as well and corresponded to severity of tumour. Liver cell morphology was also increasingly disrupted, correlating with severity of liver overgrowth and period of doxycycline treatment. Doxycycline removal resulted in recovery and tumour regression, as evidenced by liver size reduction. Cell morphology returned to near normal concurrently. No significant EGFP-LC3 puncta were observed during tumour initiation and progression. EGFP-LC3 puncta re-appeared in the fifth time point, corresponding to recovery. Dotted lines demarcate liver tissue from surrounding tissue. Fish were representative of three different individual fish that were sampled per time point. Confocal images were taken at 1,000X magnification. 119 Results Figure 33 Immunoblot analyses of Xmrk over-expressing fish showing increased free EGFP. Increased free EGFP were observed in double transgenic fish during tumour regression, suggesting either increased autophagic flux or increased lysosomal degradation. Double transgenic fish also had visibly increased amounts of EGFP-LC3 compared to controls. There were no significant changes in p62 levels throughout the entire experimental duration. Changes in LC3-I and LC3-II as well as EGFP-LC3 levels remain to be investigated. Fish liver lysates were used for immunoblotting. Numbers refer to time point at which samples were collected. L: EGFP-LC3, X: Xmrk, D: doxycycline. 120 Results 121 Results Figure 34 Tumour progression in fish over-expressing cmyc, manifesting as liver overgrowth and disruption to cell morphology. Fish were grouped according to sampling time point. Liver overgrowth was monitored as treatment time lapsed, with significant overgrowth by the third time point. Liver cell morphology was disrupted, although not to the extent of fish over-expressing Xmrk. Doxycycline removal resulted in recovery and tumour regression, as evidenced by liver size reduction and return of near normal cell morphology. No significant EGFP-LC3 puncta were observed in any of the fish sampled. Dotted lines demarcate liver tissue from surrounding tissue. Fish were representative of three different individual fish that were sampled per time point. Confocal images were taken at 1,000X magnification. 122 Results Figure 35 Immunoblot analyses of cmyc over-expressing fish showing increased free EGFP and reduced p62 levels during tumour regression. Increased free EGFP and reduced p62 were observed in double transgenic fish during tumour regression, suggesting either increased autophagic flux or increased lysosomal degradation. Changes in LC3-I and LC3-II as well as EGFP-LC3 levels remain to be investigated. Changes in protein levels in L+M+D+ recapitulated those of L+M-D-, however L+M+D- and L+M-D+ did not recapitulate this trend, suggesting that doxycycline as well as genotype may play subtle roles in autophagy regulation. Fish liver lysates were used for immunoblotting. Numbers refer to time point at which samples were collected. L: GFP-LC3, M: Cmyc, D: doxycycline. 123 Discussion Discussion 124 Discussion Autophagy is an evolutionarily conserved cellular process that involves the degradation of a cell’s cytoplasm, long-lived proteins as well as organelles in the lysosome; it enables cells to survive periods of stress such as starvation, and in the case of cancer cells, metabolic and therapeutic stress (Klionsky, 2007; Mizushima et al., 2008). As such, it is central for cells to be able to regulate autophagy for their own survival during periods of cellular stress, and for cancer cells to co-opt autophagy for its survival and subsequent proliferation and metastasis. Although autophagy was first observed in mammalian hepatocytes, studies on autophagy have been carried out mainly in yeast, until recently where more focus is directed towards its involvement in diseases including cancer in humans (Klionsky, 2007). Autophagy has been found to be evolutionary conserved from yeast to humans (Kourtis and Tavernarakis, 2009; Rubinsztein et al., 2007), and in this study, we extended our analyses of autophagy to the zebrafish, a model organism where essentially no studies on autophagy have been done. We first characterised the autophagic machinery in zebrafish and found it to be highly conserved in terms of function like its mammalian counterparts. We further generated transgenic lines with constitutively liver-specific egfp-lc3 and mrfp-egfp-lc3 expression in order to study autophagic flux in the liver. In addition, with our newly developed zebrafish models of liver cancer, we can, at the same time study the involvement of autophagy in liver cancer caused by oncogene over-expression, and expand our knowledge on how these oncogenes affect the regulation of autophagy and how dysregulation of autophagy impact on tumour development. 1. Conserved autophagy machinery in zebrafish To study autophagy in zebrafish, we first investigated the atg gene expression in the zebrafish liver using transcriptomic data generated by deep sequencing (Table 125 Discussion 4). We found that the zebrafish indeed expresses numerous atg genes that encode the core proteins required for functional autophagy. However, we failed to detect any transcripts corresponding to atg7 and atg12, genes that are required for the autophagic process. We hypothesise that lack of studies on autophagy in zebrafish may have been the cause of this and propose the existence and nucleotide sequences of these two genes be verified. We also found that the zebrafish expresses Atg proteins that have high degrees of homology to human Atg proteins when their amino acid sequences were compared (Table 5). We also found that Atg7 and Atg12 protein sequences were derived from predicted genomic sequences, corroborating with our earlier findings that zebrafish atg7 and atg12 genes were still currently uncharacterised. Studies can be done to verify and characterise these proteins in detail after their nucleotide sequences have been verified. We further investigated the degree of similarity of LC3 and Atg5 proteins across different mammalian and non-mammalian species, and found them to be highly conserved in terms of amino acid sequences as well as types of amino acids at each position (Figures 9 and 10). Phylogeny analyses further showed the close evolutionary relationships between zebrafish and mammalian LC3 and Atg5, suggesting that the function of zebrafish Atg proteins may be well-conserved compared to their counterparts in other organisms. Thus, the zebrafish can be used as a complementary organism to study autophagy as zebrafish develops externally and this feature of the zebrafish facilitates such developmental studies involving autophagy. It can also be used as a complementary organism to study the relationship between autophagy and cancer. We next set out to study the temporal lc3 and atg5 gene expression pattern during development. We found that both lc3 and atg5 were maternally deposited as 126 Discussion explained by the presence of their respective mRNA transcripts at the one-cell stage (Figure 11), suggesting that autophagy may be involved in as early as during the cleavage and blastula stages. This observation was also consistent with findings from Tsukamoto et al. that autophagy is triggered by fertilisation and up-regulated in early mouse embryos (Tsukamoto et al., 2008). More studies need to be performed to confirm if the same occurs during zebrafish development. However, He et al. disputed this hypothesis by providing evidence that other atg genes such as ulk1a, ulk1b, atg9a and atg9b that are also crucial for autophagy are not expressed until later during development (He et al., 2009). Hence our findings of maternal deposition of lc3 and atg5 may also point to some unknown function or requirement for the presence of these transcripts during development and this may be worthy of investigation. Morpholino knockdown of lc3 and atg5 may shed light on the need for maternal deposition of these two genes as well as their roles in embryonic development. Furthermore, the presence of these two proteins in addition to Ulk1 and Atg9 in the first 24 hpf should also be confirmed by immunoblotting. Armed with this knowledge, we further investigated the temporal expression of LC3 protein during development. We are also interested in finding out if zebrafish LC3 undergoes conjugation with phosphatidylethanolamine (PE) like its mammalian counterparts. We found that LC3-I was expressed in 1 dpf embryos, while PE conjugation occurred only from 2 dpf onwards (Figure 12). This provided evidence that LC3 undergoes identical post-translational modification like its mammalian counterparts, and that LC3-I and LC3-II can be used as readout for autophagic flux under appropriate experimental conditions. We also observed an increase in the levels of LC3-II from 2 dpf to 4 dpf, highly suggesting autophagy induction and increased autophagic flux during this period of time. Autophagy up-regulation may be required 127 Discussion for larval development from 2 dpf to 4 dpf as zebrafish larvae undergo developmental changes such as fin formation, pigment cell differentiation and circulatory system formation among others from 2 dpf to 4 dpf (Kimmel et al., 1995). This observation corroborated with findings that autophagy plays an important role in mammalian development and cellular differentiation (Mizushima and Levine, 2010). We further observed that LC3-II levels were greatly reduced at 5 dpf. This may be due to an increased rate of lysosomal degradation or a reduction in the autophagic flux at 5 dpf, and we hypothesised that this may be due to autophagic flux returning to a basal level as larval morphogenesis comes to an end. Alternatively, this observation can be interpreted as a blockage at the degradation step in the autophagic process from 2 dpf to 4 dpf, resulting in an increase in LC3-II levels with a corresponding increase in lysosomal degradation, leading to reduced LC3-II levels on 5 dpf. More studies are proposed to understand the dynamics of autophagy during larval development, especially during the transition from 4 dpf to 5 dpf. We next assessed the conservation of autophagy function in the zebrafish larvae using chemical inducers and inhibitors of autophagy (Figure 13). Increases in LC3-II were observed after subjecting wild type larvae to rapamycin treatment, indicative of autophagy induction, while co-incubation with chloroquine further increased LC3-II amounts, which is consistent with published observations (Mizushima et al., 2010). Chloroquine, a lysosomotropic agent, inhibits lysosomal degradation, which includes degradation of materials sequestered and delivered to the lysosome via autophagy as well as LC3-II that is found in the inner and outer leaflets of the autophagosomes, resulting in increases in LC3-II levels. The use of chloroquine alone can effect autophagy inhibition, which is the basis of our future set of experiments that will be discussed in detail later. 128 Discussion We also assessed the pro-survival function of autophagy by continuously exposing wild type zebrafish larvae to chloroquine in the presence and absence of food (Figure 14). We observed that continuous exposure to chloroquine lead to early deaths regardless of whether the larvae were fed or starved, whereas larvae that were not exposed to chloroquine had significantly higher survival rates. This observation may be explained by chloroquine inhibiting lysosomal degradation, thereby disrupting the intracellular turnover of amino acids and other nutrients and building blocks required for cell survival. This set of experiments also recapitulated the observation of reduced survival time of neonates deficient in atg5 by Kuma et al., where they observed starved neonates deficient in atg5 did not survive as long as starved wild type neonates (Kuma et al., 2004). They also observed reduced amino acid concentrations in the plasma and tissues as well as signs of energy depletion in these atg5-deficient neonates, suggesting that autophagy is required for the maintenance of energy homeostasis during neonatal starvation. Furthermore, Tsukamoto et al. published the finding that autophagic degradation in the early mouse embryos was necessary for preimplantation development in mammals and that deficient autophagy may impair protein recycling in these embryos (Tsukamoto et al., 2008). Thus this leads us to hypothesise that autophagy is required for cellular turnover during embryonic development as well as survival during early neonatal starvation regardless of nutritional availability. Furthermore, this published finding can also be correlated to the finding that atg5 transcripts were found in 1-cell stage embryos via maternal deposition, although we did not have evidence suggesting autophagic activity during the cleavage stage. However we cannot refute the possibility of toxic side effects that chloroquine exerts that can cause their early demise, thus we propose that similar experiments using lower concentrations of chloroquine should be tried in future 129 Discussion experiments in order to eliminate the possibility of chloroquine poisoning. Experiments using other autophagy inhibitors can also be performed although these inhibitors must be used with care as they are not specific for autophagy and may have other off-target effects. Similarly, morpholino knockdown can be performed to determine the requirement of autophagy genes, such as atg5 for zebrafish larval development. 2. Establishment and characterisation of Tg(fabp10:egfp-lc3) transgenic line We further went on to generate a transgenic zebrafish line with liver-specific egfp-lc3 expression. We made use of the construct which was a kind gift from Kabeya et al. in the generation of our transgenic line (Kabeya et al., 2000). As their construct utilised rat lc3 gene, we first compared the degree of similarity between zebrafish and rat LC3 amino acid sequence to ensure that rat LC3 will be correctly recognised and processed by zebrafish Atg4. We then proceeded to sub-clone the egfp-lc3 fragment into the plasmid containing the liver-specific fabp10 promoter so that egfp-lc3 expression is controlled by the fabp10 promoter. Finally, we utilised the Ac/Ds transposition system to increase transgenesis rate, generating the plasmid as shown in Figure 15. We found that the larvae had transient liver-specific EGFP expression after microinjecting the plasmid, suggesting that the plasmid worked and that expression was indeed restricted to the liver through the use of the fabp10 promoter. We then coinjected Ac mRNA and the plasmid and reared the larvae with EGFP-LC3 expression (Figure 17). The use of the transposition system greatly enhanced transgenesis rates, as five out of 11 potential founders produced larvae with EGFP expression (Figure 18). We further went on to verify the sites of EGFP-LC3 expression and found that only one founder (named Line 1) produced progeny with liver-specific EGFP-LC3 130 Discussion expression, while the other four founders expressed EGFP-LC3 in the gut or had nonspecific EGFP-LC3 expression in tissues surrounding the liver. This may be due to the enhancer trap effect that is commonly observed when using transposition systems such as the Tol2 transposition system (Kawakami et al., 2004; Parinov et al., 2004). Thus we selected Line 1 for further characterisation. Genotyping PCR using primers that amplify a DNA fragment containing egfp verified germline transmission of the fabp10:egfp-lc3 expression cassette, while immunoblotting using anti-GFP antibodies confirmed the expression of GFP-LC3 fusion protein in the transgenic line (Figure 19), both evidences suggesting the successful integration and expression of the transgene in the fish. We next subjected the transgenic progeny to starvation and successfully visualised EGFP-LC3 puncta formation in starved larvae, while the fed larvae serving as control had diffused cytoplasmic EGFP-LC3 distribution (Figure 20). As zebrafish larvae are able to survive without external nutrient supply for an extended period of time, we deliberately starved them from 3 dpf to 12 dpf in order to exhaust the nutrient supply found in the yolk to induce starvation, while the control larvae were fed from 3 dpf onwards with egg yolk suspension. We observed fewer puncta induced in the livers of zebrafish larvae when compared to cell culture conditions and this may be due to obvious differences between in vitro and in vivo model systems. We also observed that different larvae induced different degrees of EGFP-LC3 puncta formation and attributed this to individual differences. In addition, our observations were also consistent with those of Mizushima et al. using their GFP-LC3 transgenic mice, where starvation-induced autophagic activity differed among individual mice (Mizushima et al., 2004), suggesting individual differences and responses to the same stimulus. 131 Discussion To further prove that our transgenic line was also amendable to chemical manipulation by autophagy inducers and inhibitors, we exposed transgenic larvae to everolimus and chloroquine (Figure 21). Everolimus is an analogue of rapamycin with similar potency and has been used in treating advanced renal cell carcinoma (Garnock-Jones and Keating, 2009; Schuler et al., 1997). Thus, we are interested to find out if everolimus can induce autophagy in vivo, and if we can adopt its use in future experiments involving the liver cancer models. As expected, we observed changes in intracellular EGFP-LC3 distribution when transgenic larvae were exposed to everolimus, as well as a combination of everolimus and chloroquine for 48 hours. We found that the EGFP-LC3 puncta that were induced in the hepatocytes were not as numerous than those induced in in vitro systems, again reflecting the differences between in vivo and in vitro systems. We further observed that the EGFP-LC3 puncta that were induced by chemical treatment were significantly smaller than those observed under starvation conditions. We hypothesised that the involvement of the upstream AMPK pathway during starvation resulted in larger feedback to the mTOR pathway, resulting in larger GFP-LC3 puncta formation. We further hypothesised that as everolimus acts mainly on the mTORC1 complex, a smaller feedback may result from such an inhibition, resulting in smaller EGFP-LC3 puncta formation. We also hypothesised that this difference in puncta size may be due to the application of different stimuli, with physiological stimuli like starvation resulting in larger puncta being formed and smaller puncta formation in response to chemical stimuli. Surprisingly, we did not observe any significant EGFP-LC3 puncta formation in larvae treated with chloroquine. Chloroquine being a lysosomotropic agent inhibits lysosomal degradation by increasing lysosomal pH, rendering lysosomal enzymes inactive; hence the application of chloroquine should inhibit autophagic degradation 132 Discussion and should result in an increase in EGFP-LC3 puncta observed. However, we did not observe this increase and thus we put forth the hypothesis that exposure to chloroquine alone may be insufficient to induce significant GFP-LC3 puncta formation, although we observe increases in EGFP-LC3 puncta formation in larvae treated with the combination of everolimus and chloroquine. We turned to immunoblotting for LC3 proteins in an attempt to explain these observations (Figure 22). We saw a significant increase in both LC3-I and LC3-II in larvae exposed to both everolimus and chloroquine, suggesting that inhibition of the lysosomal degradation by choloroquine and autophagy induction by everolimus were successful. Although we saw an increase in LC3-II levels in larvae exposed to everolimus alone, the slight increase can be explained by lysosomal degradation of LC3-II in the autolysosomes. Interestingly, larvae exposed to chloroquine alone had slightly increased amounts of LC3-II; however this does not explain why we were unable to observe any significant GFP-LC3 puncta in the same group of larvae. More studies need to be performed to provide a reasonable explanation for this observation. We also further characterise our transgenic line with atg3 knockdown (Cui J, unpublished), and proved that GFP fluorescence were absent in morphants, even when exposed to both everolimus and cholorquine, while GFP fluorescence were still observable in larvae microinjected with control morpholino. This set of experiments further proved the value of our transgenic line in studying autophagy and the relationship between different Atg proteins in the zebrafish. Our transgenic line, with liver-specific EGFP-LC3 protein expression, finds value in studying the relevance of autophagy in various liver diseases and pathologies. 133 Discussion 3. Establishment and characterisation of Tg(fabp10:mrfp-egfp-lc3) transgenic line We also generated a transgenic zebrafish line with constitutively liver-specific mrfp-egfp-lc3 expression using the pDS-FABP10-mRFP-EGFP-LC3 plasmid construct shown in Figure 23. Microinjection of both the plasmid and Ac mRNA also yielded larvae with liver-specific mRFP and EGFP expression (Figure 24), and two founders producing transgenic progeny with germline transmission of fabp10:mrfpegfp-lc3 were isolated (Figure 25). One of the founders (Line 2) was randomly selected for characterisation for its ability to measure autophagic flux under both starvation and chemical treatment. Starvation represents a cellular stressor that up-regulates autophagy, thus increases in both the number of red puncta corresponding to autolysosomes and yellow puncta corresponding to autophagosomes should be observed in the livers of transgenic larvae under starvation conditions. We observed increased numbers of red puncta, but we did not observe equivalent increases in the number of yellow puncta, which is contrary to what is expected (Figure 26). We hypothesised that what we observed may be due to increased rates of autophagosome-lysosome fusion and maturation into autolysosomes and increased rates of lysosomal degradation, leading to reduced numbers of autophagosomes observed. On the other hand, autophagy levels returning to basal levels after prolonged periods of starvation may also result in lesser number of autophagosomes compared to autolysosomes. Mizushima et al. using their GFP-LC3 transgenic mice showed that more GFP-LC3 puncta were observed in the livers following 24 hours starvation, whereas a lesser number of GFP-LC3 puncta were observed in the livers after 48 hours starvation (Mizushima et al., 2004). We hypothesised that zebrafish, similar to mice, limits the duration of autophagy upregulation to ensure that cellular integrity is not affected during starvation. This 134 Discussion limitation may also apply to other autophagy-inducing stimuli as well. To prove this, we intend to perform daily observations to find out if this applies to zebrafish larvae as well. We also subjected the transgenic larvae to identical chemical treatment conditions to find out if they can be used to measure autophagic flux. Identical conditions were applied to the transgenic larvae and we observed that transgenic larvae treated with everolimus alone displayed increased numbers of red puncta corresponding to autolysosomes without any concomitant increases in the number of yellow puncta (Figure 27). Again, we hypothesised that autophagy levels may have returned to basal levels after prolonged periods of induction, causing fewer autophagosomes to be observed, or that there was an increased rate of autophagosome-lysosome fusion and maturation, leading to reduced numbers of yellow puncta observed. We further observed that larvae treated with chloroquine alone displayed more yellow puncta compared to red puncta. This is indicative of blocked maturation of autophagosomes into autolysosomes. We also observed equivalent increases in the number of red and yellow puncta in transgenic larvae treated with a combination of both everolimus and chloroquine, indicative of increased autophagic flux. We intend to perform experiments involving shorter incubation periods with everolimus alone to find out the cause behind the reduced numbers of yellow puncta formation observed. 4. Establishing the role of autophagy in liver cancer Our lab successfully generated two transgenic lines with inducible oncogene over-expression, culminating in the development of liver tumours. In the Tg(fabp10:TA; TRE:Xmrk; krt4:GFP) transgenic line, Xmrk over-expression resulted 135 Discussion in the development of hepatocellular carcinoma (HCC), while cmyc over-expression in the Tg(fabp10:TA; TRE:cmyc; krt4:RFP) transgenic line resulted in the development of hepatocelluar adenoma (HA) (Figure 28). Thus we have on hand two different transgenic lines that will result in the development of different types of liver tumours and these two models are perfect to study the role of autophagy in these two tumours. These two transgenic models further allow us to study the interaction between Xmrk and cmyc over-expression and regulation of autophagy. As oncogene over-expression in both lines are inducible by the addition of doxycycline, both transgenic lines further allow us to study the course of tumour development and how autophagy may have been affected at different time points in the course of tumour development. Similarly, removal of doxycycline causes tumour regression in our liver cancer models, and we can also study the relationship between autophagy and tumour regression at the same time. Our early study revealed that activation of the TOR pathway may be chronic in Xmrk over-expressing fish as a result of Xmrk over-expression. We observed that transgenic fish over-expressing Xmrk had elevated levels of phosphorylated S6 (p-S6) protein, which serves as read out of TOR pathway activation, hence our conclusion that TOR pathway may be activated in transgenic fish over-expressing Xmrk (Figure 29). Xmrk, an orthologue of the EGF receptor, was found to activate intracellular signalling pathways such as the ras/raf/MEK/MAPK pathway, and can activate ERK (Li et al., 2011). It has also been reported that EGFR signalling will directly and indirectly activate the PI3K/Akt signalling pathway (Citri and Yarden, 2006) and that the PI3K/Akt signalling pathway is a central regulator of the TOR complex and therefore, coupled to the TOR pathway (Fabregat et al., 2007). Our results here demonstrated that Xmrk over-expression also has an activating effect on the 136 Discussion PI3K/Akt/TOR pathway, and this may have led to an inhibition on autophagy. We also observe elevated levels of LC3-I in the same samples with elevated levels of p-S6; however, we were unable to determine LC3-II levels as they seem to be non-existent in our immunoblotting experiments, as a result, we were unable to determine if autophagy has been inhibited in these same samples. This may be due to rapid turnover of LC3-II in vivo. Thus more thorough investigation needs to be carried out to validate if TOR signalling indeed has been activated and that autophagy is indeed been inhibited in Xmrk over-expressing transgenic fish. To determine if autophagy has been inhibited in these transgenic fish when the oncogenes were over-expressed, we crossed the newly generated Tg(fabp10:egfp-lc3) transgenic line with the two oncogene over-expressing transgenic lines to obtain progeny that will express EGFP-LC3 both with and without oncogene expression. We further separated them into groups with and without addition of doxycycline to induce liver tumour formation, while trying to dissect the autophagic flux via confocal microscopy. We observed severe overgrowth of liver tissue manifesting as enlarged bellies in fish treated with doxyxycline to induce Xmrk and cmyc over-expression. Xmrk over-expressing fish were found to develop increasingly larger liver tissue that was gradually discoloured, probably due to enhanced proliferation and increasing severity of the tumour developed (Figure 32). Accordingly, we observed increasingly disrupted cell morphology when observing under the confocal microscope, again corresponding to severity of the tumour. We did not find any EGFP-LC3 puncta developing in the hepatocytes during doxycycline induction, instead, we found EGFPLC3 to be freely distributed in the cytoplasm, suggesting that autophagy may have been inhibited. Upon doxycycline withdrawal, the size of the livers shrank to almost their normal size, and the colouration returned to normal as recovery proceeded. 137 Discussion EGFP-LC3 puncta re-emerged as recovery proceeded suggesting that autophagy may have been re-activated once Xmrk over-expression was repressed. Our lab found that apoptosis may have played an important role during tumour regression by eliminating HCC cells; we also hypothesised that the newly generated normal hepatocytes may have replaced these HCC cells and direct reversion of HCC cells to normal hepatocytes may have occurred concurrently during tumour regression (Li et al., 2011). We hypothesise that autophagy may have also been up-regulated during tumour regression by both processes, to either aid apoptosis or aid cellular differentiation to normal hepatocytes, although more studies are required to prove this. We next examined LC3-I, LC3-II, p62, EGFP-LC3 as well as free EGFP levels in the protein lysates (Figure 33). We observed increased amounts of EGFPLC3 in L+X+D+ protein lysates compared to controls (L+X+D-, L+X-D+ and L+XD-) throughout the entire experimental duration from tumour initiation through tumour regression except at the last time point (corresponding to three weeks after regression was induced). This observation may suggest that there may be an inhibition of autophagy flux, especially at the later stages of autophagy involving lysosomal degradation, therefore manifesting as significantly increased amounts of GFP-LC3 as observed in immunoblots. This increase in EGFP-LC3 was also evident from confocal microscopy imaging, where L+X+D+ livers displayed diffused intracellular GFP-LC3 distribution. The reduction in EGFP-LC3 levels three weeks after tumour regression was induced corroborated with the re-appearance of EGFP-LC3 puncta, suggesting the blockage of autophagic flux may have been relieved and that autophagy may have been involved in recovering from tumourigenesis and generation of normal hepatocytes. However, we did not observe any reduction in free EGFP in L+X+D+ lysates during tumour initiation and progression suggesting blocked autophagic 138 Discussion degradation of EGFP-LC3, thus we were unable to validate if there was indeed a blockage in autophagy caused by Xmrk over-expression. Increases in p62 levels are usually associated with blocked autophagic flux since p62 is a substrate for autophagy; however, we failed to detect any significant changes in p62 levels across samples and across time points to further verify if autophagy was indeed blocked. We also failed to detect LC3-II accurately in most of our samples, thus we could only hypothesise that autophagy may have been blocked during tumour initiation and progression. More work needs to be done in order to verify our hypothesis of autophagy being an anticancer mechanism and that it is inhibited during tumourigenesis. We also observed severe liver tissue overgrowth manifesting as enlarged bellies in cmyc over-expressing fish, although we did not detect any tissue discolouration, suggesting that the tumours that developed were not as malignant as those that arise from Xmrk over-expression, although the liver tissue overgrowth was due to hyper-proliferation of hepatocytes (Figure 34). This was confirmed by histopathological analyses where liver tumours that arise from cmyc over-expression were indeed HA (Figure 28). Confocal imaging did not reveal any differences in the intracellular distribution of EGFP-LC3 in the hepatoytes either, thus we were unable to determine the autophagy status in these tumours. We also carried out immunoblotting for several proteins, including EGFP, to determine autophagic flux in these fish (Figure 35). We observed increased p62 levels during tumour induction, highly suggestive of autophagy repression. Elevated p62 levels could be a driving force for tumourigenesis by enhancing oxidative stress and genomic instability as well as activating the non-canonical NF-κB pathway, suggesting that autophagy downregulation may be favoured during tumourigenesis in order to elevate p62 levels (Mathew et al., 2009). Accordingly we observed reduced p62 levels and increased 139 Discussion levels of free EGFP in cmyc over-expressing fish during recovery, suggesting autophagy inhibition may have been relieved to promote recovery. However, the role of autophagy during cellular recovery remains to be investigated; although we hypothesise that enhanced autophagy may serve to rid the cell of damaging superfluous protein aggregates thereby relieving cells of oxidative stress. It may also be involved in activating tumour-suppressing pathways as well as cellular remodelling or simply aiding apoptosis in riding tumourigenic hepatocytes. Immunoblotting for LC3-I and LC3-II were not very successful as we were unable to detect them accurately, thus we were unable to further furnish supporting evidence involving changes in LC3-II levels for this phenomenon. Furthermore, we found difficulty in detecting LC3-I and LC3-II levels in immunoblots, and we attributed this to the dynamic nature of the autophagic flux as a whole. We suggest adding chloroquine 24 hours prior to sacrificing the fish in an attempt to visualise and quantify LC3-II levels. We also hope to make use of the Tg(fabp10:mrfp-egfp-lc3) transgenic line to visualise the autophagic flux and hope that both transgenic lines will complement each other. 5. Future work and conclusions In this study we attempted to characterise autophagy in zebrafish and found that some autophagy genes like lc3 and atg5 were maternally deposited although no detectable autophagic activity was detected at the early stages of embryonic development. This presents an opportunity to study the relevance of these two autophagy genes during early embryonic development. This also provides an opportunity for us to learn if these proteins have other functions apart from being involved in autophagy. Our experiments further showed that autophagy inhibition via 140 Discussion chloroquine led to catastrophic consequences regardless of larvae being fed or starved, further proving that autophagy plays a very important role in cell survival whether nutrients are available, although we have to prove that larval deaths were due to defective cellular turnover and not due to chloroquine toxicity. We successfully generated and characterised the transgenic line Tg(fabp10:egfp-lc3) with liver-specific EGFP-LC3 fusion protein expression in order to visualise autophagic flux in real time as well as to provide supporting evidence for our immunoblotting findings. We further made use of this transgenic line to study the role of autophagy in liver tumourigenesis by crossing this transgenic line with existing transgenic lines with inducible oncogene over-expression. We found evidence that autophagy may be inhibited during tumourigenesis, regardless of which oncogene was being over-expressed, and that its inhibition may be relieved once oncogene overexpression ceased. To further prove our hypothesis we need to investigate LC3-I and LC3-II levels in more detail, as we were unable to detect with accuracy LC3-I and LC3-II levels in our samples. Our next step also entails further validation of p-S6 levels in an attempt to validate TOR signalling pathway activation, as well as to find out which pathways were activated and how these pathways cross-talk with the autophagy machinery to invoke autophagy inhibition during tumourigenesis. We also intend to manipulate autophagy during liver tumourigenesis using chemical inducers and inhibitors. To further prove our hypothesis that autophagy may be inhibited during liver carcinogenesis, we intend to induce autophagy by exposing the double transgenic fish to everolimus to find out if such an induction will lead to slower tumour initiation and progression. We also generated the transgenic line Tg(fabp10:mrfp-egfp-lc3) in order to visualise the autophagic flux more accurately. This is the first transgenic model 141 Discussion utilising differences in the propensity of EGFP and mRFP to fluoresce in acidic conditions in order to visualise the autophagic flux from autophagosome formation to maturation into autolysosomes. Although more in-depth characterisation is still required to validate the functionality of our model, this model will in the future serve the function of verifying our data on autophagy inhibition in our liver cancer models as well as on studying autophagy in various liver diseases. We believe that our study provides an impetus for more research to be carried out in this area and that our data shed more light on the role of autophagy in liver tumourigenesis. Moreover, our study provides initial clues that will help improve current therapeutic options by designing combinatorial chemotherapy for liver cancer in humans using autophagy inducers or inhibitors at the appropriate stages of treatment together with established chemotherapeutic drugs. 142 References References 143 References Amati, B. (2001). Integrating Myc and TGF-beta signalling in cell-cycle control. Nat Cell Biol 3, E112-113. Amati, B., Alevizopoulos, K., and Vlach, J. (1998). Myc and the cell cycle. Front Biosci 3, d250-268. Amatruda, J.F., and Patton, E.E. (2008). Genetic models of cancer in zebrafish. Int Rev Cell Mol Biol 271, 1-34. Amatruda, J.F., Shepard, J.L., Stern, H.M., and Zon, L.I. (2002). Zebrafish as a cancer model system. Cancer Cell 1, 229-231. Apel, A., Herr, I., Schwarz, H., Rodemann, H.P., and Mayer, A. (2008). Blocked autophagy sensitizes resistant carcinoma cells to radiation therapy. Cancer Res 68, 1485-1494. Avila, M.A., Berasain, C., Sangro, B., and Prieto, J. (2006). New therapies for hepatocellular carcinoma. Oncogene 25, 3866-3884. Bampton, E.T., Goemans, C.G., Niranjan, D., Mizushima, N., and Tolkovsky, A.M. (2005). The dynamics of autophagy visualized in live cells: from autophagosome formation to fusion with endo/lysosomes. Autophagy 1, 23-36. Barbazuk, W.B., Korf, I., Kadavi, C., Heyen, J., Tate, S., Wun, E., Bedell, J.A., McPherson, J.D., and Johnson, S.L. (2000). The syntenic relationship of the zebrafish and human genomes. Genome Res 10, 1351-1358. Berasain, C., Castillo, J., Prieto, J., and Avila, M.A. (2007). New molecular targets for hepatocellular carcinoma: the ErbB1 signaling system. Liver Int 27, 174-185. Berasain, C., Perugorria, M.J., Latasa, M.U., Castillo, J., Goni, S., Santamaria, M., Prieto, J., and Avila, M.A. (2009). The epidermal growth factor receptor: a link between inflammation and liver cancer. Exp Biol Med (Maywood) 234, 713-725. Berghmans, S., Murphey, R.D., Wienholds, E., Neuberg, D., Kutok, J.L., Fletcher, C.D., Morris, J.P., Liu, T.X., Schulte-Merker, S., Kanki, J.P., et al. (2005). tp53 mutant zebrafish develop malignant peripheral nerve sheath tumors. Proc Natl Acad Sci U S A 102, 407-412. Blackwood, E.M., and Eisenman, R.N. (1991). Max: a helix-loop-helix zipper protein that forms a sequence-specific DNA-binding complex with Myc. Science 251, 12111217. Blommaart, E.F., Krause, U., Schellens, J.P., Vreeling-Sindelarova, H., and Meijer, A.J. (1997). The phosphatidylinositol 3-kinase inhibitors wortmannin and LY294002 inhibit autophagy in isolated rat hepatocytes. Eur J Biochem 243, 240-246. Boya, P., Gonzalez-Polo, R.A., Casares, N., Perfettini, J.L., Dessen, P., Larochette, N., Metivier, D., Meley, D., Souquere, S., Yoshimori, T., et al. (2005). Inhibition of macroautophagy triggers apoptosis. Mol Cell Biol 25, 1025-1040. 144 References Brech, A., Ahlquist, T., Lothe, R.A., and Stenmark, H. (2009). Autophagy in tumour suppression and promotion. Mol Oncol 3, 366-375. Breuhahn, K., Longerich, T., and Schirmacher, P. (2006). Dysregulation of growth factor signaling in human hepatocellular carcinoma. Oncogene 25, 3787-3800. Calvisi, D.F., and Thorgeirsson, S.S. (2005). Molecular mechanisms of hepatocarcinogenesis in transgenic mouse models of liver cancer. Toxicol Pathol 33, 181-184. Castillo, J., Erroba, E., Perugorria, M.J., Santamaria, M., Lee, D.C., Prieto, J., Avila, M.A., and Berasain, C. (2006). Amphiregulin contributes to the transformed phenotype of human hepatocellular carcinoma cells. Cancer Res 66, 6129-6138. Chan, E.Y., Kir, S., and Tooze, S.A. (2007). siRNA screening of the kinome identifies ULK1 as a multidomain modulator of autophagy. J Biol Chem 282, 25464-25474. Chen, N., and Debnath, J. (2010). Autophagy and tumorigenesis. FEBS Lett 584, 1427-1435. Chen, S., Rehman, S.K., Zhang, W., Wen, A., Yao, L., and Zhang, J. (2010). Autophagy is a therapeutic target in anticancer drug resistance. Biochim Biophys Acta 1806, 220-229. Citri, A., and Yarden, Y. (2006). EGF-ERBB signalling: towards the systems level. Nat Rev Mol Cell Biol 7, 505-516. Corradetti, M.N., and Guan, K.L. (2006). Upstream of the mammalian target of rapamycin: do all roads pass through mTOR? Oncogene 25, 6347-6360. Corradetti, M.N., Inoki, K., Bardeesy, N., DePinho, R.A., and Guan, K.L. (2004). Regulation of the TSC pathway by LKB1: evidence of a molecular link between tuberous sclerosis complex and Peutz-Jeghers syndrome. Genes Dev 18, 1533-1538. Dang, C.V. (1999). c-Myc target genes involved in cell growth, apoptosis, and metabolism. Mol Cell Biol 19, 1-11. Dang, C.V., O'Donnell K, A., and Juopperi, T. (2005). The great MYC escape in tumorigenesis. Cancer Cell 8, 177-178. Davidson, A.E., Balciunas, D., Mohn, D., Shaffer, J., Hermanson, S., Sivasubbu, S., Cliff, M.P., Hackett, P.B., and Ekker, S.C. (2003). Efficient gene delivery and gene expression in zebrafish using the Sleeping Beauty transposon. Dev Biol 263, 191-202. Degenhardt, K., Mathew, R., Beaudoin, B., Bray, K., Anderson, D., Chen, G., Mukherjee, C., Shi, Y., Gelinas, C., Fan, Y., et al. (2006). Autophagy promotes tumor cell survival and restricts necrosis, inflammation, and tumorigenesis. Cancer Cell 10, 51-64. Ding, Z.B., Shi, Y.H., Zhou, J., Qiu, S.J., Xu, Y., Dai, Z., Shi, G.M., Wang, X.Y., Ke, A.W., Wu, B., et al. (2008). Association of autophagy defect with a malignant 145 References phenotype and poor prognosis of hepatocellular carcinoma. Cancer Res 68, 91679175. Dooley, K., and Zon, L.I. (2000). Zebrafish: a model system for the study of human disease. Curr Opin Genet Dev 10, 252-256. Driever, W., Solnica-Krezel, L., Schier, A.F., Neuhauss, S.C., Malicki, J., Stemple, D.L., Stainier, D.Y., Zwartkruis, F., Abdelilah, S., Rangini, Z., et al. (1996). A genetic screen for mutations affecting embryogenesis in zebrafish. Development 123, 37-46. Eilers, M. (1999). Control of cell proliferation by Myc family genes. Mol Cells 9, 1-6. Eilers, M., and Eisenman, R.N. (2008). Myc's broad reach. Genes Dev 22, 2755-2766. Emelyanov, A., Gao, Y., Naqvi, N.I., and Parinov, S. (2006). Trans-kingdom transposition of the maize dissociation element. Genetics 174, 1095-1104. Esclatine, A., Chaumorcel, M., and Codogno, P. (2009). Macroautophagy signaling and regulation. Curr Top Microbiol Immunol 335, 33-70. Fabregat, I., Roncero, C., and Fernandez, M. (2007). Survival and apoptosis: a dysregulated balance in liver cancer. Liver Int 27, 155-162. Feitsma, H., and Cuppen, E. (2008). Zebrafish as a cancer model. Mol Cancer Res 6, 685-694. Feng, H., Langenau, D.M., Madge, J.A., Quinkertz, A., Gutierrez, A., Neuberg, D.S., Kanki, J.P., and Look, A.T. (2007). Heat-shock induction of T-cell lymphoma/leukaemia in conditional Cre/lox-regulated transgenic zebrafish. Br J Haematol 138, 169-175. Fujita, N., Itoh, T., Omori, H., Fukuda, M., Noda, T., and Yoshimori, T. (2008). The Atg16L complex specifies the site of LC3 lipidation for membrane biogenesis in autophagy. Mol Biol Cell 19, 2092-2100. Fung, C., Lock, R., Gao, S., Salas, E., and Debnath, J. (2008). Induction of autophagy during extracellular matrix detachment promotes cell survival. Mol Biol Cell 19, 797806. Garnock-Jones, K.P., and Keating, G.M. (2009). Everolimus: in advanced renal cell carcinoma. Drugs 69, 2115-2124. Geisler, R., Rauch, G.J., Baier, H., van Bebber, F., Bross, L., Dekens, M.P., Finger, K., Fricke, C., Gates, M.A., Geiger, H., et al. (1999). A radiation hybrid map of the zebrafish genome. Nat Genet 23, 86-89. Giansanti, V., Torriglia, A., and Scovassi, A.I. (2011). Conversation between apoptosis and autophagy: "Is it your turn or mine?" Apoptosis 16, 321-333. 146 References Glucksmann, A. (1965). Cell death in normal development. Arch Biol (Liege) 76, 419-437. Gossen, M., and Bujard, H. (1992). Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci U S A 89, 5547-5551. Haffter, P., Granato, M., Brand, M., Mullins, M.C., Hammerschmidt, M., Kane, D.A., Odenthal, J., van Eeden, F.J., Jiang, Y.J., Heisenberg, C.P., et al. (1996). The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio. Development 123, 1-36. Hara, T., and Mizushima, N. (2009). Role of ULK-FIP200 complex in mammalian autophagy: FIP200, a counterpart of yeast Atg17? Autophagy 5, 85-87. Hara, T., Takamura, A., Kishi, C., Iemura, S., Natsume, T., Guan, J.L., and Mizushima, N. (2008). FIP200, a ULK-interacting protein, is required for autophagosome formation in mammalian cells. J Cell Biol 181, 497-510. He, C., Bartholomew, C.R., Zhou, W., and Klionsky, D.J. (2009). Assaying autophagic activity in transgenic GFP-Lc3 and GFP-Gabarap zebrafish embryos. Autophagy 5, 520-526. He, C., Song, H., Yorimitsu, T., Monastyrska, I., Yen, W.L., Legakis, J.E., and Klionsky, D.J. (2006). Recruitment of Atg9 to the preautophagosomal structure by Atg11 is essential for selective autophagy in budding yeast. J Cell Biol 175, 925-935. Hemelaar, J., Lelyveld, V.S., Kessler, B.M., and Ploegh, H.L. (2003). A single protease, Apg4B, is specific for the autophagy-related ubiquitin-like proteins GATE16, MAP1-LC3, GABARAP, and Apg8L. J Biol Chem 278, 51841-51850. Her, G.M., Chiang, C.C., Chen, W.Y., and Wu, J.L. (2003). In vivo studies of livertype fatty acid binding protein (L-FABP) gene expression in liver of transgenic zebrafish (Danio rerio). FEBS Lett 538, 125-133. Higashijima, S., Okamoto, H., Ueno, N., Hotta, Y., and Eguchi, G. (1997). Highfrequency generation of transgenic zebrafish which reliably express GFP in whole muscles or the whole body by using promoters of zebrafish origin. Dev Biol 192, 289299. Huang, X., Nguyen, A.T., Li, Z., Emelyanov, A., Parinov, S., and Gong, Z. (2011). One step forward: The use of transgenic zebrafish tumor model in drug screens. Birth Defects Res C Embryo Today 93, 173-181. Hukriede, N.A., Joly, L., Tsang, M., Miles, J., Tellis, P., Epstein, J.A., Barbazuk, W.B., Li, F.N., Paw, B., Postlethwait, J.H., et al. (1999). Radiation hybrid mapping of the zebrafish genome. Proc Natl Acad Sci U S A 96, 9745-9750. Ichimura, Y., Imamura, Y., Emoto, K., Umeda, M., Noda, T., and Ohsumi, Y. (2004). In vivo and in vitro reconstitution of Atg8 conjugation essential for autophagy. J Biol Chem 279, 40584-40592. 147 References Ichimura, Y., Kirisako, T., Takao, T., Satomi, Y., Shimonishi, Y., Ishihara, N., Mizushima, N., Tanida, I., Kominami, E., Ohsumi, M., et al. (2000). A ubiquitin-like system mediates protein lipidation. Nature 408, 488-492. Ivics, Z., Kaufman, C.D., Zayed, H., Miskey, C., Walisko, O., and Izsvak, Z. (2004). The Sleeping Beauty transposable element: evolution, regulation and genetic applications. Curr Issues Mol Biol 6, 43-55. Jacinto, E., Loewith, R., Schmidt, A., Lin, S., Ruegg, M.A., Hall, A., and Hall, M.N. (2004). Mammalian TOR complex 2 controls the actin cytoskeleton and is rapamycin insensitive. Nat Cell Biol 6, 1122-1128. Jung, C.H., Jun, C.B., Ro, S.H., Kim, Y.M., Otto, N.M., Cao, J., Kundu, M., and Kim, D.H. (2009). ULK-Atg13-FIP200 complexes mediate mTOR signaling to the autophagy machinery. Mol Biol Cell 20, 1992-2003. Kabeya, Y., Kamada, Y., Baba, M., Takikawa, H., Sasaki, M., and Ohsumi, Y. (2005). Atg17 functions in cooperation with Atg1 and Atg13 in yeast autophagy. Mol Biol Cell 16, 2544-2553. Kabeya, Y., Mizushima, N., Ueno, T., Yamamoto, A., Kirisako, T., Noda, T., Kominami, E., Ohsumi, Y., and Yoshimori, T. (2000). LC3, a mammalian homologue of yeast Apg8p, is localized in autophagosome membranes after processing. EMBO J 19, 5720-5728. Kabeya, Y., Mizushima, N., Yamamoto, A., Oshitani-Okamoto, S., Ohsumi, Y., and Yoshimori, T. (2004). LC3, GABARAP and GATE16 localize to autophagosomal membrane depending on form-II formation. J Cell Sci 117, 2805-2812. Kamada, Y., Funakoshi, T., Shintani, T., Nagano, K., Ohsumi, M., and Ohsumi, Y. (2000). Tor-mediated induction of autophagy via an Apg1 protein kinase complex. J Cell Biol 150, 1507-1513. Karantza-Wadsworth, V., Patel, S., Kravchuk, O., Chen, G., Mathew, R., Jin, S., and White, E. (2007). Autophagy mitigates metabolic stress and genome damage in mammary tumorigenesis. Genes Dev 21, 1621-1635. Kawakami, K., Shima, A., and Kawakami, N. (2000). Identification of a functional transposase of the Tol2 element, an Ac-like element from the Japanese medaka fish, and its transposition in the zebrafish germ lineage. Proc Natl Acad Sci U S A 97, 11403-11408. Kawakami, K., Takeda, H., Kawakami, N., Kobayashi, M., Matsuda, N., and Mishina, M. (2004). A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev Cell 7, 133-144. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., and Schilling, T.F. (1995). Stages of embryonic development of the zebrafish. Dev Dyn 203, 253-310. 148 References Kimura, S., Noda, T., and Yoshimori, T. (2007). Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy 3, 452-460. Kirisako, T., Baba, M., Ishihara, N., Miyazawa, K., Ohsumi, M., Yoshimori, T., Noda, T., and Ohsumi, Y. (1999). Formation process of autophagosome is traced with Apg8/Aut7p in yeast. J Cell Biol 147, 435-446. Kirisako, T., Ichimura, Y., Okada, H., Kabeya, Y., Mizushima, N., Yoshimori, T., Ohsumi, M., Takao, T., Noda, T., and Ohsumi, Y. (2000). The reversible modification regulates the membrane-binding state of Apg8/Aut7 essential for autophagy and the cytoplasm to vacuole targeting pathway. J Cell Biol 151, 263-276. Klionsky, D.J. (2005). The molecular machinery of autophagy: unanswered questions. J Cell Sci 118, 7-18. Klionsky, D.J. (2007). Autophagy: from phenomenology to molecular understanding in less than a decade. Nat Rev Mol Cell Biol 8, 931-937. Klionsky, D.J., Abeliovich, H., Agostinis, P., Agrawal, D.K., Aliev, G., Askew, D.S., Baba, M., Baehrecke, E.H., Bahr, B.A., Ballabio, A., et al. (2008). Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes. Autophagy 4, 151-175. Komatsu, M., and Ichimura, Y. (2010). Physiological significance of selective degradation of p62 by autophagy. FEBS Lett 584, 1374-1378. Kourtis, N., and Tavernarakis, N. (2009). Autophagy and cell death in model organisms. Cell Death Differ 16, 21-30. Kuma, A., Hatano, M., Matsui, M., Yamamoto, A., Nakaya, H., Yoshimori, T., Ohsumi, Y., Tokuhisa, T., and Mizushima, N. (2004). The role of autophagy during the early neonatal starvation period. Nature 432, 1032-1036. Kuma, A., Mizushima, N., Ishihara, N., and Ohsumi, Y. (2002). Formation of the approximately 350-kDa Apg12-Apg5.Apg16 multimeric complex, mediated by Apg16 oligomerization, is essential for autophagy in yeast. J Biol Chem 277, 1861918625. Lam, S.H., and Gong, Z. (2006). Modeling liver cancer using zebrafish: a comparative oncogenomics approach. Cell Cycle 5, 573-577. Lam, S.H., Wu, Y.L., Vega, V.B., Miller, L.D., Spitsbergen, J., Tong, Y., Zhan, H., Govindarajan, K.R., Lee, S., Mathavan, S., et al. (2006). Conservation of gene expression signatures between zebrafish and human liver tumors and tumor progression. Nat Biotechnol 24, 73-75. Langenau, D.M., Feng, H., Berghmans, S., Kanki, J.P., Kutok, J.L., and Look, A.T. (2005a). Cre/lox-regulated transgenic zebrafish model with conditional myc-induced T cell acute lymphoblastic leukemia. Proc Natl Acad Sci U S A 102, 6068-6073. 149 References Langenau, D.M., Jette, C., Berghmans, S., Palomero, T., Kanki, J.P., Kutok, J.L., and Look, A.T. (2005b). Suppression of apoptosis by bcl-2 overexpression in lymphoid cells of transgenic zebrafish. Blood 105, 3278-3285. Langenau, D.M., Keefe, M.D., Storer, N.Y., Guyon, J.R., Kutok, J.L., Le, X., Goessling, W., Neuberg, D.S., Kunkel, L.M., and Zon, L.I. (2007). Effects of RAS on the genesis of embryonal rhabdomyosarcoma. Genes Dev 21, 1382-1395. Langenau, D.M., Traver, D., Ferrando, A.A., Kutok, J.L., Aster, J.C., Kanki, J.P., Lin, S., Prochownik, E., Trede, N.S., Zon, L.I., et al. (2003). Myc-induced T cell leukemia in transgenic zebrafish. Science 299, 887-890. Law, J.M. (2001). Mechanistic considerations in small fish carcinogenicity testing. ILAR J 42, 274-284. Legakis, J.E., Yen, W.L., and Klionsky, D.J. (2007). A cycling protein complex required for selective autophagy. Autophagy 3, 422-432. Levine, B., and Kroemer, G. (2008). Autophagy in the pathogenesis of disease. Cell 132, 27-42. Levine, B., and Yuan, J. (2005). Autophagy in cell death: an innocent convict? J Clin Invest 115, 2679-2688. Li, Z., Huang, X., Zhan, H., Zeng, Z., Li, C., Spitsbergen, J.M., Meierjohann, S., Schartl, M., and Gong, Z. (2011). Inducible and repressable oncogene-addicted hepatocellular carcinoma in Tet-on xmrk transgenic zebrafish. J Hepatol. Liang, X.H., Jackson, S., Seaman, M., Brown, K., Kempkes, B., Hibshoosh, H., and Levine, B. (1999). Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature 402, 672-676. Lockshin, R.A., and Williams, C.M. (1965a). Programmed Cell Death--I. Cytology of Degeneration in the Intersegmental Muscles of the Pernyi Silkmoth. J Insect Physiol 11, 123-133. Lockshin, R.A., and Williams, C.M. (1965b). Programmed cell death. IV. The influence of drugs on the breakdown of the intersegmental muscles of silkmoths. J Insect Physiol 11, 803-809. Lockshin, R.A., and Williams, C.M. (1965c). Programmed cell death. V. Cytolytic enzymes in relation to the breakdown of the intersegmental muscles of silkmoths. J Insect Physiol 11, 831-844. Loewith, R., Jacinto, E., Wullschleger, S., Lorberg, A., Crespo, J.L., Bonenfant, D., Oppliger, W., Jenoe, P., and Hall, M.N. (2002). Two TOR complexes, only one of which is rapamycin sensitive, have distinct roles in cell growth control. Mol Cell 10, 457-468. 150 References Long, Q., Meng, A., Wang, H., Jessen, J.R., Farrell, M.J., and Lin, S. (1997). GATA1 expression pattern can be recapitulated in living transgenic zebrafish using GFP reporter gene. Development 124, 4105-4111. Longo, L., Platini, F., Scardino, A., Alabiso, O., Vasapollo, G., and Tessitore, L. (2008). Autophagy inhibition enhances anthocyanin-induced apoptosis in hepatocellular carcinoma. Mol Cancer Ther 7, 2476-2485. Lum, J.J., Bauer, D.E., Kong, M., Harris, M.H., Li, C., Lindsten, T., and Thompson, C.B. (2005). Growth factor regulation of autophagy and cell survival in the absence of apoptosis. Cell 120, 237-248. Maiuri, M.C., Le Toumelin, G., Criollo, A., Rain, J.C., Gautier, F., Juin, P., Tasdemir, E., Pierron, G., Troulinaki, K., Tavernarakis, N., et al. (2007). Functional and physical interaction between Bcl-X(L) and a BH3-like domain in Beclin-1. EMBO J 26, 2527-2539. Maiuri, M.C., Tasdemir, E., Criollo, A., Morselli, E., Vicencio, J.M., Carnuccio, R., and Kroemer, G. (2009). Control of autophagy by oncogenes and tumor suppressor genes. Cell Death Differ 16, 87-93. Marino, G., Salvador-Montoliu, N., Fueyo, A., Knecht, E., Mizushima, N., and Lopez-Otin, C. (2007). Tissue-specific autophagy alterations and increased tumorigenesis in mice deficient in Atg4C/autophagin-3. J Biol Chem 282, 1857318583. Mathew, R., Karantza-Wadsworth, V., and White, E. (2007a). Role of autophagy in cancer. Nat Rev Cancer 7, 961-967. Mathew, R., Karp, C.M., Beaudoin, B., Vuong, N., Chen, G., Chen, H.Y., Bray, K., Reddy, A., Bhanot, G., Gelinas, C., et al. (2009). Autophagy suppresses tumorigenesis through elimination of p62. Cell 137, 1062-1075. Mathew, R., Kongara, S., Beaudoin, B., Karp, C.M., Bray, K., Degenhardt, K., Chen, G., Jin, S., and White, E. (2007b). Autophagy suppresses tumor progression by limiting chromosomal instability. Genes Dev 21, 1367-1381. Matsuura, A., Tsukada, M., Wada, Y., and Ohsumi, Y. (1997). Apg1p, a novel protein kinase required for the autophagic process in Saccharomyces cerevisiae. Gene 192, 245-250. McCallum, C.M., Comai, L., Greene, E.A., and Henikoff, S. (2000). Targeting induced local lesions IN genomes (TILLING) for plant functional genomics. Plant Physiol 123, 439-442. McClintock, B. (1950). The origin and behavior of mutable loci in maize. Proc Natl Acad Sci U S A 36, 344-355. Meijer, A.J., and Codogno, P. (2011). Autophagy: regulation by energy sensing. Curr Biol 21, R227-229. 151 References Mizushima, N., Kuma, A., Kobayashi, Y., Yamamoto, A., Matsubae, M., Takao, T., Natsume, T., Ohsumi, Y., and Yoshimori, T. (2003). Mouse Apg16L, a novel WDrepeat protein, targets to the autophagic isolation membrane with the Apg12-Apg5 conjugate. J Cell Sci 116, 1679-1688. Mizushima, N., and Levine, B. (2010). Autophagy in mammalian development and differentiation. Nat Cell Biol 12, 823-830. Mizushima, N., Levine, B., Cuervo, A.M., and Klionsky, D.J. (2008). Autophagy fights disease through cellular self-digestion. Nature 451, 1069-1075. Mizushima, N., Noda, T., Yoshimori, T., Tanaka, Y., Ishii, T., George, M.D., Klionsky, D.J., Ohsumi, M., and Ohsumi, Y. (1998a). A protein conjugation system essential for autophagy. Nature 395, 395-398. Mizushima, N., Sugita, H., Yoshimori, T., and Ohsumi, Y. (1998b). A new protein conjugation system in human. The counterpart of the yeast Apg12p conjugation system essential for autophagy. J Biol Chem 273, 33889-33892. Mizushima, N., Yamamoto, A., Hatano, M., Kobayashi, Y., Kabeya, Y., Suzuki, K., Tokuhisa, T., Ohsumi, Y., and Yoshimori, T. (2001). Dissection of autophagosome formation using Apg5-deficient mouse embryonic stem cells. J Cell Biol 152, 657668. Mizushima, N., Yamamoto, A., Matsui, M., Yoshimori, T., and Ohsumi, Y. (2004). In vivo analysis of autophagy in response to nutrient starvation using transgenic mice expressing a fluorescent autophagosome marker. Mol Biol Cell 15, 1101-1111. Mizushima, N., Yoshimori, T., and Levine, B. (2010). Methods in mammalian autophagy research. Cell 140, 313-326. Mizushima, N., Yoshimori, T., and Ohsumi, Y. (2002). Mouse Apg10 as an Apg12conjugating enzyme: analysis by the conjugation-mediated yeast two-hybrid method. FEBS Lett 532, 450-454. Nakatogawa, H., Ichimura, Y., and Ohsumi, Y. (2007). Atg8, a ubiquitin-like protein required for autophagosome formation, mediates membrane tethering and hemifusion. Cell 130, 165-178. Nguyen, A.T., Emelyanov, A., Koh, C.H., Spitsbergen, J.M., Lam, S.H., Mathavan, S., Parinov, S., and Gong, Z. (2011). A high level of liver-specific expression of oncogenic KrasV12 drives robust liver tumorigenesis in transgenic zebrafish. Dis Model Mech. Noda, T., Kim, J., Huang, W.P., Baba, M., Tokunaga, C., Ohsumi, Y., and Klionsky, D.J. (2000). Apg9p/Cvt7p is an integral membrane protein required for transport vesicle formation in the Cvt and autophagy pathways. J Cell Biol 148, 465-480. Notte, A., Leclere, L., and Michiels, C. (2011). Autophagy as a mediator of chemotherapy-induced cell death in cancer. Biochem Pharmacol 82, 427-434. 152 References Obara, K., Sekito, T., and Ohsumi, Y. (2006). Assortment of phosphatidylinositol 3kinase complexes--Atg14p directs association of complex I to the preautophagosomal structure in Saccharomyces cerevisiae. Mol Biol Cell 17, 1527-1539. Panaretou, C., Domin, J., Cockcroft, S., and Waterfield, M.D. (1997). Characterization of p150, an adaptor protein for the human phosphatidylinositol (PtdIns) 3-kinase. Substrate presentation by phosphatidylinositol transfer protein to the p150.Ptdins 3-kinase complex. J Biol Chem 272, 2477-2485. Parinov, S., Kondrichin, I., Korzh, V., and Emelyanov, A. (2004). Tol2 transposonmediated enhancer trap to identify developmentally regulated zebrafish genes in vivo. Dev Dyn 231, 449-459. Patton, E.E., Widlund, H.R., Kutok, J.L., Kopani, K.R., Amatruda, J.F., Murphey, R.D., Berghmans, S., Mayhall, E.A., Traver, D., Fletcher, C.D., et al. (2005). BRAF mutations are sufficient to promote nevi formation and cooperate with p53 in the genesis of melanoma. Curr Biol 15, 249-254. Pelengaris, S., Khan, M., and Evan, G. (2002). c-MYC: more than just a matter of life and death. Nat Rev Cancer 2, 764-776. Perlmutter, D.H. (2009). Autophagic disposal of the aggregation-prone protein that causes liver inflammation and carcinogenesis in alpha-1-antitrypsin deficiency. Cell Death Differ 16, 39-45. Petiot, A., Ogier-Denis, E., Blommaart, E.F., Meijer, A.J., and Codogno, P. (2000). Distinct classes of phosphatidylinositol 3'-kinases are involved in signaling pathways that control macroautophagy in HT-29 cells. J Biol Chem 275, 992-998. Postlethwait, J.H., Yan, Y.L., Gates, M.A., Horne, S., Amores, A., Brownlie, A., Donovan, A., Egan, E.S., Force, A., Gong, Z., et al. (1998). Vertebrate genome evolution and the zebrafish gene map. Nat Genet 18, 345-349. Qu, X., Yu, J., Bhagat, G., Furuya, N., Hibshoosh, H., Troxel, A., Rosen, J., Eskelinen, E.L., Mizushima, N., Ohsumi, Y., et al. (2003). Promotion of tumorigenesis by heterozygous disruption of the beclin 1 autophagy gene. J Clin Invest 112, 1809-1820. Reggiori, F., Tucker, K.A., Stromhaug, P.E., and Klionsky, D.J. (2004). The Atg1Atg13 complex regulates Atg9 and Atg23 retrieval transport from the preautophagosomal structure. Dev Cell 6, 79-90. Rubinsztein, D.C., Gestwicki, J.E., Murphy, L.O., and Klionsky, D.J. (2007). Potential therapeutic applications of autophagy. Nat Rev Drug Discov 6, 304-312. Saunders, J.W., Jr. (1966). Death in embryonic systems. Science 154, 604-612. Scarlatti, F., Granata, R., Meijer, A.J., and Codogno, P. (2009). Does autophagy have a license to kill mammalian cells? Cell Death Differ 16, 12-20. 153 References Schmelzle, T., and Hall, M.N. (2000). TOR, a central controller of cell growth. Cell 103, 253-262. Schuler, W., Sedrani, R., Cottens, S., Haberlin, B., Schulz, M., Schuurman, H.J., Zenke, G., Zerwes, H.G., and Schreier, M.H. (1997). SDZ RAD, a new rapamycin derivative: pharmacological properties in vitro and in vivo. Transplantation 64, 36-42. Seglen, P.O., and Gordon, P.B. (1982). 3-Methyladenine: specific inhibitor of autophagic/lysosomal protein degradation in isolated rat hepatocytes. Proc Natl Acad Sci U S A 79, 1889-1892. Shen, H.M., and Codogno, P. (2011). Autophagic cell death: Loch Ness monster or endangered species? Autophagy 7, 457-465. Shepard, J.L., Amatruda, J.F., Finkelstein, D., Ziai, J., Finley, K.R., Stern, H.M., Chiang, K., Hersey, C., Barut, B., Freeman, J.L., et al. (2007). A mutation in separase causes genome instability and increased susceptibility to epithelial cancer. Genes Dev 21, 55-59. Shepard, J.L., Amatruda, J.F., Stern, H.M., Subramanian, A., Finkelstein, D., Ziai, J., Finley, K.R., Pfaff, K.L., Hersey, C., Zhou, Y., et al. (2005). A zebrafish bmyb mutation causes genome instability and increased cancer susceptibility. Proc Natl Acad Sci U S A 102, 13194-13199. Spencer, C.A., and Groudine, M. (1991). Control of c-myc regulation in normal and neoplastic cells. Adv Cancer Res 56, 1-48. Spitsbergen, J.M., Tsai, H.W., Reddy, A., Miller, T., Arbogast, D., Hendricks, J.D., and Bailey, G.S. (2000a). Neoplasia in zebrafish (Danio rerio) treated with 7,12dimethylbenz[a]anthracene by two exposure routes at different developmental stages. Toxicol Pathol 28, 705-715. Spitsbergen, J.M., Tsai, H.W., Reddy, A., Miller, T., Arbogast, D., Hendricks, J.D., and Bailey, G.S. (2000b). Neoplasia in zebrafish (Danio rerio) treated with N-methylN'-nitro-N-nitrosoguanidine by three exposure routes at different developmental stages. Toxicol Pathol 28, 716-725. Stack, J.H., DeWald, D.B., Takegawa, K., and Emr, S.D. (1995). Vesicle-mediated protein transport: regulatory interactions between the Vps15 protein kinase and the Vps34 PtdIns 3-kinase essential for protein sorting to the vacuole in yeast. J Cell Biol 129, 321-334. Stanton, M.F. (1965). Diethylnitrosamine-Induced Hepatic Degeneration and Neoplasia in the Aquarium Fish, Brachydanio Rerio. J Natl Cancer Inst 34, 117-130. Stern, H.M., and Zon, L.I. (2003). Cancer genetics and drug discovery in the zebrafish. Nat Rev Cancer 3, 533-539. Stoletov, K., and Klemke, R. (2008). Catch of the day: zebrafish as a human cancer model. Oncogene 27, 4509-4520. 154 References Stoletov, K., Montel, V., Lester, R.D., Gonias, S.L., and Klemke, R. (2007). Highresolution imaging of the dynamic tumor cell vascular interface in transparent zebrafish. Proc Natl Acad Sci U S A 104, 17406-17411. Takamura, A., Komatsu, M., Hara, T., Sakamoto, A., Kishi, C., Waguri, S., Eishi, Y., Hino, O., Tanaka, K., and Mizushima, N. (2011). Autophagy-deficient mice develop multiple liver tumors. Genes Dev 25, 795-800. Tanida, I., Sou, Y.S., Minematsu-Ikeguchi, N., Ueno, T., and Kominami, E. (2006). Atg8L/Apg8L is the fourth mammalian modifier of mammalian Atg8 conjugation mediated by human Atg4B, Atg7 and Atg3. FEBS J 273, 2553-2562. Teter, S.A., Eggerton, K.P., Scott, S.V., Kim, J., Fischer, A.M., and Klionsky, D.J. (2001). Degradation of lipid vesicles in the yeast vacuole requires function of Cvt17, a putative lipase. J Biol Chem 276, 2083-2087. Tsukamoto, S., Kuma, A., Murakami, M., Kishi, C., Yamamoto, A., and Mizushima, N. (2008). Autophagy is essential for preimplantation development of mouse embryos. Science 321, 117-120. Turcotte, S., Chan, D.A., Sutphin, P.D., Hay, M.P., Denny, W.A., and Giaccia, A.J. (2008). A molecule targeting VHL-deficient renal cell carcinoma that induces autophagy. Cancer Cell 14, 90-102. Urlinger, S., Baron, U., Thellmann, M., Hasan, M.T., Bujard, H., and Hillen, W. (2000). Exploring the sequence space for tetracycline-dependent transcriptional activators: novel mutations yield expanded range and sensitivity. Proc Natl Acad Sci U S A 97, 7963-7968. Vennstrom, B., Sheiness, D., Zabielski, J., and Bishop, J.M. (1982). Isolation and characterization of c-myc, a cellular homolog of the oncogene (v-myc) of avian myelocytomatosis virus strain 29. J Virol 42, 773-779. Volinia, S., Dhand, R., Vanhaesebroeck, B., MacDougall, L.K., Stein, R., Zvelebil, M.J., Domin, J., Panaretou, C., and Waterfield, M.D. (1995). A human phosphatidylinositol 3-kinase complex related to the yeast Vps34p-Vps15p protein sorting system. EMBO J 14, 3339-3348. Wang, C., Tao, W., Wang, Y., Bikow, J., Lu, B., Keating, A., Verma, S., Parker, T.G., Han, R., and Wen, X.Y. (2010). Rosuvastatin, identified from a zebrafish chemical genetic screen for antiangiogenic compounds, suppresses the growth of prostate cancer. Eur Urol 58, 418-426. Wang, C.W., and Klionsky, D.J. (2003). The molecular mechanism of autophagy. Mol Med 9, 65-76. Westerfield, M. (2000). The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio). 4th edn (University of Oregon Press, Eugene). 155 References White, E., and DiPaola, R.S. (2009). The double-edged sword of autophagy modulation in cancer. Clin Cancer Res 15, 5308-5316. Winnemoeller, D., Wellbrock, C., and Schartl, M. (2005). Activating mutations in the extracellular domain of the melanoma inducing receptor Xmrk are tumorigenic in vivo. Int J Cancer 117, 723-729. Wittbrodt, J., Adam, D., Malitschek, B., Maueler, W., Raulf, F., Telling, A., Robertson, S.M., and Schartl, M. (1989). Novel putative receptor tyrosine kinase encoded by the melanoma-inducing Tu locus in Xiphophorus. Nature 341, 415-421. Wittbrodt, J., Lammers, R., Malitschek, B., Ullrich, A., and Schartl, M. (1992). The Xmrk receptor tyrosine kinase is activated in Xiphophorus malignant melanoma. EMBO J 11, 4239-4246. Wixon, J. (2000). Featured organism: Danio rerio, the zebrafish. Yeast 17, 225-231. Xie, B.S., Zhao, H.C., Yao, S.K., Zhuo, D.X., Jin, B., Lv, D.C., Wu, C.L., Ma, D.L., Gao, C., Shu, X.M., et al. (2011). Autophagy inhibition enhances etoposide-induced cell death in human hepatoma G2 cells. Int J Mol Med 27, 599-606. Yan, J., Kuroyanagi, H., Kuroiwa, A., Matsuda, Y., Tokumitsu, H., Tomoda, T., Shirasawa, T., and Muramatsu, M. (1998). Identification of mouse ULK1, a novel protein kinase structurally related to C. elegans UNC-51. Biochem Biophys Res Commun 246, 222-227. Yang, Z., Huang, J., Geng, J., Nair, U., and Klionsky, D.J. (2006). Atg22 recycles amino acids to link the degradative and recycling functions of autophagy. Mol Biol Cell 17, 5094-5104. Yang, Z., and Klionsky, D.J. (2009). An overview of the molecular mechanism of autophagy. Curr Top Microbiol Immunol 335, 1-32. Yang, Z., and Klionsky, D.J. (2010). Eaten alive: a history of macroautophagy. Nat Cell Biol 12, 814-822. Yen, W.L., Legakis, J.E., Nair, U., and Klionsky, D.J. (2007). Atg27 is required for autophagy-dependent cycling of Atg9. Mol Biol Cell 18, 581-593. Yin, X.M., Ding, W.X., and Gao, W. (2008). Autophagy in the liver. Hepatology 47, 1773-1785. Yousefi, S., Perozzo, R., Schmid, I., Ziemiecki, A., Schaffner, T., Scapozza, L., Brunner, T., and Simon, H.U. (2006). Calpain-mediated cleavage of Atg5 switches autophagy to apoptosis. Nat Cell Biol 8, 1124-1132. Yue, Z., Jin, S., Yang, C., Levine, A.J., and Heintz, N. (2003). Beclin 1, an autophagy gene essential for early embryonic development, is a haploinsufficient tumor suppressor. Proc Natl Acad Sci U S A 100, 15077-15082. 156 References Zhang, J., Sattler, M., Tonon, G., Grabher, C., Lababidi, S., Zimmerhackl, A., Raab, M.S., Vallet, S., Zhou, Y., Cartron, M.A., et al. (2009). Targeting angiogenesis via a c-Myc/hypoxia-inducible factor-1alpha-dependent pathway in multiple myeloma. Cancer Res 69, 5082-5090. Zhivotovsky, B., and Orrenius, S. (2010). Cell cycle and cell death in disease: past, present and future. J Intern Med 268, 395-409. Zhou, F., Yang, Y., and Xing, D. (2011). Bcl-2 and Bcl-xL play important roles in the crosstalk between autophagy and apoptosis. FEBS J 278, 403-413. 157 [...]... for mutations that are present in a gene-ofinterest Progeny identified to carry the desired mutations are then propagated to maintain the line to study the function of the gene -of- interest Furthermore, fluorescent proteins can be expressed in specific organs in order to study their development over time by placing the gene encoding fluorescent proteins under the control of organ-specific promoters (Higashijima... pathways, extrinsic and intrinsic, with the extrinsic pathway activated via receptor binding of molecules belonging to the Tumour Necrosis Factor (TNF) family, and the intrinsic pathway activated by stimuli converging and acting on the mitochondria Both pathways culminate in the activation of caspases, which serves to degrade cellular components at the final step of apoptosis, following which the degraded... only the expression of the gene, but also the degree of expression by controlling Dox concentration 1.3.2 The use of transposable elements to generate transgenic zebrafish The standard method of generating transgenic zebrafish is via the injection of DNA constructs containing gene expression cassettes consisting of a promoter, a transgene and polyA signals into one-cell stage embryos in the hope that the. .. advances in sequencing technologies, the zebrafish genome sequence is also made available as well for positional cloning In spite of the large evolutionary distance between zebrafish and human, the syntenic relationship between their genomes further allow the use of the zebrafish in the identification of gene functions of unknown genes in both zebrafish and humans (Barbazuk et al., 2000) In all, these... to be active in zebrafish and human cells, eliminating the requirement for host cell-specific factors for transposition to occur Furthermore, our laboratory recently reported the successful use of the Ac/Ds transposable element in the generation of transgenic zebrafish expressing zebrafish kRASV12 under the control of the liver- specific fabp10 promoter in modelling oncogene-driven liver cancer (Her... stimulated during inflammation in the liver following injury and that tumourigenesis is favoured in the backdrop of inflammation Further studies implicated the over-expression of EGFR and its ligands in human HCC (Avila et al., 2006; Berasain et al., 2007; Breuhahn et al., 2006; Castillo et al., 2006) Thus it would be interesting to uncover the role chronic Xmrk over-expression plays in liver carcinogenesis... expression inducing vessel permeabilisation and vessel remodelling to aid the intravasation of cancer cells into the remodelled blood vessels (Stoletov et al., 2007) The technique of generating transgenic zebrafish with the ability to control the expression of various genes in the study of gene function can also be extended to the study of oncogenes and tumour suppressors Through the use of tissue-specific... driven liver cancer models Our laboratory has also successfully made use of the Tet-on system to generate transgenic zebrafish with inducible, liver- specific expression of oncogenes to model liver cancer In our zebrafish, a liver- specific fabp10 promoter was used to drive the constitutive expression of the rtTA transactivator, which will activate transcription of the oncogene found downstream of the minimal... found to harbour activating mutations in the extracellular domain, explaining its tumourigenicity in vitro and in vivo (Winnemoeller et al., 2005) The suitability of using Xmrk to drive liver carcinogenesis was further supported by studies implicating dysregulated EGFR signalling in human hepatocellular carcinoma or HCC (Berasain et al., 2009) It is noteworthy to point out that the EGFR pathway is chronically... attributes of the zebrafish has positioned it to overcome the shortcomings of mouse models as well as made it relevant in developmental biology and genetics studies 3 Introduction 1.2 The use of zebrafish as a model organism in cancer studies Fish has been used to study cancer since the early 1900s (Stern and Zon, 2003) Early studies involving Xiphophorus, or swordtail fish, ascertained the existence of an ... transgenic line Establishing the role of autophagy in liver cancer 4.1 Preliminary data implicating autophagy inhibition in liver cancer 4.2 Establishing double transgenic lines with liver- specific... as a model organism in the study of human diseases including cancers, we aim to investigate the role of autophagy in liver cancer using our established zebrafish liver cancer models As little study.. .ELUCIDATING THE ROLE OF AUTOPHAGY IN ZEBRAFISH MODELS OF LIVER CANCER SIM HUEY FEN, TINA (B.Sc.(Hons.)), NUS A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE DEPARTMENT OF BIOLOGICAL