Expression of Anti-Viral Shrna Molecules in Transgenic Zebrafish

DEAKIN UNIVERSITY

Expression of anti-viral shRNA molecules in transgenic zebraﬁsh

by Brian Clarke BSc(Hons)

A thesis submitted in fulﬁllment for the degree of Doctor of Philosophy

in the Molecular Medicine Research Faculity School of Medicine

April 2015

Acknowledgements

Work detailed in this dissertation was completed at the CSIRO Australian Animal Health Laboratory (AAHL) Geelong, Australia and at Deakin University School of Medicine, at the Molecular and Medical Research Strategic Research Centre. CSIRO Livestock Industries provided scholarships for this research.

I would like to express my sincere appreciation to my principal Deakin University supervisor Professor Alister Ward for his guidance, encouragement and friendship throughout the duration of this study. The support that I received from Alister was both admirable and irreplaceable, and I am truly thankful for all of his kindness, advice, and friendship throughout my Ph.D.

I am overwhelmingly grateful for the efforts of my supervisors at CSIRO AAHL. Dr. Ken McColl was instrumental in the planning and analysis of experimental data and in the critical reviewing of all writing associated with this study and I am grateful for his constant support, mentorship, and kindness. Professor Tim Doran offered continual support and mentorship throughout this study as well and provided an irreplaceable contribution to experimental design, shRNA design and significantly contributed to the development of my critical thinking and I am grateful for his continued friendship and kindness. Dr. Brian Meehan, whose hu- mour and support during my undergraduate degree first drew me to virology, I am forever grateful for your support, advice and friendship throughout my undergraduate and post-graduate studies “Thanks Mate”.

The assistance provided by the staff at AAHL and Deakin MMR SRC was out- standing. In particular, the efforts of Terry Wise, Scott Tyack, Kristy Jenkins, David Cummins, Kathleen McLachlen, Mark Tizard, Mark Crane, Clifford Leek, Monique Trengove, and Matthew Bruce were greatly appreciated.

Lastly, I would like to express my gratitude to my family and friends for their continued support and to Steven “Phyro” Wadeson for my continued sanity. In particular, I am forever grateful to Marg Moore and Alan Clarke for their uncon- ditional thoughtfulness, love, support, and advice. iv v Publications arising from this thesis

1. Clarke, BD., McColl, K., Doran, TJ., & Ward, AC. shRNA expression from pol III promoters prevents normal zebraﬁsh development. Transgene Research. 2015. In Preparation

2. Clarke, BD., McColl, K., Ward, AC., & Doran, TJ. Inhibition of viral hemorrhagic septicemia virus in Zebraﬁsh ZF-4 cells using short hairpin RNAs targeting the G and L genes. Anti-viral Research. 2015. In Preparation

3. Clarke, BD., Cummins, DM., McColl, K., Ward, AC., & Doran, TJ. Char- acterisation of zebraﬁsh polymerase III promoters for the expression of short- hairpin RNA interference molecules. Zebraﬁsh. 2013. doi:10.1089/zeb.2012.0782.

4. McColl, K., Clarke, BD., & Doran, TJ. Genetically engineered animals – the arguments for and against. Australian Veterinary Journal. 2013.

5. Prabandono, K., McColl, KA., Clarke, BD., Barnes, A., Sunarto, A., Pal- lister, J., & Doran, TJ. The role of RNA interference in KF-1 cell lines against cyprinid herpesvirus 3 (CyHV-3). Anti-viral Research. 2014. Submitted “So long and thanks for all the ﬁsh.”

Douglas Adams. Hitchhikers guide to the galaxy. DEAKIN UNIVERSITY

Abstract

Molecular Medicine Research Faculity School of Medicine

Doctor of Philosophy

by Brian Clarke BSc(Hons)

Aquaculture provides a significant proportion of the global food supply. Ma- rine fish aquaculture, like intensive farming of many other species, is vulnerable to significant losses due to pathogens, particularly viral infections. Novirhab- doviruses, such as Viral Haemorrhagic Septicaemia virus (VHS virus) are important pathogens of cold water fishes, and to date there are no effective vaccines or other treatments. Quarantine and physical separation are the only effective preventative measures available against VHS virus.

Since the discovery of RNA interference (RNAi) in 1998, there has been significant development of RNAi as an antiviral treatment. In particular, short-hairpin RNA (shRNA) and RNAi-based transgenesis has been demonstrated to effectively reduce viral titres in VHS virus-challenged fish. Additionally, cassettes and vectors that produce multiple small interfering RNA can potentially target positions across the entire viral transcript providing a useful strategy to inhibit viral escape. The zebrafish is an important model for aquaculture as it is the only teleost species to have a complete and well-annotated genome sequence available. This makes the zebrafish suitable for testing an anti-viral strategy for future application in aquaculture.

The work in Chapter 3, identified and isolated a functional zebrafish RNA polymerase III promoter, H1 , and a new U6 (U6-4) promoter, along with a variant of the latter. shRNAs were be expressed from these, and three other previously characterised zebrafish U6, promoters in a zebrafish cell line (ZF-4), but not in the mammalian West African monkey kidney (Vero) cells. The converse was also demonstrated, ie, two control polymerase III promoters from mice and chicken allowing expression of shRNAs in Vero cells but not the ZF-4 cell line. This Chapter also showed that the newly-identified U6 promoter showed similar levels of activity to previously characterised zebrafish U6 promoters, and that the H1 promoter showed a lower level of activity as measured by activity of an enhanced green fluorescent protein (EGFP) reporter. Finally, it was demonstrated that species-matched promoters, but not necessarily homologous promoters, are required for effective expression of, and knockdown by, shRNAs. This supports previous evidence that there are inherent differences between mammalian and avian U6 promoters compared with their teleost counterparts.

Chapter 4 details the design and evaluation of VHS virus-specific shRNAs, targeting 21 nucleotide sequences of either the VHS virus glycoprotein (G) gene or the VHS virus polymerase (L) gene. To test their function, shRNAs were expressed in ZF-4 cells utilising the U6-2 promoter (identified as the strongest), and the cells were then inoculated with VHS virus at Multiplicity of Infections (MOI) between 101 -10−5 MOI. Five of the six shRNA molecules successfully reduced VHS virus replication between 2 and 4 logs in titre relative to an irrelevant control shRNA at all MOIs and also reduced viral mRNA as measured by RT-qPCR. The shRNA molecule that did not induce RNAi silencing formed an improper hairpin when analysed in silico providing an explanation for its observed lack of activity. To ensure that observed reductions in viral titre were dependent on shRNA silencing and not interferon induction induced non-specifically by the dsRNA shRNA molecules, the interferon response was analysed by RT-qPCR. Only the defective shRNA induced an interferon response as measured by fold change of both zebrafish interferon and Mx.

Chapter 5 describes the generation of zebrafish harbouring potential anti-VHS virus transgenes. However, following microinjection of single-cell embryos with Tol2 constructs expressing shRNAs a significant dose-dependent decrease in the number of viable zebrafish embryos, and an increase in developmental defects was observed, This was associated with a significant decrease in the levels of native microRNAs, leading to the hypothesis that these changes were due to the introduction of excessive shRNA molecules, above the level sustainable by the endogenous miRNA machinery. Over-expression of human Exportin-5 (hXpo-5) effectively increased the toxicity of the shRNA vectors, but this effect was rescued by human Argonaute 2 (hAgo2) co-injection, leading to the hypothesis that the Xpo5-mediated export mechanism may act as a brake on the levels of silencing RNAs in the cytoplasm. Decreasing the dose of shRNAs by rational deletion of elements of polymerase III promoters effectively rescued both embryo viability and normal development of the zebrafish. Utilisation of the H1 and modulated U6 promoters did eventually generate fertile adult transgenic zebrafish providing evidence that the approach was sound.

The results presented in this thesis further support the hypothesis that efficient and safe use of RNAi, either directly as a therapeutic or indirectly as a virus-specific transgene, is dependent on (i) effective and specific shRNA design to avoid IFN- type responses or partial or complete binding of siRNA RISC to unintended mRNA targets, and (ii) the dosage of interfering molecules which should be carefully controlled to limit the effects on the native miRNA pathway. Contents

Access to Thesis ii

Declaration iii

Acknowledgements iv

Publication List vi

Abstract viii

List of Figures xvii

List of Tables xix

Abbreviations xxi

1 Literature review 1 1.1 General introduction ...... 1 1.2 The zebrafish model ...... 3 1.3 Viral diseases of zebrafish ...... 6 1.3.1 The zebrafish as a model for viral infection ...... 7 1.4 Viral haemorrhagic septicaemia ...... 9 1.5 RNA interference (RNAi) ...... 12 1.5.1 The RNAi pathway and history of RNAi-induced silencing .13 1.5.2 The endogenous mechanisms of RNAi ...... 18 1.6 Synthetic inducers of the RNAi pathway ...... 21 1.6.1 siRNA ...... 21 1.6.1.1 siRNA delivery in fish ...... 22 1.6.2 Polymerase III promoters for shRNA expression ...... 24 1.6.3 Off-target effects of synthetic RNAi in fish ...... 28 1.7 Saturation of the endogenous microRNA pathway by exogenous dsRNA molecules ...... 29 1.7.1 Saturation of Argonaute ...... 31

xi Contents xii

1.7.2 Viral saturation and inhibition of the RNAi pathway .... 33 1.7.2.1 Antiviral RNAi in ﬁsh ...... 34 1.8 Research rationale and summary ...... 36

2 Materials and Methods 39 2.1 Materials ...... 39 2.1.1 Suppliers ...... 39 2.1.2 Bacterial strains, plasmids and media ...... 41 2.1.3 Primers, probes and oligonucleotides ...... 41 2.2 DNA methods...... 42 2.2.1 Transformation of E. coli ...... 42 2.2.1.1 Using electrocompetent E. coli ...... 42 2.2.1.2 Using chemically competent E. coli ...... 42 2.2.2 Plasmid purification ...... 43 2.2.2.1 Small-volume isolation of plasmid DNA from E. coli 43 2.2.2.2 Large-volume isolation of plasmid DNA from E. coli 43 2.2.3 Nucleic acid manipulation ...... 43 2.2.3.1 Restriction endonuclease digestion ...... 43 2.2.3.2 Manipulation of restriction endonuclease digestion fragments ...... 43 2.2.3.3 Separation of nucleic acid fragments by agarose gel electrophoresis ...... 44 2.2.3.4 Extraction of DNA from agarose gels ...... 44 2.2.3.5 DNA ligations ...... 45 2.2.4 Polymerase Chain Reaction amplification of DNA ...... 45 2.2.5 Cloning of PCR products ...... 45 2.2.6 PCR cleanup ...... 46 2.2.7 DNA sequencing ...... 46 2.2.8 Determination of nucleic acid concentration ...... 47 2.2.9 Tol 2 insertion site analysis PCR ...... 47 2.2.9.1 Genomic DNA (gDNA) extraction ...... 47 2.2.9.2 Cassette Method ...... 47 2.3 RNA Methods ...... 48 2.3.1 RNA isolation and purification ...... 48 2.3.1.1 Total RNA extraction from cells or zebrafish optimised for small RNAs using TrizolTM ...... 48 TM 2.3.1.2 Total RNA extraction using RNeasy ...... 49 2.3.2 Extraction of viral RNA ...... 49 2.4 RNase protection assay (RPA) ...... 49 2.4.1 Radioactive labeling of probe and marker RNA ...... 49 2.4.2 Separation of RNA on denaturing agarose gels and autoradiography ...... 50 2.4.3 Quantitation of RNA ...... 51 Contents xiii

2.4.3.1 DNase treatment of RNA samples for cDNA synthesis...... 51 2.4.3.2 First-strand cDNA synthesis ...... 51 2.4.3.3 Quantitative PCR of shRNA and microRNA expression ...... 52 2.4.4 Quantitative reverse-transcriptase analysis of mRNA expression ...... 52 2.4.5 Data analysis for qRT-PCR ...... 53 2.4.6 shRNA design and analysis ...... 54 2.4.6.1 shRNA design and construction ...... 54 2.5 Tissue culture methods ...... 55 2.5.1 Immortalised eukaryotic cultured cell lines ...... 55 2.5.2 Harvesting adherent cell lines ...... 55 2.5.3 Transfection of eukaryotic cells ...... 56 2.5.3.1 Electroporation of plasmid DNA ...... 56 2.5.3.2 Chemical transfection of plasmid DNA ...... 56 2.5.4 Flow cytometry ...... 56 2.6 Virological methods ...... 57 2.6.1 Viruses and infections ...... 57 2.6.2 TCID50 determination ...... 57 2.7 Zebrafish methods ...... 58 2.7.1 General zebrafish care and breeding ...... 58 2.7.2 Microinjection of zebrafish ...... 58 2.7.3 Fluorescent microscopy ...... 59 2.7.4 Fixation of zebrafish embryos and larvae ...... 59 2.7.5 Whole-mount in-situ hybridisation of zebrafish embryos (WISH) 59 2.7.5.1 Digoxigen-labelling of nucleic acids ...... 59 2.7.5.2 Hybridisation and Staining ...... 60 2.8 Protein methods ...... 61 2.8.1 Western blot analysis ...... 61 2.8.1.1 Sample collection and preparation ...... 61 2.8.1.2 Bradford Assay ...... 61 2.9 Bioinformatics ...... 62 2.9.1 Data mining ...... 62 2.9.2 Nucleotide sequence data management ...... 62 2.9.3 Image Analysis ...... 62

3 Expression of shRNA molecules in zebrafish ZF-4 cell culture 65 3.1 Introduction ...... 65 3.2 Characterisation of and identification of zebrafish U6 and H1 snRNA promoters for the expression of shRNAs ...... 68 3.2.1 U6 promoter identification ...... 68 3.2.2 H1 Promoter identification ...... 69 3.2.3 shRNA expression vector construction ...... 72 Contents xiv

3.2.4 Optimisation of plasmid DNA transfection in zebrafish ZF-4 cells ...... 74 3.2.5 The zebrafish U6 and H1 promoters express shRNAs .... 79 3.2.6 Zebrafish polymerase III promoters can decrease eGFP expression in native zebrafish cells but not mammalian Vero cells ...... 81 3.3 Utilisation of the chicken Pank-1 Intron 5 for the delivery of shRNAs in zebrafish ZF-4 cells ...... 83 3.3.1 Splicing analysis of Pank 1 Intron 5 and delivery of shRNAs into ZF-4 cells ...... 87 3.4 Discussion ...... 89

4 Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 93 4.1 Introduction ...... 93 4.2 shRNA design and construction ...... 95 4.2.1 shRNA screening ...... 96 4.3 Growth of VHS virus in ZF-4 cells ...... 100 4.3.1 Viral haemorrhagic septicaemia virus growth kinetics in ZF- 4 cells...... 100 4.3.2 The effect of nucleofection on VHS virus growth in zebrafish cells ...... 101 4.4 shRNA mediated VHS virus suppression ...... 105 4.4.1 Transfection of shRNA plasmids reduces VHS virus replication105 4.4.2 Effect of shRNA expression on MxA and IFNφ1 expression. 106 4.5 Intron-mediated knockdown of VHS virus in vitro...... 111 4.6 Discussion ...... 114

5 Creation and characterisation of transgenic zebrafish 117 5.1 Introduction ...... 117 5.2 Construction and characterisation of Tol-2 expression plasmids ...121 5.3 Full strength pol III promoters lead to high levels of mutation and mortality in zebrafish...... 123 5.3.1 Microinjection of zebrafish embryos ...... 123 5.3.2 shRNA overexpression increases mortality and developmental defects in zebrafish larvae ...... 123 5.3.3 shRNA over-expression inhibits zebrafish development ....124 5.3.4 shRNA overexpression causes increased mortality and developmental abnormalities in zebrafish in a sequence independent manner ...... 129 5.3.5 Mortalities and developmental defects induced by shRNA overexpression is not dependent on p53 ...... 130 5.3.6 shRNA overexpression inhibits miRNA expression in developing zebrafish ...... 132 5.4 Modification of pol III promoters to modulate shRNA expression . . 135 Contents xv

5.5 Over-expression of human Argonaute 2 and Exportin-5 in zebraﬁsh larvae partially rescues developmental abnormalities and mortality from over-expression of shRNA molecules ...... 137 5.6 Discussion ...... 143

6 General discussion 151 6.1 Overview ...... 151 6.2 Recommended future directions ...... 157 6.2.1 shRNA expression ...... 157 6.2.2 Tol2 and transgenic production ...... 160 6.2.3 Multi-shRNA cassettes ...... 161 6.2.4 The zebrafish model ...... 163 6.2.5 Specific recommendations for in vivo shRNA expression in zebrafish ...... 164

A Synthetic Oligonuclotides used in this study 217

B Plasmids used or generated in this study 223

C Accession number of VHS virus glycoprotien and polymerase used in sequence conservation analysis 227

List of Figures

1.1 Structure of VHS virus ...... 10 1.2 VHS virus genome ...... 11 1.3 The natural inducers of the RNAi pathway ...... 16 1.4 The structure of RNA polymerase III promoters ...... 26

3.1 Induction of RNAi response from polymerase III promoters or intronic vectors ...... 67 3.2 Polymerase III promoter element alignment ...... 70 3.3 Alignment of Zebrafish U6 snRNA promoters ...... 72 3.4 The RPPH1 ribonuclease P RNA component H1 gene and promoter 73 3.5 Cloning and construction of shRNA expression vectors ...... 75 3.6 Alignment of zebrafish U6 4 variant promoters ...... 76 3.7 Transfection of ZF-4 Cells ...... 78 3.8 RNase Protection Assay for sheGFP ...... 80 3.9 shRNA knockdown of eGFP using zebrafish polymerase III promoters in mammalian Vero cells...... 82 3.10 shRNA knockdown of eGFP using zebrafish polymerase III promoters in ZF4 Cells...... 84 3.11 ClustalW2 pairwise alignment of miR107 ...... 86 3.12 Transfection of chicken Pank 1 intron in ZF4 cells and shRNA and GFP expression...... 89

4.1 Relative conservation of shRNA targets...... 97 4.2 Structural organisation of the VHS virus genome and location of shRNAs...... 98 4.3 Targets of shRNAs on the VHS virus mRNA secondary structure .99 4.4 VHS virus replication in ZF-4 cells ...... 102 4.5 Preliminary infection of ZF-4 cells with VHS virus ...... 103 4.6 Zebrafish ZF-4 transfection with pEGFP-Max and infection with VHS virus ...... 104 4.7 Cytopathic effect of VHS virus on ZF-4 cells is inhibited by specific shRNA transfection ...... 107 4.8 shRNA molecules targeting VHS virus glycoprotein (G) gene inhibit VHS virus replication ...... 108 4.9 shRNA molecules targeting VHS virus polymerase (L) gene have a range of effects on VHS virus replication ...... 109

xvii List of Figures xviii

4.10 The eﬀect of poly(I:C) on VHS virus replication in ZF-4 cells ....110 4.11 The eﬀect of poly(I:C) and shRNA expression on IFN and MxA in ZF-4 cells at 15 ◦C ...... 112 4.12 Pank1 intron is not a suitable method for introducing shRNA expression in ZF-4 cells at 15 ◦C ...... 114

5.1 Tol-2 transposable elements and basic shRNA vector ...... 119 5.2 Transgenesis in zebrafish ...... 121 5.3 Construction and validation of pTol2-EF1α:GFP-pol III:shRNA expression plasmids ...... 123 5.4 Expression of shRNAs from pol III type 3 promoters results in reduced survival and increased developmental defects in zebrafish larvae and adults ...... 125 5.5 Expression of shRNAs in developing zebrafish increases overt morphological abnormalities ...... 127 5.6 shRNA overexpression inhibits zebrafish development ...... 129 5.7 shRNA induced developmental defects are independent of shRNA sequence ...... 131 5.8 Over-expression of shRNA does not induce p53 related developmental defects in developing zebrafish ...... 133 5.9 Over-expression of shRNA reduces miRNA levels in developing zebrafish ...... 134 5.10 Summary of the modification U6 and H1 promoters ...... 136 5.11 Relative quantitation of shRNA expression following promoter modification ...... 138 5.12 Alignment of human and zebrafish Argonaute 2 and Exportin 5 proteins ...... 141 5.13 Characterisation of hXpo5 and hAgo2 over-expression in zebrafish larvae ...... 143 5.14 Expression of hAgo2 and hXpo5 can reduce mortality and developmental abnormalities in zebrafish larvae expressing shRNA molecules145 5.15 Expression of hAgo2 and hXpo5 restores normal miRNA expression following shRNA overexpression ...... 147

6.1 Summary of the modiﬁed U6 and H1 promoters ...... 156 6.2 Summary of the RNAi pathway, including potential external medi- ators of saturation of individual pathway components ...... 158 List of Tables

1.1 Developmental stages of the zebraﬁsh ...... 5 1.2 Viruses that have an established zebraﬁsh model ...... 8

2.1 Typical PCR conditions ...... 46 2.2 PCR conditions for insertion site characterisation ...... 48 2.3 PCR conditions for miRNA / shRNA quantitation ...... 52 2.4 Fold Change calculations ...... 53 2.5 Bradford assay dilutions ...... 62 2.6 Image J Batch processing ...... 63

3.1 Clustal W2 analysis of zebraﬁsh U6 promoters ...... 69

4.1 shRNA Sequences ...... 96 4.2 Sequence of shRNA Scramble and G0906 ...... 114

xix

Abbreviations

AAHL Australian Animal Health Laboratory Ago & Ago2 Argonaute (2) ATCC American Type Culture Collection ATP Adenosine triphosphate AVONA Analysis of variance BF-2 Blue Fin-2 Cells bp base pair BSA Bovine Serum Albumin CHIKV Chikungunya virus CHS Chalcone synthase CHSE Chinook Salmon Epithelial Cells CLIP Cross-linked immuno-precipitation CMV Cytomegalovirus CPE Cytopathic Eﬀect CRISPR Clustered regularly interspaced short palindromic repeats CSIRO Commonwealth Scientiﬁc & Industrial Research Organisation DdsRNA Dicer substrate double stranded RNA DIG Digoxigen dpf days post fertilisation dpi days post infection dsRNA double stranded RNA DSE Distal Sequence Element EDTA Ethylene Diamine Tetraacetic Acid EPC Epithelioma Papulosum Cyprini xxi Abbreviations xxii

FCS Foetal Calf Serum FDA Food & Drug Administration FHV Flock House Virus flt floating head GGlycoprotein gDNA genomic DNA GFP Green Fluorescent Protein GGNNV Greasy Grouper Nervous Necrosis Virus hAgo2 human Argonaute 2 HCV Hepatitis Cvirus HEK293 Human embryonic kidney cells HeLa Henrietta Lacks cells hpi hours post infection HIV-1 Human immunodeficiency virus type-1 HPLC High Performance Liquid Chromatography HTA Head to Trunk Angle hXpo5 Human Exportin 5 IFN Interferon IHNV infectious hematopoietic necrosis virus IPNV Infectious pancreatic necrosis virus ISG Interferon Stimulated Gene ISKNV Infectious spleen and kidney necrosis virus L Polymerase protein LDV Lymphocystis disease virus LNA Locked Nucleic Acid NNucleoprotein NBT NitroBlue Tetrazolium ncRNA non coding RNA

nfH2Onuclease free H2O NNV Nervous necrosis virus noi No isthmus Abbreviations xxiii

ntl No tail NV Non Virion MMatrix protein MCP Methyl-accepting chemotaxis protein MDA5 Melanoma Differentiation-Associated protein 5 MFI Mean Flourescence Intensity miRNA micro RNA MOI Multiplicity of Intesity mRNA messenger RNA OCT Octamer OiE World organisation for Animal Health PPhosphoprotein PAM-URP Poly-Adenylated Modified-Universal Reverse Pank 1 Pantothenate Kinase 1 PAR-CLIP Photoactivatable ribonucleoside- enhanced CLIP PAZ Piwi; Argonaute and ZwilePinhead P-Body Processing Body PB Influenza Polymerase Basic protein PBS Phosphate-buffered saline PCR Polymerase Chain Reaction PFA Paraformaldehyde PFV-1 Primate foamy virus type-1 piRNA Piwi-interacting RNA PKR Protein kinase RNA-activated Pol Polymerase PSE Proximal Sequence Element PTU 1-phenyl-2-thio-urea qPCR quantitative PCR Rig-I Retinoic acid-inducible gene 1 RISC RNA iinduced silencing complex RLR RNA- like receptors Abbreviations xxiv

RNAi RNA interference RPM Revolutions per minute RT-PCR Reverse Transcriptase-Polymerase Chain Reaction SEM Standard Error of the Mean shRNA short hairpin RNA SHRV snakehead rhabdovirus siRNA small interfering RNA SJNNV Striped Jack nervous necrosis virus SNAPC snRNA activating protein complex SPH SphI post-octamer homology StdDev Standard Deviation SV40 Simian Virus 40 SVCV Spring Viraemia of Carp virus TALE Transcription activator-like eﬀector nuclease TCID Tissue Culture Infectious Dose TFIIIB Transcription factor IIIB TFV Tiger Frog Virus tiRNA Transcription initiation RNA TLR Toll like receptor T(L / R)E Tol2 Left / Right Element US United States VA V irus Associated VHS Viral Hemorrhagic Septicemia wpf weeks post fertilisation WT Wild Type Xpo5 Exportin 5 ZF-4 Zebrafish embryo epithelial cells ZIFN Zebrafish Information Network Chapter 1

Literature review

1.1 General introduction

Aquaculture, both wild caught and farmed, is the fastest growing food industry globally, producing between 157-175 million tonnes annually, which is estimated to provide about 16% of total protein consumed worldwide. The majority of this production comes from wild caught fish reserves; however, the proportion of farmed fish is rapidly increasing, particularly in Asia. The annual production increase in aquaculture products has remained steady at around 7% per annum since 1994[98,398]. However, even with this growth, farmed aquaculture still only amounts to less than a quarter of the total fisheries harvest. Farmed fish stocks will only become more and more important in the coming years as many wild fish stocks are considered to be overexploited[98,234,264] or declining[283]. Indeed, the use of so-called “super-trawlers” has been the subject of significant debate and legislation in the Australian Parliament within the last 12 months because of this[275].

By further increasing aquaculture, wild schools and stocks can be preserved and protected whilst also providing the necessary food resources for increasing populations. However, before aquaculture can continue to expand to meet current demand a number of hurdles, both technical and social, must be overcome. The

1 Chapter 1. Introduction 2 nature of fish farms, like many high density animal farms, makes them susceptible to many pathogens, such as the parasite Gyrodactylus salaris affecting salmonoid species[166], a number of significant bacterial agents, particularly the potentially zoonotic Vibrio parahaemolyticus and vulnificus,[19] the fish viruses koi herpesvirus[120,355] and spring viraemia of carp virus (SVCV)[116,120],andsig- nificantly for this project viral haemorrhagic septicaemia virus (VHS virus) of fish[141,233,337].

The use of transgenic animals to improve productivity, as well as animal and human health, is hotly debated and polarising within both the scientiﬁc community and the lay population. The press and popular response to the AquAdvan- tageTMgrowth hormone-enhanced salmon developed by AquaBounty provides a direct example of this as the US FDA invitation for comment on the approval of these salmon for production in the USA provoked a great deal of debate (for example as described in Ledford et al. 2013[200]). However, the use of a minimal transgene that induces a response which,results in improvements in animal or human welfare and food security may prove more palatable than a transgenic where productivity and proﬁtability are the major motivating factors[105,241,398].

The zebrafish is a model organism of unique benefit for the aquaculture industry. While the zebrafish is genetically distinct from most fish species important in aquaculture - with the exception of carp, which, like zebrafish, belong to the family Cyprinidae - it is significantly closer than any other available model organism with the major separation event theorised to be approximately 150 million years ago (reviewed in[83]). In addition, the much anticipated recent publication of the zebrafish genome sequence[153], continuing the analysis of the genome of the zebrafish initiated by the Welcome Trust Sanger Institute in 2001 and first made public in 2002, suggests that approximately 70% of the human protein-coding genes have a zebrafish orthologue[153]. The subsequent annotation of the sequences will only enhance the zebrafish utility as a model, especially when the carp and trout sequencing projects are completed. Chapter 1. Introduction 3

The zebrafish is among the most widely utilised model organisms for the study of developmental genetics, diseases and embryonic development as well as being an excellent model for aquaculture research[83]. This chapter will focus on the zebrafish as a model of infectious disease and immunity for aquaculture species, and VHS virus will be of particular focus. It will also discuss the utility of the zebrafish to be modified by RNA interference (RNAi) as well as providing an overview of mechanisms and inducers of the RNAi response.

1.2 The zebraﬁsh model

The overwhelming majority of zebrafish research has focused upon vertebrate development, however, this is rapidly expanding to include studies of disease, including cancer and infectious disease. The Zebrafish Information Network (ZFIN)is thriving as a community of researchers and provides invaluable resources for the husbandry and laboratory utilisation of the zebrafish. The primary advantages which, separate the zebrafish from other laboratory animals of note is that development occurs ex vivo with the developing zebrafish being transparent, thus allowing real-time, whole body visualisation and staining. Zebrafish possess an almost complete complement of the immune functions and lymphoid organs found in mammals[246,267,397]; however, as with many phylogenetically ‘lower’ organisms, zebrafish depend more heavily on the innate immune response to infection than ‘higher’ organisms[4,344]. Zebrafish development can be split into a number of distinct stages broadly summarised in Table 1.1. For excellent detailed descriptions of zebrafish embryo development, the publication by Kimmel et al 1995 [186] is un- surpassed. Zebrafish innate immunity is additionally summarised and reviewed by Sullivan and Kim in 2008[344].

The zebrafish immune system develops in a number of distinct stages in the embryonic and juvenile zebrafish. Macrophages are present in the yolk sac from one day post fertilisation (dpf)[142] and the intermediate cell mass[88] from the same time. However, lymphoid cells only appear from 3 dpf with the adaptive immune Chapter 1. Introduction 4 system not considered to be completely active until 4 - 6 weeks post fertilisation (wpf) when functional lymphocytes fully mature[88,267,381]. Aswellastemporal separation of innate and adaptive immunity, the temperature of challenge also has significant impact on the expression of immune genes in zebrafish[93]. Chapter 1. Introduction 5 6mm 9mm 3mm . . . 14 mm 9mmto1 6mmto2 3mmto10mm 9mmto3 . . . . ≥ Developmental stages of the zebrafish 33 h 50% epibody - Tail bud . 25 h 128 cells - 30% epiboly 25 h 2 - 64 cells . . C) Stage Length . Timing of developmental stages is approximate and heavily dependent on ◦ Table 1.1: ZFIN 33 h to 24 h 1 somite - 26 somites 0 90 Sexual maturity . 25 h to 5 75 h to 2 25 h to 10 . . . and ≥ ] 186 [ NameZygoteCleavage Time (28 Blastula 0 min to 45 0 min 1 cell 2 GastrulaSegmentation 10 5 Pharyngula 24 h to 48 h Prim-5 - Rudimentary pec fin 1 JuvenileAdult 30 d to 90 d 10 mm to 14 mm HatchlingLarval 48 h to 72 h 4 d to Elongated 30 d pec fin - Fully formed pec fin 1 3 Modified from outside factors including temperature and water quality. Observational staging is generally preferred. Chapter 1. Introduction 6

1.3 Viral diseases of zebraﬁsh

At the time of writing there have been no natural viruses of zebrafish identified[75,224,267,344]. However, there have been a number of publications describing laboratory viral infection of zebrafish (summarised in Table 1.2), with VHS virus, snakehead rhabdovirus (SHRV), SVCV, and infectious haematopoietic necrosis virus (IHNV) being the most well-characterised in this model.

Despite the description of a viral infection of zebrafish with VHS virus dating back over thirty years[326], the use of zebrafish as a model of viral infection remains in its infancy when compared to rodent models or to the use of zebrafish as a model of development and genetics. Indeed, the majority of virology and immunology researchers work in mammalian models, mostly with small laboratory rodents, while fish immunologists and virologists have traditionally focused upon rainbow trout (Oncorhynchus mykiss), Atlantic salmon (Salmo salar), common carp (Cyprinus carpio) and other commercially important farmed fish (reviewed excellently in[245,266,267,336]). While these studies are important, the zebrafish offers researchers a unique model for exploring the innate immune system independent of the adaptive immune system. The wealth of data on the development of the zebrafish immune systems (as reviewed recently in[246,267,344,402] allows for the study of immune responses at specific levels of immune development. In addition, there is a high level of conservation between the zebrafish and human immune systems. For fish immunologists and virologists, the zebrafish offers a model, which, is far more easily housed than most commercially-important fish, and genetic tools and methodologies are available which, dwarf those validated or tested in any other fish species. The zebrafish are a particularly good model for diseases of carp that belong to the same family (Cyprinidae)). Despite these advantages, there has been relatively little consideration of the zebrafish as a model for viruses of aquatic organisms. Chapter 1. Introduction 7

1.3.1 The zebraﬁsh as a model for viral infection

The zebrafish is perhaps an odd choice for the study of human viral diseases, as despite the ubiquitous nature of viral infections, viruses are frequently extremely specific, often to the species level, due to the requirement of specific receptor proteins or other systems in the host for successful replication. This specificity is most apparent in pathogens of higher vertebrates[26], where influenza A is perhaps the most well known exception with humans, pigs, dogs, waterfowl and other birds, a variety of small laboratory mammals and most recently bats hosting a variety of influenza A subtypes; but influenza A virus remains the exception rather than the rule[24,311,348,358]. However, among lower vertebrates, it is more common for a virus to have a variety of hosts and there are a number of studies exploiting zebrafish as a model for viruses from hosts as diverse as frogs and humans including herpes simplex type 1 (HSV-1)[55], Chikungunya virus[277] (CHIKV), and Tiger Frog virus[228]. Chapter 1. Introduction 8 ] ] ] ] ] ] ] ] ] ] ] 9 [ 55 [ 224 388 326 198 347 220 228 224 289 , , , [ [ [ [ [ , 6 [ 198 369 265 [ [ , 93 [ Adult Larval Adults Adults Adults Adults Adults ZF4 cells 72 h Larvae, and Adults 24 h Embryos, Juveniles, and Adults 24 h Embryos - 29 d Juveniles, and Adults Alphaviridae Herpesviridae Rhabdoviridae Birnaviridae Iridoviridae Iridoviridae Nodaviridae Rhabdoviridae Rhabdoviridae IridoviridaeRhabdoviridae In Vitro Viruses that have an established zebrafish model Table 1.2: Tiger frog virus (TFV) Viral haemorrhagic septicemia virus (VHS virus) Nervous necrosis virus (NNV) Spring viraemia of carp virus (SVCV) Snakehead rhabdovirus Infectious spleen and kidney necrosis virusLymphocystis (ISKNV) disease virus (LCD) Herpes simplex virus type 1 Infectious hematopoietic necrosis virus (IHNV) Infectious pancreatic necrosis virus (IPNV) VirusChikungunya virus (CHIKV) Family Age at infection Reference(s) Chapter 1. Introduction 9

As mentioned previously, much of the power of the zebrafish model comes from their transparency and relative ease of generation of mutants and transgenics. Re- cently, a number of groups, spearheaded by Professor Jean-Pierre Levraud of the Pasteur Institute, have been exploiting established zebrafish methods and combining them with modern and classical virology and bacteriology to establish the zebrafish as a unique and critically important model which, can set the zebrafish apart from established rodent or other small laboratory mammals. In their 2011 publication in the journal PloS Pathogens[224] they describe the “whole-body” infection of zebrafish larvae with a heat-adapted IHNV virus. This publication relied on both the transparency of the zebrafish and the availability of unique relevant fluorescent mutants to track the infection throughout the larvae. This publication is significant as it is among the first to utilise the power of the zebrafish model for imaging, and to combine it with an infection model. Live imaging, fluorescent viruses, whole-mount in situ hybridisation (WISH), and transgenic knock-in or knock-out zebrafish will hopefully become a more widely-used tool to elucidate the patterns of viral infection in this important model[208,224].

1.4 Viral haemorrhagic septicaemia

Viral haemorrhagic septicaemia (VHS) is a disease of ﬁsh caused by the viral haemorrhagic septicaemia virus (VHS virus). Previously VHS and VHS virus have been known as Egtved disease and Egtved virus, respectively, but this has fallen into disuse. VHS virus has a negative-sense single stranded RNA genome of around 11 kb and belongs to the order Mononegavirales, the family Rhabdoviridae,andthe genus Novirhabdovirus. The virus is contained within an enveloped, bullet-shaped virion 180 nm × 50 nm and is covered in glycoprotein peptomers of between 5 - 15 nm (Figure 1.1).

VHS virus infection of farmed fish species causes significant and long term losses, both in loss of fish stocks as a direct result of disease, as well as the loss of market share caused by restrictions on the movement and sale of fish caught or farmed Chapter 1. Introduction 10

A B

Figure 1.1: Structure of VHS virus under the electron microscope[406]. A VHS virus particles maturing on the surface of RTG-2 cells. × 262,000. B Negatively stained VHS virus particle × 300,000 C Cross-section of VHS virus particle × 617,000.

in waters known to be infected by VHS virus[106,258]. It is, therefore, considered to be a major threat to a number of ecologically and economically signiﬁcant ﬁsh species, particularly the major farmed species of salmon and trout.

VHS virus, like the related IHNV, consists of 6 open reading frames encoding the viral genes in the order 3-N-P-M-G-NV-L-5 (Figure 1.2)[320]. The genes encode the nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), non-structural viral protein (NV), and the polymerase (L) genes. As with most negative-sense single stranded RNA viruses, VHS virus gene expression is greatest closest to the 3 end of the RNA strand and reduces towards the 5 (Figure 1.2). Chapter 1. Introduction 11

3’ 5’ NPMGNV L

RNA Expression Level

Figure 1.2: VHS virus genome. The VHS virus genome consists of 6 genes: the nucleoprotein (N), phosphoprotein (P),matrixprotein(M), glycoprotein (G), non-structural viral protein (NV), and the polymerase (L) genes. RNA expression of VHS virus is greatest closest to the 3 end of the genome and decreases in a linear manner towards the 5 end.

The NV gene is specific to the novirhabdoviruses and is among the most variable regions within the genome[115,148]. The NV gene is a non-structural protein that is not found within the mature VHS virus virion but only within infected cells. The presence of the NV gene has been shown to be crucial for pathogenicity in some species, particularly rainbow trout[319,357]; however, the specific regions that differentiate highly pathogenic VHS virus from less virulent VHS virus have not been determined[57]. The G gene has also been identified as important in determining the virulence of VHS virus[12]. The highly pathogenic Great Lakes strain of VHS virus (M1030GL) differs significantly from the other type IV VHS virus isolates[12,57].

Clinically, VHS virus causes rapid onset of mortality, which is preceded by lethargy, darkening of the skin, exophthalmia, anaemia (pale gills), distended abdomen, haemorrhages and redness at the base of the fins, gills, eyes and skin, as well as behavioral changes, including abnormal swimming, such as flashing and spiraling, as well as disorientation[272]. Mortality rates following infection with VHS virus vary depending on the genotype of virus and environmental factors including the temperature of the water. Small fry infected with VHS virus often show 100% mortality, while older rainbow trout, salmon and pacific herring show mortalities between 5 % and 90 % under experimental conditions[236,272]. Chapter 1. Introduction 12

Diagnosis of VHS virus in the laboratory is accomplished via a number of methods including electron microscopy, where it appears as a typical bullet shaped rhabdovirus between 60-70 nm in diameter (Figure 1.1)[104]. For VHS virus infection to be definitively established, the virus must be rescued and grown in cell culture in either BF-2 (bluegill sunfish (Lepomis macrochirus [384]), or EPC (Epithelioma papulosum cyprini [from fathead minnow (Pimephales promelas] cells. It should be noted in passing that EPC cells that were originally characterised as carp epi- dermal cells and are considered to be contaminated at the lowest available passage with fathead minnow cells[382]) cells. BF-2 or EPC cells should be examined for CPE every 24 - 48 hours[253]). If CPE is detected, identification via antibody methods such as neutralisation immunostaining or enzyme-linked immunosorbent assay (ELISA), or molecular methods such as reverse transcription polymerase chain reaction (RT-PCR) are required, using antibodies and RT-PCR primers, respectively, defined by the World Organisation for Animal Health (OiE) within the Manual of Diagnostic tests for Aquatic Animals[272]. Typical CPE of rhabdovirus infection of EPC and CHSE (Chinook Salmon Epithelial Cells) includes cell lysis, rounding of cells, syncytia formation, and detached cells which, tend to clump and bloat into grape-like formations[106,161,272]. Importantly, there are a number of quantitative RT-PCR (RT-qPCR) primer sets using either SYBR greenTM, or TAQ-man based probes, which, can then be used to determine the relative amounts of virus present [58,81,151,236]. These RT-qPCR reagents have not been certified for the diagnosis of VHS virus by the OiE; however, they are widely used for research purposes[272].

1.5 RNA interference (RNAi)

RNAi is an ancient and highly conserved cellular process that appears to be ubiquitous in eukaryotes. Evidence of functional RNAi pathways exists in plants[262]; worms such as Caenorhabditis elegans [C. elegans] [110]; insects such as fruit-fly Drosophila melanogaster [178] and crickets Acheta domesticus[194]; other invertebrates such as shrimp, Penaeus monodon, as well as fish including zebrafish[371], Chapter 1. Introduction 13 and Nile tilapia, Oreochromis niloticus [46]; and mammals, including humans[102] and mice[23], along with many other species. Endogenously, the microRNA (miRNA) pathway exists as both a regulatory [349] and anti-viral[285] mechanism. To regulate transcription and degradation of messenger RNA (mRNA), the RNAi pathway utilises small, mostly non-coding, miRNAs[314]. The following sections will discuss the RNAi pathway, mechanisms and inducers of RNAi silencing as well as limitations and off-target effects of RNAi, with a particular focus on teleosts, and specifically zebrafish, research.

1.5.1 The RNAi pathway and history of RNAi-induced silencing

The initial discovery of dsRNA-induced gene silencing in plants occurred in 1990 through efforts to enhance the purple colour of petunias by over-expressing via the injection of the chalcone synthase gene that synthesises this pigment[262]. However, plants with the introduced gene produced either white and/or patterned flowers instead of the hypothesised purple colour[262]. Unfortunately for these researchers, this discovery was treated as a curiosity and its significance not realised until almost 10 years later when, in 1998, two independent studies demonstrated that gene silencing was initiated by dsRNA molecules in the nematode Caenorhabditis elegans (C. elegans) [110] and again in tobacco plants[373]. The mechanism of action remained elusive until Fire and colleagues demonstrated that by injecting a combination of RNA sense and antisense strands into C. elegans, more efficient gene knockdown was achieved compared to either one of these strands independently. Prior to this experiment, modulation of gene expression had been described in C. elgans using either sense or antisense RNAs. However, the introduction of dsRNA vastly increased potency of the gene knockdown .

The natural functions of RNAi and its related processes were first suggested to involve protection of the genome against mobile genetic elements such as integra- tive viruses and transposons. More recently, this theory has been expanded to include a broader role that RNAi inducers regulate the expression of genes via Chapter 1. Introduction 14 the degradation of, or transcriptional blockage of, endogenous mRNA via endogenous RNAi inducing molecules called microRNAs (miRNAs)[61,293,395]. The first miRNA, lin-4, was discovered in the early 1990s[201,380], and miRNAs are now known to control gene expression in numerous fundamental cellular processes, including cellular differentiation, apoptosis, developmental timing, and regulation of transposable elements. The functions and actions of miRNAs have been the subject of massive research interest since the discovery of lin-4 and are covered frequently in reviews[53,136,181,290]. The role of RNA silencing is critically important throughout the lifecycle of the organism and it has been estimated that miRNAs acting via the RNAi pathway, may regulate up to a third of all human genes [209] (Figure 1.3)[[110]].

From the initial discovery of RNAi, it was suggested that this pathway plays an important role in anti-viral immunity. This was first demonstrated in plants when it was illustrated[294] that transgene-induced viral immunity[30,213] was initiated by the expression of viral mRNA. Following infection, all viruses may produce dsRNA that can potentially be recognised by the endogenous RNAi machinery (Figure 1.3). In addition, a number of viruses encode and utilise factors that act via the RNAi pathway. In particular, miRNAs are expressed by a number of viruses including Simian virus 40 (SV40)[345], Herpesvirus family members[60,288,309] and human immunodeficiency virus type-1 (HIV-1)[33,273]. In addition, host-encoded miRNAs are also known to directly or indirectly affect the replication of many viruses, such as retrovirus primate foamy virus type-1 (PFV-1)[199], and hepatitis C virus[168,255,312], while it is also believed that saturation of host RNAi pathway components, to prevent viral degradation, is achieved by the adenoviral-expressed virus associated (VA) RNAs[130]. Human cytomegalovirus expresses a number of miRNAs that specifically target genes to facilitate persistence and replication in cell culture[123,249], and additionally encodes an inhibitory molecule specific to the human miRNA miR-27[212]. Understanding the underlying processes by which, the RNAi pathways function in cells and organisms are fundamental to our ability to harness this pathway to specifically modulate genes of interest for both research and therapeutic applications. Chapter 1. Introduction 15

MicroRNA Precursor miRNA Gene A

RAN+GTP Xpo5

Nucleus

B Cytoplasm

Viral RNA

Xpo5 MicroRNA Precursor

Exogenous dsRNA

RISC Enzymatic Cleavage

siRNA

Perfect complementary Imperfect complentary Binding Binding siRNA RISC miRNA RISC

mRNA Cleavage

mRNA Sequestion into P-Body Chapter 1. Introduction 16

The principle of externally introduced RNAi with genes in a sequence specific manner was first identified in petunias [262], when plasmids designed to over-express chalcone synthase (CHS) proteins in petals resulted instead in the reversible degradation of CHS mRNA and a distinctive petal phenotype[99,262]. This study suggested that the introduction of anti-sense RNA was the catalyst for the observed knockdown of mRNA, a phenomenon that was also observed in fungus, where it was termed “quelling”[303]. However, it was the identification of dsRNA as a “potent and specific inhibitor”[110] of mRNA transcription in C. elegans that truly sparked the scientific interest in RNAi. In the landmark study, injection of purified sense and anti-sense RNAs corresponding precisely to a region of the unc-22 gene produced a phenotype of characteristic behavior and mottling identical to that observed in the unc-22 knockout. Crucially, this study also identified that the introduction of both sense and anti-sense RNAs was several orders of magni- tude more effective at inducing knockdown of the gene than the injection of either strand alone[110].

Following this study, long dsRNAs were also used to successfully knockdown genes in plants[373],inD. melanogaster [178] and also in fish[371]. In the latter study, zebrafish were injected with long dsRNA designed against three characterised genes: floating head (flt), no tail (ntl) and pax2.1/no isthmus (noi). These produced

Figure 1.3 (preceding page): The natural inducers of the RNAi pathway. A Endogenous miRNA genes express miRNA precursors that are processed and exported from the nucleus by the Xpo5 RanGTP complex. B Virus encoded microRNAs (miRNAs) are expressed upon viral infection to form miRNA precursors. These miRNA precursors can be either encoded by the virus, target cellular processes and repress native genes, or be cleaved by cytoplasmic enzymes and repress viral mRNA as a primitive anti-viral response. These viral or endogenous-derived miRNA precursors are enzymatically processed into mature miRNAs that are then incorporated into RISC. C Transposon or pathogen exogenous dsRNA is recognised within the cell and processed into active siR- NAs. Both siRNA and miRNA are then further processed within RISC where unwinding of the RNAi inducer and passage to the corresponding cellular mRNA occurs. The RNAi-inducing molecules then cause the nucleolytic degradation of the target or repression of translation by either sequestering mRNA within p-bodies or by blocking translation by proteases. Chapter 1. Introduction 17 effective interference that was dependent on the concentration of the dsRNA injected[371]. This paper was followed by a flurry of research into the effectiveness of RNAi as a method of gene knockdown in fish[211,232,269,371].These studies were inconclusive as to the specificity of the knockdown; in each of these studies, phenotypes which, were either highly variable or significantly different from the knockout animal were observed. The authors suggested that any phenotypes resulting from the specific interference caused by injected dsRNA were being swamped by non- specific effects related to the injection of the RNA[269,323,404]. It seems likely that the inconclusiveness of these findings coupled with the nearly simultaneous development of additional genetic tools for gene knockdown[97,346] and rescue in the zebrafish model have resulted in RNAi tools being less developed in zebrafish[323].

More recently, a number of explanations have been proposed for the seemingly contradictory nature of these results using RNAi. Firstly, differences in the microinjection methods used and the amount of dsRNA injected have likely played a role in inducing non specific defects[111]. However, the results from titration experiments, makes it clear that the concentration of RNA alone is unlikely to be responsible for the lack of specificity of the interference[232,269,323]. Double- stranded RNA has long been understood to activate the type 1 interferon (IFN) immune response via the Toll like receptor 3 (TLR3) pathway as well as pathways that detect intracyctoplasmic dsRNA such as the retinoic acid-inducible gene 1 (Rig-I) like receptors (RLRs) as well as the protein kinase regulated by RNA (PKR)[163,238]. Indeed, synthetic dsRNAs such as polyI:C are used deliberately to induce such a response[89]. It has, therefore, been additionally proposed that the “non-specific” defects noted in these papers following the introduction of dsRNA could be the result of an interferon response[323]. Zebrafish and other fish species have functional interferon present at 18 - 24 hours of embryonic development with poly:IC shown to induce an upregulation of IFN production[66,207,300,323].

More recent studies using dsRNA in zebrafish[2,154] and Xenopus[261] have shown that reproducible and specific defects can be observed in fish. One study used both morpholinos and injected dsRNA targeting the 3-end of the Myostatin gene and was able show that the same phenotype was produced by both methods of Chapter 1. Introduction 18 transient knockdown[2]. In this case, significantly less dsRNA was used ( 1fg/L), despite targeting a gene encoding a protein ubiquitously expressed throughout the organism[2,111].

1.5.2 The endogenous mechanisms of RNAi

Primary miRNA transcripts (pri-miRNAs) are transcribed within the nucleus by RNA polymerase II transcription. These short 100 - 300 nt transcripts can be produced either from independent genes or alternatively spliced from the introns of protein-coding genes, from either the 3 or 5 untranslated regions[62].Ineach case the pri-microRNAs are further processed in the nucleus by the Drosha/ DRG8 protein complex into 70 nt - 100 nt pre-miRNAs[202,399].

Pre-miRNAs form hairpins that are transported from the nucleus into the cytoplasm by Exportin-5[44,50,130,226,391] (Xpo-5). Until recently this was presumed to be the only role for Xpo-5 within the microRNA pathway. However, in a recent study[34] ablation of Xpo-5 expression using siRNA molecules simultaneously reduced Dicer but not Drosha or other associated miRNA pathway members. It also demonstrated that the over-expression of Xpo-5 increased Dicer protein level. Most importantly for this study, they also demonstrated that expression of large quantities of pri-mir30 from the pol II CMV promoter saturated the Xpo-5 protein that concomitantly reduced Dicer protein availability in the cytoplasm[34]. This is among the ﬁrst evidence that the relative abundance of individual microRNAs has a cross-regulatory eﬀect on the expression of RNAi pathway members, and provides further evidence of Xpo-5 as a limiting factor in miRNA abundance within the cytoplasm.

The ﬁrst step of the cytoplasmic RNAi mechanism is based on the cleavage of dsRNA of greater than 30 nt, processed into 21 - 23 nt active siRNAs by the RNase III enzyme Dicer[102]. Like all RNase III enzymes, Dicer cleaves dsRNA and leaves 3 overhangs of either two or three nucleotides with 5-phosphate and 3- hydroxyl termini[102,164]. Dicer activity is dependent on the adenosine triphosphate Chapter 1. Introduction 19

(ATP)-dependent translocation of the enzyme along its target prior to dsRNA cleavage[38]. Dicer has four distinct domains: an amino-terminal helicase domain, dual RNase III motifs, a dsRNA-binding domain, and a PAZ domain (named after the proteins Piwi, Argo, and Zwile/Pinhead)[64,349].

The cleaved strand of the duplex selected for incorporation into the RNA Induced Silencing Complex (RISC) is most frequently the strand which, is the least thermodynamically stable, which, is dependent on the nucleotides surrounding the 5 end[179,321]. This strand is colloquially known as the guide or targeting strand, and is responsible for the sequence-speciﬁc interaction of RISC with the target messenger RNA (mRNA). The complementary strand, known as the passenger strand, is less frequently incorporated into RISC[374], and is typically degraded[135].Ifthe guide strand is completely identical in sequence to the target mRNA then the mRNA is cleaved and degraded, which, throughout the remainder of this document will be described as siRNA degradation. If there are a small number of mismatches between the target mRNA and the miRNA seed region, located at nucleotides 2-7 from the 5 end of the miRNA, then the mRNA is sequestered within P-bodies in the cell resulting in translational repression (miRNA repression)[94,179,188]. Where miRNAs have two eﬀective strands, the strands partner with separate RISC entities and can both result in mRNA degradation or inhibition of translation[299]. As well as microRNAs and synthetic siRNAs and shRNAs, endogenous dsRNA from repetitive sequences such as transposons and exogenous dsRNA from pathogens, particularly of viral origin, additionally trigger the RNAi pathway. Post-Dicer cleavage, the siRNA then binds to a multiprotein complex to form the RNA-induced silencing complex (RISC). Activation of RISC is dependent on unwinding of the siRNA duplex via Argonaute-mediated cleavage of the unincorporated passenger strand[122,237,291], and the antisense strand is guided to the target mRNA. Binding of the homologous siRNA strand to the target mRNA results in the RISC-mediated cleavage of the target mRNA.

Most recently, deep sequencing has elucidated whole additional families of short ncRNAs. Some are processed by the canonical RNAi pathway, others interact with one or more members, frequently either Dicer or Argonaute proteins with or Chapter 1. Introduction 20 without the RISC complex with an accompanying post-transcriptional targeting and mRNA modification. Along with the piRNAs, tiRNAs and other families of short ncRNAs, RNA-seq and other sequencing technologies have recently identified scores of additional siRNA molecules within both the cytoplasm and the nucleus. These additional families of ncRNAs have exposed flaws within the models of RNAi silencing proposed so far. Among the most accepted of new hypotheses from these studies are suggestions that there are a number of RISCs and that the downstream consequences of RNAi silencing are dependent upon the RISC within which the siRNA, miRNA, piRNA or other potential substrate is incorporated. The specific Argonaute protein that the RISC is built around may be responsible for the differences in outcome from the same substrate.

The original model of RNAi suggested that siRNA/shRNA and miRNA molecules shared key components of the miRNA pathway (see[351] for a review). For the transport and processing molecules Xpo-5[34,129,391] and Dicer[65,113,126,276,298],respectively, this appears largely to be the case. There are certain limited exceptions, however, with molecules including mir-451 having Dicer; independent,[70,280,389] Argonaute; dependent processing requirements. As well as, shRNAs of 18-19 nt having RNAi silencing ability thought to be Dicer independent [118,218,334]. However, these views are limited to three separate publications at this time, and this thesis will treat shRNA molecules as Dicer-dependent unless otherwise noted speciﬁ- cally. However, there is an alternative theory that siRNAs and perfect match miRNAs enter an siRNA-dependent siRISC, deﬁned by its “slicing” activity and by a miRISC where the outcome of miRNA-target mRNA binding is sequestration of target mRNA within cytoplasmic P-bodies[215,304,327].

While both siRNA strands have the potential to facilitate selection during the transition of the siRNA/miRNA duplexes into RISC[374], the end which is the more weakly paired thermodynamically is preferentially selected to enter RISC, and to bind target mRNA. Upon the final binding of the unwound mature siRNA or microRNA within RISC to the target homologous mRNA, cleavage occurs at a single site in the centre of the duplex region, 10 nt from the 5-end of the siRNA[102]. Although this process is not ATP dependent[268], successive rounds of Chapter 1. Introduction 21 mRNA cleavage are more efficient in the presence of ATP[156]. The final loaded RISC complex catalyses hydrolysis of the target RNA phosphodiester linkage and is reminiscent of the reaction where Dicer generates siRNA from dsRNAs[235,322].

1.6 Synthetic inducers of the RNAi pathway

1.6.1 siRNA

Synthetic interfering RNA molecules are artiﬁcially-created 19 - 23 bp duplexes designed with 3 overhangs to facilitate entry into the endogenous miRNA pathway [179]. siRNA molecules are designed with guide and passenger strands: the guide strand is identical to the mRNA sequence to be silenced and the passenger stand is the exact complement. In order for speciﬁc silencing to occur the passenger strand enters the RISC complex and binds to the mRNA. The siRNA is constructed so that there are 2 uracil bases overhanging at the 3 end. This overhang assists in guiding the correct strand into the RISC complex[16,363].

As previously mentioned, the design of siRNA molecules is crucial for the efficiency of the knockdown, and there are a number of published algorithms available online for the purposes of designing siRNAs[102]. As miRNAs are frequently not completely complementary to the sequence being targeted and bulges and slight mis-matches are common, it is possible for a single miRNA to act on multiple targets. Thus, if the introduced siRNA is partially complementary to an undesired target, particularly around the 5-12 bp seed region, the siRNA may bind to an undesired mRNA, which could then be targeted via RISC for cleavage or incorporated into a P-body and sequestered[217]. Therefore, specific use of siRNA, and RNAi in general, absolutely requires that the genome sequence of the organism into which the interfering molecule is being introduced be known, and that multiple controls including scramble/irrelevant interfering molecules are included as part of any experimental design. An alternative to synthetic production of Chapter 1. Introduction 22 siRNAs is to synthesise longer, 27 bp transcripts as the base molecules to be introduced into the cell. These longer dsRNA molecules enter the RNAi pathway downstream of the RISC complex and are cleaved into Dicer substrates, which are potentially more efficiently loaded into RISC and induce effective silencing [181]. dsRNA Dicer substrates (DsiRNA) have been used effectively in Chinook Salmon Embryo (CHSE-214) cells to specifically silence VHS virus[43] without silencing the heterologous virus IHNV. Whether longer 23 b to 30 b DsiRNA molecules or shorter 17 – 21 bp siRNA are more effective silencers is only one small consideration when RNAi silencing molecules are being designed. The additional costs involved in synthesising longer molecules and the ready availability of commercial siRNA is likely to ensure that DsiRNA molecules are less widely utilised.

1.6.1.1 siRNA delivery in ﬁsh

When properly designed, siRNAs are an effective method of knocking down mRNA levels within a cell. However, due to the short half-life of RNA within the cell, they are only effective for a short period, generally for between 3 - 5 days[102,108]. This transience has led to the development of vector-based approaches to allow more sustained production of siRNA molecules in vivo [51]. There has been considerable proof of principle work to show that RNAi molecules can be effective in cell culture as siRNAs or in transgenic animals as shRNAs[221]. However, before effective RNAi treatments can be developed for control of disease, robust delivery systems are required[221,305]. Many of the RNAi methods were developed to achieve high levels of transfection in mammalian cell culture using reagents specifically designed to transfect at the temperature and conditions which suit mammalian cells[139,297]. Unfortunately, transfection of fish cell lines is particularly difficult, with many standard chemical and lipid-based transfection reagents being ineffective at trans- porting nucleic acids across cell membranes[85] at the lower temperatures required for the healthy growth of fish cell cultures. Cell lines stably expressing plasmid derived RNAi molecules have been produced in EPC (ATCC CRL2875)[183,306,307] and BF-2 (ATCC CCL-91)[85,328] cells, although, again, these studies have shown mixed results. Electroporation of the zebrafish cell line ZF4 (ATCC CRL-2050) Chapter 1. Introduction 23 has proven to be an effective method of introducing plasmid DNA, however this requires specific expensive equipment (Invitrogen Nucleofector TM) that makes this method less common[137,362].

The first study utilising siRNA molecules in fish was the knockdown of eGFP in rainbow trout embryos using siRNA molecules targeting eGFP but not with scrambled (irrelevant) siRNA molecules[45]. The same siRNA molecules were also directly injected into the F3 strain of transgenic rainbow trout and were effective at inducing knockdown of a stably expressed eGFP transgene. Approximately 10% of the injected embryos were found to have non-specific and frequently lethal phenotypes, and the number of defects observed increased rapidly when more than 5 ng of siRNA was used[323]. A significant thrust of research in fish involves the generation of antiviral RNAi, using plasmid based shRNA molecules to target a viral gene either in vivo or in vitro.

RNAi molecules targeting viral genes to induce resistance in the host have been used in a number of fish species and cell lines to mixed effect. siRNAs targeted to the MCP gene of the Iridovirus-tiger frog virus were transfected into FHM cells. The siRNA molecules were successful at reducing viral titre and delaying CPE for up to 36 hours[387]. This successful early result was contrasted by another study targeting the G gene of VHS virus published in the following year. In this study, three siRNA molecules targeting the G protein of VHS virus and 3 control siRNAs with between 1 and 4 mismatches were produced and transfected into EPC cells using the TransIT-TKO siRNA transfection reagent. These molecules were capable of protecting the cell monolayers from infection for up to 96 hours post transfection. However, there was also significant induction of interferon as measured by real- time PCR of the MxA gene. These siRNA molecules also protected against the related fish rhabdovirus spring viraemia of carp virus, which led the authors to conclude that the protection against the virus was the result of up-regulation of the interferon response, rather than specific silencing from the siRNA molecules[307,324]. As with most fish cell culture studies, transfection efficiency was relatively poor, and the authors recommended finding alternatives to the transient knockdown provided by the siRNAs. Chapter 1. Introduction 24

1.6.2 Polymerase III promoters for shRNA expression

As mentioned, siRNA molecules have only a transient effect on the cell before they are cleared and destroyed. In order to overcome this limitation, Brummelkamp et al[51] designed a plasmid vector containing, in order: a polymerase III promoter, such as the H1 promoter; 21 bases exactly complementary to the target sequence; a region of 9 bases, which forms a loop when transcribed; 21 reverse complementary bases, which will become the guide strand of the siRNA molecule; and, finally, a termination sequence[51]. This structure mimics naturally encoded Pri-miRNA molecules, and it enters the RNAi pathway in the nucleus of the cell before being exported to the cytosol by Xpo-5, in the same manner as miRNA molecules[51]. These synthetic short-RNA hairpins are termed shRNA. Following this study, there has been a vast array of designs and types and terms for these synthetic interfering molecules AmiRNA and SmiRNA, artificial microRNA and synthetic microRNA, respectively, are among the most common alternative terms utilised. For the purposes of brevity and simplicity, this document will use the term shRNA for all short RNA hairpins, which enter the endogenous RNAi pathway, are exported from the nucleus via Xpo-5, are processed into siRNA molecules by either Argonaute-2 or Dicer and are incorporated into RISC.

Small nuclear RNA genes are frequently transcribed by RNA polymerase III in eukaryotes. Polymerase I (pol I) synthesises large ribosomal RNA (rRNA), pol II synthesises mRNA, snRNA[143] and miRNA[203] and pol III synthesises 5S rRNA, transfer RNA (tRNA), 7SL RNA, U6 snRNA and a number of other small stable RNAs notably the H1 RNA component of RNase P[22,282]. Pol III promoters can be further categorised into type 1, 2 and 3 RNA pol III promoters that are diﬀerentiated by their alternative promoter structures (Figure 1.4). However, the majority of these promoters share the unusual feature of the requirement for sequence elements downstream of the +1 position of the transcriptional startsite region. The 5S rRNA gene is a prime example featuring this typical structure.

The human U6-1 promoter is the best characterised type 3 RNA pol III promoter and comprises a promoter core region featuring a TATA box between -30 and -25, Chapter 1. Introduction 25

Type 1 Promoter Xenopus Laevis 5s RNA gene Internal Control Region Tn A Block IE C Block +1 +50 +64 +70 +80 +97 +120

Type 2 Promoter Mouse tRNA gene

Tn A Block B Block +1 +8 +19 +52 +62 +73

Type 3 Promoter Human u6-1 snRNA gene -222 -202 TTTTT OCT SPH PSE TATA +1 -234 -228 -74 -27

Zebrafish u6-2 snRNA gene TTTTT OCT SPH TATA +1 -224 -216 -48 -24

Distal Sequence Element Proximal Sequence Element

Hybrid Promoter Saccharomyces u6-1 snRNA Gene Tn TATA A Block B Block +1 -30 -23 +21 +31 +113 +234 +244 Chapter 1. Introduction 26 and a Proximal Sequence Element (PSE) between -66 and -47. The TATA box is characteristic of type 3 promoters, and is located at a fixed distance downstream of the PSE[239]. The TATA box is recognised by the TATA-binding protein (TBP), which is one of three polypeptides in a multi-subunit complex, transcription factor IIIB (TFIIIB)[317,356]. The PSE is recognised by a multi-subunit protein complex known as the snRNA activating protein complex (SNAPC)[143,308]. The binding of SNAPC and TBP, and their subsequent interaction with the DNA strand, leads to the recruitment of RNA pol III[69]. Located further upstream, the distal sequence element (DSE) generally functions as an enhancer for the activation of transcription from the promoter core region[143]. This region commonly consists of an octamer element (OCT) and a staf motif known as the SphI post-octamer homology (SPH) element[193,259,313,318]. The OCT motif is bound by Oct-1[144],andthe SPH recruits staf, a septamer zinc-finger transcriptional factor[313]. The zebrafish U6 promoters differ significantly from the mammalian U6 promoters in that the SPH element for the PSE is translocated and the DSE appears to consist of the OCT motif alone[46,71,133]. There currently has been no research comparing the SNAPC and TBP binding proteins from human and zebrafish models to identify whether the zebrafish TFIIIB complex is significantly different to the mammalian orthologue. As the pol III type 3 promoters consist of elements entirely prior to the start of transcription, they are uniquely suited for use to specifically and efficiently express different shRNA molecules[51].

A number of the most recent studies in ﬁsh cells using vector-based shRNA molecules targeted either the G or L genes of VHS virus. Firstly, Ruiz et al and Dang et al in 2008, and then Kim and Kim in 2010 successfully created stable cell lines transfected with various plasmids expressing shRNA molecules[85,183,307]. Ruiz

Figure 1.4 (preceding page): The structure of RNA polymerase III promoters. Examples of each of the three types of RNA pol III promoters are illustrated with the various promoter elements and the locations of the promoter elements relative to the start of transcription. Transcription start sites are indicated by the arrow and +1 and the termination site of transcription is indicated by Tn for type I, type II and the hybrid promoters, and by TTTTT for type III promoters. Figure is modified from[282,316] and includes additional data from[71]. Chapter 1. Introduction 27 et al found that the three plasmids expressing different siRNA molecules targeting the L gene provided different levels of protection against VHS virus. However, this protection was also not specific to VHS virus, with the shRNA molecules additionally able to provide protection against IHNV, a related novirhabdovirus.One of the stably generated cell lines showed increased levels of unhealthy cells prior to infection, including cells showing poor adherence to the flask and each other, and also increased numbers of bloated and floating cells. As no increased IFN response was observed by real-time PCR of the MxA gene, the authors suggested that the shRNA molecules produced by the particular expression plasmid may be interfering with the endogenous miRNAs present in the cells[307]. This observation is supported by a number of papers that have suggested that fish may be particularly sensitive to the expression of additional shRNA or siRNA molecules by interfering with the natural miRNA pathway[112,216]. However, another study using the same method of generating stable cell lines, but using a different promoter to express the interfering hairpins, showed specific silencing of VHS virus and complete protection of EPC cells against VHS virus, with little induction of interferon and no cross protection against IHNV. By using a weaker fish U6 promoter, the Fugu U6 promoter[401], rather than the enhanced CMV promoter used by others[307], no unhealthy cells or potential off-target effects were observed[183]. There remains a pressing need for further studies in alternative cell lines and using alternative methods of expressing hairpins to confirm the impact of using a weaker promoters on reducing such off-target effects.

In 2012, two publications in the journal Zebrafish[71,298], one published from the work performed in Chapter 3 of this thesis[71] and a third in early 2013 in Ge- netics[96], offered alternatives for the expression of shRNA molecules in both cell culture[71] and transiently transgenic zebrafish[96,298].Clarkeetal[71] utilised native zebrafish polymerase III promoters and characterised the relative strength of these promoters, and Dong et al[96] used a modified intron to deliver an shRNA utilising a miRNA backbone to the cells and zebrafish. Both of these methods may be of significant importance in over-coming off-target effects of over-expression of shRNA molecules in fish, firstly, by providing alternative native polymerase III promoters, Chapter 1. Introduction 28 including the weaker H1 promoter[71] and secondarily by coupling an shRNA to a native intronic miRNA and controlling the dosage of the shRNA to a known tolerated level[96,298]. Due to the effectiveness of alternative methods of gene ablation, in particular the TALE nucleases used by Professor Stephen Ekkers group[32],and the exciting developments in clustered, regularly interspaced, short palindromic repeat (CRISPR) - Cas9 RNA nuclease combinations, excellently reviewed in[310], the major utility of RNAi in fish remains to target exogenous mRNA, particularly from viruses.

1.6.3 Off-target effects of synthetic RNAi in fish

A number of papers have used EPC cells to either transfect siRNA molecules or shRNA plasmids to generate stable cell lines producing siRNA molecules targeting VHS virus. However, there have been discordant results[183,307,324,325].One potential explanation for this is the recent report that all known lineages of EPC cells were contaminated with cells from fathead minnow, Pimephales promelas [382]. One obvious consequence of the misidentification of EPC cells is that any scan- ning or trawling of the draft Carp genome or available carp sequences by Ruiz et al for potential off target effects by the shRNA molecules was performed against the wrong species. This might possibly explain the large induction of interferon and consequential cross protection of cells against IHNV by some of the molecules used by Ruiz et al, through binding unknown endogenous mRNA from what were in fact minnow cell lines

Sequence independent off-target effects have been observed in detail both in mice, where adenoviral vectors expressing shRNAs from U6 pol III promoter induced significant liver damage, and in Xenopus, where polymerase II and III vectors induced significant developmental deformities and mortality. This dose dependent inhibition of endogenous miRNAs offers an alternative explanation for the differences in the results observed by Ruiz et al[306] and Kim and Kim (2010)[183]. Kim and Kim used a native fish (Fugu U6) promoter to express their shRNA Chapter 1. Introduction 29 molecules. This promoter is likely weaker than the commercially available human cytomegalovirus CMV enhanced promoter used by Ruiz et al, as described in Zenke 2008[401]. By utilising a weaker promoter Kim and Kim may have avoided interfering with the endogenous miRNA pathways. However, significantly more research is required into the effects of different levels of siRNA (shRNA) molecules on zebafish if the development of an RNAi transgenic fish free from non-specific defects is to be achieved. The saturation and inhibition of the native miRNA pathway by exogenous dsRNA molecules will be discussed in further detail (Section 1.7).

To begin to study the non-specific inhibitory effect of siRNA molecules on miRNAs in more detail, Fjose and Zhoa[112,403] knocked down Dicer in zebrafish embryos using anti-sense morpholinos. This blocked the miRNA pathways with no miRNA specific-silencing being observed[403] leading to marked and terminal non-specific mutations. There was a marked effect on one specific miRNA, mir-430 that has been identified as a crucial regulator of the removal of maternal mRNA early in embryonic development. Non-targeted (irrelevant) siRNAs were injected into the developing embryo, and similar disruptions in mRNA, and specifically to mir-430, were observed in a dose dependent manner[112,403].

1.7 Saturation of the endogenous microRNA pathway by exogenous dsRNA molecules

The concept of the saturation of the endogenous miRNA pathway following expression of high doses of synthetic interfering dsRNA molecules has been champi- oned by Grimm et al[47,126–129]. They showed that high doses of shRNA molecules delivered from an adenoviral vector into mice livers led to hepatocellular death and cancers[128]. Supported by utilising two independent publications[390,391],they hypothesised that Xpo-5 was the likely cause of the problem. This concept has important implications for the development of shRNA molecules that can potentially minimise interactions with vulnerable components of the microRNA pathway. For Chapter 1. Introduction 30 instance, modulating shRNA design to facilitate the bias of shRNA molecules away from incorporation into miRISC if possible, may allow for the maximization of the numbers of molecules without interference.

Xpo-5 was first identified as a RanGTP-dependent binding protein that strongly associated with dsRNA, and the association of dsRNA was a limiting factor in its efficacy of translocation between the nucleus and the cytoplasm[50]. The following year the connection between a dsRNA binding translocation protein and miRNAs was established and Xpo-5 was confirmed to be the protein responsible for translocation of miRNAs and shRNAs from the nucleus into the cytoplasm[44,130,391] in mammalian cells. However, the significance of the latter was established by the observation that over-expression of Xpo-5 in the human immortalised T-cell line (HEK293 T) [ATCC CRL-11268] significantly enhanced the effectiveness of observed silencing by both endogenous levels of miRNAs and by transfection of Pol III delivered shRNAs. Interestingly, no additional pathogenicity was observed in cells with these vectors and the vast amount of research utilising high doses of shRNA molecules in cell culture suggest that shRNA dose is less important in immortalised cells than in in vivo models.

Expression of discrete shRNAs from either transgenic constructs or plasmid vectors is dependent upon the endogenous RNAi pathway during export from the nucleus. It is logical to assume that this single protein-dependent step will be the most vulnerable to saturation and repression by the expression of exogenous shRNA molecules and this is precisely what was observed in the original studies. Xpo-5 has previously been shown to preferentially bind to a mini-helix structure in ncR- NAs[130] and this mini-helix or hairpin structure is a characteristic of both shRNAs and miRNAs, with the RNA-binding domain of the of the Xpo-5 protein binding to the RNA hairpin in a RanGTP-dependent manner[130]. Whilst there have been a number of publications which have identified that Xpo-5 is a limiting factor in either miRNA and / or shRNA transport[34,44,50,119,126,128–130,134,177,225,227,390,391,403], saturation of downstream cytoplasmic molecules such as the argonaute proteins have additionally been identified as responsible for some of the off-target phenotypes observed. Chapter 1. Introduction 31

1.7.1 Saturation of Argonaute

The Argonaute (Ago) proteins are catalytic proteins which are present in the RNA induced silencing complex (RISC)[214,270] as well as involved in non-canonical cleavage of pre-miRNAs into mature microRNAs[65,70,389]. Research into the saturation of Ago proteins has focused upon the presence of Ago within RISC and not on its role as a precursor cleavage RNase. However, while a number of studies have demonstrated that suppression of the most well known non-canonical miRNA, miR-451[65,70,389] is required for normal erythroid development in mice, zebraﬁsh and mammalian cell culture[218,280,281,359], it remains possible that this is one of the underlying mechanisms observed in early embryo toxicity induced by RNAi. This potential additional mechanism will have implications on the development of even shorter Ago-dependent, Dicer-independent silencing molecules as these molecules will utilise Ago2 as both a dependent RNase cleavage protein and during RISC incorporation[218].

As well as the potential of shRNAs and exogenous hairpin RNA molecules to out- compete endogenous miRNAs for access to Dicer and prevent or inhibit translocation from the nucleus into the cytoplasm, it has additionally been reported that both saturation of Xpo-5 and knockdown of Xpo-5 both significantly reduce Dicer mRNA levels in the cytoplasm, preventing its export and translocation [34]. This prevention of translocation is directly related to the presence of small RNAs that contain either a hairpin structure or a similar RNA-helix [34]. This suggests that the presence of Dicer within the cell is dependent on Xpo-5 availability and that there may be a secondary mechanism of action occurring following the over-expression of an XPo-5 substrate preventing normal processing of miRNAs. This Dicer-Xpo-5 binding competition is proposed to be additionally relevant in adenoviral infection where the adenoviral VA1 ncRNA has been repeatedly demonstrated to have an inhibitory effect on the RNAi pathway[34,37,90,130,132,223,225,391]. Prior to the connection between Xpo-5 availability and Dicer being made, it had been assumed that the VA1 RNA was acting as a competing Dicer substrate [130,223,225]. However, the presence of micro-helix structures within the VA1 RNA suggest that the Dicer Chapter 1. Introduction 32 inhibition maybe a downstream side effect of the Xpo-5 binding, and Dicer is generally not considered to be a primary checkpoint in the generation of miRNAs in the presence of high-doses of shRNAs[129] (reviewed in[126]).

Interestingly, current evidence suggests that the two RNase III enzymes involved in the miRNA pathway, Drosha and Dicer, appear free from saturation by excessive shRNA doses[129]. However, in the case of Drosha, all of the investigations to this time have utilised distinct shRNA hairpins expressed from either pol II or III promoters that enter the miRNA / RNAi pathway downstream of Drosha interaction. So, it remains possible that the expression of longer transcripts of pri-miRNAs that do require Drosha for processing and nuclear export by Xpo-5 will induce a similar effect. Resistance by Dicer to saturation is completely consistent with reports that over-expression of Dicer does not increase the efficacy of shRNAs or miRNAs[92] or siRNAs[367] in cell culture. These observations are additionally consistent with the delay in mortalities of both Dicer -\- mice[206,248,342,389] or zebrafish[31,70,250] compared to either Argonaute knockout[237] or miRNA saturation by shRNA expression[129]. However, a number of studies have suggested that Dicer is inhibited by expression of longer viral ncRNA transcripts, which can act as either Dicer RNase III substrates or potentially be inhibiting the RNAi pathway via Xpo-5 saturation. The Dicer inhibition is then a downstream consequence of Xpo-5 saturation and inefficient processing of miRNA molecules into the cytoplasm[34,37].

Most recently, stable integration of human Ago2 into three human cell lines HeLaP4 (a HeLa derivative overexpressing primary and secondary HIV-1 receptors), 293T (human embryonic kidney cells modified with the SV40 large T antigen) and Huh7.5 (a hepatocarcinoma cell line subclone with increased permissiveness for HCV)[47] resulted in increased RNAi efficiency and reduced off target effects[47].In addition, the addition of Ago2 expression cassette to lentiviral vectors expressing shRNAs, alleviated previously observed deleterious effects on miRNA expression and liver function and disease[47,128]. Chapter 1. Introduction 33

1.7.2 Viral saturation and inhibition of the RNAi pathway

Viral saturation of RNAi pathway members by expression of small RNA molecules was first observed following Adenovirus infection of HEK293 cells[14,223] as well as Hela cells[223], Adenovirus expresses two ncRNA molecules, VA RNAI and VA RNAII. Expression of the VA RNAI transcript from a CMV promoter resulted in ablation of mature miR-30 expression and to a lesser extent synthetic miR-30 expressed from a similar construct[223]. This affect was demonstrated to be relevant in viral replication in a following study where the effect on the synthetic and endogenous miR-30 were replicated using live virus[14]. However, the same VA RNAI ncRNA additionally perturbed the interferon response, and so the efficacy and importance of viral RNAi pathway suppression in vertebrates remains a relatively controversial topic.

It was originally theorised that viruses which solely infect vertebrate hosts may not encode suppression of RNA silencing (SRS) molecules, since interferon induction following accumulation of dsRNA within the cytoplasm generates a rapid and powerful inhibitory response on viral and host gene expression via MxA and protein kinase PKR signaling (reviewed in[54,352,368]). There is an array of viral re- pressors of the interferon system and these methods are again excellently reviewed in[292].The effectiveness of the interferon system at modulating gene expression following dsRNA accumulation has led to the relevance of the mammalian RNAi system in response to viral infection becoming a controversial topic within this field. So while the RNAi system is conserved, the absence of an amplification step generated from an RNA-dependent RNA polymerase (RdRp) - such as that conserved within invertebrates[176,222,365,394] and plants[84,254] - in mammalian cells has led to suggestions that any Dicer-derived RNAi response will be unimportant when compared to the interferon system.

This theory is, however, slowly bowing to weight of evidence that the RNAi response is important in vertebrate anti-viral responses. In particular, back-to-back publications in Science in October 2013 by Maillard et al [230] and Li et al [210] as well as additional work by Li Seo et al [329] suggest that RNAi interference of viruses is Chapter 1. Introduction 34 fundamental to the normal anti-viral response in primary undifferentiated mouse cells as well as hamster cells and suckling mice, respectively. These publications have come under significant criticism from tenOever et al [78]. However, these potential responses must be taken into account when discussing modulation of the RNAi pathway to express artificial RNAs that may interfere with the endogenous RNAi pathway.

1.7.2.1 Antiviral RNAi in ﬁsh

At this time, there is a paucity of data about the importance of the RNAi response to viral infection in fishes. These data may prove crucial in determining whether RNAi transgenes and therapeutics have a future as potential treatments for viral disease in aquaculture, particularly when the reduction of efficacy in low temperatures on the interferon response in teleost fishes is accounted for[93]. However, there has been a single publication dealing with a member of the betanodaviridae family, Greasy grouper nervous necrosis virus (GGNNV). The betanodaviridae [107] viruses encode a small non-structural B2 protein that appears to inhibit the RNAi response by preventing Dicer cleavage. This has been additionally demonstrated in the familial virus Nodamura Virus(NoV)[167] and viruses from the related Alphan- odaviridae family, including Flock House virus (FHV) that also encode a conserved B2 nonvirion protein.

Dios et al demonstrated that accumulation of the viral RNA1 and RNA2 RNAs within the cell correlated with reduced eﬃcacy of shRNA induced knockdown of a co-transfected eGFP marker[107]. This publication was almost immediately followed by a second supporting paper in which a second betanodavirus, Striped jack nervous necrosis virus (SJNNV), was also shown to express a B2 protein. Viral extracts containing the B2 protein in anAgrobacterium-mediated transient expression system which was inﬁltrated into leaves of Nicotiana benthamiana (N. benthamiana) demonstrated comparable silencing of eGFP reporter proteins to the siRNA control[160]. Chapter 1. Introduction 35

These two publications provided significant supporting evidence that the B2 protein is a strong suppressor of RNAi silencing in fish. However, to the knowledge of the author there has been no follow up work in fish viruses. The second publication in particular leaves significant room for improvement. The differences between plant and vertebrate RNAi responses, specifically the presence of multiple plant Dicers, and the intercellular nature of plant RNAi responses and intercellular small RNA and RNAi signaling in plants, make these results interesting but inconclusive as to the physiological relevance of these anti-RNAi interactions in teleosts.

Saturation of the RNAi system by over-expression of shRNA molecules has pheno- typically similar outcomes to the effect of a number of strategies employed by an array of viruses to avoid this pathway. While these strategies are most common in viruses with invertebrate hosts, there is significant evidence that vertebrate viruses employ similar strategies in conjunction with interferon avoidance and immune modulation to avert the host immune system. While there is minimal evidence for the relevance of the RNAi system as a potent anti-viral response in teleosts, the presence of a conserved nodavirus B2 protein with an anti-Dicer effect suggests that the response is important enough for viruses to have developed and retained defenses against this ancient regulatory system. Coupled with deep sequencing evidence that siRNAs, along with miRNAs, piRNAs and other small ncRNAs are generated by Dicer proteins and have physiologically relevant effects within the cell, this demonstrates the need for further research within this field. This research becomes important for deeper understanding of the anti-viral response, and host-virus interactions as well as being an obligatory requirement before RNAi transgenes or treatments could potentially become a reality as a treatment for disease in finfish.

To summarise, it has become apparent that there is variation in the observed effects of shRNA / siRNA molecules upon the endogenous RNAi following transfection and expression. Observations in vitro and in vivo systems suggest that there is unlikely to be a single answer for the generation of RNAi transgenics free of side-effects and off-target effects. The complexities of the RNAi pathway, shRNA sequence effects, changing loops and including pri-miRNA type sequences Chapter 1. Introduction 36

3 and 5 as well as changing and varied dependence and sensitivities of different cell types for normal levels of miRNAs all need to be considered. The export proteins Xpo-5 as well as the catalytic Argonaute proteins have been proposed as being the primary molecules vulnerable to saturation by RNAi molecules across a range of models. Viral inhibitors of the RNAi response occur in a wide variety of hosts, and act to inhibit this pathway in a similar manner by including Dicer or RISC substrates. This suggests that this is a powerful mechanism for inhibiting an RNAi response, and that it can prevent normal host-processes including viral defenses from acting upon the virus. To avoid these effects, dose-modulation of shRNAs by varying promoter choice, incorporating shRNAs as miRNAs or alongside endogenous miRNAs without additional promoters, and modifying the shRNAs to include specific loops and sequences have each been utilised with varying degrees of success at reducing off-target effects from shRNAs. However, despite this progress, further work is needed before shRNA silencing molecules can become palatable to regulators and the general public as safe and effective therapeutics.

1.8 Research rationale and summary shRNAs represent a powerful approach to elicit an anti-viral response. However, it is clear that the levels at which shRNAs are transcribed within cells is also fundamental to efficient and safe gene knockdown. As a consequence, the development of an optimal expression unit is crucial to the development of potential transgenes and treatment utilising RNAi, and the identification of effective species-specific promoter sequences are critical. Considering that currently the use of RNAi in the zebrafish is both controversial and poorly-developed compared to other species, it will be advantageous to develop this technology for application in zebrafish and teleosts more broadly. The goal of the current study was to develop an optimised expression system for the expression of shRNAs in zebrafish cells, and then to utilise these optimised expression constructs to test novel shRNA molecules, first in cell culture and then in transgenic zebrafish expressing antiviral constructs. Chapter 1. Introduction 37

Analysis of the literature suggests that there is a paucity of data comparing the effectiveness of pol III promoters in native and non-native cells. The use of a zebrafish promoter for shRNA expression therefore represents an opportunity to mediate enhanced and controlled delivery. Since the use of type 3 RNA pol III promoters for shRNA expression has become widespread and extremely successful, one aim of this study was to characterise previously identified pol III type 3 promoters in zebrafish, as well as to identify any new promoters of this subtype for the expression of shRNAs in zebrafish cells. Such native promoters may result in enhanced target gene suppression compared to native type 3 RNA pol III promoters, and may indeed be required for the transgenic application of RNAi in zebrafish.

RNAi can be a highly eﬀective technique for the suppression of virus replication and infection. However, the high mutation rate of RNA viruses in particular, such as rhabdoviruses, allows escape from RNAi strategies targeting only a single sequence[36,189,372] due to the high sequence speciﬁcity of RNAi. Therefore, a strategy where multiple sequences are targeted is required before the generation of transgenic animals resistant to viral infection can be considered. A successful inhibition strategy involves careful selection of the molecules that will be incorporated into the construct and also the type of construct that will be used to express these molecules. This study also aims to further the knowledge of the use of RNAi for the targeting of genes in the context of creating antiviral transgenes capable of simultaneously targeting multiple viral genes and therefore preventing the risk of viral escape.

Finally, in vitro systems act as models and guides to test expression and effectiveness of individual components of more complex systems. The incorporation of the identified endogenous zebrafish promoters along with shRNA sequences derived in vitro were combined into Tol2 vectors and microinjected into zebrafish embryos at the 1 cell stage[186]. These original vectors resulted in universally lethal non-specific mutations to all fish. Following the hypothesis that saturation of the miRNA pathway was causing these non-specific mutations, weaker H1 promoters Chapter 1. Introduction 38 were utilised and the previously characterised promoters further modified to ablate expression by specific deletion of motifs. These modified promoters showed a significant relative reduction of lethal and abnormal phenotypes as well as a corresponding reduction in shRNA expression. To test the theory that saturation of one or more RNAi pathway components was responsible for the observed phenotypes, two components of the RNAi previously identified as vulnerable to saturation; Argonaute and Xpo-5 were over-expressed and this over-expression also significantly reduced embryo lethality and abnormal morphology. Chapter 2

Materials and Methods

Laboratory work was split over two sites, CSIRO AAHL and Deakin University Medical School, and there were diﬀerences in standard laboratory protocols and available equipment at each site. Therefor to limit diﬃculties with analysis, each individual experiment were performed at a single site and there were no changes of equipment used within individual experiments; in particular, all RT-qPCR were performed at CSIRO AAHL.

2.1 Materials

2.1.1 Suppliers

The following companies supplied reagents, equipment and software utilised in this study:

Adobe Systems Software Australia Ltd, Sydney, NSW, Australia Aglient Technologies, Santa Clara, CA, USA Ambion, Austin, TX, USA Amersham Biosciences, Piscataway, NJ, USA Addgene, Cambridge, MA, USA

39 Chapter 2. Materials and Methods 40

Applied Biosystems, Foster City, CA, USA Astral Scientific, Gymea, NSW, Australia BD Biosciences, North Ryde, NSW, Australia Beckman Coulter, Fullerton, CA, USA Becton Dickinson Labware, Franklin Lakes, NJ, USA Bio-Rad Laboratories, Hercules, CA, USA Boehringer-Mannheim, Mannheim, Germany Eppendorf, Hamburg, Germany GE Healthcare, Chalfont St Giles, Bucks, United Kingdom Geneworks, Hindmarsh, SA, Australia Graph Pad Software, La Jolla, CA, USA Interpath, West Heidelberg, VIC, Australia Invitrogen, Mount Waverly, VIC, Australia Leica Microsystems GmbH, Wetzlar, Germany Life Technologies, Rockville, MD, USA Merck-Calbiochem, Darmstadt, Germany Millipore, Billerica, MA, USA Minitab Inc, PA, USA Nalge Nunc International, Roskilde, MA, USA New England Biolabs, Beverly, MA, USA Olympus, Shinjuku, Tokyo, Japan Pacific Laboratory Products Pty Ltd, Blackburn, VIC, Australia Promega, Madison, WI, USA Qiagen, Germantown, MD, USA Roche Diagnostics, Castle Hill, NSW, Australia Sigma-Aldrich, Castle Hill, NSW, Australia Thermo Fisher Scientific, Waltham, MA, USA Whatman, Middlesex, United Kingdom Chapter 2. Materials and Methods 41

2.1.2 Bacterial strains, plasmids and media

All cloning and vector ampliﬁcation were performed in Escherichia coli (E. coli) DH5a Top10 electrocompetent cells (Invitrogen) or XL10-Gold Ultracompetent chemically competent cells (Stratagene). All E. coli cells were grown in Luria Bertani (LB) broth or on LB agar supplemented with the following antibiotics and supplements when required: 100 μg/ml ampicillin (Sigma), 25 μg/ml kanamycin (Sigma), 100 μg/ml X-gal (Promega) and 1 nM Isopropyl-β-D-1-thiogalactopyranoside (IPTG) [Sigma]. Liquid cultures were shaken at 250 rpm at 37 ◦C in a shaker and aliquots of broth cultures were preserved at −80 ◦C in 25% (v/v) glycerol (Sigma). E. coli plated onto solid media were allowed to grow over night at 37 ◦Cinair before colonies were picked.

Plasmids pEGFP-N1 and pGEM-T Easy (Promega) were obtained from Promega. pUC118 were a kind gift from Dr. Scott Tyack (CSIRO, [AAHL]). The plasmids pCMV-Tol2 and pMini-Tol[25] were donated as absorbed ﬁlter paper dots by Professor Stephen Ekker (Dept. Biochemistry and Molecular Biology. Mayo Clinic Cancer Center, Rochester, Minnesota, USA). The shRNA expression plasmids pCh-U64-shEGFP and pMU-shEGFP were provided by Dr. David Cummins (CSIRO, AAHL, Geelong, Australia) and the pFugu-shEGFP[401] shRNA expression plasmid were generously donated by Professor Ki Hong Kim (Department of Aquatic Life Medicine, Pukyong National University, Korea). Plasmids pQE60- Exp5[50] and pIRESneo-FLAG/HA Ago2[247] were purchased from Addgene.

2.1.3 Primers, probes and oligonucleotides

A list of all primers and oligonucleotide probes used in this study is included in Appendix A. Polymerase chain reaction (PCR) primers were designed using Clone Manager 9.0 software (SciEd Central) and obtained from Geneworks. PCR primers were designed according to the following rules: 18 nt to 24 nt, 40-60% GC content, 50 ◦Cto65◦C melting temperature, 1 nt to 2 nt GC clamp, maximum polynucleotide repeats (N) = 4. All primers were received as lyophilised DNA Chapter 2. Materials and Methods 42 and were re-suspended in nuclease free water (nfH2O) to a stock concentration of 100 μM. Working primer stocks were prepared at 5 μM for quantitative reverse transcriptase PCR (qRT-PCR). One step PCR primers used for construction of shRNAs (Figure 2.4.6.1) were additionally puriﬁed by High Performance Liquid Chromatography (HPLC). Locked nucleic acid (LNA) probes were obtained unla- beled from Sigma-Aldrich.

2.2 DNA methods.

2.2.1 Transformation of E. coli

2.2.1.1 Using electrocompetent E. coli

A40μl aliquot of electrocompetent E. coli cells were thawed on ice and 2 μlof either ligation reaction or diluted plasmid DNA was added. The mixture was then transferred to a cold 0.2 cm electroporation cuvette (BioRad) and pulsed using a Gene PulserTM Transformation Apparatus (Bio-Rad) under the following conditions: 25 μF capacitance, 2.25 kV and the Pulse Controller set to 200 Ω. 1 ml of warmed (37 ◦C) LB broth were added to the transformation mix immediately and the total volume transferred into a 10 ml falcon tube. The suspension were agitated for 1 h at 37 ◦C and then approximately 100 μl (if circular plasmid) or 100 μl and 900 μl (if ligation) suspension were plated onto LB agar plates with the appropriate antibiotics and incubated overnight at 37 ◦C.

2.2.1.2 Using chemically competent E. coli

An 100 μl aliquot of Stratagene XL10-Gold Ultracompetant E. coli cells were thawed on ice and between 5 μlto10μl of either a ligation reaction or plasmid DNA was added and incubated on ice for a further 15 min before a 45 s heat-shock at 42 ◦C. Shocked cells were added to a pre-warmed 800 μl aliquot of LB broth, and the cells were incubated at 37 ◦C for 1 h with shaking at 200 rpm. Cells were Chapter 2. Materials and Methods 43 plated onto LB agar plate containing appropriate antibiotic and once dry were incubated overnight at 37 ◦C.

2.2.2 Plasmid puriﬁcation

2.2.2.1 Small-volume isolation of plasmid DNA from E. coli

Small-volume plasmid DNA extraction from E. coli were performed using the QIAprepTMSpin Miniprep Kit (Qiagen) as per the manufacturer’s instructions.

2.2.2.2 Large-volume isolation of plasmid DNA from E. coli

For large-volumes of plasmid DNA, the PureYieldTMPlasmid Maxiprep system (Promega) were used according to the manufacturer’s instructions.

2.2.3 Nucleic acid manipulation

2.2.3.1 Restriction endonuclease digestion

Digestion of DNA into fragments for cloning or mRNA transcription were accomplished using a variety of restriction endonucleases (Promega, New England Biolabs or MBI Fermentas) following the speciﬁc manufacturer’s instructions for the enzyme(s). For the most part, digestions were performed at 37 ◦C in volumes of between 10 μland20μl, which contained between 0.5 μgto2.0 μg of DNA, and were incubated for 1 h to 3 h or overnight. The reactions were stopped if recommended by a further incubation of 10 min to 30 min at 65 ◦Cto80◦C.

2.2.3.2 Manipulation of restriction endonuclease digestion fragments

To facilitate blunt-end cloning a number of restriction endonuclease digestions re- cessed 3 -terminal ends were ﬁlled in using the DNA Polymerase I Large (Klenow) Chapter 2. Materials and Methods 44

Fragment (Promega) following the manufacturer’s instructions. To improve selection of correctly inserted DNA fragments and to prepare DNA fragments for further manipulation removal of 5 phosphate groups was performed using Tem- perature Sensitive Alkaline Phosphatase (Promega) following the manufacturer’s instructions.

2.2.3.3 Separation of nucleic acid fragments by agarose gel electrophoresis

For analytical and preparative purposes, DNA and RNA samples were separated on 1% to 3% (w/v) agarose gels in 1 x TAE buﬀer (40 mM Tris (pH 8.0), 20 mM acetic acid, and 1 mM EDTA) to aid visualisation of nucleic acid under ultraviolet light 1 μl of either GelRedTM (Biotium) or SybrSafeTM (Invitrogen) per 100 ml of TAE agarose solution was added immediately before pouring the gel. To each nucleic acid sample, 6 x DNA loading dye (Fermentas) or 2× RNA loading dye (Fermentas) was added before separation by electrophoresis at between 60 V to 120 V for between 30 min to 120 min. To achieve separation of large fragments a number of samples were run overnight at 50 V. Six μl of preprepared molecular weight marker were loaded in a separate lane to indicate the size of the separated products. The markers used were, 1 kb GeneRulerTM, 1 kb Plus DNA Ladder (Fermentas) for all PCR and plasmid DNA, and RiboRulerTM High Range RNA Ladder for RNA. After separation, the resolved DNA were visualised using a UV- transilluminator.

2.2.3.4 Extraction of DNA from agarose gels

Isolation of DNA from agarose gels following electrophoresis was performed using the QIAquick Gel Extraction Kit (Qiagen) following the manufacturer’s directions. DNA were eluted with 30 μl of EB buﬀer (10 mM Tris-HCl, pH 8.5). Chapter 2. Materials and Methods 45

2.2.3.5 DNA ligations

In general, ligations were performed with a 3:1 molar ratio of insert to vector DNA. Ligation reactions consisted of one unit of T4-DNA ligase (Promega) or Quick Ligase (New England Biolabs) and 1 × ligase buﬀer supplied by the manufacturer, they were incubated at room temperature for 2 h to 4 h or at 16 ◦C overnight.

2.2.4 Polymerase Chain Reaction ampliﬁcation of DNA

Polymerase Chain Reaction (PCR) were used for the amplification of DNA regions from template DNA sequences using sequence-specific anti-parallel oligonucleotides (primers) flanking the amplified region and either Thermus aquaticus (Taq) or Pyrococcus furiosus (Pfu) DNA polymerase. PCR amplification were performed using Promega GoTaqR Green HotStart Master Mix (Promega) or New England Biolabs Phusion High-Fidelity DNA Polymerase Master Mix. Generally 10 n g to 100 n g of cDNA, gDNA or plasmid DNA were added to 10 μlof2× Master

Mix with 1 μl of 10 mM speciﬁc primers and made up to 20 μlwithnfH2O. Where larger quantities of product were required the components were increased propor- tionally. Where standard PCR conditions did not result in expected products, gradient PCR, touchdown PCR or PCR additives Dimethyl sulfoxide (DMSO), and/or Bovine Serum Albumin (BSA) or a combination of the above were used to optimise reactions. In particular ampliﬁcation of large ≥ 3 kb products with Pfu were performed with the addition of 0.6 μl DMSO per 20 μl reaction.

2.2.5 Cloning of PCR products

Taq DNA polymerase has terminal deoxynucleotidyltransferase activity, that causes the addition of an extra deoxyadenosine (dA) at the 3-end of the PCR product. This 3-dA overhang facilitates direct cloning of PCR fragments into pGEMR - T Easy (Promega A1360) as this vector has a complementary 3-deoxythymidine (dT). Complementary endonuclease cleaved ends were used to facilitate cloning Chapter 2. Materials and Methods 46

Taq PCR conditions Pfu PCR conditions

95 ◦C for 2 min 98 ◦C for 30 s

⎫ ⎫ ◦ ⎪ ◦ ⎪ 95 C for 15 s ⎬⎪ 98 C for 15 s ⎬⎪ 55 ◦Cto62◦C for 30 s 30 Cycles 60 ◦C for 15 s 30 Cycles ⎪ ⎪ ⎭⎪ ⎭⎪ 72 ◦C for 1 min/kb 72 ◦C for 15 s/kb

72 ◦C for 10 min 72 ◦C for 10 min 4 ◦C 4 ◦C

Table 2.1: Typical PCR conditions into vectors other than pGEMR -T Easy. All ligation reactions were performed according to the manufacturer’s instructions as detailed in 2.2.3.5 on 45.

2.2.6 PCR cleanup

Cleanup of PCR products were performed using the QIAquick PCR cleanup kit (Qiagen) following the manufacturer’s directions. PCR products were eluted with 30 μl of EB buﬀer (10 mM Tris-HCl, pH 8.5).

2.2.7 DNA sequencing

Puriﬁed plasmid DNA and PCR products were sequenced prior to further use. All infectious samples were sequenced using the ABI PRISM 377XL automated DNA Sequencer (Applied Biosystems) at CSIRO, AAHL Geelong. Non-infectious samples were sequenced at Micromon DNA sequencing (Monash University) on a capillary-platform Genetic Analyser (Applied Biosystems). Chapter 2. Materials and Methods 47

2.2.8 Determination of nucleic acid concentration

The concentration of nucleic acid samples was determined using the NanoDropR ND-1000 Spectrophotometer coupled to an Optiplex GX280 computer (Dell). The machine were blanked using distilled and autoclaved water, and the concentration of a 2 μl aliquot of either nuclease-free water or EB buﬀer, as appropriate, were measured to ensure a low level of background before 2 μl aliquots of nucleic acid samples were measured.

2.2.9 Tol 2 insertion site analysis PCR

2.2.9.1 Genomic DNA (gDNA) extraction

TM gDNA was extracted from zebraﬁsh using the QuickExtract DNA Extraction kit (EpiBio, illumina) as per the manufacturer’s protocol.

2.2.9.2 Cassette Method six individual restriction enzyme digests were setup as follows: 5 μg gDNA, 50

U restriction enzyme, 5 μl10× buffer, 5 μl 100 × BSA if required, nfH2Oto 50 μl. Digestions were incubated overnight at 37 ◦C then stabilised with 5 μl3M Sodium Acetate (pH 5.2) and precipitated with 137.5 μl cold EtOH and incubated at −20 ◦C for 1 h. Precipitate were rinsed again with −20 ◦C EtOH and the pellet ◦ left to dry before reconstitution with 10 μl55CnfH2O. Ligation of restriction enzyme and enzyme specific endcassettes were performed using NEB Quick Ligase as per 2.2.3.5 at a ratio of 1:2 digested gDNA to cassette. 1 μl ligated gDNA ◦ was added to 33.5 μlnfH2O and dissolved at 94 C for 10 min and amplified using GoTaq Hotstart PCR with Cassette Primer 1 and Tol2 Primer 1.

To perform the secondary PCR 1 μlofcleaned2.2.6 PCR product was diluted

1:10,000 with nfH2O and a second PCR performed using Cassette Primer 2 and Tol2 Primer 2 using identical PCR conditions as the previous PCR. Chapter 2. Materials and Methods 48

95 ◦C for 2 min

⎫ ◦ ⎬ 98 C for 20 s ⎭ 30 Cycles 68 ◦C for 12 min

72 ◦C for 10 min 4 ◦C

Table 2.2: PCR conditions for insertion site characterisation

2.3 RNA Methods

2.3.1 RNA isolation and puriﬁcation

2.3.1.1 Total RNA extraction from cells or zebraﬁsh optimised for small RNAs using TrizolTM

Confluent cell monolayers were washed twice in PBSA, followed by addition of 1 ml of TRIzol Reagent (Invitrogen). Whole zebrafish larvae upto 7 days post fertilisation were homogenised by freezing in 1 ml TRIzol reagent prior to the addition of 200 μl glass beads and homogenisation by bead beating on a Precellys homogeniser for 15 min at 6000 rpm. All subsequent steps were identical for either whole zebrafish or cells. To separate the RNA,DNA and protein, 200 μl chloroform was added to each sample, and mixed by shaking for 15 s. Samples were then incubated at room temperature for 3 min followed by centrifugation at 12 000 g at 4 ◦C for 15 min. The aqueous phase, containing the RNA, was transferred to a clean RNasefree tube to which 10 μg UltraPure Glycogen (Invitrogen) and 500 μl isopropanol were added. Samples were mixed and incubated for 10 min at room temperature, followed by centrifugation at 12 000 g at 4 ◦C for 15 min the supernatant were removed, and the pellet containing the RNA were washed with 1 ml 80% (v/v) ethanol with vortexing, followed by centrifugation at 7500 g at 4 ◦C for 5 min. Ethanol was removed and the pellet was partially air-dried before ◦ resuspension in 30 μl nucleasefree H2O(nfH2O) preheated to 55 C, incubated at Chapter 2. Materials and Methods 49

55 ◦C for 5 min before being placed on ice. RNA samples were further puriﬁed by addition of 90 μl 100% ethanol, 3 μl sodium acetate, and 10 μg UltraPure Glycogen, mixed by pipetting and incubated for 24 h at -20 ◦C with subsequent precipitation and washes performed as above.

TM 2.3.1.2 Total RNA extraction using RNeasy

Isolation of RNA from adherent cells, or homogenised zebrafish or zebrafish organs or fin clips (where small RNAs optimisation and enrichment was not required) was performed using the RNeasy Mini Kit (Qiagen) following the manufacturer’s directions. RNA was eluted with 30 μl of EB buffer (10 mM Tris-HCl, pH 8.5) by centrifugation.

2.3.2 Extraction of viral RNA

Isolation of viral RNA from adherent cell supernatants, zebrafish or zebrafish organs or fin clips were performed using the ViralRNA Mini Kit (Qiagen) following the manufacturer’s directions. Homogenisation of zebrafish organs or tissue were accomplished as per 2.3.1.1 RNA were eluted with 30 μl of EB buffer (10 mM Tris-HCl, pH 8.5) by centrifugation.

2.4 RNase protection assay (RPA)

2.4.1 Radioactive labeling of probe and marker RNA

RNA oligonucleotides complementary to the sense strand of shEGFP[27] were syn- thesised, and diluted to 1 pmol/μl with nuclease-free water. These were end- labeled with 32Pγ-dATP (Amersham Biosciences) using the mirVana Probe & Marker Kit (Ambion) following the manufacturer’s instruction. Brieﬂy, 1 μlofthe diluted probe were end-labeled with 3.34 pmol of 32Pγ-dATP 10 mCi/ml, 10 Units Chapter 2. Materials and Methods 50 of T4 PNK, 1× kinase buﬀer, and water up to 10 μl. In addition to the shEGFP probes, the miRNA miR-16 RNA probe included as a positive control were also end-labelled with 32Pγ-dATP using an identical reaction. For an indication of the size of separated RNA fragments by electrophoresis, the Decade Marker (mirVana Probe & Marker Kit, Ambion) was similarly end-labelled.

2.4.2 Separation of RNA on denaturing agarose gels and autoradiography

Electrophoretic separation of RNA was performed on denaturing 7 M Urea, 15% (w/v) acrylamide gels. These were prepared as follows: 12.6 g of Urea, 3 ml of 10 x TBE buffer (0.9 M Tris base, 0.9 M Boric acid, 20 mM EDTA pH 8.0), and 11.25 ml of 40% (w/v) acrylamide/bis (19:1) made up to 30 ml with nuclease-free water. With stirring to dissolve the urea, and then 150 μl 10% (w/v) ammonium persul- phate (APS) and 20 μl of N, N, N”, N”-tetramethylethylene diamine (TEMED) were added. Gels were left at room temperature for approximately 1 h and then run in a Vertical Gel Electrophoresis System (Bethesda Research Laboratories) at 40 mA for 1 hour in 1 × TBE prior to loading of the RNA samples. All RNA samples were mixed with Gel Loading Buffer II (mirVana Probe & Marker Kit, Ambion) and heated at 95 ◦C prior to loading. Separation was achieved by running gels at 40 mA for approximately 2 h. Gels were then enclosed with one sheet of Hy- perfilm ECL (Amersham Biosciences) in an EC-AWU Cassette (Fuji) and stored at -80 ◦C for the required exposure time. The exposed film were then removed and developed using an X-ray Processor FPM-100A (Fuji). Chapter 2. Materials and Methods 51

2.4.3 Quantitation of RNA

2.4.3.1 DNase treatment of RNA samples for cDNA synthesis.

Total RNA was used for cDNA synthesis prior to qRT-PCR. Before cDNA synthesis, all RNA samples were treated with DNaseI (RQ1, Promega) to remove potential genomic DNA contamination. Reactions were performed using 8 μl RNA, 1 μlof10× RQ1 Reaction Buﬀer (Promega), and 1 μl RQ1 DNaseI (Promega). Then incubated at 37 ◦C for 30 min, then 1 μl of RQ1 STOP solution (Promega) were added before a further 10 min incubation at 65 ◦C. DNaseI treated samples were used immediately for cDNA synthesis, or stored at −20 ◦C.

2.4.3.2 First-strand cDNA synthesis

DNAse treated RNA samples were poly-adenylated according to the approach described previously[332]. Poly-adenylated reactions contained 8 μl (approximately 1 μg) of DNAseI treated total RNA, 0.25 μl (150 U) of yeast Poly(A) Polymerase (PAP) (USB), 4 μl5× PAP Reaction Buffer and 1 μlof10nMrATP(Ambion) and nfH2O (Promega) in a final volume of 20 μl. Reactions were incubated at 37 ◦C for 30 min then at 95 ◦C for to 5 min terminate the reaction. They were used immediately for first-strand cDNA synthesis. First strand cDNA synthesis were performed using Superscript III (SuperscriptTMIII First Strand Synthesis Su- permix [Invitrogen]) according to manufacturer’s instructions for oligo-dT primed RNA. For qRT-PCR, cDNA synthesis was primed using the modified oligo-dT primer miR-PTA. The annealing reaction contained 4 μl of poly-adenylated RNA, 3 μlof25μM miR-PTA and 1 μl of Annealing Buffer and were incubated at 65 ◦C for 5 min. First-strand cDNA synthesis reactions primed with miR-PTA were incubated at 50 ◦C for 50 min, and inactivated by incubation at 85 ◦C for 5 min. Chapter 2. Materials and Methods 52

2.4.3.3 Quantitative PCR of shRNA and microRNA expression

Each qRT-PCR was performed using SYBR Green detection reagents and the StepOnePlus real-time PCR system (reagents, machine and software, Applied Biosystems). Reactions were analysed in triplicate in 96-well format (MicroAmp PCR plates, Applied Biosystems). According to optimised parameters, cDNA were used at either 1:30 or 1:50 dilution and all primers were used at a final concentration of 200 n M. Every reaction contained 2 μl cDNA, 10 μl SYBR Green PCR Master Mix (Applied Biosystems), 0.8 μl of forward target sequence specific primer, 0.8 μl of PAM-URP and nfH2O to a final volume of 20 μl using the following cycle conditions.

◦ 95 C for 10 min ⎫ ◦ ⎬ 95 C for 15 s ⎭ 40 Cycles 60 ◦C for 1 min

Melt curve analysis

⎫ ◦ ⎬ 95 C for 15 s ⎭ 40 Cycles 60 ◦C for 1 min ramp +0.3 ◦C

95 ◦C for 15 s 4 ◦C

Table 2.3: PCR conditions for miRNA / shRNA quantitation

2.4.4 Quantitative reverse-transcriptase analysis of mRNA expression qRT-PCR analysis of mRNA were performed as per 2.4.3.3 with the substitution of a second specific primer for the PAM-URP primer and modification of the Chapter 2. Materials and Methods 53 annealing temperature as required by specific primer pairs. Zebrafish 5S RNA (5S) was used as the reference control for normalising expression levels.

2.4.5 Data analysis for qRT-PCR

Each qRT-PCR analysis were performed using the comparative Ct method, also known as the 2-ΔΔCt method[219,315]. Which, generates an expression profile for the target that is normalised against a reference gene. For miRNA qRT-PCR results in this thesis, expression levels are shown as a fold difference relative to the level of a selected reference sample, where the fold difference of the reference sample has a value of 1.

Data analysis Ct values for qRT-PCR were calculated by the StepOnePlus software (Applied Biosystems) based on a set Ct threshold of 0.1. For 2-ΔΔCt calculations, ampliﬁcation data were then downloaded in tab delimited format from StepOne- Plus software and imported into Microsoft Excel (Microsoft Corporation). The following calculations were then performed for each target per sample:

Step Calculation Excell formulae 1 Mean Ct target/ref. gene =AVERAGE (cell range for triplicate Ct values) 2 ΔCt sample/ref. sample =(mean Ct target)-(mean Ct ref. gene) 3 ΔΔCt sample =(ΔCt sample)-(ΔCt ref. sample) 4 FD value =POWER(2,-(ΔΔCt sample))

Table 2.4: Fold Change calculations

Mean fold differences (± standard error of the mean, SEM) were calculated for each target from 3 biological replicates per sample, each replicate obtained from an independent transfection or injection. Within each experiment two or three replicate reactions were performed per sample. Statistical comparisons between samples (i.e. control versus treatment) were performed using an un-paired student t test or a one-way ANOVA, where the threshold for statistical significance was set at 0.05 (Minitab v16). Graphs were generated using Prism 5 software (GraphPad) Chapter 2. Materials and Methods 54 depicting sample/treatment on the X axis and relative expression (fold) on the Y axis. Asterisks indicate the level of statistical significance ( p ≤ 0.05).

2.4.6 shRNA design and analysis

2.4.6.1 shRNA design and construction

All oligonucleotides used in the construction of shRNAs are described in Table 1. Six shRNAs, three against each of G and L genes, were designed against VHSV strain (23.75) [FN665788.1] were designed using the Dharmacon siRNA Design tool. Sequences were designed using the formula (NN) N21 nt and a total GC content of less than 50%. To generate eﬀective shRNAs, siRNA sequences were further screened using the criteria outlined previously[353]. Brieﬂy potential siRNAs were scored for the stability of the duplex formed between nucleotides 6 - 11 on the sense and 14 - 9 on the antisense strands and siRNAs with a central duplex ΔG of ≥−12.9 were discarded. ΔG scores were calculated as described in[114] Table 4. Remaining potential siRNAs were screened for sense 5 GC and 3 AT content, and high scoring siRNAs were further screened for the presence of conserved regions across strains of VHSV to minimise the chances viral escape. siRNA sequences were also screened against the related virus IHNV. siRNA pairs were joined with the loop sequence 5 TTCAAGAGA 3 [51], and followed by 5 tyrosine (T) bases and on XbaI restriction site. shRNA expression plasmids were constructed using the one-step PCR method as described[71,383] Chapter 2. Materials and Methods 55

2.5 Tissue culture methods

2.5.1 Immortalised eukaryotic cultured cell lines

Zebrafish embryo fibroblast (ZF-4) cells (ATCC CRL-2050) were maintained in 50% Dulbecco’s Minimal Essential Media / Hams F12 media (Gibco) containing 10% foetal bovine serum (FBS), 2 μM glutamine, 10 mM sodium bicarbonate, 10 mM N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (HEPES), supplemented with penicillin 100 μg/ml and streptomycin 100 μg/ml. Cells were grown ◦ (except when necessary for virus growth) at 28 C, 5% CO2 in a humidified atmosphere. African green monkey kidney (VERO) cells (ATCC CCL-81) were grown in Eagles Minimal Essential Media supplemented with 10% FBS, 2 μM glutamine, 10 mM sodium bicarbonate, 10 μM HEPES, penicillin 100 μg/ml and streptomycin 100 μg/ml. Epithelioma Papulosum Cyprini Flat-head minnow (EPC) epithelial cells were cultured in Leibovitz - 15 (L15) media, with 10% FCS, 2 μM glutamine, 10 mM sodium bicarbonate, 10 μM HEPES, penicillin 100 μg/ml and streptomycin 100 μg/ml at 25 ◦C in air in a humidified atmosphere. Cells were cultured as required in 25 cm2,75cm2 and 150 cm2 cell culture flasks (Corning) or in 6, 24 or 96 well plates.

2.5.2 Harvesting adherent cell lines

To harvest cell monolayers, growth media were removed, cells were washed 3 × using phosphate buﬀered saline-A (PBSA), and appropriate volumes of 0.25% (w/v) trypsin with 380 μg/l ethylene-diaminete-tetra-acetic acid (EDTA) [Gibco]. ZF-4 cells were harvested by adding 0.25% (w/v) trypsin without EDTA to the monolayer (Gibco). After 10 min to 30 min at the appropriate incubation temperature, the monolayer were completely detached by gently shaking the culture vessel, and the trypsin inactivated by addition of normal growth media. All cells were counted using a hemocytometer and plated as appropriate for experiments. Chapter 2. Materials and Methods 56

2.5.3 Transfection of eukaryotic cells

2.5.3.1 Electroporation of plasmid DNA

Transfection of plasmid DNA into ZF-4 and EPC cells were accomplished via electroporation using the Nucleofector 2b electroporator (Amaxa CD-MN029) following the manufacturer’s instructions. Briefly, cells were harvested by trypsinization as described in section 2.5 and aliquots of 2 × 106 cells were pelleted using a bench top centrifuge at 2000 g . Cell pellets were washed twice in PBSA, resuspended in 100 μl of electroporation solution V and with 4 μg of plasmid DNA, gently flick mixed and pipetted into electroporation cuvettes. ZF-4 cells were pulsed using the preset program T-27. Cell suspensions for each were then mixed with 900 μlof warm complete media, dispensed into appropriate culturing vessels and incubated at 37 ◦C. Transfected ZF-4 cells for shRNA detection was dispensed into 6-well plates (Nunc) and cells for VHSV infection were dispensed into 25 cm2 flasks.

2.5.3.2 Chemical transfection of plasmid DNA

Transfection of plasmid DNA into Vero cells were achieved using Lipofectamine 2000 (Invitrogen 11668019) following the manufacturer’s instructions. Cells were grown to between approximately 70% and 80% conﬂuence in 25 cm2 ﬂasks for shRNA expression. For each, 5 μg of plasmid DNA were complexed with 10 μlof Lipofectamine 2000 in 1 ml of OptimemR .

2.5.4 Flow cytometry

Analysis of cells transfected with plasmids expressing fluorescent markers by fluorescence- activated cell sorting, (FACS) were performed by harvesting cells as described in section 2.5 prior to being washed in PBSA twice and pelleted at 2000 g before resuspension in 200 μl FACS buffer (0.01% sodium azide and 2% FCS in PBSA) and analysed on a FACScalibur (LSR II Becton Dickinson). Data analysis was performed using CELLQuest software (Becton Dickinson) and mean fluorescence Chapter 2. Materials and Methods 57 intensity (MFI) values obtained. All FACS analysis was performed on duplicate cell populations following triplicate independent transfections. To analyse change in MFI control fluorescence (pEGFP-N1) were designated as 100% and change in fluorescence following transfection with shEGFP expression plasmids calculated as a percentage of the control transfection.

2.6 Virological methods

2.6.1 Viruses and infections

The 23.75 strain of viral haemorrhagic septicemia virus (FN665788.1)wereob- tained from AAHL stocks and were used for all viral experiments. Infections to test the eﬃciency of anti VHSV shRNA were performed on approximately 70% - 80% conﬂuent cells at a multiplicity of infection (MOI) of between 1 and 10-5 in T25 plates 72 hours post-transfection in 1 ml complete media. Inoculums were absorbed onto cells for 1 h at 15 ◦C and then replaced with complete media.

2.6.2 TCID50 determination

Supernatants from VHSV infected ZF4 cells were prepared by centrifugation of the media at 4000 g for 30 min to remove any cell debris. Titrations of clariﬁed supernatants were performed in 96 well plates on 90% conﬂuent EPC cells using a starting concentration of 20 μl diluted to 200 μl in complete media and 8 10- fold serial dilutions. Supernatants were incubated at 15 ◦C for 1 h. Each assay was performed with 10 replicates per dilution and in triplicate for all transfection conditions. Viral titres were calculated via the Reed & Muench method[295]. Chapter 2. Materials and Methods 58

2.7 Zebraﬁsh methods

2.7.1 General zebraﬁsh care and breeding

Larval and adult zebrafish were housed in an Aquatic Habitats aquarium system at 28.5 ◦C on a 14 h/10 h light/dark cycle and fed twice daily. Embryos were manu- ally spawned and maintained in the incubator at 28.5 ◦C in a petri dish containing aquarium water and 0.00005% (w/v) Methylene blue (Sigma). If imaging was required 8 hours post fertilization (hpf), methylene blue was replaced with aquarium water containing 0.003% (w/v) 1-phenyl-2-thio-urea (PTU) to inhibit pigmenta- tion and to enhance embryo transparency. Embryos were collected at appropriate time points, and anesthetised with 0.4 μg/μl benzocaine before fixation with 4% (w/v) paraformaldehyde (PFA) in phosphate-buffered saline (PBS). If embryos were used for breeding or adult experiments, feeding began at day 4 post fertilisation with 1 feed daily of approximately 500 μlofParamecia multimicronucleatum. All experiments were performed under Deakin University Animal Welfare Com- mittee and Office of Gene Technology Committee, approvals and A73/10 and LBC 10-2011, respectively.

2.7.2 Microinjection of zebraﬁsh

Microinjection needles were drawn on a Flaming brown micropipette puller model P-87 (Sutter Instruments) using a heat setting of 339, pull 45, velocity 80 and time 150, from a 1.0 mm capillary (SDR Clinical Technologies). Needles were bent into a ‘Z’ shape by ﬂaming and loaded using pipette tips with either 20 ng/μl to 30 ng/μl mRNA, 40 ng/μlto50ng/μl plasmid DNA, or a mixture of the two. All nucleic acid were diluted in 1×Danieau buﬀer (58 mM NaCl, 0.7mMKCl,0.4mM

MgSO4,0.6mMCaCl2,5.0 mM mM HEPES; pH 7.6) containing 1% (w/v) phenol red. Needles were mounted in a Narishige MN-151 micromanipulator attached to a Nikon SMZ 645 dissecting microscope. Embryos at the 1 cell stage were aligned along the end of a microscope slide and the needle positioned so as to Chapter 2. Materials and Methods 59 enter the yolk and then the cell. Typical injection conditions included a pulse duration of between 0.3s to0.6 s and a gas pressure of 400 kPa. Injection settings were adjusted as required to maintain an injection bead of approximately 0.1mm containing approximately 500 pl of injection mix. Embryos were subsequently allowed to develop on a warming tray prior to incubation.

2.7.3 Fluorescent microscopy

Embryos were visualised using an MVX10 microscope (Olympus) with stage light- ing provided using LG-PS2 ﬁbre optics light source (Olympus). Digital images were recorded using CellSens Dimension 1.6 software (Olympus) and DP72 camera (Olympus).

2.7.4 Fixation of zebraﬁsh embryos and larvae

Embryos were euthanised using 0.4 μg/μl benzocaine (Sigma) prior to ﬁxation in 4% (w/v) paraformaldehyde (PFA) in PBSA with 0.1% (v/v) (PFA/PBS). Fixed embryos were stored at 4 ◦C for at least 1 day. Embryos were subsequently dehydrated in 100% (v/v) methanol for long-term storage at −20 ◦C.

2.7.5 Whole-mount in-situ hybridisation of zebraﬁsh embryos (WISH)

2.7.5.1 Digoxigen-labelling of nucleic acids

To produce digoxigenin (DIG)-labeled RNA probes for in situ hybridization, 1 μl to 2 μl linearized DNA or PCR products, were mixed with 2 μl10× DIG labeling mix (Roche), 2 μl10× Transcription buffer, 1 μl RNase inhibitor, 2 μlT7orSP6 polymerase, and the volume was adjusted to 20 μl by adding sterile nuclease-free water (Amresco). Following incubation at 37 ◦Cfor2h,2μl of DNase I (10 U/μl) were added and the sample incubated at 37 ◦C for a further 15 min to destroy Chapter 2. Materials and Methods 60 the DNA template. The reaction was terminated by the addition of 2 μlof0.2M EDTA. The RNA was further purified using G-50 gel filtration exclusion micro- columns (GE Health) as per the manufacturer’s directions.

2.7.5.2 Hybridisation and Staining

Fixed embryos (section 2.7.4) were rehydrated via a decreasing series of 5 minwashes in methanol and PBS-T (PBS plus 0.1% Tween-20) 100% (v/v) , 50%, and 30% (v/v) before refixing for 20 min in 4% (v/v) PFA/PBS followed by 3 washes in PBS-T. Embryos older than 26 hpf were then incubated at room temperature in PBSA containing proteinase K, 24-36 hpf with 10 μg/μl proteinase K for 10 min, 36 hpf to 7 dpf with 20 μg/μl proteinase K for 30 min. Hybridization involved digoxigenin-labelled antisense RNA probes transcribed from appropriately linearized plasmids or LNAs. Embryos were incubated in 500 μl of HYB- (50% (v/v) formamide, 5 x sodium chloride/sodium citrate (SSC), 0.1% (v/v) Tween20) for 5 min at 70 ◦C and replaced with 500 μl of HYB+ (HYB- with 5 μg/μl ribonucleic acid, 0.05 μg/μl heparin (Sigma)) which were incubated at 70 ◦C for at least 3 h. Embryos were incubated at 70 ◦C overnight with 150 μl to 300 μl of probe solution 1:50 to 1:300 diluted in HYB+. Embryos were washed in 50% (v/v) formamide/2 × SSC-T for 30 min and 60 min at 70 ◦C. Embryos were washed for 15 min in 2× SSC-T followed by 2 × 15 min washes in 0.2 × SSC-T also at 70 ◦C. This was followed by 3 room temperature washes in PBS-T followed by incubation in blocking solution (0.02% (w/v) zebrafish serum albumin (BSA), 5% (v/v) foetal calf serum (FCS) in PBS-T) for 2 h at room temperature. Following the incubation embryos were placed into blocking solution with (0.15 U/μl) of alkaline phosphatase (AP) conjugated anti-DIG antibody (Roche) and incubated at 4 ◦C overnight. Embryos were washed 6 × 15 min in PBS-T at room temperature with gentle rocking and then 3 × 5 min washes in staining buffer (0.1 M Tris pH 9.5, 0.05 M MgCl2, 0.1 M NaCl, 0.1% (v/v) Tween 20) containing 25 μg/ml levamisol. Following this, embryos were stained in staining buffer containing (0.225 mg/ml nitroblue tetrazolium (NBT) and 0.175 mg/ml 5-bromo,4-chloro,3-indolyl phosphate (BCIP) at Chapter 2. Materials and Methods 61 room temperature in the dark for 1 h to 48 h. Embryos were finally rinsed in PBS-T and fixed in 4% PFA/PBS when stained to an appropriate extent.

2.8 Protein methods

2.8.1 Western blot analysis

2.8.1.1 Sample collection and preparation

Zebrafish embryos or larvae were dechorionated if necessary and placed in extraction buffer (10 mM Tris, pH 7.4, 2% Triton X 100,1 mM PMSF,1 mM aprotinin,1 mM leupeptin, 1 mM trypsin inhibitor). Samples were thawed and pelleted by centrifugation at max speed for 2 min and homogenised before boiling for 5 min and centrifugation at 13 000 g for 2 min and the supernatant removed and frozen in SDS buffer at −20 ◦C (0.63 ml 1M Tris-HCl, pH 6.8, 1 ml glycerol,0.5ml β- mercaptoethanol, 1.75 ml 20% SDS, 6.12 ml H2O). .

2.8.1.2 Bradford Assay

Total protein concentration were determined with the addition of 2 μl supernatant,

100 μlnfH2O, and 1 ml Bradford reagent, and concentration determined by extrap- [48] olation to a linear curve generated by A595 concentration of a Bradford assay . Samples were prepared by adding 1 μl of lysate sample to 1 ml of 1 × protein assay solution. Both samples and BSA controls were mixed by inversion and left at room temperature for 5 min. Absorbance of prepared standard BSA concentration samples were read at 595 n m and a standard curve were generated in GraphPad Prism 6.0. Chapter 2. Materials and Methods 62

BSA 1 μg/ml 8 μl6μl4μl2μl0μl Bradford Assay reagent 1 ml 1 ml 1 ml 1 ml 1 ml Final BSA Concentration 0.8 μg/μl0.6 μg/μl0.4 μg/μl0.2 μg/μl0μg/μl

Table 2.5: Bradford assay dilutions

2.9 Bioinformatics

2.9.1 Data mining

Homology searches were conducted using the BLAST-N server at the National Centre for Biotechnology Information (NCBI), National Library of Medicine, Bethesda, Maryland, USA[10]. To identify the zebrafish U6 and H1 promoter sequences, the zebrafish whole genome shotgun sequences deposited in the genome survey sequence (GSS) section of GenBank and the zebrafish genome sequencing project (ZFIN) Zv9[340] were searched using the human and mouse U6 snRNA sequences [RnU6] (GenBank accession numbers M14486 and M10329.1) using BLAST-N or Megablast with default parameters and a threshold e-value of 0.01. The zebrafish ribonuclease P RNA component H1 (H1)[100]had previously been identified (NR 036645.1).

2.9.2 Nucleotide sequence data management

All nucleotide sequence data was managed using the Clone manager 6 (Version 9.00) and Serial Cloner (Version 2.50) [Serial Basics].

2.9.3 Image Analysis

Zebrafish larvae were photographed using cellSens Dimension 1.6 software (Olym- pus) and DP72 camera (Olympus) at 400 × combined magnification and files saved as .tiff. to determine area and standard length (STL) of the zebrafish larvae Image J analysis software were used with the following protocol. The new macro Image J Chapter 2. Materials and Methods 63 macro zebrafishlength.ijm were developed to partially automate zebrafish embryo batch processing. A brief protocol is included below (see 2.6). This technique requires clear images of zebrafish embryos with sufficient bright light from directly above the embryo to eliminate shadows, which may obscure the binary mask.

Step Action Image J process 1 Greyscale Image → Type → 8-bit 2 Set Scale Draw a line over the scale bar 3SetScale Analyze → Set Scale Known distance = xxxx 4 Determine Area Process → Binary → Make Binary 5 Determine Area Analyze → Analyze Particles 6 Export to Excel Save to clipboard and Paste to Excel 7 Determine Length Edit → Selection → Make Bounding box 8 Export to Excel Save to clipboard and Paste to Excel

Table 2.6: Image J Batch processing. Bold text indicates command path. Italic text indicates inputs within commands.

Chapter 3

Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture

The publication “Clarke, B.D., Cummins, D.M., McColl, K., Ward, A.C., & Do- ran, T.J. Characterisation of zebraﬁsh polymerase III promoters for the expression of short-hairpin RNA interference molecules. Zebraﬁsh. 2013 (In Press).” includes results from this chapter.

3.1 Introduction

This results chapter describes the identification and characterisation of the zebrafish H1 snRNA promoter and U6 snRNA promoters for the expression of shRNAs as well as the utilisation of the chicken (Gallus gallus) pantothenate kinase 1 (Pank 1) [NC 006093.3] intron 5 construct to introduce shRNA molecules into zebrafish cell culture as illustrated in Figure 3.1. The zebrafish promoters were identified based on their downstream proximity to the snRNA genes and by the presence of conserved snRNA polymerase III type 3 elements motifs. The Pank-1 intron was originally generated to facilitate the introduction of miRNA based shRNA molecules into chicken eggs by Dr. Kristie Jenkins (CSIRO AAHL)

65 Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 66

[Unpublished data] and is used in this study as an alternative method of introduction of shRNAs. To validate these promoters shRNA expression vectors were constructed and the shRNA expression driven by the respective promoters tested by co-transfection and reduction of intensity of the reporter gene, eGFP.

Following the initial descriptions of the application of human snRNA pol III type 3 promoters to transcribe shRNAs[51,276,393], expression of shRNAs from a variety of pol III type 3 promoters has an invaluable tool for gene knockdown and reverse- genetics in cell culture. In many cases the promoters of choice have been the U6 or H1 promoters, however, the 7SK promoters[27,127,195] have also been popular. The U6 and H1 promoters have been by far the most frequently used for the expression of shRNA for gene suppression. However, there have been only a pair of studies that have directly compared the activity of the U6 and H1 promoters for the expression of shRNAs. Both studies were performed using the human promoters and both concluded that the U6 was the stronger promoter [42,231].

While there has been significant work performed in the identification and characterisation of mammalian pol III promoters[51,195,260,302,316,383] and to a lesser extent the chicken promoters[27,79,383]. Promoters identified as potentially suitable for use in teleosts are limited to the zebrafish or fugu U6 promoters[46,182,370,401] and considering the controversies relating to RNAi based technologies in zebrafish[269,323,370,403] and implications that the dose of interfering RNA molecules maybe critical for avoiding significant non-specific results from injections of siRNA[403] therefor the characterisation of the strength of the zebrafish U6 promoters and the identification and cloning for the first time of the zebrafish H1 promoter begin the construction of a potential rational tool kit for the utilisation of polymerase III promoters in zebrafish to express shRNAs.

Intronic miRNAs were first identified in C. elegans [11] and have since been determined to be among the most populous of all miRNAs [11,251,392]. Synthetically generated intronic miRNAs have been used successfully in mice[251], chicken (Un- published Data) and zebrafish[95,298]. Intronic shRNA molecules enter the RNAi pathway earlier than either siRNA or polymerase III expressed shRNAs Figure 3.1 Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 67

+1 GFP Intron CMV Promoter Expression Plasmid

Pre-mRNA

Splicing + +1 Splicesosome

Polymerase III Promoter shRNA Intron Expression Plasmid

DROSHA

Protein

shRNA miRNA / shRNA Excision

Dicer / RISC

Mature miRNA

Transcriptional Repression / Regulation

Figure 3.1: Induction of RNAi response from polymerase III promoters or intronic vectors Schema of induction of RNAi mechanisms through pol III expressed shRNA or intronic miRNA/shRNA. The intronic miRNA is transcribed by the Polymerase II CMV promoter and expressed in the intron 4 region of the Pank 1 gene transcript (pre-mRNA). After RNA splicing in the splisosome and further additional processing, the spliced out intron then functions as a pre-miRNA for intronic miRNA generation and is then excised by Drosha RNases to form a hairpin-like pre-miRNA template and then exported to the cytoplasm for further processing by Dicer to generate mature miRNAs / shRNAs. The shRNAs generated from either intron or polymerase III promoters are finally incorporated into a RNA-induced silencing complex (RISC) leading to disruption of target mRNA. Figure modified from[392]. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 68 and it is hypothesised that by including shRNA molecules into an intron which already expresses an endogenous miRNA that the dose of the shRNA will be both tolerated and sufficient for effective repression of targeted mRNA.

This chapter describes the identiﬁcation and characterisation of the zebraﬁsh sequence with predicted polymerase III promoter activity based on its proximity to either a U6 or H1 snRNA sequence and the presence of typical type 3 RNA pol III promoter element motifs. The validation of this sequence for shRNA expression was achieved by the construction and testing of plasmid vectors targeting an ex- ogenously expressed reporter gene, eGFP and by RT-PCR and RNase protection assay to qualify RNA expression.

3.2 Characterisation of and identiﬁcation of ze- braﬁsh U6 and H1 snRNA promoters for the expression of shRNAs

3.2.1 U6 promoter identiﬁcation

To identify native U6 promoters the zebrafish genome was analysed using BLAST for sequences homologous to the previously identified U6 sequences (U6 1-3)[133], from both zebrafish and fugu[401]. This search yielded an additional novel zebrafish U6 promoter located on chromosome 9 (U6 4)[CU651612.3][59].

These promoters were screened for the presence of known polymerase III promoter elements such as a TATA-box[86], Octamer motifs[257], PSE like[316],SPHand STAF[313] elements. All four zebrafish U6 promoters were found to possess a SPH element immediately upstream of the TATA box, which is different to the previously characterized mouse and chicken U6 promoters (Figure 3.2). The U6 promoters were not highly conserved between the individual promoters (Figure 3.3) andClustalW2[67] multiple sequence alignment clearly analysis identified the U6 1 promoter as clearly different to the other zebrafish U6 promoters, as summarised Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 69 in Table 3.1 and (Figure 3.3), only the U6 4v1 promoter was included in this analysis as a variant U6 4 promoter was amplified from a single clone and likely results from a single animal, pool or PCR amplification artifact and sequence analysis of relevant genomic DNA amplified from adult zebrafish and from the zebrafish ZF4 cell line confirmed that the Octamer, SPH and TATA elements were highly conserved in all promoters against previously published consensus sequences (Figure 3.2).

Table 3.1: Clustal W2 analysis of ze- braﬁsh U6 promoters

U6 1 U6 2 U6 3 U6 4 U6 1 100 U6 2 57 100 U6 3 51 72 100 U64526878100

Numbers indicate % identity

3.2.2 H1 Promoter identiﬁcation

The zebrafish snRNA component of RNase P H1 has also been previously identified[100]. The H1 promoter was identified by analysis of the sequence 1 kb up-stream of the RNase P gene transcriptional start site for polymerase III type 3 promoter elements. The identified promoter elements as expected begin with a distal sequence element consisting of an Oct-1 and STAF motifs and a proximal sequence element consisting as expected of a SPH binding site and TATA box. The H1 promoter was also more conserved than the zebrafish U6 promoters in comparison to orthologous H1 promoters and between the identified promoter elements and the published consensus sequence (Figure 3.4 B). The H1 promoter showed higher conservation each of the identifiable promoter elements was conserved when compared to the consensus sequences and the order of the specific elements was also conserved (Figure 3.4). In each of the mammalian species the RPPH1 gene is located on the antisense strand of the PARP1 gene in the first exon in the zebrafish Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 70

A OCT SPH TATA Consensus ATTTGCAT YYWCCCRVMATSCMYYRC R TATA

Zebrafish U6-1 (-339)ATCGTTTg(-331) (-66)CCACC-ACCtTGgAggAAC------TAcATAT(-23) Zebrafish U6-2 (-224)tTTTGCAT(-216) (-48)aTACCCAGAAGtCCCTtgG-TATATA(-24) Zebrafish U6-3 (-218)ATTTGCag(-210) (-47)aTACCCAGAAGtCCCGg GTATATA(-23) Zebrafish U6-4v1 (-214)TACGcTTA (-208) (-47)aTACCCAGAAGtCACTG GTATATA(-23) Zebrafish U6-4v2 (-214)TAAgtGTA (-208) (-47)aTACCCAGAAGtCACTGGTATATA(-23)

OCT SPH Consensus ATTTGCAT YYWCCCRVMATSCMYYRC R

Mouse U6 (-235)ATTTGCAT(-228) (-222)TTTCCtAGtAaCtATaGa G(-202) Chicken U6-4 (-194)ATGCAAAg(-187) (-206)CTTCCCGCCgTGCACCGC G(-216) Human U6-8 (-234)ATGCAAAg(-228) (-247)tTTACCCAgggTGCccgG g(-227) Fugu U6 (-247)TTTCCCACAGCCCCCT GGC(-227)

PSE TATA Consensus STSACCGTGWSTGTRAA R(0-3)TG TATA

(-74) (-27) Mouse U6 CTCACCcTaACTGTAAA G(5)TG------TATAAATAT (-66) (-24) Chicken U6-4 CTCACCGctACTtaAAAA T(2)TG------TAAATA (-68) (-26) Human U6-8 GTCACCGTaAGTaGAAt A(2)TG------TTATAA Fugu U6 (-85)CCAACCATGACCCAGA G------(-62) (-31)TATATATA(-24)

U6 snRNA Mouse U6-1 snRNA Promoter Tn OCT SPH PSE TATA +1

U6 snRNA Zebrafish U6-1 snRNA Promoter Tn OCT SPH TATA +1

Figure 3.2: Polymerase III promoter element alignment A Sequence alignment of the U6 promoter elements. The conserved U6 promoter elements, the distal sequence element (DSE) comprising SPH and OCT, the proximal sequence element (PSE) and TATA motif are shown for each zebrafish U6 promoter. B The equivalent human, mouse, fugu and chicken U6 promoters for comparison. Nucleotides that matched the consensus sequence for SPH, OCT, PSE and TATA sequence are shown in uppercase or lowercase to identify mismatches. Underlined letters represent mismatches observed between variant U6 4 clones. Dashes (–) indicate spacing between PSE elements. Nucleotide positions are shown at the relative to the predicted transcription start site (+1). C Graphical representation of interchanged promoter motifs of the zebrafish and mouse U6 1 promoters. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 71

1 --CTAGT------TATAGTT TCG--AT------GGGCC- --GTTT-GCT GGCGTCTGAG 39 1 --CTA------TATAGCT CTG--CTCTA GTGAGAGCA- --GCTT-CCT TGTGTCGGAG 44 1 TTCTATAG-- ---TATA-CT CGGTCATTCA GTAAGACGA- --GTTT-CCT GCTCTCGGCG 50 1 --CTATATAT AGATATAGAT TCCAAATGGT GCATGAGGAT GTGTTTAGCA AATCGTGGGA 58 *** **** * * * * ** * *

40 CATGTTGTCC TCTGGTC--T GCGGCGGGGG AA--GGTCCG CGCATGCCTG -TGGGGGAAG 94 45 TGTTTTGTGC TCTTATC--C AAGGCAGGGG G---GATCTG CGCATGCCTG -TGGGGGGAG 98 50 TCTTTTGT-- TCTGGTCATC AAGGAGGGGG GAATGTTCTG CGCATGCCTG -TGGGGGAGG 107 59 GGTGCTGC-- -TGGGTC--- --AACTGGAA AA----TCTA --CATGCATG GTGAAAACGA 104 * ** ** ** ** ***** ** **

95 GCGAAGGACA CGAGGACAAA AGATCAGTTA CGTCACA--- GAAG------AACTCATCCA 145 99 GAGAAGGACA CGTGAACAAA AGCTCCTCGA TGTCACACAG GAAGTTCAGG AACTCATCCA 158 108 GAGAAGGACA CGTCACTGAA AACGTCCCTG CATCACACC- GAGA------CACCCA 156 105 ATGTGGGATG CTTTGATCG- ---TTTGCTT TGTTGTG------CAACCA 143 * *** * * ** ***

146 AT-CACACTA GCCTAAAA-C CAGCTTC------CTCTGC -ATATGCTTT ACAGCT-CAA 194 159 AT-CACTCTA AAGAAAAGGC CTGTTTC------CTTCGC -ATACGCTT- ACAGCT-ACA 207 157 AT-CACTCAA GCCGAGA--C CAGATAA------TTTTGC -ATATGCTTT ACAGTTTGAA 205 144 ATGCACTT-- -CCAAAGCCC TTGCTTTAGT GGGACTTTGC CATAGACTAT CCTGACACTA 200 ** *** * * * * * ** *** ** * * *

195 AAATACTACG GTAAACCTCC ACA-AAACTG CTGGTTT-CA A-ACTTTGAA G-GTTTTAAG 250 208 AAACTCTACG GTAAACCTAC ATA-AA-CTG CTGGTTTTCA A-ATTTTAAA G-AATTTAAG 263 206 AAATACCACA GTAATCCCTC ACACAAACT- CTGGATTTGA G-ATCCTTCA G-GTTTTATC 262 201 ATTTCCCCCT GT------CC ATA-AACTTG GTGACCTACT GCATCAATCA GTGTGCCTAG 253 * * * ** * * * ** * ** * * * *

251 CATTTGCAGG TTTTCTCTGA ---AGAGGTT TACTGTCAT GTTTGAG-GT AAATGAGTTGA 306 264 GGTTTACAGG TTTACTACTA CACAGTGATT TACTGACAC ATGTAAGTGT AAATGAGTTGA 323 263 AGTTTGCAGG TTTA-TGTCA C--CATGATA TAGGGTCAG ACTTGAT-CT AAGGGAGCTGA 318 254 CACTGACAGG GCCAAAGCTC T---GTAGTG CACGG-CAT AC------AAGTGG---GA 298 * **** * * * ** ** * **

307 ATAAGTAGGT TTATCCACTT ACCACATGGC AGAAACATAC CCAGAAGTC CCGG-GTATAT 365 324 ATAAGTAAGT A-AGCCATAT ACCACACATG AAACACATAC CCAGAAGTC ACTG-GTATAT 381 319 ATAAGT-GGT TTAGTCACTC ACCACCTCCC AAAAACATAC CCAGAAGTC CCTG-GTATAT 376 299 ACATGGCTCT CCAGGT-CCC ACCACCTTGG AGGAACATTG CTGGGACTC AAGGCGTACAT 357 * * * * * ***** * **** * * * ** * *** **

366 ATAGCAGTCC TCCAGGCTCC TAGT-TCA-C AATC- 397 382 ATAGCCGTCC TCCAGACTCC CAGC-TCAAT AGTC- 414 377 ATAGCTCTCC CTCCAGCTCT TGGT-TCGAA CATC- 409 358 AT-GCCCTGA GCCTGACTGC TGATCTTATA CCTCG 391 ** ** * * ** * **

ZF U61

ZF U62

ZF U63

ZF U64 Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 72 the RPPH1 gene has been translocated to the sense strand and occurs in the ﬁnal exon of the antisense METTL17 gene which is approximately 10 kb downstream of the PARP1 gene (Figure 3.4 A).

3.2.3 shRNA expression vector construction

To validate the predicted promoter activity of the polymerase III promoters, shRNA expression vectors were constructed encoding these promoter sequences upstream of shRNA molecules that, target eGFP. The eGFP shRNA sequence used was previously validated and was found to be the most effective of all molecules tested[27]. The loop sequence between the shRNA stem sequences was a sequence used by many other researchers, the “Brummelkamp loop” 3-TTC AAG AGA- 5 [51]. The termination sequence consisted of a stretch of five thymidine residues which has previously been described to cause efficient pol III termination[51,242], with an additional termination sequence 5-GGA A-3 added to the end of this sequence was included to promote termination and efficient silencing[51].

In a one-step PCR, the forward primer specific to the promoter was paired with a reverse primer comprising the last 20 nt of the polymerase promoter, along with all other shRNA components as well as a XhoI restriction endonuclease site (Figure 3.5). A clear band consistent with the expected size (between 382 - 464 bp depending on construct) was gel purified and ligated into pGEM-T Easy. The XhoI site facilitated screening for full-length inserts by digestion of recombinant plasmids with this enzyme. Sequencing of linearised plasmids revealed two clones that contained slightly different promoter sequences for the U6 4 promoter (Figure 3.6) while, all other clones returned appropriate sequences for the plasmids of interest. The zebrafish U6 4 promoter showed some variation to the remaining

Figure 3.3 (preceding page): Alignment of zebrafish U6 snRNA promoters Sequence alignment of complete zebrafish U6 promoters. Dashes () Indicates a gap and asterisks () indicates a conserved base across all promoters. Phyloge- netic tree was generated utilising a Clustal Omega multiple sequence alignment and is indicative of the relative conservation of the U6 promoters Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 73 A DRE 37565145 37585260 CH 02 SDR39UL RPPH1 PARP2 LOC100333868 METTL17 HSA 20757290 20881579 SNORD126 CH 14 TTC5 RPPH1 TEP1 CCNB1IP1 PARP2 MMU 50765415 50870554

CH 14 TTC5 RPPH1 TEP1 CCNB1IP1 PARP2 CFA 17673687 17748435

CH 15 CCNB1IP1 RPPH1 TEP1 LOC100687224 PARP2 LOC100687294 B

Figure 3.4: The RPPH1 ribonuclease P RNA component H1 gene and promoter. A Synteny analysis of zebrafish (ZF NC 007113.5),human (NC 000014.8[22], mouse (NC 000080.6)[8]and dog (NC 006597.3[28]. RPPH1 genes. In each of the mammalian species the RPPH1 gene is located on the antisense strand of the PARP1 gene in the first exon. In the zebrafish the RPPH1 gene has been translocated to the sense strand and occurs in the final exon of the antisense METTL17 gene which is approximately 10 kb downstream of the PARP1 gene. B Sequence of the zebrafish H1 promoter region. The bolded sequences indicate the consensus OCT, STAF, PSE and TATA motifs with mismatches represented by lowercase letters[71]. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 74 zebrafish U6 promoters in both the PSE and SPH elements, but most significantly a variant U6 promoter was identified (U6 4v2) with three base substitutions within the OCT motif (Figure 3.6).

For all promoter shRNA expression constructs, the first nucleotide of the predicted shRNA was located at the +1 position directly after the promoter sequence and was a guanine (G) residue, corresponding to the first nucleotide of the native U6 snRNA. All final shRNA expression constructs consisted of either the full-length zebrafish U6 or H1 promoter, a shRNA sense sequence, a loop sequence 3-TTC AAG AGA-5, an shRNA antisense sequence, and a termination sequence 3-TTT TTT GGA ATC TAG A-5 and were cloned into pGEM-T Easy.

3.2.4 Optimisation of plasmid DNA transfection in zebraﬁsh ZF-4 cells

Fish cells in culture are difficult to transfect using typical lipid or polymer transfection reagents and there is comparatively little research into how different transfection reagents work in the cell lines. Only a single paper had previously described transfection of ZF-4 cells using the Amaxa nucleofector[137], which achieved over 70% transfection efficiency. Therefore, a study of ten different transfection methods using eGFP and sheGFP plasmids was performed to identify the most effective. In addition to the chemical reagents the Amaxa Nucleofector and Invitrogen Neon Transporator were used to transfect both ZF4 and EPC cell lines to determine the suitability of these methods. The effectiveness of the transfection methods was estimated by the observation of eGFP expression under a fluorescence microscope and by flow cytometry to count eGFP positive cells and trypan blue staining to estimate necrotic and apoptotic cells. Seven widely used chemical transfection agents were tested: Lipofectamine, Lipofectamine 2000, Lipofectamine RNAi (In- vitrogen), X-tremeGENE (Roache), Polyfect (Qiagen), Fugene, and Fugene HD (Promega) following the manufacturers recommendations. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 75

One-step PCR

Forward Primer SpeI

Polymerase III Promoter

shRNA XbaI Reverse Primer

Promoter Construct

Hairpin XbaI Polymerase III Promoter SpeI Passenger Guide Strand Strand

Ligation

oter om sh Pr RN II A l I o P

X b a I

A I e pGEM-T Easy T p

S Promoter Vector

A T

Figure 3.5: Cloning and construction of shRNA expression vectors The one-step PCR method used to generate Polymerase III promoters. PCRs used forward primers paired with long unique reverse primers comprising all shRNA components. All ﬁnal PCR products consisted of a polymerase III promoter either U6 or H1 shRNA sense (Passenger) , hairpin loop, shRNA antisense (Guide), termination sequence 3-TTT TTT GGA A-5 and XbaI site. Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 76

1 CTATTATAGC TCTGCTCTAG TGAGAGCAGC TTCCTTGTGT CGGAGTGTTT 50 # 1 CTA-TATAGC TCTGCTCTAG TGAGAGCAGC TTCCTTGTGT CGGAGTGTTT 49

51 TGTGCTCTTA TCCAAGGCAG GGGGGATCTG CGCATGCCTG TGGGGGGAGG 100

50 TGTGCTCTTA TCCAAGGCAG GGGGGATCTG CGCATGCCTG TGGGGGGAGG 99

101 AGAAGGACAC GTGAACAAAA GCTCCTCGAT GTCACACAGG AAGTTCAGGA 150

100 AGAAGGACAC GTGAACAAAA GCTCCTCGAT GTCACACAGG AAGTTCAGGA 149

151 ACTCATCCAA TCACTCTAAA GAAACGGCCT GTTTCCTTCG CATACGcTTA 200 # ## 150 ACTCATCCAA TCACTCTAAA GAAAAGGCCT GTTTCCTTCG CATAaGtgTA 199 TACGTTTA 201 CAGCTCCAAA ACTCTACGGT AAACCTACATA AACTGCTGG TTTTCAAATT 250

200 CAGCTACAAA ACTCTACGGT AAACCTACATA AACTGCTGG TTTTCAAATT 249

251 TTAAAGAATT TAAGGGTTTA CAGGTTTACT ACTACACAGTG ATTTACTGA 300

250 TTAAAGAATT TAAGGGTTTA CAGGTTTACT ACTACACAGTG ATTTACTGA 299

301 CACATGTAAG TGTAAATGAG TTGAATAAGT AAGTAAGCTA TATACCACACA 350

300 CACATGTAAG TGTAAATGAG TTGAATAAGT AAGTAAGCCA TATACCACACA 349 Y 351 TGAACCACAT ACCCAGAAGT CACTGGTATA TATAGCCGTC CTCCAGACTCC 400 350 TGAAACACAT ACCCAGAAGT CACTGGTATA TATAGCCGTC CTCCAGACTCC 399 YWCCCRVMAT SCMYYRCR T ATA 400 CCAGCTCAA- AGTCGTA 417 # # 401 CCAGCTCAAT CGTCGAA 417

Figure 3.6: Alignment of zebrafish U6 4 variant promoters Clustal W2[67] sequence alignment of U6 4v1 and U6 4v2 variant promoters. Overall similarity 407/405 bp 98.07%. snRNA promoter elements and consensus sequences are indicated in Bold. OCT Motif occurs in the 3 to 5 orientation. Mismatches between promoters are indicated by #. Mismatches between zebrafish U6 4 promoters and consensus sequences indicated by lowercase letters. U6 snRNA gene / Hairpin transcription start site designated as +1. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 77

Analysis of chemically transfected cells 72 hours post-transfection by fluorescence microscopy showed that a very low level of transfection was achieved for each agent (Data not shown) and confirmed by flow cytometry (Figure 3.7). The highest level of eGFP expression was evident in the cells transfected using Fugene HD, however the inconsistency of the transfection and the low levels of transfection suggested that it would be very difficult to accurately and consistently differentiate eGFP expression levels between various conditions. The various chemical transfection reagents were also quite toxic (Figure 3.7) as measured by cell counts following trypan blue staining and the chemical transfection methods were judged to be not suitable for analysing suppression of eGFP in ZF-4 cells. However lipofectamine 2000 transfection of Vero cells is a well established method and would be used for all transfections of mammalian cells.

In an attempt to increase transfection rates of eGFP in ZF4 cells, electroporation efficiencies were tested using a Nucleofector machine (Amaxa). Three different electroporation solutions were provided by the manufacturer; solutions R, V, and T. A number of different programs are additionally available to optimise transfection, and a combination of four electroporation programs which have been successfully used in our laboratory in other cell lines were trialed with each solution and screened for GFP expression (Figure 3.7) and toxicity. Each program was tested using 2.5 μg of plasmid DNA per 1×106 cells, as suggested by the manufacturer (Figure 3.7).

Analysis of electroporated cells for eGFP expression showed that the transfection efficiency varied significantly depending on both the solution and program used as well as the consistency and the toxicity observed in the ZF-4 cells following transfection. The combination of solution T and program T-20 resulted in the most consistent transfection efficiency and the lowest toxicity. The transfection efficiency was similar to that observed with Fugene and significantly below that previously reported[137]. It seems likely that discrepancies in transfection efficacies between the published studies and the observed efficacies maybe related to either the passage number or growth conditions of the ZF-4 cells used. However a transfection efficiency of 52 ± 6 percent is sufficient for the purposes of this study. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 78

Chemical transfection reagents

Untransfected Lipofectamine * Lipofectamine 2000 * Polyfect Oligofectamine Fugene Fugene HD

X-tremeGene * 0

80 60 40 20 100 eGFP-N1 postitive (%) Trypan Blue positive (%)

Nucleofection solution and program

Untransfected Solution T * T-20 T-27 G-16 A-23 Solution R T-20 T-27 G-16 * A-23 Solution V T-20 * T-27 G-16 *

A-23 0

80 60 40 20 100 eGFP-N1 postitive (%) Trypan Blue positive (%)

Figure 3.7: Transfection of ZF-4 Cells Chemical and nucleofection methods of transfection of plasmid DNA into zebrafish ZF-4 cells. eGFP expression left hand graph measured via flow cytometry. Toxicity of transfection reagent estimated by Trypan Blue staining and manual cell counting right hand graph via haemocytometer. Error bars are indicative of the standard error of the mean (SEM) calculated from duplicate FACS analysis of or cell counts from three separate transfection experiments. Statistical significance was measured by ANOVA with Tukey’s post tests *P≤0.05 was considered significantly different to untransfected. n = 3 Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 79

3.2.5 The zebraﬁsh U6 and H1 promoters express shRNAs

To validate the ability of the zebrafish U6 and H1 promoters to express functional shRNAs, the putative sequences were used to construct six vectors pZFU6- 1sheGFP, pZFU6-2 sheGFP,pZFU6-3 sheGFP, pZFU6-4v1sheGFP, pZFU6-4v2sheGFP and, pZF-H1sheGFP, in which the relevant polymerase III promoter mediates expression of a shRNA molecule targeting eGFP. Additional constructs using chicken and mouse promoters (pCH-U6sheGFP and pM-U6sheGFP, respectively) have previously been shown to strongly express shRNAs in a wide variety of cells[27,195,383], while the construct pFUGU-sheGFP has been shown to be an effective silencer of eGFP in Bluefin gill (BF-2) cells[401] and more recently Epithelioma papulosum cyprini (EPC) cells (Fathead Minnow) and Chinook salmon embryonic (CHSE-214) cells[182] and were therefore used as controls. Two irrelevant control vectors; pZFU6-2 Irrelevant and pCHU6-4 NP, were used to express an irrelevant shRNA as negative controls in ZF-4 and Vero cells, respectively. The U6-2 promoter was used to drive the irrelevant sequence as preliminary work showed that the U6-2 promoter was the most active in the cell line tested (Data not shown).

RNA extracted from transfected ZF4 cells at 72 hours and Vero cells at 48 hours was probed for the expression of sheGFP using an RNase protection assay. A 19 nt band was detected in RNA samples from ZF4 cells transfected with all zebrafish U6 and H1 plasmids and the fugu positive control. No sheGFP expression was detected in RNA samples from the negative control or from either the pCh- U6sheGFP or pM-U6sheGFP (Figure 3.8). Detection of miR-16 as a loading control in all transfected and control RNA samples confirmed the presence of small RNAs in each case. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 80

Vero shEGFP 19 nt

mir-16 21 nt

CHU6 shEGFPMU6 shEGFP MU6 shEGFP ZFU6-1 shEGFPZFU6-2 shEGFPZFU6-3 shEGFP ZF H1 shEGFP CHU6 shEGFP ZFU6-4v1 shEGFP Fugu U6 shEGFP ZFU6-4v2 shEGFP ZFU6-2 shIrrelevant

Figure 3.8: RNase Protection Assay for the quantitation of sheGFP expression in cell culture Zebraﬁsh ZF4 cells and Vero cells were co-transfected by nucleofection and co- transfected with lipofectamine 2000 respectively with the shRNA expression vectors and peGFP-N1 as labeled. Small RNAs were isolated by TrizolTMat 72 h post-transfection and hybridised with a complementary sheGFP γ-P32- ATP radiolabelled RNA oligonucleotide. In addition, a γ-P32-ATP radiolabelled RNA oligonucleotide corresponding to miR-16 was used as a loading control across all samples. Representative of n = 2 Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 81

3.2.6 Zebraﬁsh polymerase III promoters can decrease eGFP expression in native zebraﬁsh cells but not mammalian Vero cells

To analyse the activity of the zebrafish polymerase promoters for shRNA expression, plasmids pZFU61sheGFP, pZFU64v2sheGFP, pZFH1-sheGFP, pFUGU- sheGFP, pMU6-sheGFP and pCHU6-sheGFP were co-transfected with peGFP- N,1 and the expression of eGFP monitored by fluorescence microscopy and flow cytometry. Analysis of Vero cells by fluorescence microscopy showed there was no eGFP expressed in the untransfected cells (Not shown), and that cells transfected with peGFP-N1 alone expressed eGFP at a similar level to cells transfected with both peGFP-N1 and pCHU6-Irrelevant (Irrelevant) [Figure 3.9 A]. Detailed analysis by flow cytometry identified the previously characterized control mouse and chicken expression plasmids strongly and significantly p≤0.01 reduced eGFP MFI - pMU6sheGFP (13.05 ± 6.55,p≤ 0.01) and pCHMU6sheGFP (26.55 ± 6.05; p ≤ 0.01) (Figure 3.9B). Additionally the zebrafish pZFH1sheGFP construct also significantly reduced eGFP MFI (69.13 ± 11.65; p ≤ 0.05). None of the other zebrafish or the fugu promoters showed any decrease in eGFP expression in Vero cells.

The promoter shRNA constructs were then tested in the homologous zebrafish ZF-4 cell line for comparison that revealed strikingly different results to the corresponding analysis in Vero cells (Figure 3.10). No eGFP expression was detected from the untransfected cells (Not Shown), and the cells transfected with peGFP- N1 alone expressed eGFP at a very similar level to the cells co-transfected with peGFP-N1 and pZF4-Irrelevant. However, fluorescence microscopy indicated that eGFP knockdown was comparable between pZFU6-1sheGFP, pZFU6-2sheGFP, pZFU6-3sheGFP and pZFU6-4v1sheGFP but no silencing was observed by either the pMU6sheGFP or pCHU6sheGFP plasmids (Figure 3.10 A). Statistical analysis of the Mean Fluorescence Intensity (MFI) [Figure 3.10 B] indicated that in ZF4 cells there was a significant reduction in eGFP MFI following co-transfection with Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 82

A Vero a Irrelevant pZF-U6-4v2 shEGFP

pZF-U6-1 shEGFP pZF-H1 shEGFP

pZF-U6-2 shEGFP pFuguU6shEGFP

pZF-U6-3 shEGFP pMouse U6 shEGFP

pZF-U6-4v1 shEGFP pCH-U6-4 shEGFP

Figure 3.9: shRNA knockdown of eGFP using zebrafish polymerase III promoters in mammalian Vero cells A Fluorescent microscope images of eGFP expression in Vero West African Monkey Kidney cells following co-transfection with shRNA constructs and peGFP-N1. An irrelevant non-silencing vector was used as a control. Im- ages are representative of triplicate transfections. B eGFP Knockdown in Vero cells. Flow cytometry results for eGFP knockdown in co-transfected Vero cells. eGFP knockdown was measured as the relative percentage decrease in mean fluorescence intensity (% MFI), normalised to the MFI of the respective irrelevant controls (100%) following transfection. Error bars are indicative of the standard error of the mean (SEM) calculated from duplicate FACS analysis of three separate transfection experiments. Statistical significance was measured by ANOVA with Tukey’s post tests P≤0.05 was considered significant. n = 3 Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 83 pU6-1sheGFP (48.19±5.92; p ≤ 0.01), pU6-2sheGFP (39.95±4.91; p ≤ 0.01), pU6- 4v1sheGFP (38.01±4.67; p ≤ 0.01), as well as a lesser but still significant reduction of eGFP MFI following co-transfection with pU6-3sheGFP (32.76±4.02; p ≤ 0.01) and with pFUGU-sheGFP (25.15 ± 3.09; p ≤ 0.05) p ≤ 0.05. No other constructs induced significant silencing in ZF4 cells. Finally, the variant U6 4 promoter U64v2 showed significantly less efficient knockdown of eGFP than the U64v1 promoter at (23.61 ± 5.24; p ≤ 0.05).

3.3 Utilisation of the chicken Pank-1 Intron 5 for the delivery of shRNAs in zebraﬁsh ZF- 4 cells

The chicken Pank 1 (pantothenate kinase 1) fifth intron was originally cloned and spliced into eGFP by Dr. Kristie Jenkins in our laboratory as a vector for the introduction of shRNAs into chicken cells and eggs. As this vector was available and preliminary experiments suggested that the splicing of the intron out of eGFP was effective in zebrafish cells (Figure 3.12 A). It was utilised in this study as an additional method of introduction of an shRNA into zebrafish ZF4 cells. The Pank 1 gene was chosen for this purpose as it natively encodes the miRNA miR107 and by utilising this intron it is hoped to eventually use nuclease-mediated homologous end joining approaches to modify the intron to contain an shRNA expressed in concert with miR107. miR107 is homologous to miR103 and has been predicted to occur or experimentally confirmed in over 30 species according the current miR- Base database[124,125,190]. In the zebrafish the Pank-1 gene has been duplicated and occurs on chromosomes 6 Pank 1a and 12 Pank 1b and miR107 is contained within both fifth introns, the Pank 1 b gene in zebrafish is considered the homologue to the chicken and mammalian Pank 1 genes[366] as it is most conserved. Alignment of the miRNAs (Figure 3.11) reveals a single nucleotide shift in the predicted mature siRNA sequence[339]. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 84

pZF-U6-4v2 shEGFP Irrelevant

pZF-U6-1 shEGFP pZF-H1 shEGFP

pZF-U6-2 shEGFP pFuguU6shEGFP

pZF-U6-3 shEGFP pMouse U6 shEGFP

pZF-U6-4v1 shEGFP pCH-U6-4 shEGFP

Figure 3.10: shRNA knockdown of eGFP using zebrafish polymerase III promoters in ZF4 cells. shRNA knockdown of eGFP using zebrafish pol III promoters. A Fluores- cent Microscope images of eGFP expression in zebrafish ZF4 cells following co-transfection with shRNA constructs and peGFP-N1. Irrelevant represents non-silencing vector. Images are representative of triplicate transfections. B eGFP Knockdown in ZF4 cells. Flow cytometry results for eGFP knockdown in co-transfected ZF4 cells. eGFP knockdown was measured as the relative percentage decrease in mean fluorescence intensity (% MFI), normalised to the MFI of the respective irrelevant controls (100%) following transfection. Error bars are indicative of the standard error of the mean (SEM) calculated from duplicate FACS analysis of three separate transfection experiments. Statisti- cal significance was measured by ANOVA with Tukey’s post tests P≤0.05 was considered significant. n = 3 Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 85

To test expression of intronic miRNAs in the zebrafish ZF-4 cells the CMV promoter was replaced with the Xenopus EF1α promoter and the two constructs compared to test whether the artificially inserted intron would be spliced out as predicted in the zebrafish ZF-4 cells. Following transfection the expression of eGFP was determined by flow cytometry compared to the expression from unmodified peGFP-N1 plasmids and plasmids pCMVeGFP-Intron, pEF1αeGFP- Intron and those plasmids encoding irrelevant shRNAs pCMVeGFP-Intron-PB and pEF1αeGFP-PBFigure 3.12. Generation of the pCMVeGFP-intron and pCMVeGFP- Intron-PB was performed by Dr. Jenkins and modification of the promoters and all analysis and experimental work was performed by Brian Clarke.

To construct the Pank 1 intron vectors used in this study the original pCMVeGFP- Intron construct generated by Dr. Jenkins was used as the template to extract the entire GFP-intron construct via endonuclease digestion using the HpaI and SalI endonucleases. The HpaI nuclease site occurs within the conserved poly A region in both template and receiving vectors and maintains the in frame connection of the eGFP-intron construct. The Xenopus EF1α promoter region was extracted from the pT2-002 vector kindly provided by Dr. Scott Tyack, utilising a XhoI digest removing the entire promoter region. The EF1α and Intron constructs were simultaneously ligated into a pUC-018-sv40polyA vector pre-digested with SalI and HpaI again provided by Dr. Tyack. Conﬁrmation of correct insertion and insert direction was accomplished utilising digestion with ScytometryalI,as both SalI sites in the inserts and vector will be deleted only if both inserts are incorporated into the vector and by digestion of inserts with EcoRI and BamHI to generate a unique 1789 base product followed by sequencing using the M13 primer combination. This combination of digests was used for both the shRNA positive and negative constructs. Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 86

gga-mir-107 -CUCUUUGCUUUCAGCUUCUUUACAGUGUUGCCUUGUGGCAUGGAGUUCAAGCAGCAUUG 59 dre-mir-107a UCUGUGUGCUCUGAGCUUCUUUACAGUGUUGUCUUGUGGCAUGGAGAUCAAGCAGCAUUG 60 dre-mir-107b UCUGGCCACUCUGGGCUUCUCUACAGUGUUGCCUUGUAGCCUGGUGAUCAAGCAGCAUUG 60 ** ** * ****** ********** ***** ** *** * *************

gga-mir-107 UACAGGGCUAUCAAAGCA---UGGA---- 81 dre-mir-107a UACAGGGCUAUCACAGCACACUGAACAGC 89 dre-mir-107b UACAGGGCUUUCAGCGUGUACAGAAC--- 86 ********* *** * * *

Figure 3.11: Alignment of miR107 from zebrafish and chicken ClustalW2[67] sequence alignment of miR107 a and b from zebrafish and the chicken miR107. Matches are indicted by * and the predicted mature siRNA sequence indicated by red font and the mismatch within the mature sequence highlighted by blue. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 87

3.3.1 Splicing analysis of Pank 1 Intron 5 and delivery of shRNAs into ZF-4 cells

Transfection of the pCMVeGFP-Intron vector into ZF-4 cells resulted in the strong expression of eGFP in transfected cells (Figure 3.12 A). Analysis of relative mean fluorescence intensity by flow cytometryindicated that the four eGFP intron plasmids pCMVeGFP-Intron, pEF1αeGFP-Intron, shRNAs pCMVeGFP-Intron-PB and pEF1αeGFP-PB showed similar levels of eGFP expression in ZF-4 cells and each of the plasmids resulted in slightly lower eGFP expression than the unmodified peGFP-N1 (Figure 3.12 B). Using the primers eGFPintronF and eGFPintronR to amplify a 583 bp fragment of eGFP spanning the spliced intron was confirmed (Figure 3.12 B) along with a longer approximately 4 kb amplicon corresponding to unspliced plasmid DNA this larger amplicon occurred irrespective of reverse transcription and sequencing confirmed it to correspond to the unspliced GFP and is a likely artifact of incomplete DNase digestion of circular plasmid DNA.

Utilising a two step method of artificially poly-A tailing non coding RNAs including miRNAs and shRNAs, followed by qRT-PCR the relative expression of miR107 encoded from plasmids with or without the addition of the shRNA into the intron as determined. Addition of an extra shRNA into the Pank 1 intron resulted in a slight decrease (Figure 3.12 E) in the observed levels of miR107 following transfection however the presence of native miR107 potentially masking this drop as the low transfection efficiency of ZF-4 cells results in only a small yet significant increase in miR107 post transfection of plasmids containing the Pank 1 intron relative to peGFP-N1 transfections. Expression of mature shPB transcripts resulted in the predicted observation of an approximately 80 nucleotide product in constructs containing the shRNA and no product in either untransfected, peGFP-N1 or pCMVeGFP-Intron and pEF1αeGFP-Intron constructs (Figure 3.12 C). Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 88

A pEGFP-N1 pCMVEGFP-Intron pEF1αEGFP-Intron pCMV EGFP-Intron PB pEF1α EGFP-Intron PB

B D FermentasBlank 1kb(+) pEGFP-N1 FermentaspCMVEGFP-Intron 1kb(+)pEF1αEGFP-IntronpCMV EGFP-IntronRT (-) PB

5000 bp 5000 bp

1000 bp

500 bp

pEGFP-N1

pCMVEGFP-IntronpEF1αEGFP-Intron pCMV EGFP-Intron PB C FermentasRT 1kb(+) (-) E pEF1α EGFP-Intron PB

5000 bp 1000 bp

500 bp 200 bp 75 bp

pEGFP-N1

pCMVEGFP-IntronpEF1αEGFP-Intron pCMV EGFP-IntronpEF1α EGFP-Intron PB PB Chapter 3. Expression of shRNA molecules in zebraﬁsh ZF-4 cell culture 89

3.4 Discussion

The research performed in this chapter has characterized zebrafish polymerase III promoters. Specifically, a number of native zebrafish U6 promoters and the native H1 promoter were tested for their ability to silence an eGFP reporter by expression of an shRNA against this gene in both zebrafish ZF4 cell and Vero West African monkey kidney cell line. For comparison, mouse, fugu and chicken U6 promoters were tested in parallel.

The expression of shRNA molecules targeting eGFP from zebrafish U6 1, U6 2, U6 3 and U6 4 v2 promoters all strongly, and significantly, reduced eGFP expression in ZF4 cells. In contrast, expression of those shRNAs from either the mouse and chicken U6 promoters did not reduce the expression of eGFP in this cell line. The shRNAs produced from the zebrafish H1 promoter and the U6-4 promoter each reduced the expression of eGFP slightly, although the reduction was not significant. In the mammalian Vero cell line, the reverse was observed, with shRNAs expressed by the teleost U6 promoters unable to reduce expression of eGFP but both the mouse and chicken U6 promoters produced sufficient shEGFP

Figure 3.12 (preceding page): Transfection of Gallus Gallus Pank 1 intron in ZF4 cells and shRNA and GFP expression. A Transfection of Pank1 GFP intron constructs ultiising either EF1α results in GFP expression in ZF-4 cells. B RT-PCR using primer pair GGAPank1GFP F and GGAPank1GFP F results in clear amplification of both spliced GFP mRNA and amplification of plasmid DNA. C Polyadenalation of small nuclear RNA fraction and RT-PCR using the PAM-urp reverse and shPB-1207 F primer pair identify expression of PB-1207 shRNA only in constructs containing the additional hairpin. D Flow cytometry results for eGFP expression in transfected ZF4 cells. eGFP expression was measured as the relative percentage of in mean fluorescence intensity (% MFI), normalised to the MFI of the peGFP-N1 control (100%) following transfection. E RT-qPCR of miR107 levels post transfection of peGFP-N1 and peGFP intron plasmids. ΔΔCT is calculated relative to the small nuclear RNA 5S and the level of miR107 calculated relative to peGFP-N1 (1) . In all graphs error bars are indicative of the standard error of the mean (SEM) calculated from duplicate FACS or RT-qPCR analysis of three separate transfection experiments. Statistical significance was measured by ANOVA with Tukey’s post tests P≤0.05 was considered significant. indicates significant difference to peGFP-N1. # indicates significant difference between pCMVeGFP-Intron and pCMVeGFP-IntronPB. † indicates significant difference between pEF1αeGFP-Intron and pEF1αeGFP-IntronPB. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 90 molecules to strongly suppress eGFP expression. However, shRNAs produced by the zebrafish H1 promoter weakly, but significantly, reduced eGFP expression in Vero cells.

The zebrafish H1 promoter was identified and compared to those in mouse, chicken and human. The zebrafish H1 promoter contained each of the conserved sequence elements identified in the other H1 promoter sequences, with all promoter elements present in the same order in each species. The zebrafish H1 promoter was the only promoter able to reduce eGFP expression in both the zebrafish and mammalian cell cultures. Two previous comparisons of mammalian U6 and H1 promoters firstly in mice brains and endothelial cells[231] and secondly in Vero cell cultures[42], highlighted that the U6 promoter was a much more effective promoter than the H1 promoter. Our research indicates that the zebrafish is no exception at least in ZF4 cells.

This study identified an additional zebrafish U6 promoter (U6 4) to the three previously characterized in this organism. Genomic analysis indicated the presence of a common variant (U6 4v2) with 3 nucleotide alterations in the OCT motif, which was much less effective than the U6 4v1 promoter at reducing eGFP expression. This result is consistent with previous work that indicated that either the complete or partial removal of the OCT motif resulted in reduced expression of the H1 snRNA gene in mice and HeLa cells[260]. This information may be useful if attenuation of the promoters is required for future applications.

Interestingly, both pMU6 sheGFP and pCHU6-4 sheGFP appeared to significantly increase the expression of eGFP above the level observed in the pZFIrrelevant control in ZF4 cells. As neither of these constructs showed evidence of expression of shRNAs following co-transfection, it remains possible that the pZFIrrelevant construct is causing a moderate non-specific knockdown of expression from the peGFP-N1 plasmid, consistent with other studies suggesting that high levels of RNAi molecules have a non-specific inhibitory affect on the normal cellular functions[112,128,129,403]. Chapter 3. Expression of shRNA molecules in zebrafish ZF-4 cell culture 91

The chicken Pank 1 intron was utilised as an additional method of introducing potentially silencing shRNAs into zebrafish cells, as it allow entry into the RNAi silencing pathway earlier that expression of distinct shRNAs from polymerase III promoters. Using the eGFP-intron constructs It was observed that splicing of the intron was preserved in the non-native ZF-4 cell line and that miR107 encoded within the same intron was preserved and excised as predicted. The addition of an shRNA into the intron preserved the splicing of the eGFP gene and did not disrupt the correct secondary folding of the intron for Drosha family proteins to excise both miR107 and the PB hairpins from the mRNA and the progression of the hairpins into the miRNA pathway and the production of mature miRNAs. Whilst these experiments have not demonstrated silencing from either the artificial miR107 or the non-targeting PB the Pank-1 intron system will be further tested by the addition of shRNAs targeting VHSV in following chapters of this document. Utilisation of the Pank 1 intron allows us to potentially couple an shRNA with a ubiquitously expressed miRNA, miR107 by the utilisation of endonuclease mediated homologous end joining and the GFP-pank1 constructs used in this study represent the first steps in generating these vectors and transgenics.

In this study a functional zebrafish H1 and a new zebrafish U6 promoter, along with a variant of the latter have been identified. Expression of shRNAs from these and three other previously characterized zebrafish U6 promoters has been shown in a zebrafish cell line. The newly identified U6 promoter showed similar levels of activity to previously characterised zebrafish U6 promoters and the H1 promoter showed much lower efficacy as measured by eGFP MFI reduction following co-transfection with peGFP-N1. Importantly the requirement for species- matched promoters was demonstrated for effective expression of, and knockdown by, shRNAs, supporting previous evidence that there are inherent differences between mammalian and avian U6 promoters compared with their teleost counterparts and that the chicken Pank 1 intron 4 when cloned into eGFP is additionally an effective method of introduction of shRNAs into zebrafish cell culture.

Chapter 4

Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro

4.1 Introduction

Chapter 3 detailed the comparison of zebrafish polymerase III promoters, which resulted in the identification of the zebrafish H1 promoter, a novel U6 promoter, and a “variant” U6 promoter (U6-4). This work led to the construction of efficient plasmids for short-hairpin (shRNA) expression. Chapter 3 also characterised the relative strength of each promoter resulting in the use of the U6-2 promoter throughout the remainder of the work described in this chapter. Additionally the chicken pantothenate kinase 1 (Pank 1) intron construct was successfully used to express shRNA molecules in zebrafish ZF-4 cells. To further explore the potential for developing RNAi-based transgenic antiviral strategies, the zebrafish U6-2 promoter was used, in combination with intronic-derived shRNAs, to target VHS virus.

93 Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 94

VHS virus is an important disease of marine finfish, in particular salmoid species worldwide[336] including farmed rainbow trout, salmon, turbot and Japanese flounder[158]. Production losses as a result of direct mortalities can be between 50% and 100% in high-density fish farms. The effect of VHS virus infection is, however, not limited to the farms since it also infects a wide array of wild marine and freshwater wild fish[121]. VHS virus is a 11 - 12 kb negative-sense RNA virus of the genus novirhabdoviridus and the family rhabdoviridae [320]. The viral genome consists of genes encoding the structural nucleoprotein (N), a polymerase-associated phospho-protein (P), a matrix protein (M), a transmembrane glycoprotein (G), a polymerase (L) protein along with a single non-structural nonvirion (NV) protein. Importantly for this work, a zebrafish model of infection was previously established at a variety of developmental stages[265]. Moreover, the transmembrane glycoprotein and the viral polymerase genes have previously been shown to be effectively knockeddownbysiRNAs[43,325], shRNAs[184], and long-dsRNA molecules[182].

The use of shRNA molecules targeting viral genes has been an extremely successful method of inhibiting viral replication in a wide variety of cell culture systems[184,376]. In addition, they have been used with some success in vivo following lentiviral transduction[162] or nasal inoculation[40]. The shRNAs induce effective knockdown by inducing the endogenous RNA interference process[51,110].Tran- scribed shRNAs are exported from the nucleus by Exportin-5 and processed by the RNase-III enzyme Dicer into small dsRNA molecules of between 19 and 23 bp, called siRNAs. One strand of the processed siRNA; the complementary “guide” strand, is incorporated into the RNA-induced silencing complex (RISC) and mediates the sequence-specific cleavage of the target mRNA[129,134]. Using RNAi sequence-specific molecules to target viral genes allows for the targeting of genes and gene regions inaccessible by conventional methods. Application of complexes containing multiple shRNAs that target numerous genes and different regions of the same genes dramatically reduces the likelihood of viral escape [35,132].

This project used the zebrafish U6-2 promoter to generate constructs expressing shRNAs, targeting different regions of each of the G and L genes (3 each). These constructs were transfected into zebrafish ZF-4 cells, and the effect on G and L Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 95 gene expression and viral proliferation determined. To test the sequence-specificity of the inhibition, the expression of IFNφ1 and the IFN-induced MxA genes was also analysed following transfection. The most successful shRNA targeting the G gene of VHS virus was additionally cloned into the pCMVEGFP-intron construct and transfected into infected ZF-4 cells. Viral loads and IFNφ1 genes were again tested.

4.2 shRNA design and construction

Six shRNAs, 3 targeting each of the G and L genes, were designed against VHS virus strain (23.75) [Accession number FN665788.1] using the Dharmacon siRNA Design tool. Sequences were 21 nt long with a total GC content of less than 50%. To identify effective shRNAs, siRNA sequences were further screened using the criteria outlined previously[353]. Briefly, potential siRNAs were scored for the stability of the duplex formed between nucleotides 6 - 11 on the sense, and 14 - 9 on the antisense, strands. Those siRNAs with a central duplex ΔG of ≥−12.9 kcal/mol were discarded, ΔG scores were calculated as described in[114] shRNA sequences and ΔG scores are shown for selected shRNAs in Table 4.1.shRNA expression plasmids were constructed using the one-step PCR method described previously 3.5. The U6-2 forward PCR primer and specific shRNA reverse primer were used to amplify the zebrafish U6-2 promoter region and attach the specific shRNA at the start of transcription (+1). Completed constructs were ligated into the pGEM-T easy vector system. The control shRNA expression plasmid pZF- U6-2shEGFP was constructed as described in[71] and Section 3.2.3. The empty vector plasmid control was created by cutting the pGEM-T vector with EcoRI Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 96 and religating.

Table 4.1: shRNA Sequences

shRNA Name Sequence Score ΔG (kcal/mol) G1207 TT GGA GAC AAA CAC AAT GCT C 4 -12.4 G0906 AT GTT GGT GAT GAG ATG GTT G 4 -13.1 G1076 TG ACT ACT GCC TCA TAG TCA C 4 -14.8 L3082 AT TTG GTG ACG AAC ATT TCG G 5 -13.5 L0832 TC AGA GAA CCG GTT AAA CAA G 4-14 L1005 TC ACT GAC CTC TTG AAT TGC C 4 -13.6

The shRNA target regions of the VHS virus G and L genes were screened for conservation using rVista[240] (Appendix D). A high level of conservation was observed across all included VHS virus sequences and the related viruses IHN virus (Gen- bank NC 001652) and Snakehead rhabdovirus (Genbank NC 000903). (Figure 4.1). The predicted secondary structure of the G and L genes of VHS virus strain 23.75 was determined using mFOLD[405] to provide an estimate of the relative availability of sites for RISC interactions. shRNA molecules were preferentially selected based on regions of low complexity (Figure 4.3) and target availability, since previous analyses of shRNA and siRNA target efficiency have identified that these molecules are significantly more effective when they target regions where the ΔG is ≤ -19 to ≥ -10 kcal/mol[330,375]. Finally, the complete G gene, as well as relevant regions of the L gene, was sequenced to ensure the VHS virus inoculum had not undergone single point mutations in regions of the genome to be targeted.

4.2.1 shRNA screening

As well as ensuring that the shRNA speciﬁcally targeted conserved regions of the VHS virus genome, it was also crucial that shRNAs be designed to avoid interactions with the zebraﬁsh genome since these molecules can be incorporated within miRISC and function as potential miRNAs[52,80,94]. The increasing sophistication Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 97

3 2

shG906 Bits 1

05' 3' 3 2 shG1076 Bits 1

05' 3' 3 2 shG1207 Bits 1

05' 3' 3 2

shL0832 Bits 1

05' 3'

3 2 shL1005 Bits 1

05' 3' 3 2

shL3082 Bits 1

05' 3'

Figure 4.1: Relative conservation of shRNA targets. Sequence conservation was determined by MUSCLE alignment and sequence display generated with weblogo[76]. Relative conservation of individual residues in the mRNA sequence target by the individual shRNAs is indicated by letter height. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 98

shG1207 shL1005 shL3082

-168 -1382 -2268 -2873 -4557 -4925 3` 5` N P M G NV L

1481- 2149- 2959- 5053- 4482- 11007-

shG0906shG1076 shL0852

Viral mRNA transcription level

Figure 4.2: Structural organisation of the VHS virus genome and locations of sequences targeted by the indicated shRNAs. The VHS virus genome consists of a single open reading frame that encodes the various viral proteins, including the nucleocapsid protein (N), polymerase- associated phospho-protein (P), matrix protein (M), surface glycoprotein (G), the non-virion protein (NV), and viral polymerase (L). The start and ﬁnish of each gene are indicated numerically on the genome. As with all (-) ssRNA viruses, VHS virus mRNA is more heavily expressed towards the 3 end of the genome as indicated by the gradient. Positions of shRNA targets on the Glycoprotein (G) and viral polymerase (L) genes are indicated with shRNA molecules named for the 5 most base pair. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 99

C A

U G G C C G C shG1076 C G A U C U C U A A C C A A A 760 U A U G U C C C G A G C G G A G A G C A U U A C G C G C G G C U G G U A A U 1080 C G G C 400 A U U A A A U G U G U C A A G C G A C A

C C A 720 U A U G U A G C 800 A U C G C C C A G A C G U G A U G C U A G A G C U U A U A C G C G A A C A C A G G A A C A G A G U C A C A A A G C G C G G C G A G A A U C U C U A G C C G 840 C C C U A U A U A A U A A G G C A G C C U G C G A A U U C C C G C G G A 640 C G 680 C A U G G A G U G C U A C G G U C U C G U A G C C C A G U G A G C C C A C G G C G C U G A A C C G A U C G 600 U G A A C G U U C G C A U C G C G U G A U G C G U U C G G A C C G G G U C U G C A

C C C A U G A C A A A C A A U A C U G U G C U A U C C U A G U G A G A A C A 560 U A G A A G G C A U G U A G A G A A U C A C G G 880 U A U A C U G A C C G C G A A A U U A A C G C G U A C C C U U A U A A A A C A C A A C AG A G C U G 520U U G C G A C A G C C A U A G UA C U C A U C C A G C C A A U A A A G A U U A G U A C G C G U A U G C A G U U U U C A U U G C U U G G A C A U G A U C U C G G C C C G C U 920 A G A C A G A A G C C C G C C G C C A U G C G U C A A U C A C A C A A C A 480 G A U 960 U A C G C C U G G C C A A U A C U G C A U G A U C C G C G C A C G C U U A G U C A G C A U G A C G G A C A U U A A U G A C G G U A U C C C U G A G A 1000 A C A C C G A G A C A C A A G U U G A U A A G U C U G C G G C C C A C 440 A A A C U A U C U C G C G U A C G C G U A C C U U C G A C U A C U A UG G C 1040 U A U A U A C G G C U G A G C C G G C C U C A C C C G G A G C G C A C U C G A G G A U C C A C U A A C U C A C A U G C 400 1080 A U A U A C A U G C C G U A A C C G C G A U C A A A A A G A G U A G G G C U U U C C U C C U U C G A G C G A U A A U U C C G G G G C 360 C G U G U U C G C U G C G 320 C C C G G G G C A A A C A A C A G U G U C G A U G A C A A G G U C C U C U U C G A A C U A G C A 1120 U A C G G G A U U G A A G U G C C C U A A A G G U U C GA C A U AC CA AU AAC U U G A A U shG1207 G 1200 C C G A G C U C C A A U C A A C A C C A C A U G C 1160 A A U C A G A C G C

U U A C G C 280 A A C A U A C G A A A U G A C A A A C A A G U U G A G A C A C A A G C shG0906 G C G A U C A U A A C G C G C 1200 G A U C A A G C A A A C C A C C C A U G A

U U G C C G A C G C U U C G C U A C A A 200 C G G G C G A A C A G C C G G C G A U U G G C G C A G A U A G U C C A U C U A C A A U G A A U G U U A C G C C A A A G G U U G A C C C U A U C G U G A U C G G G A U G G U A A C U A U U U G C G G 240 G U G C C U A C U C C A U A A A A C C C G G C G U A G A U G C G G A G U 1240 U A A U U C C C A U A U A U C A G A C C G A U A G G C C A A C A G U C U G C G A C A C U C U G A G A U C A G G U 160 C A U C A A C U U U A U C A A A A C A 120 G G C A A C A C A G U G U G A C U U G A C G C G C C A C U U G C U A G C G U U G A A U A U C U C G G 1280 G A A U G U G A G G U C A U A C U U U G A C A G A G C C C C U U A C U A A C G U A C A A A C G G G G C A U A A G G C C C G U G U A 1240 G CG G U A 80 A A U C A A A C U U U A A C U A A C U G G A U U G G C C A U A U C A G U U A C A A U C A U U U U G U C U U U U G U C U U C G U A G U C C C A C A C A A U G C U C A U A U U U A U A A U 1320 C 1360 G G A C U U C G C A A A U U U C G U C A 40 C A U C G U U C G U A C A G U C A U A C G C C G U A U U A C G U C U G A A A A A U A C U C C C U C U U A G G U G A A C U A C C A U U A U C A U1400 A C C A U C C G U C G C G G C U G A G C U G C C U C A C G U G C A U U U 920 A U G C ҋ G G A C A C AA C G U G C C A U A G A A ҋ A U U A C G G C G A A A A U A A A G C C A C U A G A A U C C C A U A G G G U A 1600 A U C C U G C G U A A G U G U G C A U 1440 U C G U G C G C G U G U U A U A G C U A G C G C G A

U A A U G U A C C U G G C C A G C U U G C G C U C U C G C G C U A A C A C C G U 1560 A C G U U C G U 1480 C G U G G C U A U G

C G A C U G C G U G A U U A A G C C G C G U A A C U A U A G G C G C C U G C U A C 1520 G C G C C G U C C C A C U A U U A U A A C C

Figure 4.3: Targets of shRNAs on the VHS virus glycoprotein mRNA secondary structure The folding of the VHS virus G gene RNA was predicted by thermodynamic energy rules included in the MFOLD software[405] using the sequence VHS virus 23.75 (Genebank accession number U28799.1). shRNA target sites are boxed and highlighted. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 100 of miRNA target analysis has shown that miRNA targeting is complex and so three separate tools were utilised to identify potential interactions with the zebrafish genome as shRNA / siRNA, or miRNA. Both strands of the shRNAs, in their totality and the potential seed regions 5 (bases 2 - 13), and the loop sequences were screened independently against the entire Danio Rerio Refseq database using the BLASTN[10], optimising the search for short sequences. shRNAs that had less than 4 mismatches within the seed region or 12 consecutive bases matching regardless of the base positions were discarded (data not shown). Two additional search tools, RNAhybrid[296] and Targetscan[209], were also used. These calculate RNA binding efficiency across the whole shRNA, and the seed region, respectively. Any potential shRNA that had the potential to function as a miRNA capable of binding either non-specifically to the VHS virus genome or the zebrafish genome, was discarded.

4.3 Growth of VHS virus in ZF-4 cells

4.3.1 Viral haemorrhagic septicaemia virus growth kinetics in ZF-4 cells.

Zebrafish ZF-4 cells are typically grown at 28 ◦C. However, VHS virus growth is highly temperature dependent[103,169,187] with virus replication completely inhibited at temperatures greater than 15 ◦C. Therefore, the temperature was lowered to room temperature overnight (approximately 21 ◦C) and the cells then infected with virus and incubated at 15 ◦C.Not surprisingly, ZF-4 cells grown at 15 ◦C showed significantly slower growth than at higher temperatures. There was some evidence of cell death in the days immediately following incubation at the lower temperature (Figure 4.4) although overall the cells appeared healthy. Following infection with VHS virus, the viability of the ZF-4 cells was assessed by qualitatively estimating the confluence of the monolayer (Figure 4.4 A). A significant decrease in cell viability compared to uninfected cells was observed at 1 day post infection Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 101

(pi) following inoculation at an MOI of 1, at 2 day pi with an MOI of 10-1,and at 5 day pi with an MOI of 10-2. Extracellular virus particles were titrated daily pi, with maximum values of around 108 TCID50. Based on these experiments, it was concluded that MOIs of less than 10-1 would be most useful for subsequent experiments, with cells and supernatants assayed at 5 day pi for CPE and viral load. Typical ZF-4 uninfected and infected cells and CPE are shown in Figure 4.5.

4.3.2 The eﬀect of nucleofection on VHS virus growth in zebraﬁsh cells

It was next necessary to determine whether the stress of nucleofection and relatively rapid temperature changes would alter either viral replication or plasmid- mediated expression. Utilising the transfection methods determined in the previous chapter 3.7, ZF-4 cells were transfected with pEGFP-Max and pZFU6- 2shEGFP and allowed to recover and attach at 28 ◦C for 6 h to 24 h. Cells were then moved to room temperature (21 ◦Cto23◦C) for a further 3 h to 12 h before final transfer to 15 ◦C(Figure4.6). EGFP expression, as assessed by flow cytometry and was found to be reduced by around 30%., However, the efficiency of transfection as determined by the number of EGFP positive cells was not altered by the change in temperature (Figure 4.6 A&B). Cell viability was assessed by Trypan blue staining at 24 h post incubation at 15 ◦C and showed that there was no significant decrease in cell viability following insulation at lower temperature (data not shown).

To determine whether transfection of non-specific shRNA (shEGFP) or plasmid would affect virus growth, transfected cells were inoculated with a range of MOI of VHS virus, and supernatants harvested to determine virus titres. Nucleofection alone, or with either plasmid, reduced ZF-4 susceptibility to VHS virus infection slightly although this effect was not statistically significant, and there was no difference between transfection with shRNA delivery plasmid compared to pEGFP-N1 Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 102

108 MOI 1 107 MOI 10-1 MOI 10-2 6

50 10

5 TCID 10

104 (Limit of Detection) 103 1 2 3 4 5 6 7 8

Days post infection

Figure 4.4: VHS virus replication in ZF-4 cells Supernatant was harvested from infected ZF-4 cells, at 24 h intervals and virus titres determined by serial dilution and end point TCID50 titration using the Reed & Muench method following VHS virus infection[295]. Virus growth was determined from 2 independent experiments. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 103

Uninfected ZF-4

VHS virus MOI ZF-4

Figure 4.5: Preliminary infection of ZF-4 cells with VHS virus Uninfected and infected ZF-4 cells following incubation at 15 ◦C. Natural blank areas of uninfected ZF-4 monolayer are indicated with a black arrow. This is a normal phenomenon of ZF-4 cells. Grey arrows indicate remaining monolayer beneath the lifted infected cells. White arrows indicate typical VHS virus CPE consisting of ﬂoating, rounded, and bunched cells. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 104

AB 50000 100

40000 80

30000 60

20000 40

10000 (+) (%) Mean GFP 20

0 0 Mean Flourescence Intensity Flourescence Mean x x

pEGFP-Max pEGFP-Max pEGFP-Ma pEGFP-Ma Untransfected Untransfected Untransfected Untransfected 28$C15$C CD28$C15$C 8 10 2.0 Untransfected 107 pEGFP-Max 1.5 6 50 10 1.0 105 TCID pEGFP-Max

Fold Change 0.5 104 Untrasfected

103 0.0 1 2 3 4 5 6 7 A N Mx Days post infection IF Figure 4.6: Zebrafish ZF-4 transfection with pEGFP-Max and infection with VHS virus A Relative mean fluorescence intensity of cells either untransfected, or transfected with pEGFP-Max at the indicated temperature B Flow cytometry results for EGFP described in panel A, expressed as a percentage of sampled cells. C Supernatant virus titres determined by TCID50 assay for indicated ZF-4 cells post inoculation with VHS virus. D Expression of IFNφ1 or MxA analysed by real-time PCR in indicated zebrafish ZF-4 cells. The results are shown as mean of two independent experiments and error bars indicate SEM. βActin served as the reference gene. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 105 plasmid (Figure 4.6). This suggests the nucleofection and temperature reduction strategy determined was suitable for effective shRNA expression in ZF-4 cells.

Viral loads post nucleofection were greater than viral loads observed in cells not treated by nucleofection (Figure 4.6 C) at equivalent time points post infection. It may be that the combined stress of the nucleofection and repeated temperature changes over the days immediately prior to challenge decreased the endogenous resistance of the cells to the viral infection. There was however, no diﬀerence in the expression of MxA or IFNα in transfected and untransfected cell layers (Figure 4.6 D).

4.4 shRNA mediated VHS virus suppression

4.4.1 Transfection of shRNA plasmids reduces VHS virus replication

ZF-4 cells were qualitatively analysed post transfection and infection for the presence of viral cytopathic eﬀects (CPE). Some evidence of viral CPE was observed in all infected cell layers at all MOIs. CPE was consistent with typical observations of VHS virus and rhabdoviral CPE observed in other studies[13,116]. ZF-4 cells became rounded and swollen, and readily detaching from the monolayer into ﬂoating clumps of rounded cells that then clustered into typical rhabdoviral grape cluster formations[185]. Observations of cell layers are included from the highest MOI 10-1 as an example (Figure 4.7) as this infection level showed the greatest variation between the shRNA plasmids[21,71].

To accurately quantify the reduction of viral replication following transfection and infection, supernatants were titrated in EPC cells (Figures 4.8 and 4.9). EPC cells form a consistent monolayer and readily grow the virus at 15 ◦C. Transfection of the non-speciﬁc shEGFP did not signiﬁcantly reduce VHS virus titres relative to transfection with the empty vector control. All three shRNAs targeting the glycoprotein (G) gene (shG0906, shG1076, and shG1207) reduced viral titres at Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 106 all MOIs tested relative to both the empty vector and shEGFP controls. All three G gene shRNAs prevented virus production completely at inoculation of 10-5 MOI, and shG0906 and shG1207 also completely inhibited virus growth at 10-4 MOI. Of the three shRNAs targeting the polymerase (L) gene, shL0852 and shL3082 reduced viral titres at MOIs between 10-2 and 10-5 compared to both control transfections. However, only shL3082 could completely prevent virus production at 10-5 MOI.

4.4.2 Eﬀect of shRNA expression on MxA and IFNφ1ex- pression.

Interferon stimulation by dsRNA species, including shRNAs and siRNAs,[43,307,324,325] and resulting activation of the anti-viral pathways has been previously shown to impact on VHS virus replication and pathogenicity in vivo [171,350]. Therefore, experiments were performed to ensure that the observed reductions of CPE, virus titre, and viral genomic RNA were a direct result of the transfection of shRNAs and not an indirect anti-viral eﬀect.

Poly(I:C) is a dsRNA mimic, which added to cell culture media induces a strong IFN response primarily through TLR-like receptors (TLRs) after endocytosis as well as later via RLRs after export to the cytoplasm in mammals. It can be assumed to be the same in zebrafish, and was included as a positive control. Stimulation of ZF-4 cells with poly(I:C) 6 hours prior to inoculation with MOI 10-1 or 10-2 virus, led to a significant reduction in virus titres relative to ZF-4 cells transfected with both an empty vector and vector expressing an shRNA targeting eGFP (shEGFP) [Figure 4.10]. pZFU6-2shEGFP did not inhibit virus growth or CPE (data not shown) indicating that properly folded shRNAs do not induce significant IFN responses.

To further ensure that the use of shRNA molecules did not induce an interferon response, two zebraﬁsh immune genes, interferon and the down-stream interferonβ stimulated gene (ISG) MxA were quantiﬁed by qRT-PCR at 40 hours post-transfection, Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 107

UninnoculatedEmpty Vector shEGFP

shG0906 shG1076 shG1207

shL1207 shL0805 shL3082

Figure 4.7: Cytopathic effect of VHS virus on ZF-4 cells is inhibited by specific shRNA transfection Light microscopy demonstrating cytopathic effects produced by inoculation of VHS virus onto ZF-4 cells (3 days post-inoculation at 10-1 MOI) transfected with the empty vector or the indicated shRNA plasmids compared to uninoculated cells. Natural blank areas of ZF-4 monolayer are indicated with a black arrow. This is a normal phenomenon of ZF-4 cells. Grey arrow indicates remaining monolayer beneath the lifted infected cells. White arrows indicate typical VHS virus CPE, floating, rounded, and bunched cells. Scale bar = 100 μm Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 108

1×109 Empty Vector 1×109 shEGFP 1×108 shEGFP 1×108 shG0906 1×107 1×107 1×106 1×106 50 1×105 1×105 ** ** 1×104 1×104 ***

TCID 3 3 1×10 1×10 Limit of detection 1×102 1×102 1×101 1×101 *** *** 1×100 1×100 -1 -2 -3 -4 -5 -1 -2 -3 -4 -5 10 10 10 10 10 10 10 10 10 10

1×109 shEGFP 1×109 shEGFP 1×108 shG1076 1×108 shG1207 1×107 ** 1×107 1×106 1×106 50 1×105 1×105 1×104 1×104 *** ** ***

TCID 3 3 1×10 1×10 Limit of detection 1×102 1×102 1×101 1×101 *** 1×100 1×100 *** *** -1 -2 -3 -4 -5 -1 -2 -3 -4 -5 10 10 10 10 10 10 10 10 10 10 Multiplicity of infection Multiplicity of infection Figure 4.8: shRNA molecules targeting VHS virus glycoprotein (G) inhibit VHS virus replication ZF-4 cells were grown under standard culture conditions and transfected as previously described with shRNA (shEGFP) vector or speciﬁc G gene shRNAs. Cells were infected with the indicated MOI and supernatant harvested at 72 hours and assayed by TCID50. Graphs indicate mean (n = 2) and error bars indicate SEM. All statistical analysis was performed by ANOVA with Tukey’s post-test for multiple comparisons. Experiments on all shRNAs were conducted together and analysed by ANOVA as a single experiment. * p ≥ 0.05 *** p ≥ 0.01 *** p ≥ 0.001 relative to shEGFP plasmid. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 109

9 1×10 Empty Vector 1×109 shEGFP 8 1×10 shEGFP 1×108 shL0852 7 1×10 1×107 6 1×10 1×106

50 5 1×10 1×105 4 1×10 1×104 TCID 1×103 3 1×10 Limit of detection 2 1×10 1×102 1 1×10 1×101 0 1×10 1×100 -1 -2 -3 -4 -5 -1 -2 -3 -4 -5 10 10 10 10 10 10 10 10 10 10

1×109 shEGFP 1×109 shEGFP 1×108 shL1005 1×108 shL3082 1×107 1×107 1×106 1×106 * 50 1×105 1×105 1×104 1×104

TCID 3 3 1×10 1×10 Limit of detection 1×102 1×102 1×101 1×101 1×100 1×100 *** *** -1 -2 -3 -4 -5 -1 -2 -3 -4 -5 10 10 10 10 10 10 10 10 10 10 Multiplicity of infection Multiplicity of infection Figure 4.9: shRNA molecules targeting VHS virus polymerase (L) have a range of eﬀects on VHS virus replication ZF-4 cells were grown under standard culture conditions and transfected as previously described with shRNA (shEGFP) vector or speciﬁc L gene shRNAs. Cells were infected with the indicated MOI and supernatant harvested at 72 hours and assayed by TCID50. Graphs indicate mean (n = 2) and error bars indicate SEM. All statistical analysis was performed by ANOVA with Tukey’s post-test for multiple comparisons. Experiments on all shRNAs were conducted together and analysed by ANOVA as a single experiment. * p ≥ 0.05 *** p ≥ 0.01 *** p ≥ 0.001 relative to shEGFP plasmid. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 110

1×109 Empty Vector shEGFP 1×108 1×107 Poly:IC 1×106 1×105 1×104 3 1×10 Limit of detection 1×102 1×101 1×100 -1 -2 10 10 Figure 4.10: The effect of poly(I:C) on VHS virus replication in ZF-4 cells ZF-4 cells were grown under standard culture conditions and transfected as previously described with either a control (empty vector) or shRNA (shEGFP) vector, or incubated with poly(I:C) (10 g/ml) for 6 hours. Cells were infected at the indicated MOI and supernatant harvested at 72 hours and virus titrated by TCID50. Graphs show means (n = 2) and error bars indicate SEM. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 111 the same time point as when infection was performed. Only one shRNA plasmid, shL1005 induced a significant increase in either MxA or IFN (Figure 4.11). Inter- estingly, this plasmid was also the sole shRNA that failed to suppress VHS virus replication in these cells. Further investigation and RNA stability and folding analysis revealed this shRNA showed significantly decreased RNA stability when analysed by mFOLD[405]. These data suggest that properly designed and folded shRNAs do not induce a significant immune response in zebrafish and that the observed reduction of VHS virus (as measured by CPE, titre and mRNA) is the direct result of RNA interference (Data not shown).

4.5 Intron-mediated knockdown of VHS virus in vitro.

As an alternative to the pol III shRNA transfection method, the Pank1 5b GFP intron methodology was trialed. Relevant constructs of the most effective shRNA identified in the previous section 4.4 (shG0906) as well as an shRNA scramble control (shScramble) were prepared (see Table 4.2 for shScramble and shG09096 sequences). Utilising the same nucleofection protocol as for the shRNA pol III expression plasmids, the GFP intron plasmids were successfully transfected into the ZF-4 cells. However, upon temperature reduction to 15 ◦C, expression of GFP was significantly reduced compare to at 28 ◦C (see 4.12 A). Transfected cells were inoculated with VHS virus at MOIs of 10-1 and 10-2. Transfection of neither the GFP - shScramble nor GFP - shG0906 plasmids resulted in a reduction of VHS virus titre. qRT-PCR analysis of MxA and IFNβ in ZF-4 cells post transfection showed that neither of the transfected plasmids induced significant levels of either mRNA (data not shown).

To analyse the reduction of GFP expression, relative expression at 28 ◦Cand15◦C was determined by FACS see Figure 4.12 B). In addition, the relative expression Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 112

A 60 50 40 30 10

5 Fold Change

shEGFP shG0906 shG1076 shG1207

Poly:IC ControlUntransfected B MxA 60 50 IFN-ĳ 40 30 10

5 Fold Change

sL1005 shEGFP shL0805 shL3082

Poly:IC ControlUntransfected

Figure 4.11: The eﬀect of poly(I:C) and shRNA expression on IFN and MxA in ZF-4 cells at 15 ◦C ZF-4s were grown under standard culture conditions (Untransfected) or incubated with poly(I:C) (10 μg/ml) or transfected with appropriate shRNA expression vectors, and the expression of IFNφ1or MxA was analysed by real-time PCR. The results are shown as the mean of two independent experiments and error bars indicate SEM. βActin served as the reference gene. A shRNA molecules targeting the G gene. B shRNA molecules targeting the L gene. Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 113 Fold Change Fold Change Fold Fold Change Fold Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 114 of both the inserted shRNAs as well as the native miR-107 was determined (Figure 4.12 C and

Table 4.2: Sequence of shRNA Scramble and G0906 sequences

shRNA Name Sequence Score ΔG (kcal/mol) G0906 AT GTT GGT GAT GAG ATG GTT G 4 -13.1 Scramble TG GTT GGT GAT GAA TTG GTA GA 5 -14.7

4.6 Discussion

The primary aims of both Chapter 3 and 4 were to develop an optimised shRNA expression system for use in zebrafish, by the identification and comparison of novel zebrafish type 3 RNA pol III promoters. This Chapter described the identification of 5 functional shRNA sequences, which proved to be effective for the reduction of VHS virus titre compared to either an shRNA targeting eGFP or to a virus specific shRNA subsequently shown to fold incorrectly. As part of an important further assessment of alternatives to pol III promoters, this Chapter also extended this work to insert the most effective shRNA into a previously generated eGFP Pank 1 intron. However, the temperature required for VHS virus replication inhibited the effectiveness of the intron to express shRNAs. Therefore, the opportunity to inhibit VHS virus replication was impaired.

This study has identified 5 shRNA constructs that can strongly and significantly reduce VHS virus replication in zebrafish cells. Targeting the glycoprotein gene

Figure 4.12 (preceding page): Pank1 intron is not a suitable method for introducing shRNA expression in ZF-4 cells at 15 ◦C A Relative mean fluorescence intensity of eGFP expression at 28 ◦Cand15◦Cre- spectively. B Percentage transfection efficiency of cells transfected with pEGFP- max or pEGFP-intron plasmids. C shScramble expression D shG906 expression E miR107 expression. The results are shown as the mean of two independent experiments, and error bars indicate SEM. U6 snRNA served as the reference gene for all qRT-PCR analysis. All statistical analysis was performed by ANOVA with Tukey’s post-test for multiple comparisons. * p ≥ 0.05. ** p ≥ 0.01. *** p ≥ 0.001at 15 ◦C relative to the relevant sample at 28 ◦C Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 115 was more effective than targeting the viral polymerase, particularly at higher levels of viral infection. shRNA effectiveness also appears to be specific for VHS virus. One identified shRNA was shown to fold incorrectly which not only prevented its action on the VHS virus genome, but also induced a small interferon response. In addition, the Pank 1 intron inserted into GFP with either scramble or the most effective shRNA, shG0906, did not reduce VHS virus titre significantly as the reduction in temperature from 28 ◦Cto15◦C appears to significantly modify the secondary folding of the intron such that either the intron is no longer efficiently removed from the GFP mRNA or the shRNA and endogenous miRNA are no longer excised.

As is discussed elegantly in[324], an ideal control to demonstrate specificity of action of the shRNAs would be the use of a heterologous virus control. However, preliminary infection experiments of ZF-4 cell with Infectious Hematopoietic Necrosis virus (IHN virus), a closely related member of the novirhabdovirus family, showed no evidence of CPE or viral RNA (data not shown). Despite the absence of this additional control, the lack of significant induction of either MxA or IFNβ following transfection of shRNAs, along with the spread of relative knockdown between different shRNAs tested, provided strong evidence of a specific response.

The in silico identification of microRNA targets is very complex and, despite significant research, remains a poorly understood field. The rapid development of small RNAseq and tools for cross-linked immuno-precipitation (CLIP), such as individual nucleotide-resolution CLIP (iCLIP) and photoactivatable ribonucleoside- enhanced CLIP (PAR-CLIP), have rapidly increased the resolution of miRNA target investigation. Application of these tools has led to the in vitro and in vivo validation of in silico predictions of miRNA target and seed site predictions. Unfortunately this had led to the realisation that in many cases miRNA target prediction algorithms were flawed or simply incorrect.

One major concern about the use of RNAi-based therapies for treatment of viral infections is the potential for virus escape mutants that would then be resistant. This is especially important in the case of RNA viruses as they are notoriously Chapter 4. Knockdown of viral haemorrhagic septicaemia virus using shRNAs in vitro 116 prone to mutation during replication often enabling escape from immune system defence mechanisms[35]. Considering such mutations may also help escape attack by sequence-dependent RNA silencing, highly conserved regions of the glycoprotein and polymerase genes were chosen as primary target areas. The viral L gene has previously been shown to be a difficult target for RNAi inhibition [307] requiring constitutive expression of three independent shRNAs to produce a significant decrease in the number of infected cells by FACS. However, since that study used different shRNAs, it was difficult to say definitively if the difficulty of targeting the viral gene is related to the folding of the L mRNA, the selection of the shRNAs or the requirement of the virus for lower levels of polymerase to produce functional infectious particles. VHS virus is typical of rhabdovirus species in that the further the gene is from the 5 end of the genome the lower the expression of the gene[109,159].

The application of RNAi has been of signiﬁcant impact in probing gene function and is now promising to provide new strategies in the development of novel therapeutics. The versatility of RNAi makes this technology particularly attractive in the pursuit of therapies directed against viral diseases, including the many viral diseases of livestock. In the present study, it was shown that VHS virus, like other single-stranded RNA viruses, was susceptible to degradation via the RNAi pathway. Replication of VHS virus was eﬃciently suppressed using pol III-driven shRNA-mediated RNAi, thereby providing a novel avenue for research into the future prevention of VHS and other virus infections. Chapter 5

Creation and characterisation of transgenic zebraﬁsh

5.1 Introduction

The investigation of novel zebrafish type 3 RNA pol III promoters for the expression of shRNAs in zebrafish resulted in the characterisation of the zebrafish H1 promoter and a novel zebrafish U6 promoter and the construction of highly efficient plasmids for shRNA expression (see Chapter 3). These plasmids were used to drive the expression of shRNAs designed to target either the glycoprotein (G) or polymerase (L) genes of VHS virus. Following transfection, five of the six tested shRNA molecules effectively reduced VHS virus replication 10 fold or more across a range of MOIs. The identified promoters and shRNAs enable generation of expression cassettes utilising the medaka fish Tol-2 transposon system for generating shRNA expressing strains of zebrafish (Figure 5.3).

The medaka ﬁsh Tol-2 element is an autonomous transposon that encodes a transposase. This Tol-2 transposase can act upon any sequence bounded by the 200 and 150 base pairs of DNA from the left (TLE) and right ends (TRE) of the Tol-2 sequence, respectively (Figure 5.1). The TLE and TRE Tol-2 sequences contain essential terminal inverted repeats and subterminal sequences, with DNA inserts

117 Chapter 5. Creation and characterisation of transgenic zebrafish 118 of up to 11 kb able to be inserted between the TLE and TRE without reducing transpositional activity [361]. The Tol-2 transposon system has been shown to be active in many different species, including zebrafish, Xenopus, chicken, mouse, and human both in vitro and in vivo [173,361].

In the Tol-2-based method of transgenic zebrafish generation, a transposon-donor plasmid and synthetic in vitro transcribed mRNA (Figure 5.1) encoding the transposase will be microinjected into fertilized eggs. The transposase protein will be translated from the injected mRNA and excise the transposon construct from the donor plasmid, leading to stable integration of the excised DNA into the genome[173–175,361]. The injected mRNA and the transposase protein will gradu- ally degrade, and, after transposase activity has dissipated, the Tol-2 insertions become stable. The integration events occur during early stages of embryonic development with, some occurring in cells destined to produce primordial germ cells. By outcrossing the injected founder (F0) fish, germline transmission of Tol-2 insertions can be selected for in the F1 generation (Figure 5.2)[173–175,361]. While transgenic integration of RNAi inducing molecules has been established in a range invertebrate models, for example, Drosophila, aedes aegypti mosquitos, and Bombix mori silkworms[3,157,165], the integration of transgenic RNAi constructs in vertebrates has not been as simple to accomplish. As has been discussed in depth previously, the delivery of shRNA molecules via retroviral vectors to induce tissue specific integration of the RNAi cassette resulted in the non-specific inhibition of the endogenous miRNA pathway and increased morbidity and hepatic disease in mice[119,126–129]. Attempts to generate transgenic knockdown zebrafish have met with mixed success. Recently, however, there has been a resurgence in attempts to create heritable transgenic zebrafish which express shRNA or miRNA like hairpins that target genes of interest[15,96,333]. A reliable constitutively-expressed shRNA system for zebrafish has still yet to be realised, with the results discussed in this chapter providing insight into the non-specific and varied effects of RNAi previously observed in zebrafish. Chapter 5. Creation and characterisation of transgenic zebrafish 119

A A AA AA

1 4682

B TTTTT

200 bp 150 bp TLE ()њ eGFP polyA U6 shRNA TRE

Figure 5.1: Tol-2 transposable elements and basic shRNA vector. A Structure of full-length Tol-2, with coding regions of the Tol-2 gene are indicated by solid lines (exons) and dotted lines (introns). Grey boxes and white boxes show coding regions and untranslated regions, respectively. Black arrow- heads in boxes at both ends indicate 12 bp terminal inverted repeats (TIRs). B Structure of the Tol-2 vector with the green fluorescent protein (GFP) expression cassette and pol III expressed shRNA. The minimal Tol-2 vector contains 200 and 150 bp of DNA from the left and right ends, which include TIRs (black arrows) and subterminal regions (open boxes). Modified from[173–175,361]. Chapter 5. Creation and characterisation of transgenic zebrafish 120

T7 Promoter pTransposase

T7 Megascript Transcription

TLE

XhoI E F 1 a

e T G

F a

n A P

Transposase A s AA + pTol2-eGFP:shRNA p

5` AAA o

A s

N A

B TRE Chapter 5. Creation and characterisation of transgenic zebraﬁsh 121

5.2 Construction and characterisation of Tol-2 expression plasmids

To facilitate the generation of an appropriate transposon plasmid for the production of transgenic zebrafish the pT2-002 was kindly provided by Dr Scott Tyack (CSIRO, Geelong, Australia) (Figure 5.3 A). This plasmid contained the chicken β-actin and human cytomegalovirus immediate early promoters (CAGGS)[147],to drive expression of eGFP in chickens, as well as the Tol-2 TLE and TRE elements. The Xenopus EF1α promoter was amplified from pEF-eGFP plasmid using the primers EF1αBglII-F and EF1αBamHI-R and inserted into the BglII and BamHI sites of the pT2-002 plasmid to generate pTol2-EF1α:GFP. Insertion of the pol III type 3 shRNA construct was achieved by XhoI and SalI restriction enzyme digestion of the pZFU6-2shG906 or pZFH1shG906 plasmids and ligation into pTol2- EF1α:GFP-U62:shG0906 and pTol2-EF1α:GFP-H1:shG0906 respectively (see Fig- ure 5.3). To test for effective expression the plasmids were transfected into ZF-4 cells and expression of EGFP observed by fluorescence microscopy (Figure 5.3 B), while, expression of shRNAXenopus was measured by qRT-PCR (Figure 5.3 C).

Figure 5.2 (preceding page): Transgenesis in zebrafish. A Transposase gene is transcribed in vitro to produce 5 capped and poly-A tailed mRNA. The mRNA and Tol-2 construct containing the Xenopus EF1α promoter eGFP and the relevant pol III type 3 promoter expressing either the control shRNA or shG0906 are co-injected into fertilized zebrafish eggs. B Single-cell zebrafish eggs injected with the Tol-2 constructs become the F0 generation. Germ cells of the FO fish are mosaic, and crossing with wild-type fish yields nontransgenic fish and transgenic fish heterozygous for the Tol-2 insertion. In this figure chimeric larvae are represented by random scattering of green fluorescence and heterozygous F1 fish by fully green adult fish. Figure modified from[173–175,361]. Chapter 5. Creation and characterisation of transgenic zebrafish 122

BglIII

E 1 F Tol2

LHS BamHI EF 1

SpeI

r T

P r a M n

e A

s G

pTol2 EF1 p

I I

Tol2 N A XhoI RHS SalI

pTol2 EF1 pTol2 EF1 pTol2 EF1

C 25000

20000

15000

10000 Fold Change 5000

0 e

U6 shG0909 H1 shG0906 U6 scrambl H1 Scramble Chapter 5. Creation and characterisation of transgenic zebraﬁsh 123

5.3 Full strength pol III promoters lead to high levels of mutation and mortality in zebraﬁsh.

5.3.1 Microinjection of zebraﬁsh embryos

The Tol-2 transposon plasmids and transposase mRNA was introduced into zebrafish by microinjection into fertilized eggs collected from wild-type mating zebrafish pairs at the single cell stage. Efficiency of transposon incorporation and required doses of Tol-2 system components were determined by titration of transposase and control pTol2-EF1α:eGFP plasmid, with a 3:1 ratio of plasmid DNA to transcribed capped mRNA at a final concentration of 120 ng/μl determined to be the most effective solution. Increasing the DNA:RNA ratio past this concentration resulted in increased mortality of embryos following expression of control plasmid with no effective increase in EGFP positive larvae (data not shown).

5.3.2 shRNA overexpression increases mortality and developmental defects in zebraﬁsh larvae

U6 and H1 shRNA expression plasmids were microinjected using the optimised DNA:RNA rations and concentrations. Zebraﬁsh were injected with pTol2-EF1α:GFP- U6:shRNA, and pTol2-EF1α:GFP-H1:shRNA were screened at 72 hpf for survival,

Figure 5.3 (preceding page): Construction and validation of pTol2- EF1α:GFP-pol III:shRNA expression plasmids. A The CCAGS promoter was excised from pT2-002 with BglII and BamHI and the vector backbone used as the basis for the zebrafish transposon plasmid. The EF1α promoter was amplified by PCR and ligated with the same enzymes to form pTol2 EF1α:eGFP. The relevant pol III promoter and shRNA cassette was also amplified by PCR and ligated into the XhoI site using XhoI and SalI. The resultant plasmid was named pTol2-EF1α:GFP- pol III:shRNA. B pTol2-EF1α:GFP, pTol2-EF1α:GFP-U6:shRNA, and pTol2- EF1α:GFP-H1:shRNA were transfected into ZF-4 cells as indicated and EGFP expression was monitored 72 hours post-transfection by fluorescence microscopy. Specific details of plasmids are included in Appendix B. C Expression of shRNA hairpins from relevant U6 and H1 promoters. Expression was determined from 2 independent transfection and error bars indicate SEM. Chapter 5. Creation and characterisation of transgenic zebrafish 124

EGFP expression and normal morphology. Both pTol2-EF1α:GFP-U6:shRNA and pTol2-EF1α:GFP-H1:shRNA transgenes induced significantly decreased survival of zebrafish at 3 dpf and 7 dpf, as well as over the longer-term (Figure 5.4 A). No zebrafish that tested positive for expression of pTol2-EF1α:GFP-U6:shRNA, survived more than 7 weeks post fertilisation (wpf) (Figure 5.4 B). pTol2-EF1α:GFP- H1:shRNA injected zebrafish also showed decreased survival at all time points compared to the controls(Figure 5.4 A & B) however, they did show significantly greater survival than pTol2-EF1α:GFP-U6:shRNA injected fish. In addition to decreased survival pTol2-EF1α:GFP-U6:shG0906, and pTol2-EF1α:GFP- H1:shG0906 and transgenics were less common, and showed increased development deformities than either control zebrafish cells injected with either phenol red only (WT), or pTol2-EF1α:eGFP compared to the latter (Figure 5.4 A & B).

Developmental deformities observed following transgene incorporation represented a wide range of phenotypes (Figure 5.5 A - G), which were spread relatively evenly across the animal (Figure 5.5 H). This pattern of phenotypes matched well with very early investigations of the eﬀects of RNAi-inducing long-dsRNA in zebraﬁsh described in[269,371] (Figure 5.5 A).

5.3.3 shRNA over-expression inhibits zebraﬁsh development

The zebrafish is a popular model of development and as such the stages of zebrafish development are extremely well characterised. Determining the staging of zebrafish development by features is the a preferred method of determining the stage of zebrafish development[49,186,278] rather than simply counting the hours post fertilization, as a large number of environmental and genetic factors can influence the rate of zebrafish development. In particular, water temperature[278], chemical additives such as phenyl 2-thiourea (PTU) that is commonly added to maintain the transparency of zebrafish larvae, and manipulation can cause changes to the timing of zebrafish development. Chapter 5. Creation and characterisation of transgenic zebrafish 125

EGFP

EGFP EGFP EGFP

Figure 5.4: Expression of shRNAs from pol III type 3 promoters results in reduced survival and increased developmental defects in zebrafish larvae and adults. A Relative survival of zebrafish. Surviving zebrafish were counted weekly following three independent rounds of injections of each construct; EGFP, U6, and H1 indicate pTol2-EF1α:GFP, pTol2-EF1α:GFP-U6:shRNA, and pTol2-EF1α:GFP-H1:shRNA respectively, WT indicates zebrafish injected with Danieu buffer and phenol red dye only. Mean percentage survival was calculated from zebrafish surviving at 7 dpf. B Effects on zebrafish up to 7 dpf. Mean percentage of GFP positivity, survival, and nonspecific developmental defects are shown for transgene cassettes included in A Numbers of injected zebrafish in each group are included below. * p ≤ 0.05 compared to WT, † p ≤ 0.05 compared to EGFP.

Name n WT 400 EGFP 212 U6 461 H1 385 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 126

A B E

C D F

H 100

80 WT EGFP 60 U6 H1 40

20 Percentage Affected Percentage

Head Trunk Chapter 5. Creation and characterisation of transgenic zebraﬁsh 127

In addition to the obvious morphological abnormalities and increased mortality, at 3 dpf it was apparent that embryos injected with shRNA expression constructs were smaller, less developed, and more likely to remain within the chorion (Fig- ure 5.6 A). Standard length of zebrafish embryos (SL) is a consistent measure of zebrafish development (Figure 5.6 B), which has been used to estimate developmental stage of zebrafish larvae[186,278]. Standard length (SL)of larvae is considered in this project to be the minimum straight line distance from the tip of the snout to the posterior tip of the notochord[278] (Figure 5.6). In addition the head to trunk angle (HTA) , which determines the angle, formed between a line connect- ing the snout and midpoint of the head and anther between this point and the caudal tip of the tail was assessed at 3 dpf (Figure 5.6 C). To quantify this slow- ing of normal development the mean HTA and SL of zebrafish larvae groups was determined at 3 dpf and 7 dpf respectively. pTol2-EF1α:GFP-U6:shRNA, and pTol2-EF1α:GFP-H1:shRNA groups were significantly slower to develop relative to the two controls, pTol2-EF1α:GFP and phenol red only, at both time points and pTol2-EF1α:GFP-U6:shRNA, showed a further significant retardation slower at 7 dpf compared to pTol2-EF1α:GFP-H1:shRNA (Figure 5.6 D & E).

Figure 5.5 (preceding page): Expression of shRNAs in developing zebrafish increases overt morphological abnormalities Light microscope images at 72 hpf of A WT zebrafish head (dorsal view) B WT zebrafish head (ventral view) C WT zebrafish tail ventral view and injected embryos with developmental deformities D-G. H Comparison of the frequency of developmental defects at 72 hpf to head or tail of embryos.1 mm scale bars.

Name n WT 400 EGFP 212 U6 461 H1 385 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 128

pEF1_:eGFP pEF1_:eGFP H1:shG0906 pEF1_:eGFP U6:shG0906

B CD * 180 † 170 160 150 140 130 HTA (Degrees)HTA 120 110 WT eGFP U6 H1 E 5.0 * 4.5 # † 4.0

3.5

3.0 Standard Length (mm) 2.5

WT U6 H1 GFP Chapter 5. Creation and characterisation of transgenic zebraﬁsh 129

5.3.4 shRNA overexpression causes increased mortality and developmental abnormalities in zebraﬁsh in a sequence independent manner

To ensure that the shG0906 shRNA was not mediating a non-specific RNAi response, additional shRNA vectors were created using the U6 and H1 promoters including the shRNA scramble or a second anti-VHS virus shRNA shG1207: pTol2- EF1α:GFP-U62:shScramble, pTol2-EF1α:GFP-U62:shG1207, pTol2-EF1α:GFP- H1:shG1207, and pTol2-EF1α:GFP-H1:shScramble using the same strategy was used as for pTol2-EF1α:GFP-U62:shG0906 and pTol2-EF1α:GFP-H1:shG0906 plasmids (see Figure 5.2). To test for effective expression the plasmids were transfected into ZF-4 cells and expression of EGFP observed by fluorescence microscopy and expression of shRNAs measured by qRT-PCR. shRNA expression and EGFP fluorescence from these plasmids were equivalent to that observed from pTol2- EF1α:GFP-U62:shG0906 and pTol2-EF1α:GFP-H1:shG0906 (Data not shown).

Figure 5.6 (preceding page): shRNA overexpression inhibits zebrafish development. A Fluorescent microscope images of EGFP expression in zebrafish injected with the indicated constructs at 3 dpf. B Head to Trunk angle (HTA) in 3 dpf embryos indicated with blue lines between snout and midpoint of the head and this point and the caudal tip of the tail. The red box indicates the measured angle. C Measurement points of standard length indicated by thick red lines and total standard length indicated by blue dotted lines. D Head to Trunk angle of zebrafish larvae, measured as described previously using ImageJ at 3 dpf. n = phenol red (WT) (30), pTol2-EF1α:eGFP (EGFP) (30), pTol2- EF1α:eGFP U6:shG0906 (U6) (30) and pTol2-EF1α:eGFP H1:shG0906 (H1) (30). E) Standard length of zebrafish larvae, measured as described previously using ImageJ at 7 dpf, using the same constructs as above. All figures are representative of at least 3 independent inoculations. * p ≤ 0.05 relative to the Phenol Red control (WT), † p ≤ 0.05 indicates difference to pTol2-EF1α:GFP (EGFP), # p ≤ 0.05 indicates difference to pTol2-EF1α:GFP-H1:shG0906 (H1), pTol2-EF1α:GFP-U6:shG0906 (U6).

Name n WT 400 EGFP 212 U6 461 H1 385 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 130

The various shScramble and shG1207 plasmids were also introduced into zebrafish embryos by microinjection and subjected to the same measurements as the original shRNA vectors (Figure 5.7). The new transgene vectors induced no significant increases in survival, or decreased non-specific developmental abnormalities relative to the equivalent shG0906 shRNA constructs (Figure 5.7 A - D), suggesting that the observed development abnormalities and mortality are unrelated to sequence- specific responses.

5.3.5 Mortalities and developmental defects induced by shRNA overexpression is not dependent on p53

Expression of shRNAs from pol III promoters appears to induce dose-dependent developmental defects and phenotypes in zebraﬁsh embryos. However, anti-sense nucleic acid analogues with a morpholine backbone, called morpholinos (MO) have previously been shown to induce tumor-protein p53 (p53) related mortality in zebraﬁsh[301], which are similar to those observed in this study[39,197,301]

MOs bind to target mRNA via Watson-Crick base pairing in a similar manner to RNAi inducing molecules however the modifications in the chemical makeup of the oligonucleotides prevents the MOs inducing RISC or RNase H mediated degradation of mRNA, with MO’s acting by inhibiting translation or normal splicing[39,97,101,138,340,346]. However, MO inhibition of translation has been shown to induce non-specific neural defects in 15-20% of all MO experiments carried out in zebrafish[101,301]. These non-specific defects are proposed to derive from the activation of p53 and consequent p53-induced apoptosis[39,101,301]. A MO that targets the translation start site of p53 is effective at inhibiting non-specific phenotypes following injection of many MOs[101,301].

To ensure that non-specific p53 induced apoptosis was not responsible for the phenotypes of delayed development, phenotypic defects and increases in mortality following shRNA injection in zebrafish, p53 morpholinos were co-injected alongside Chapter 5. Creation and characterisation of transgenic zebrafish 131

ABA A A G A G A G A A C A C A C G U G U G U A U 30 A U 30 A U 30 A U G C A U 20 U A 20 A U 20 U A G C U A G C U A G C U A U A G C U A * # G C U A G C G C U A G C U A A U U A G C A U G C G C 40 A U 40 A U 40 U A U A U A 10 A U 10 G C 10 G C G C A U A U U A G C G C G C A U A U G C U A U A U A G C G C U A G C G C G C U A U A G C 50 U A 50 U A 50 5' U A 5' G C 5' G C A C 3' C 3' 3' C shG0906 shScramble shG1207 D

* # * #

Figure 5.7: shRNA induced developmental defects are independent of shRNA sequence. A shRNA sequences and predicted secondary structures B Percentage of EGFP positive zebrafish at 7 dpf following inoculation with shRNA expression plasmids. C Survival of zebrafish at 7 dpf following inoculation with shRNA expression plasmids. D HTA at 3 dpf. Graphs shown with mean from at least 3 independent microinjection rounds, and error bars represent SEM. Statistical analysis by Two-way ANOVA. * p ≤ 0.05 relative to the Phenol Red control (WT), † p ≤ 0.05 indicates difference to pTol2-EF1α:GFP (EGFP)

Name n WT 317 EGFP 381 U6 320 H1 400 Chapter 5. Creation and characterisation of transgenic zebrafish 132 shRNA and eGFP expression plasmids and transposase mRNA. Injection of p53- specific morpholinos did not rescue affect transgene incorporation but also did not result in rescue of normal development. It is therefore unlikely that p53 induced apoptosis is an explanation for the phenotypes observed following shRNA injection into developing zebrafish (See Figure 5.8).

5.3.6 shRNA overexpression inhibits miRNA expression in developing zebraﬁsh

Over-expression of shRNA molecules has previously been shown to inhibit the normal accumulation and functions of endogenous miRNAs[128]. To determine whether a similar response could be occurring in these zebrafish, qRT-PCR was used on six different miRNA targets to determine whether shRNA expression was impacting on the normal levels of these miRNAs. These miRNAs were previously identified to either be crucial or representative of the normal development, or from different miRNA biogenesis pathways. In particular, miR-451 has been identified as a dicer independent miRNA and is dependent on catalytic cleavage by Ago2 for cleavage into a mature miRNA from a pri-miRNA[65,280,389], and so may be more sensitive to Ago2 sublimation and inhibition.

Following microinjection RNA was harvested at 7 dpf from pooled embryos, and relative miRNA expression determined by qRT-PCR. The expression of individual miRNAs failed to reach statistical significance as the miRNAs changed expression by less than 50%. However, when the relative expression across all six miRNAs was examined by ANOVA, pTol2-EF1α:GFP-U6:shG0906 significantly reduced overall miRNA expression in 7 dpf zebrafish larvae (p ≤ 0.05) (Figure 5.9). Chapter 5. Creation and characterisation of transgenic zebrafish 133

DNA RNA Morpholino

BCD 100 50 4.5

80 40 4.0 60 30 3.5 40 20

Mean Survival (n) Survival Mean 3.0 20 10 Mean GFP (+) (%) (+) GFP Mean Standard Length (mm) 0 0 2.5

p53 p53 p53

U6shG906 H1shG906 U6shG906 H1shG906 Phenol Red Phenol Red U6shG906 H1shG906 xEF1a:EGFP xEF1a:EGFP Phenol Red xEF1a:EGFP

Figure 5.8: Over-expression of shRNA does not induce p53-related developmental defects in developing zebrafish. A Chemical structures of morpholino, RNA, and DNA backbones (Modified from[341]). B Mean-survival of zebrafish embryos co-injected with (Dark grey bars) or without (Light grey bars) p53 morpholino in addition to the relevant transgene and transposase. C Relative transgenisis as determined by percentage of GFP positive embryos by fluorescence microscopy at 7 dpf D Standard length of zebrafish larvae, measured as described previously using ImageJ at 7 dpf. Error bars show SEM. Statistical analysis by Two-way ANOVA. Injections were performed on at least three independent occasions for each construct.

Name n WT 278 EGFP 345 U6 300 H1 465 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 134

EGFP

Figure 5.9: Over-expression of shRNA reduces miRNA levels in developing zebrafish. Following microinjection of zebrafish embryos with the relevant Tol-2 constructs; U6 (pTol2-EF1α:GFP-U6:G0906), H1 (pTol2-EF1α:GFP-H1:G0906), EGFP (pTol2-EF1α:GFP) or the WT (phenol-red only) control pooled 7 dpf zebrafish were homogenised from three independent rounds of injection. Rel- ative expression of the 6 independent miRNAs indicated was determined via ΔΔCt compared to WT and 5S as the endogenous control.

Name n WT 368 EGFP 297 U6 314 H1 295 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 135

5.4 Modiﬁcation of pol III promoters to modulate shRNA expression

The initial aim of this study was to validate the potential for the highly effective U6 and H1 promoters identified in Chapter 3 to mediate the in vivo expression of shRNAs targeted at VHS virus. However, these promoters have proven to be inappropriate for use in vivo. Therefore shRNA expression vectors were constructed using modified U6 and H1 promoters to express the most effective shRNA sequences at potentially tolerable levels in zebrafish.

The pol III promoters were altered to remove specific promoter elements and weaken the promoters. To determine whether modification of the promoters would result in reduced expression of the shRNAs, and whether such reduced expression would result in rescue of normal development and survival, eight modified promoters were designed (Figure 5.10).

The zebrafish U6 promoter was modified to remove the Oct-1 element (U6-Oct), to shorten the promoter to approximately half its original length (U6-Half), to remove the SPH element from the PSE (U6-SPH) and to remove the TATA box (U6-TATA). The H1 promoter was similarly modified; however, as the DSE of the H1 promoter contains both an Oct-1 and a SPH element, both the SPH and Oct-1 elements were removed instead (H1-DSE) (Figure 5.10).Inthecaseofthe U6-SPH and H1-SPH constructs, the spacing between the SPH element and the +1 site of transcription was maintained by the insertion of random nucleotides. The U6-TATA and H1-TATA were constructed by the amplification of two over- lapping regions, which were subsequently joined by recombinant PCR. shG0906 and shScramble were added to the start site of transcription by one-step PCR as described in Chapter 4. Following PCR amplification the resulting promoter - shRNA constructs were inserted into the pTol2-EF1α:eGFP vector using the XhoI and SalI cloning sites.

Injection of the constructs containing the modified promoters into zebrafish embryos elicited a clear reduction in overt developmental defects and a significant Chapter 5. Creation and characterisation of transgenic zebrafish 136

U6 Promoter modification H1 Promoter modification U6 H1 Oct-1 SPH TATA Oct-1 SPH PSE TATA U6 Oct H1 Oct

SPH TATA SPH PSE TATA U6 Half H1 DSE

SPH TATA PSE TATA U6 SPH H1 SPH

Oct-1 TATA Oct-1 SPH TATA U6 TATA H1 TATA

Oct-1 SPH Oct-1 SPH PSE

Figure 5.10: Summary of the modified U6 and H1 promoters. Graphical representation of modifications made to the indicated promoter elements to ablate expression in second-generation Tol-2 constructs. Chapter 5. Creation and characterisation of transgenic zebrafish 137 increase in survival (Figure 5.11 B) This was evident for all modifications to the U6 or H1 promoters relative to the unmodified promoter. The levels of shRNA expression from the modified promoters were reduced as expected (Figure 5.11 A) which correlated to the increased survival of the zebrafish larvae to 7 dpf. Modification to remove the DSE elements from either promoter by either removing the Oct-1 motif or the Oct-1 and SPH as appropriate resulted in 86% and 45% reduction in shG0906 expression in the U6 and H1 driven promoters respectively. Removal of any part of the PSE of either the U6 or H1 promoter resulted in complete block of transcription from all constructs (Figure 5.11 A). As well as increasing overall survival of the zebrafish larvae modulation of shRNA dose additionally partially rescued zebrafish development at 3 dpf (Figure 5.11 C)and 7 dpf (Figure 5.11 D). All modifications made to the U6 promoter significantly increased developmental stage as determined by HTA relative to both eGFP only expression and unmodified U6. Only the H1 DSE modified promoter lead to a significantly increased HTA relative to the H1 promoter (Figure 5.11 C). Likewise at 7 dpf, modification of promoters to reduce shRNA expression resulted in the dose dependent rescue of zebrafish developmental delay.

5.5 Over-expression of human Argonaute 2 and Exportin-5 in zebraﬁsh larvae partially rescues developmental abnormalities and mortality from over-expression of shRNA molecules

Dose-dependent increases in mortality and deformities of zebrafish following expression of shRNAs suggested that the introduced shRNAs were having sequence specific off-target effects or that the volume of shRNA molecules being introduced was greater than could be tolerated by the organism. Considering the observed Chapter 5. Creation and characterisation of transgenic zebrafish 138

AB * * ** 200 100 † # 80 150 60 100 40 † #

Fold Change 50 20 Percentage Survival

0 0 P P F U6 H1 F U6 H1 eG eG U6 OctU6 HalfU6 SPH H1 OctH1 DSEH1 SPH U6 OctU6 HalfU6 SPH H1 OctH1 DSEH1 SPH U6 TATA H1 TATA U6 TATA H1 TATA

C † D † # # 180 5.0 * 170 4.5 4.0 160 * * 150 3.5

140 3.0

HTA (Degrees) 2.5 130 Standard Length (mm) 2.0 120 P P F U6 H1 F U6 H1 eG eG U6 OctU6 HalfU6 SPH H1 OctH1 DSEH1 SPH U6 OctU6 HalfU6 SPH H1 OctH1 DSEH1 SPH U6 TATA H1 TATA U6 TATA H1 TATA

Figure 5.11: Relative quantitation of shRNA expression following promoter modification. A qRT-PCR analysis of shG0906 expression 2 dpf in embryos injected with shRNA expression constructs. B Mean percentage survival of microinjected zebrafish embryos with modified promoters at 7 dpf. C Individual HTA at 3 dpf. D Standard lengths of zebrafish larvae at 7 dpf. For visualisation purposes a representative sample of individual values are displayed from a single injection round in C&D. However, mean and SEM displayed, were calculated from at least 3 rounds of injections and n = 30 for each construct micro-injected.* indicates P ≤ 0.05 difference to eGFP (No shRNA control). †,p≤ 0.05 difference to U6 (Full-strength U6 promoter), # p ≤ 0.05 difference to H1 (Full-strength H1 promoter). Total numbers of zebrafish injected for this experiment are included below:

Name n Name n eGFP 397 U6 267 H1 297 U6 Oct-1 240 H1 Oct-1 310 U6 Half 211 H1 PSE 209 U6 SPH 301 H1 DSE 297 U6 TATA 391 H1 TATA 414 Chapter 5. Creation and characterisation of transgenic zebrafish 139 effect was dose-dependent and that multiple shRNA molecules introduced similar patterns of abnormal development and mortality, it seemed more likely that the amount rather than the sequence of the introduced was responsible for the problems observed. Xpo-5 and Ago2 have previously been identified as rate limiting steps affected by expression of shRNAs in mice[128,129]. To investigate this, plasmids encoding Flag-tagged hAgo2 and hXpo5 (pQE60-Exp5 and pIRESneo- Flag/HA Ago2) were kindly provided by Dr Scott Tyack. Capped and poly-A tailed mRNA was transcribed from the plasmids following linerisation with XbaI and HpaI, respectively (Figure 5.2). Expression of Flag tagged proteins was confirmed by Western blot analysis (Figure 5.12 B).

Previous work has shown that Xenopus embryos injected with shRNA and siRNA constructs did not respond to RNAi induction until human argonaute2 (hAgo2) mRNA was co-injected along with the RNAi inducing molecules[113,227].Inter- estingly for this project, this work implies that RNAi functions in developing organisms might not be mature at all stages of organism development and further demonstrates that hAgo2 is functional in Xenopus. Many of the RNAi genes are highly conserved across species and MUSCLE alignment of Ago2 and Xpo-5 from zebraﬁsh and human indicated showing homology (Figure 5.12 A). The function- ality of hAgo2 in Xenopus as well as the high sequence conservation, particularly across the functional domains, suggests that the human proteins are likely to also be functional in zebraﬁsh[128,129,364].

To assess whether over-expression of either hXpo5 or hAgo2 in zebrafish produced deleterious effects, the panel of developmental assays were repeated on zebrafish embryos microinjected with mRNA encoding hXpo5 (Figure 5.13 A - D)and hAgo2 (Figure 5.13 E - H) at concentrations between 50 - 300 pg/nL. Over- expression of hXpo5 significantly inhibited zebrafish larvae growth at 3 dpf and 7 dpf at 300 pg/nl as well as at 100 pg/nL at 7 dpf. Over-expression of hAgo2 at 200 pg/nL significantly decreased zebrafish size and length at 7 dpf but did not significantly decrease rates of survival to 7 dpf or HTA at 3 dpf. There was also Chapter 5. Creation and characterisation of transgenic zebrafish 140

Human 1 MYSG---AGPALAPPAPPPPIQGYAFKPPPRPDFGTSGRTIKLQANFFE Human 1 MAMDQVNALCEQLVKAVTVMMDPNSTQRYRLEALKFCEEFKEKCPICVPCGLRLAEKTQV Zebrafish 1 MYPIGAAGATELFQGRPSSGSDVSAPASPPAPQEYVFKPPQRPDFGTMGRTIKLQANFFE Zebrafish 1 MGEQMRALCEELIKAVNVMMEAESSQTYRLEAFKFIEDFKEKSPFCVECGLQLAEKSQT

Human 47 MDIPKIDIYHYELDIKPEKCPRRVNREIVEHMVQHFKTQIFGDRKPVFDGRKNLYTAMPL Importin-beta N-terminal Zebrafish 61 MEIPKLEVYHYEIDIKPEKCPRGVNREIVEHMVQHFKTQIFGDRKPVYDGRKNLYTAMPL Human 61 AIVRHFGLQILEHVVKFRWNGMSRLEKVYLKNSVMELIANGTLNILEEENHIKDALSRIV Zebrafish 60 AVIRHFGLQILEHVVKFRWNNMAPQDKLQLKNCTMGMLSNGTHPILQEECHVKDALSRIV Human 107 PIGRDKVELEVTLPGEGKDRIFKVSIKWVSCVSLQALHDALSGRLPSVPFETIQALDVVM Zebrafish 121 PIGRDKVELEVTIPGEGKDRSFKVAIKWMSCVSLQALHEALSGRLPNIPFETIQALDVVM Exportin 1-like Human 121 VEMIKREWPQHWPDMLIELDTLSKQGETQTELVMFILLRLAEDVVTFQTLPPQRRRDIQQ Human 167 RHLPSMRYTPVGRSFFTASEGCSNPLGGGREVWFGFHQSVRPSLWKMMLNIDVSATAFYK Zebrafish 120 VEMIKREWPQQWPDMLKEMEALTALGDAQTELVMLVLLRLAEDVITFQTLPSQRRRDIQQ Zebrafish 181 RHLPSMRYTPVGRSFFTPSEGCSNPLGGGREVWFGFHQSVRPSLWKMMLNIDVSATAFYK Human 181 TLTQNMERIFSFLLNTLQENVNKYQQVKTDTSQESKAQANCRVGVAALNTLAGYIDWVSM Human 227 AQPVIEFVCEVLDFKSIEEQQKPLTDSQRVKFTKEIKGLKVEITHCGQMKRKYRVCNVTR Zebrafish 180 TLTQNMDSVFTFLLGILQLHVNEYSKMKCVAGLDLKMKAHLRVGVATLNTLAGYIDWVSL Zebrafish 241 AQPVIEFMCEVLDFKSIEEQQKPLTDSQRVKFTKEIKGLKVEITHCGQMKRKYRVCNVTR Human 241 SHITAENCKLLEILCLLLNEQELQLGAAECLLIAVSRKGKLEDRKPLMVLFGDVAMHYIL PAZ Domain Zebrafish 240 SHITSQNCRLLEILCLLLSEPELQLEAAECLLIAISRKGKLEERKPFMVLFDEAAMNYIL Human 287 RPASHQTFPLQQESGQTVECTVAQYFKDRHKLVLRYPHLPCLQVGQEQKHTYLPLEVCNI Zebrafish 301 RPASHQTFPLQQENGQTIECTVAQYFKDKYKLVLRYPHLPCLQVGQEQKHTYLPLEVCNI Human 301 SAAQTADGGGLVEKHYVFLKRLCQVLCALGNQLCALLGADSDVETPSNFGKYLESFLAFT Human 347 VAGQRCIKKLTDNQTSTMIRATARSAPDRQEEISKLMRSASFNTDPYVREFGIMVKDEMT Zebrafish 300 SAAQSS--GGIDERRYTFLKRLCQVLCALGSQVCSLVGSDVEVQVPVNLNKYLEALLAFT Zebrafish 361 VAGQRCIKKLTDNQTSTMIRATARSAPDRQDEISKLMRSANFNTDPYVREFGVMVRDDMT Human 361 THPSQFLRSSTQMTWGALFRHEILSRDPLLLAIIPKYLRASMTNLVKMGFPSKTDSPSCE Human 407 DVTGRVLQPPSILYGGRNKAIATPVQGVWDMRNKQFHTGIEIKVWAIACFAPQRQCTEVH Zebrafish 358 THPSQFLRSSTQMTWGIIFRHEILSKDPVVGQMAIKYLRATRINLVKTGFPSKNDCPGCE Zebrafish 421 EVNGRVLQAPSILYGGRNKAIATPVQGVWDMRNKQFHTGIEIKVWAIACFAPQRQCTELL Human 421 YSRFDFDSDEDFNAFFNSSRAQQGEVMRLACRLDPKTSFQMAGEWLKYQLSTFLDAG--- Human 467 LKSFTEQLRKISRDAGMPIQGQPCFCKYAQGADSVEPMFRHLKNTYAGLQLVVVILPGKT Zebrafish 418 FSRVDFDSDEDFNSSFNSSRAQQGEAVRLTCRIVPFKAFQIARDWMQYQISTPIDAGKTT Zebrafish 481 LKAFTDQLRKISRDAGMPIQGQPCFCKYAQGADSVEPMFKHLKYTYQGLQLVVVILPGKT Human 478 -SVNSCSAVGTGEGSLCSVFSPSFVQWEAMTLFLESVITQMFRTLNREEIPVNDGIELLQ Human 527 PVYAEVKRVGDTVLGMATQCVQMKNVQRTTPQTLSNLCLKINVKLGGVNNILLPQGRPPV Zebrafish 478 DNCKAVLALGTTEKGLCSPLSPSVVQWEAMTTFTENVFGQLFKILEKEKLPIDEGMALLQ Zebrafish 541 PVYAEVKRVGDTVLGMATQCVQVKNVQKTTPQTLSNLCLKINVKLGGVNNILLPQGRPLV Human 537 MVLNFDTKDPLILSCVLTNVSALFPFVTYRPEFLPQVFSKLFSSVTFETVEESKAPRTRA PIWI Domain Zebrafish 538 IAVNFETRDPLILSCVLTIVSTLFPILTHRPHFLPQVLFKIFSAITFELVDERKAPRTRA Human 587 FQQPVIFLGADVTHPPAGDGKKPSIAAVVGSMDAHPNRYCATVRVQQHRQEIIQDLAAMV Zebrafish 601 FQQPVIFLGADVTHPPAGDGKKPSIAAVVGSMDAHPSRYCATVRVQQHRQDIIQDLATMV Human 597 VRNVRRHACSSIIKMCRDYPQLVLPNFDMLYNHVKQLLSNELLLTQMEKCALMEALVLIS Zebrafish 598 VKNVRRHACSSIIRICRDYSDFMLPCFDLMYEHVKRLFSDELLLTQLEKCALMEALILIS Human 647 RELLIQFYKSTRFKPTRIIFYRDGVSEGQFQQVLHHELLAIREACIKLEKDYQPGITFIV Zebrafish 661 RELLIQFYKSTRFKPTRIIYYRDGISEGQFNQVLQHELLAIREACIKLEKDYQPGITFVV Human 657 NQFKNYERQKVFLEELMAPVASIWLSQDMHRVLSDVDAFIAYVGTDQKSCDPGLEDPCGL Zebrafish 658 NQFKDYKKQKAFLEELMAPVTALWLSEEMRSVLWDPATFLTFVGADQEISDSDTDEQMGI Human 707 VQKRHHTRLFCTDKNE------RGTSRPSHYH Zebrafish 721 VQKRHHTRLFCMDRNERVGKSGNIPAGTTVDTKITHPFEFDFYLCSHAGIQGTSRPSHYH Human 717 NRARMSFCVYSILGVVKRTCWPTDLEEAKAGGFVVGYTSSGNPIFRNPCTEQILKLLDNL Zebrafish 718 NRSRISLCVHTILGVVKRARWPADADQAKAGGFVVRTASDGTPVYRNPCAEALQALLPNL Human 733 VLWDDNRFSSDELQILTYQLCHTYVRCTRSVSIPAPAYYAHLVAFRARYHLVDKEHDSAE Zebrafish 781 VLWDDNHFTSDELQVLTYQLCHTYVRCTRSVSIPAPAYYAHLVAFRARYHLVDKEHDSAE Human 777 LALIRTHNTLYAPEMLAKMAEPFTKALDMLDAEKSAILGLPQPLLELNDSPVFKTVLERM

Zebrafish 778 LALIRTNNSLFLPENITRLSKTFARVYDITDMEKNCVLGISQVVLDSYEAAVYKNFAERM Human 793 GSHTSGQSNGRDHQALAKAVQVHQDTLRTMYFA

Zebrafish 841 GSHTSGQSNGRDQQALAKAVQIHQDTLRTMYFA Human 837 QRFFSTLYENCFHILGKAGPSMQQDFYTVEDLATQLLSSAFVNLNNIPDYRLRPMLRVFV

Zebrafish 838 QGFFSSLFENCYHVLGNVGPCLQQDFYGIEGLAEQIVGSAFNHLDSVPDHRLR------

Human 897 KPLVLFCPPEHYEALVSPILGPLFTYLHMRLSQKWQVINQRSLLCG--EDEAADENPESQ B Zebrafish 891 -PLIHILEITIFHTFAKVVF--FFFSKKQRLNFRWQIINQRASLSAQEEEEAYEENHVTQ Exportin-5Xpo5 Argonaute Ago2 2 Human 955 EMLEEQLVRMLTREVMDLITVCCVSKKGADHSSAPPADGDDEEMMATEVTP-----SAMA + _ + _ Zebrafish 948 EMVEEQLLRLVTREVMDLLSVTCITRKCPEVNANKEEADGDEEMVSMDSSQGNQVNTPSD 150 kda 100 kda Human 1010 ELTDLGKCLMKHEDVCTALLITAFNSLAWKDTLSCQRTTSQLCWPLLKQVLSGTLLADAV Zebrafish 1008 ELSDLGKCLLQSEDIYMTVLTICFNCLSWKDTVNCQRTAGVLCWTLLKQVQGGNLLPEAV

100 kda 75 kda Human 1070 TWLFTSVLKGLQMHGQHDGCMASLVHLAFQIYEALRPRYLEIRAVMEQIPEIQKDSLDQF Zebrafish 1068 TWLFASVLKGLQMHGQHEGCNVALTQLALLIYESLRPRYAELRLIMNQIPDVQADALEQF

GAPDH Human 1130 DCKLLNPSLQKVADKRRKDQFKRLIAGCIGKPLGEQFRKEVHIKNLPSLFKKTKPMLETE 37 kda Zebrafish 1128 DQKIQ-PGASKLGEKKKKEQFRRLIAGTVGKPLAQQFKKEVHIRNLPSLFKKPKPTKD-L

Human 1190 VLDNDGGGLATIFEP Zebrafish 1186 LENNEDATLISLFTPDHDRC Chapter 5. Creation and characterisation of transgenic zebrafish 141 no significant difference in miRNA expression at 3 dpf following inoculation with 100 pg/nl to 200 pg/nl mRNA relative to phenol red control (WT).

Studies next sought to determine if co-injection of zebrafish embryos with shRNA constructs and either hAgo2 or hXpo5 or both simultaneously would rescue normal development. Embryos were therefore injected with either phenol red controls (WT), pTol2-EF1α:GFP-U62:shG0906 (U6 Uninfected), pTol2-EF1α:GFP- H1:shG0906 (H1 uninfected), either 100 pg/nL hAgo2 or hXpo5 with phenol red, or with pTol2-EF1α:EGFP (GFP), pTol2-EF1α:GFP-U62:shG0906 (U6), or pTol2- EF1α:GFP-H1:shG0906 (H1) or alternatively with both hAgo2 and hXpo5 ( each 100 pg/nL) in the presence or absence of the respective test plasmids. None of the combinations of injections were able to significantly improve either survival of zebrafish larvae (Figure 5.14 A & E) or prevent gross morphological abnormalities of zebrafish larvae following overexpression of shRNAs (Figure 5.14 B & D). How- ever, increased hAgo2 expression following microinjection of mRNA significantly increased zebrafish development at 7 dpf of both pTol2-EF1α:GFP-U6:shG0906 of pTol2-EF1α:GFP-H1:shG0906 transgenic zebrafish and similarly increased hXpo5 expression rescued development of pTol2-EF1α:GFP-H1:shG0906 but not pTol2- EF1α:GFP-U6:shG0906 at 7 dpf (Figure 5.14 C & D, G & H).

In addition, co-injection of both hAgo2 and hXpo5 as well as the expression of hAgo2 alone but not hXpo5 alone was able to rescue normal miRNA expression in pooled zebraﬁsh larvae (Figure 5.15). Interestingly, as observed in Section

Figure 5.12 (preceding page): Alignment of human and zebrafish Argonaute 2 and Exportin 5 proteins. A Clustal W2[67] alignment of human Ago2 (NP 036286.2) and from Xpo5 (NP 065801.1) and zebrafish Ago2 and Xpo5 (NP 699226.2 and NP 001921422.4 respectively). Non conserved amino acids are indicated by Bold text. The Ago2 Paz domain is highlighted in Red and the Piwi domain highlighted in Blue. The Xpo5 Importin-beta N-terminal domain is highlighted in Green and the Exportin 1-like domain is highlighted in Pink. B Expression of Flag tagged Xpo5 and Ago2 protein expressed following microinjection with pQE60-Exp5 and pIRESneo-Flag/HA Ago2, respectively. Protein was harvested from pooled zebrafish larvae and expression of Xpo5 and Ago2 assessed by Flag antibody. GAPDH was included as a loading control. Chapter 5. Creation and characterisation of transgenic zebrafish 142

A B

CD Fold Change

E F 55 200 50 45 WT 180 40 300 pg/nL 35 200 pg/nL 30 160 25 100 pg/nL 20 50 pg/nL 15 Survival (n) Survival

HTA (Degrees)HTA 140 10 5 0 120 0 1 2 3 4 5 6 Days post fertilisation WT

G H 50 pg/nL 300 pg/nL 200 pg/nL 100 pg/nL 5.5 5.0 1.4 4.5 * 1.2 4.0 1.0 3.5 3.0 0.8 Fold Change 2.5 Standard Length (mm) 0.6 2.0

WT miR107miR101miR146miR181miR451miR107miR101miR146miR181miR451 50 pg/nL 300 pg/nL 200 pg/nL 100 pg/nL 200 pg\nL 100 pg\nL Chapter 5. Creation and characterisation of transgenic zebraﬁsh 143

5.3.6, there were no significant differences between individual miRNAs but overall miRNA expression observed were significantly restored.

5.6 Discussion

The work presented in thesis collectively details the construction of a Tol-2-based approach for the constitutive expression of shRNAs targeting VHS virus in transgenic zebrafish. Native zebrafish promoters were characterised in Chapter 3 and effective shRNA molecules identified in Chapter 4. However, combining these promoters and shRNAs with Tol-2 vectors in zebrafish embryos resulted in significant mortality, development defects and developmental delay. Indeed, of the 1134

Figure 5.13 (preceding page): Characterisation of hXpo5 and hAgo2 overexpression in zebrafish larvae. A Percentage survival of zebrafish embryos inoculated with 50 - 300 pg/nL pQE60-Exp5 (Xpo5) to 7 dpf. B Head to Trunk angle of zebrafish embryos inoculated with 50 - 300 pg/nL pQE60-Exp5 at 3 dpf. Mean and SEM indicated from representative samples. qRT-PCR was performed on technical triplicates from 3 independent injection sessions. C Head to trunk angle of representative (n=30) sample of zebrafish larvae at 3 dpf. Dark grey symbols are representative of single injections. Black symbols are representative of hAgo2 co-injections or of hAgo2 only and light grey symbols representative of hXpo5 only or hXpo5 co-injection. D Standard length of zebrafish larvae at 7 dpf. Symbols and samples as per (C). E Percentage survival of zebrafish embryos inoculated with 50 - 300 pg/nL pIRESneo-Flag/HA Ago2 (Ago2) to 7 dpf. F Head to Trunk angle of zebrafish embryos inoculated with 50 - 300 pg/nL pIRESneo-Flag/HA Ago2 at 3 dpf, with mean and SEM indicated from representative samples n=30 from 3 independent injection sessions. G Standard length of zebrafish larvae inoculated with 50 - 300 pg/nL pIRESneo-Flag/HA Ago2 at 7 dpf, Mean and SEM indicated from representative samples n=30 from 3 independent injection sessions H qRT-PCR determined expression of miRNAs from pooled n=10 zebrafish embryos inoculated with 50 - 300 pg/nL pIRESneo-Flag/HA Ago2 at 7 dpf, with mean and SEM indicated from representative samples. qRT-PCR was performed on technical triplicates from 3 independent injection sessions.

Name n Name n WT 762 pQE60-Exp5 pIRESneo-Flag/HA Ago2 50 pg/nL 240 50 pg/nL 310 100 pg/nL 421 100 pg/nL 321 200 pg/nL 301 200 pg/nL 234 300 pg/nL 367 300 pg/nL 214 Chapter 5. Creation and characterisation of transgenic zebraﬁsh 144

A B 40 50

30 40

30 20 20 10 10 Developmental defects (%) defects Developmental Mean surving transgenic (%) transgenic surving Mean 0 0

U6 H1 U6 H1 U6 H1 U6 H1 GFP GFP GFP GFP GFP GFP

U6 UninjectedH1 Uninjected U6 UninjectedH1 Uninjected 100 pg/nL Ago2 100 pg/nL Xpo5 100 pg/nL Ago2 100 pg/nL Xpo5 C D 5.5 180 ** * 5.0 170 † # 160 4.5 150 * 4.0 140 3.5 130 * 3.0

HTA (Degrees) HTA 120 110 2.5 Standard Length (mm) Length Standard 100 2.0

WT U6 H1 U6 H1 U6 H1 U6 H1 GFP GFP :GFP α xEF1a:G PhenolU6:shG906 RedH1:shG906 xEF1 U6 UninjectedH1 Uninjected 100 pg/nL Xpo5 100 pg/nL Ago2 100 pg/nL Ago2 100 pg/nL Xpo5

EF40 50

40 30 30 20 20 10 10 Developmental defects (%) defects Developmental Mean surving transgenic (%) transgenic surving Mean 0 0

U6 H1 U6 H1 GFP GFP GFP EGFP

GHU6 UninjectedH1 Uninjected100:100 pg/nL U6 UninjectedH1 Uninjected100:100 pg/nL 180 † # 5.5 170 5.0 † †# 160 4.5 * † * 150 * 4.0 140 3.5 130 3.0

HTA (Degrees) HTA 120 2.5

110 (mm) Length Standard 100 2.0 U6 H1 WT U6 H1 WT GFP GFP

U6 UninjectedH1 Uninjected100:100 pg/nL U6 UninjectedH1 Uninjected100:100 pg/nL Chapter 5. Creation and characterisation of transgenic zebrafish 145 overall zebrafish injected with a full strength U6 promoter driving any shRNA, none survived 7 weeks post-fertilisation. The weaker H1 promoter did not induce such dramatic mortalities; however, quantitative and qualitative measures of development demonstrate that the level of shRNA induced from this promoter also significantly increased mortality over expression of control plasmids and impacted on normal development of zebrafish larvae.

Reduction of global miRNA levels has previously been demonstrated to signiﬁ- cantly inhibit the development of zebraﬁsh and Xenopus using either morpholino

Figure 5.14 (preceding page): Expression of hAgo2 and hXpo5 can reduce mortality and developmental abnormalities in zebrafish larvae expressing shRNA molecules A Mean survival of zebrafish larvae at 7 dpf following injection with pTol2- EF1α:EGFP (GFP), pTol2-EF1α:GFP-U6:shG0906 (U6 uninjected), pTol2- EF1α:GFP-H1:shG0906 (H1 uninjected) expression plasmids only (white bars). Along with hAgo2 only or hAgo2 and shRNA expression plasmids (plain grey bars), hXpo5 only or hXpo5 and shRNA expression plasmids (patterned grey bars). B Mean developmental abnormalities observed up to 7 dpf, samples as per (A). Error bars indicate SEM. C Head to trunk angle of representative (n=30) sample of zebrafish larvae at 3 dpf. Dark grey symbols are representative of single injections. Black symbols are representative of hAgo2 co-injections or of hAgo2 only and light grey symbols representative of hXpo5 only or hXpo5 co- injection. D Standard length of zebrafish larvae at 7 dpf. Symbols and samples as per (C). E Mean surviving zebrafish larvae at 7 dpf following injection with tol 2 system plasmid and transposase only (White bars) or following co-injection with both 100 pg/nL hAgo2 and hXpo5 plasmid and transposon system (Grey bars). F Mean percentage zebrafish showing gross morphological deformities following injection. Samples as per (E). G Head to trunk angle of representative (n=30) sample of zebrafish larvae at 3 dpf. Tol-2 system only injections (light grey symbols). Co-injection of RNAi components and Tol 2 components (Black symbols). H Standard length of zebrafish larvae at 7 dpf. Samples as per (G). Mean and SEM displayed for each sample. Representative individual samples indicated for HTA and standard length. * p ≤ 0.05 relative to pTol2- EF1α:EGFP (GFP) only. † p ≤ 0.05 relative to pTol2-EF1α:GFP-U6:shG0906 (U6 uninjected). # p ≤ 0.05 relative to pTol2-EF1α:GFP-H1:shG0906 (H1 uninjected). Name n Name n WT 1262 U6 712 H1 675 hAgo2 541 hXpo5 476 hAgo2 & hXpo5 600 GFP 400 GFP 321 GFP 400 U6 301 U6 234 U6 465 H1 367 H1 214 H1 321 . Chapter 5. Creation and characterisation of transgenic zebrafish 146

A 1.5

1.0

0.5 Fold Change Fold

0.0

mir-101 mir-146 mir-181 mir-107 mir-451

GFP 100 pg/nL Ago2 100 pg/nL Xpo5 U6 Uninjected GFP GFP H1 Uninjected U6 U6 B H1 H1 1.5

1.0

0.5 Fold Change

0.0

mir-101 mir-146 mir-181 mir-107 mir-451

GFP 100:100 pg/nL U6 Uninjected GFP H1 Uninjected U6 H1 Chapter 5. Creation and characterisation of transgenic zebrafish 147 or siRNA mediated knockdown of RNAi pathway components, such as Dicer[378] and Ago[227]. In particular, siRNA inhibition of Dicer in zebrafish larvae resulted in the accumulation of pre-miRNAs, and concomitant lack of mature miRNAs, leading to complete mortality by 10 dpf. Xenopus embryos have previously been shown to be significantly more sensitive to siRNA / shRNA saturation of miRNA components. Xenopus larvae inoculated with siRNAs, required additional human Dicer to maintain normal development and siRNA enable effective activity. It follows that overexpression of shRNA molecules is likely interfering with the normal incorporation of maternal and endogenous miRNAs at crucial stages of development and that the observed developmental delays and abnormal morphology are the result of prevention of normal miRNA function.

As dose dependent non-specific effects of shRNA expression have previously been noted in zebrafish[371] and mice[119,126–129], modification of promoters to specifically introduce lower levels of expression were investigated. Modifications to the distal sequence elements of either the U6 or H1 promoters resulted in approximately 50% reduction in shRNA expression, whereas any modifications made to the proximal sequence elements completely prevented shRNA expression. Non-specific shortening of the U6 promoter did not decrease the expression of shRNA beyond that

Figure 5.15 (preceding page): Expression of hAgo2 and hXpo5 restores normal miRNA expression following shRNA overexpression. A) Expression of zebrafish miRNA panel in pooled zebrafish samples from 3 independent rounds of injection at 7 dpf with Tol-2 system plasmid and transposase only (White bars), as well as co-injection with hAgo2 and Tol-2 system (Light grey bars with black patterns) and with hXpo52 and Tol-2 system (Light grey bars with white patterns). B Expression of zebrafish miRNA panel in pooled zebrafish samples from 3 independent rounds of injection at 7 dpf with tol 2 system plasmid and transposase only (White bars), as well as co-injection with hAgo2, hXpo5, and Tol-2 system (Light grey bars with black patterns). Mean and SEM.

Name n Name n WT 1262 U6 712 H1 675 hAgo2 541 hXpo5 476 hAgo2 & hXpo5 600 GFP 400 GFP 321 GFP 400 U6 301 U6 234 U6 465 H1 367 H1 214 H1 321 . Chapter 5. Creation and characterisation of transgenic zebrafish 148 of modifying the distal sequence element, confirming the identification of the key promoter motifs in Chapter 3. The level of reductions in expression observed following promoter modification were approximately similar to those observed for the human H1 promoter[260].

Modulation of promoters to alter the toxicity observed from shRNA over-expression has not been a common strategy for the generation of transgenic RNAi. Most modifications to promoters have been to increase the amount of shRNA expression to increase the effectiveness of knockdown. However, there remains a major defi- ciency in the literature surrounding the effect of shRNA expression in vivo and in particular the effects of destabilizing the normal conditions of miRNA transcription and regulation in the early stages of development. A similar method of promoter modulation for the human U6-8 promoter, by replacing the PSE with the PSE from the U6-7 promoter, resulted in the reduction of hepatocellular toxicity in non-human primate models. In adult mammals, in particular mice[128] and cynomolgus monkeys[343], the primary site of shRNA-induced toxicity has been in the liver. The unmodified U6-8 promoter used to express shRNAs from an adenoviral vector TT-033 resulted in dose-dependent increases in alanine aminotransferase, aspartate aminotransferase and alkaline phosphatase, as well as decreases in albumin and total cholesterol [343], all markers of dose-dependent liver toxicity. This was supported by liver-bridge necrosis observed in non-human primates that correlated strongly with shRNA levels determined by qRT-PCR[343]. These results in non-human primate models are consistent with previous studies performed in rodent models[128,343].

Ago and Exportin-5 protein(s) have been demonstrated to be limiting factors for the expression of shRNA and siRNAs in other systems[63,128]. Elevated levels of Ago proteins have been shown to elevate nonspecific toxicity of siRNAs in vitro or in vivo [126,129,367], as well as increase levels of endogenous mature miRNAs and increase RNAi efficiency[92,113,126,227]. The limited availability of Ago is also proposed to be a mechanism for miRNA-dependent increases in mRNA expression by competition for available Ago protein[126,227,287]. The nuclear karyopherin Exportin-5 has been identified as an essential component of the shRNA and miRNA pathways Chapter 5. Creation and characterisation of transgenic zebrafish 149 and is therefore a prime candidate for shRNA mediated saturation. Exportin-5 has been demonstrated to be limiting within the RNAi and miRNA pathways [128]. As well as preventing homeostatic translocation of endogenous miRNA precursors, Exportin-5 saturation may also reduce Dicer expression and subsequent activity, illustrating some of the complexity of the cellular mechanisms underlying RNAi toxicity[34,126].

To isolate the RNAi components that may potentially be limiting the normal functioning of the miRNA pathway in the zebrafish larvae, Xpo5 and Ago2 were overexpressed with or without competing shRNA molecules. Overexpression of hXpo5, and hXpo5 and hAgo2, but not hAgo2 in zebrafish larvae, partially rescued normal development. shRNA saturation has previously been shown to be independent of the presence of an shRNA target suggesting an early component of the RNAi pathway was involved. However, co-injection of both hXpo5 and hAgo2 was significantly better at rescuing normal zebrafish development. To con- clusively determine all the members of the RNAi pathway involved, future work should focus on selective inhibition or overexpression of other factors shared by shRNAs/miRNAs, such as Dicer or other components of the RNA-induced silencing complex[126,129]. If the miRNA pathway can be oversaturated as has been shown by over expression of hXpo5 and hAgo2, efficient and persistent RNAi in vivo should potentially be possible with the use of minimal vector doses and/or inherently weakly expressed shRNAs. The observed rescue of zebrafish development with weaker modified pol III promoters supports this hypothesis.

Deliberately co-expressing known rate-limiting cellular RNAi factors can boost shRNA potency and reduce toxicity[129,227]. However, the long-term outcomes of this approach for the cell and organism remain to be studied. It is interesting to note in this context that a series of recent findings have indicated that essential parts of the RNAi machinery are inherently dysregulated in several cancers and many viruses specifically encode RNAi silencing suppressors[30,33,37,132]. This suggests that, along with the strategies described above, an important goal for future research should be determining the critical concentrations of RNAi components and the tolerance of these components to external stimulation. This will Chapter 5. Creation and characterisation of transgenic zebrafish 150 allow researchers to adapt and fine-tune strategies toward maximum efficiency and minimum toxicity.

In this study, Tol-2 transgenesis and resulting shRNA expression occured very early in development[25,173–175,361]. Normal miRNA expression has previously been shown to be crucial in development in an array of in vivo models including mice, chickens, Xenopus, non-human primates and zebraﬁsh. This dependence on normal functioning of the miRNA pathway to maintain cellular homeostasis is likely to be directly related to the dose-dependent toxicity observed in this project. This is the ﬁrst evidence of shRNA over expression resulting in induction of abnormal development and miRNA pathway inhibition during early development. Chapter 6

General discussion

6.1 Overview

Utilising a combination of polymerase III type 3 promoters to drive expression of shRNA molecules targeting virus specific sequences and the medaka Tol2 transposon system to integrate the expression cassette, this project aimed to create transgenic zebrafish that expressed shRNA molecules capable of eliciting anti-viral immunity. Specifically, these molecules targeted regions of the VHS virus genome in order to mediate resistance to VHS virus. This collectively represents a useful proof-of-principle demonstration of polymerase III promoter shRNA technology that can potentially be adapted to a range of fish species.

The application of vector-mediated shRNA gene silencing has necessarily depended on suitable promoters to facilitate appropriate shRNA expression. These promoters have been derived from a range of sources and have been utilised in a somewhat opportunistic manner. For example, the human, chicken and mouse U6 and H1 promoters[51,79,191,276], the human and chicken 7SK promoters[27,82], the human U1 promoter[91], and the human ubiquitin C promoter[385], as well as viral promoters such as the CMV early[386,399,400] and adenovirus E1b promoters[155] have all been used. These promoters have also been used in combination to increase potential

151 Chapter 6. General Discussion 152 shRNA expression; for example, combining the zebraﬁsh U6 promoter and the CMV early enhancer region[95].

Type 3 RNA pol III promoters have long been considered effective for shRNA transcription[51], and they have been the most commonly used promoter type for this purpose. This project began by identifying and comparing zebrafish type 3 RNA pol III promoters, since it was initially hypothesised that a zebrafish pol III promoter would likely direct higher shRNA expression in zebrafish cells compared with alternative heterologous promoters. Moreover, it was thought that utilisation of entirely native transgenes will likely prove less problematic for ethics and regulatory bodies as well as the general public in the event that transgenic RNAi models reached a commercial stage of development.

U6 promoters from a number of species have proven to be very effective for shRNA expression[51,276,393]. Therefore, the initial objective of this project was to identify and utilise zebrafish U6 promoter(s) for this purpose, as well as a potentially weaker zebrafish H1 promoter as a back-up in case the U6 promoter should overwhelm the native miRNA pathway. The identified promoters were tested for their ability to transcribe shRNAs that could suppress eGFP in comparison with shRNAs expressed by the widely used mouse U6 promoter[68,87,172,387] and chicken U6 promoter[27,79,383]. These experiments confirmed the hypothesis that native zebrafish U6 promoters were capable of mediating a higher level of gene knockdown compared to heterologous U6 promoters. Conversely, it was found that the mouse and chicken U6 promoters were significantly more effective in mammalian (VERO) cells at mediating gene knockdown than the equivalent zebrafish promoters. In addition, the zebrafish H1 promoter was demonstrated to be a much weaker promoter than the U6.

Polymerase III promoters have been considered to be significantly less complex than their polymerase II counterparts and this relative simplicity, particularly the location of all required elements upstream of the transcription start site makes them attractive for shRNA expression. However, the identification of species- specificity in their effectiveness at driving shRNA expression suggests a level of Chapter 6. General Discussion 153 regulation not previously presumed, which may be relevant in this regard [71,79,354]. Interestingly the zebrafish U6 promoters identified, contained reversed SPH and Oct-1 promoter motifs when compared with previously identified mammalian and avian U6 promoters[51,79,191,276]. Analysis of histone and chromatin interactions at the site of pol III transcription, has also demonstrated that this process is significantly more complex than previously considered[180,274,282,377]. DNA context- dependent epigenetic modulation of pol III transcription is a key issue, which indicates that pol III transcription is significantly less “plug and play” than suggested by early work[51]. Combined with the data on the various U6 and H1 promoters it suggests that the pol III promoter element motifs and the context of the potential tol 2 insertion site represent important determinants of promoter activity in homologous or non-homologous cell lines than simply the species of origin.

For the next step in the development of a potential virus-resistant zebrafish, the genome of VHS virus was targeted for RISC-mediated degradation by pol III- expressed shRNAs. The work described in Chapter 4 tested in vitro shRNAs targeting two genes of the VHS virus genome the glycoprotein (G) and polymerase (L) genes. These were targeted in three independent locations per gene prior to validation of the most efficient sequences in vitro. This demonstrated that the combination of the U6-2 polymerase III promoter and 5 of the 6 tested shRNAs were capable of significantly reducing the replication of VHS virus in zebrafish ZF-4 cells.

The determination of specificity for shRNA or siRNA molecules is dependent on a thorough understanding of the rules and processes that govern the strand selection of interfering molecules and RISC mRNA interaction. This project invested significant effort into eliminating unwanted sequence specific interactions, in particular those dependent upon the canonical seed region dependent interactions. However, recent studies have highlighted that canonical RISC mRNA seed region complexes may represent only a small fraction of sequence dependent RISC silencing[140]. The miRNA seed region has generally been considered to be a 6 - 8 base region at the 5 end of the miRNA, and these potential interactions were examined Chapter 6. General Discussion 154 in detail and used to select for shRNA molecules in this project. However, there are a number of well known exceptions to canonical seed region binding such as the earliest identified miRNA C. elegans lin4, with the lin-4:lin-14 interaction dependent on bulged nucleotides[131]. In contrast, the crucial cell-cycle and immune function regulator, human miR-24, is dependent on recognition sequences spread across the whole miRNA[279].

The anti-viral shRNAs identified in this project compare favorably with other molecules previously identified to target VHS virus. In particular, these molecules do not appear to induce a significant interferon and / or stress response like that observed by Schyth et al [324,325] and Ruiz et al [307].

The zebrafish H1 and U6 promoters identified in this study were able to be used to construct shRNA expression plasmids capable of suppressing VHS virus replication in vitro by expressing shRNAs targeting the virus. However, the introduction of these constructs into transgenic animals resulted in significant mortality and developmental defects. Indeed, no transgenic zebrafish expressing a hairpin from the U6 promoter survived past 7 weeks of age. These findings are consistent with previous work performed in mice[126,128,129] and Xenopus[113,227] using lentiviral vectors and mammalian U6 promoters to drive expression of siRNA type molecules. Similar results were found in the original studies of RNAi in zebrafish[2,269,371,396]. The best hypothesis for this is that the levels of shRNA molecules expressed are above the level that the native pathway can tolerate and still maintain normal processing of endogenous miRNAs.

To combat the significant mortality, two alternative methods were used. Firstly, the promoters were modified to reduce the expression of the shRNAs (See 6.1). Secondly, mRNAs encoding human versions of RNAi components were co-injected into the zebrafish in order to increase their levels, and thus potentially alleviate the “swamping” of this pathway by the over-expression of by shRNAs.

With respect to modifying the promoters, removal of the distal sequence element; the Oct-1, and the Oct-1 and SPH motifs in the U6 and H1 promoters, respectively, significantly reduced shRNA expression between 5 and 20 fold depending on the Chapter 6. General Discussion 155 modification. Non-specific shortening of the U6 or H1 promoters had no greater effect than removing the Oct-1 motif, and any modifications made to the proximal sequence elements or TATA box in either promoter resulted in the complete loss of shRNA transcription. Most of the research surrounding modifications to pol III promoters has focused on increasing promoter efficacy[177,184,231,298,306,370].How- ever, the characterisation of the unusually short human H1 promoter[260],aswell as the identification of the DSE and PSE regions of pol III promoters[193,259,313,318], reveals that simple modifications can also decrease promoter activity as observed in this study. Deceasing the dosage of introduced shRNA by modifying the polymerase III promoters effectively rescued both embryo viability and normal development.

Saturation of the native miRNA pathway, thereby preventing endogenous miRNA processing, is not unique to genetic engineered systems. A number of mammalian viruses, in particular adenoviruses[14,223], act to suppress the mammalian RNAi pathway by introducing large quantities of RNA with miRNA-like helices. These are then incorporated into Exportin-5 and prevent the normal uptake and export of pri-microRNAs from the nucleus. Other inducers of RNAi pathway saturation are extensively reviewed in Chapter 1 and are summarised in Figure 6.2.Our data, as well as that presented in previous work[34,128], support the hypothesis that Xpo5 saturation is a key mediator of this effect. However, when exogenous human Exportin-5 (hXpo5) was introduced into zebrafish embryos, both in the presence and absence of shRNA vectors a significant increase in toxicity and developmental abnormalities was observed. This increase in toxicity was observed in hXpo5 injected embryos only and not in either hXpo5 and human Argonaute-2 (hAgo2) co-injections or in hAgo2 only. This may imply that Xpo5 may act as a brake or limiting factor in the number of RISC incorporating molecules that enter from the nucleus into cytoplasm. If this is the case then, modulation of shRNA toxicity by over-expression of the RNAi pathway components will consistently lead to increased toxicity unless alleviated by the artificial over-expression of additional Ago2 as described in (Chapter 5) and in[47]. This leads to the hypothesis that it is the imbalance of miRNAs in the cytoplasm, rather than the specific saturation Chapter 6. General Discussion 156

U6 Promoter modification H1 Promoter modification U6 H1 Oct-1 SPH TATA Oct-1 SPH PSE TATA U6 Oct H1 Oct

SPH TATA SPH PSE TATA U6 Half H1 DSE

SPH TATA PSE TATA U6 SPH H1 SPH

Oct-1 TATA Oct-1 SPH TATA U6 TATA H1 TATA

Oct-1 SPH Oct-1 SPH PSE

Figure 6.1: Summary of the modified U6 and H1 promoters// Graphical representation of modifications made to the indicated promoter elements that ablate expression when delivered in second-generation tol 2 constructs. Chapter 6. General Discussion 157 of the individual components, that is responsible for the effects observed in this study.

These results suggest that use of RNAi as a therapeutic is dependent on: effective and specific shRNA design to avoid IFN-type responses or off-target effects, and the dosage of interfering molecules introduced should also be carefully controlled to limit its effects on the native miRNA pathway.

6.2 Recommended future directions

6.2.1 shRNA expression

While the polymerase III type 3 promoters identified in this study have proven extremely effective in cell culture systems, the complications observed in vivo suggest that they may not be the most effective option for the future generation of shRNA transgenic animals. Pol III type 3 promoters express an extremely defined hairpin which, when expressed in the manner utilised in this study, avoids Drosha/DGRG8 interaction and excision. Similar to Exportin-5 (see Chapter 5), Drosha processing of longer constructs is a potentially rate-limiting step, which could minimise further downstream consequences of shRNA expression. However, as has been repeatedly demonstrated[1,152,170,338], the expression of long and short ncRNA fragments capable of processing by Drosha within the nucleus can induce a robust interferon-dependent response both in vitro and in vivo [1,271,297,338].

As for an alternative approach, either inducible or virus specific promoters could be utilised for the generation of transgenic zebrafish. Both techniques would allow the expression of viral shRNA molecules only as required following either addition or viral infection, respectively, thus avoiding expression during developmental stages of the zebrafish potentially minimising mortality and developmental issues. However, in the case of an inducible promoter, such as a small molecule induced tet promoter it is likely to need to be induced constantly, therefore there is a likelihood of similar phenotypes being observed as with the adenoviral delivery of shRNA Chapter 6. General Discussion 158

Viral RNA Intronic shRNA

Pri-microRNA

Drosha / DRG8 Vector miRNA

shRNA Pre-microRNA

Nucleus Exportin-5 / RanGTP Cytoplasm

Pre-microRNA

Dicer

siRNA microRNA

Ago-2 Ago-1 to Ago-4

siRNA miRNA

siRISC miRISC

Target Degradation

Figure 6.2: Summary of the RNAi pathway, including potential external me- diators of saturation of individual pathway components Schematic representation of RNAi pathway showing RNA species (rounded boxes) and RNAi machinery (square boxes) within shaded boxes representing the nucleus (orange) and cytoplasm (blue). Externally derived RNAi-inducing elements; vector-encoded intronic shRNA, shRNA or miRNA, delivered siRNA and viral RNA (bullet shaped boxes) and their potential risk of saturation of the cellular RNAi pathway. Where there is experimental evidence for a specific RNAi inducer type saturating this is indicated as follows: (virally derived RNAs by hexagonal figure, miRNAs by bulged miRNA loop, shRNA by unbulged loop and siRNAs by parallel lines. Red indicates experimental evidence derived from this study. (Figure is modified from[129] with additional data from Chapter 5 as well as[14,17,37,43,63,79,113,117,119,126,129,132,150,177,227,243,263,323,325,335]) Chapter 6. General Discussion 159 molecules into adult mice[128]. In this study 49 independent shRNA constructs were delivered into adult mice, 36 of which caused significant liver damage following disruption of the normal miRNA expression levels. Therefor long-term shRNA expression using a small molecule inducible promoter or similar may not solve the issues caused by overexpression of shRNA if these doses are still above the tolerated level. In addition, pragmatic concerns regarding the long-term interventions required with an induced promoter for a potentially commercialised product also indicate that inducible promoters are likely a poor choice in this context.

Virus or viral-RNA inducible promoters, such as Mx-1, interferon-α [7,72,192],STAT- 1 and STAT-2[73], have been used to express mRNA specifically in response to virus infection. These pol II promoters could also conceivably be used to drive expression of shRNA molecules at specific times of need. However, a number of studies[284,286,312,360,379] have identified that normal miRNA propagation is required for the normal activation of both the innate and active immune responses. While the importance of the direct RNAi response has been hotly debated in the context of mammalian viral infections[20,54,77,78,230], the absence of normal miRNA processing is refractory to the normal anti-viral response [5,229,329,331]. Overexpression of shRNA molecules, as an anti-viral response ramps up, via expression of anti-viral shRNAs linked with a virally-induced protein such as interferon-α or Mx-1, and this may inhibit these important processes. Additionally, many viruses such as IHNV in zebrafish[224], as well as Ebola via the VP35 protein[29], and Dengue Virus NS2a and NS4 proteins[256], among many others (as reviewed in[149]) specifically interfere with IFN and MX production and as such would additionally inhibit shRNA expression linked to these prominent anti-viral genes.

The shRNA design strategies utilised in this study have been reliable over the course of a number of shRNA-based projects[27,195,196,353]. Quantitating whether this design strategy is the most effective or not is problematic as alternative effective strategies are rapidly commercialised and data surrounding their relative effectiveness is almost always impossible to obtain. However, since the design of the shRNAs utilised within this and earlier projects, there has been significant in- vestment of research effort in determining specific biochemical and stoichiometric Chapter 6. General Discussion 160 properties that direct RISC incorporation and specific Ago protein interactions, whist avoiding other aspects of the RNAi pathway[145]. Modification of shRNA design, perhaps to remove requirements for Dicer processing by shortening the shRNA and by the inclusion of a G - U or U - G[218] at the loop simplifies the number of steps required for shRNA processing and is an interesting possibility for potentially further minimising the likelihood of inhibiting the endogenous RNAi pathways.

6.2.2 Tol2 and transgenic production

Tol2 constructs can transpose efficiently in a variety of cell types, including in developing embryos (either differentiating somatic cells or germ cells), and integrate throughout the genome[173]. No reported preferential chromosome integration has been reported[25,173], and no consensus DNA sequence observed in the 8 bp target duplications created upon integration of Tol2. Moreover, transgenes cloned in the Tol2 vector are reliably expressed in transgenic animals and cells, and there is no evidence of gene silencing mechanisms by the host or integration site effect on the efficacy of transcription.[25,173]. However, several of the factors that facilitate effective transposon integration make the Tol2-based vectors and constructs utilised in this work problematic for introducing rational, precise and measured shRNA dosages in transgenic animals. To overcome these potential issues, precision gene targeting methods, such as nuclease-mediated homologous recombination using Talens or CRISPR proteins[18,32,41,56,74,252], are recommended for future development of transgenic RNAi delivery system.

In this study, it has been demonstrated, using an artificial EGFP-intron construct, that the expression of an intronic miRNA was not significantly perturbed by the introduction of a second hairpin into the same intron, and that the secondary hairpin was expressed at similar levels to the primary miRNA (see Chapter3). This represents an innovative approach that may provide an effective method of introducing shRNAs at tolerated levels. Precision genome targeting methods, coupled with homologous recombination again provide the mechanisms for achieving this Chapter 6. General Discussion 161 in vivo single hairpin construct, targeting either an intron, or a 5 untranslated- region (UTR) where an identified miRNA which is expressed in appropriate tissues occurs. The 5 UTR is potentially preferable as modifying the secondary structure of an intron may introduce unintended consequences with respect to correct intron processing. However, extensive in vitro and in vivo testing are required to elucidate the most effective methods. Application of such a method for generating transgenic zebrafish would also be significantly more complex than the methods utilised in the current study, and also precludes the use of a fluorescent reporter with screening of transgenics necessarily requiring molecular analysis. However, since it would introduce a minimal amount of exogenous sequence, and it does not require any additional promoters nor affects the target site, it is inherently attractive. Coupling the shRNA to an expressed tolerated miRNA should also reduce the possibility of saturation of the endogenous system.

6.2.3 Multi-shRNA cassettes

Multiple effective shRNA molecules are crucial for generating transgenics with long term viral resistance. This is because the sequence specificity of RNAi, which makes it ideal for degrading desired targets while leaving non-complementary sequences untouched, also makes it easy for viruses to evade or escape. Viruses, particularly RNA viruses, alter their genomes by producing point mutations allowing them to escape RNAi strategies that target a single sequence. Previous work using a computational model of HIV replication[204,205] has suggested that to prevent viral escape it was necessary to target multiple gene sequences with high efficiency silencing molecules. Specifically, they found that targeting a single sequence with a shRNA of average effectiveness (e.g. reducing viral mRNA expression by 70%), would prevent viral escape in 0% of trials, while four individual shRNAs of the same efficiency would prevent viral escape in 100% of trials. They further demonstrated that RNAi inducing molecules of 60% efficacy were unable Chapter 6. General Discussion 162 to prevent viral escape even when five different sequences were targeted. There- fore, to prevent VHS virus escape, multiple gene sequences must be targeted with molecules that have high suppressive activity.

The design and generation of constructs for the expression of multiple shRNAs from either single or multiple transcription units is outside the scope of this thesis. However, the panel of both shRNAs and promoters identified does collectively provides the basic elements required for such constructs, at least in vitro.There has been substantial research, into different approaches that enable the incorporation of multiple target sequences into a single construct. This may be necessary to induce a coordinated RNAi strategy, which is able to both suppress the target virus and prevent the potential evolution of viral escape mutants. The limitations of these methods include improper processing of long transcripts, and the significant risk of toxicity caused by over-expression. In addition, shRNAs within a multiple transcription unit have been found to produce a reduced level of suppression compared to identical shRNAs expressed from individual vectors[146,243]. Proposed explanations for these phenomena include positional effects, such as a reduced expression of the hairpin placed in the last position of the construct. In addition, there may be competition between hairpins expressed simultaneously for access to critical components of the RNAi pathway including Exportin-5 and RISC components[63,244], with simultaneous expression of shRNAs resulting in suppression of each hairpin in direct proportion to the number of hairpins expressed. However, the data remains inconclusive, and significant further research is needed. Multiple hairpins have been driven specifically from either pol II or pol III promoters utilising either single or multiple promoters in either a long stem or multiple individual hairpins, respectively. Utilising a miRNA cluster, such as the hsa-let-7C cluster, may provide an alternative strategy for incorporation of multiple hairpins within a single construct, such that they do not compete with each other for potentially limited nuclear or cytoplasmic processing components. Chapter 6. General Discussion 163

6.2.4 The zebraﬁsh model

The majority of virology and immunology researchers work in mammalian models, mostly small laboratory rodents, whereas fish virologists and immunologists have traditionally focused upon rainbow trout (Oncorhynchus mykiss), Atlantic salmon (Salmo salar)aswellascommoncarp(Cyprinus carpio) and other commercially important farmed fish (reviewed excellently in[245,266,267,336]). In contrast, despite the earliest description of infection of zebrafish with VHS virus dating back over thirty years [326], the use of zebrafish as a model of viral infection remains in its infancy when compared to these models, or to the use of the zebrafish as a model of development and genetics.

However, the zebrafish offers great potential as a research model for fish immunologists and virologists. It is far more easily housed than most commercially farmed fish, and the genetic tools and methodologies available to the zebrafish researcher dwarf those validated or tested in any other fish species. Zebrafish is a particularly good model for diseases of carp because they belong to the same family (Cyprinidae). A recent publication[208,224] describe the “whole-body” infection of zebrafish larvae with a heat-adapted IHN virus. Which took advantage of both the transparency of the zebrafish and the availability of relevant fluorescent strains to track the infection throughout the larvae. While showing a predictable pattern of infection, this work was significant in being among the first work to utilise the power of the zebrafish model for imaging, and to combine it with an infection model. It is envisaged that the possibilities for live imaging, and transgenic knock-in or knock-out zebrafish will provide unique tools to elucidate the patterns of viral infection in this important model[208,224]. In the context of virally-induced shRNA expression, the model may provide crucial information as to the tissues infected at early, intermediate and late stages of infection. These data would be crucial for the determination of insertion sites to “hitchhike” shRNAs to miRNA expression, and also for the selection of promoters for virally induced expression. Chapter 6. General Discussion 164

6.2.5 Speciﬁc recommendations for in vivo shRNA expression in zebraﬁsh

The requirements for shRNA expression, cassette insertion, ease of inclusion of multiple shRNA constructs, and the opportunities and resources of the zebrafish as a model are important considerations for the design of future projects for the generation of disease-resistant fish. While the shRNAs identified in this project were effective at reducing virus growth and CPE in vitro the choice of constitutive pol III promoters, as well as the use of the Tol2 system to randomly incorporate the expression cassette has prevented the successful development of a zebrafish strain which expresses an shRNA which provides specific resistance to VHS virus.

To overcome the limitations of the systems used within this project to generate an anti-viral zebrafish, one proposal would be to use the Pank1B 5th intron as an appropriate starting point for the generation of intron-derived shRNA delivery in zebrafish. Insertion of the shRNA into the site identified in this study as conserved in zebrafish provides an obvious starting point, as this delivery system worked effectively in vitro when driven as an EGFP intron by the CMV promoter. This proposal avoids a number of the issues that were observed during this project; in particular, the over-expression of the shRNAs during development that led to the observed increases of developmental deformities and mortalities in the zebrafish larvae. This method of attaching the shRNA to an already tolerated and duplicated miRNA is proposed to limit the likelihood of affecting the natural processes involving this miRNA and hopefully provides a tolerated level of expression of the shRNA. As discussed previously, an alternative to the integration within a miRNA expressing intron could be the use of virus specific promoters however these promoters would drive expression of the shRNA only after the detection of the virus by the cell and this may limit the effectiveness of the shRNA or potentially inhibit normal miRNA processing required for the endogenous virus response.

As well as replacing the promoters it is suggested that CRISPR nucleases are utilised to provide a precision cleavage point, and homologous recombination to insert the shRNA, or shRNA expression cassette into the desired location within Bibliography 165 the genome. This method of insertion provides a number of significant advantages over the Tol2 systems used in this project. By choosing the site of insertion the effect on surround genes can be minimalised as identified coding and conserved sites within the genome can be avoided. As well as this advantage if the combination of a miRNA linked intron and CRISPR insertion is utilised it would provide a minimal insertion size of 87 nt for a single shRNA. This minimal insertion size and lack of an additional promoter may additionally make this method of insertion more palatable from a regulatory / public perception perspective.

Bibliography

[1] A. Ablasser, F. Bauernfeind, G. Hartmann, E. Latz, K. A. Fitzgerald, and V. Hornung. RIG-I-dependent sensing of poly (dA: dT) through the induction of an RNA polymerase III-transcribed RNA intermediate. Nature Immunology, 10(10):1065–1072, 2009.

[2] J. Acosta, Y. Carpio, I. Borroto, O. Gonz´alez, and M. P. Estrada. Myo- statin gene silenced by RNAi show a zebraﬁsh giant phenotype. Journal of Biotechnology, 119(4):324–331, 2005.

[3] Z. N. Adelman, M. A. Anderson, E. M. Morazzani, and K. M. Myles. A transgenic sensor strain for monitoring the RNAi pathway in the yellow fever mosquito, Aedes aegypti. Insect Biochemistry and Molecular Biology, 38(7):705–713, 2008.

[4] A. Agrawal, Q. M. Eastman, and D. G. Schatz. Implications of transposition mediated by V (D) J-recombination proteins RAG1 and RAG2 for origins of antigen-speciﬁc immunity. Nature, 394:744–751, 1998.

[5] K. Ahrens and A. Karlas. Genome-wide RNAi screening to identify human host factors crucial for inﬂuenza virus replication. In RNA Interference from Biology to Therapeutics, pages 243–257. Springer, 2013.

[6] M. Alonso, C. H. Kim, M. C. Johnson, M. Pressley, and J. A. Leong. The NV gene of snakehead rhabdovirus (SHRV) is not required for pathogenesis, and a heterologous glycoprotein can be incorporated into the SHRV envelope. Journal of Virology, 78(11):5875–5882, 2004.

167 Bibliography 168

[7] T. Alonzi, D. Maritano, B. Gorgoni, G. Rizzuto, C. Libert, and V. Poli. Essential role of stat3 in the control of the acute-phase response as revealed by inducible gene activation in the liver. Molecular and Cellular Biology,21 (5):1621–1632, 2001.

[8] S. Altman, D. Wesolowski, and S. Puranam, R. Nucleotide sequences of the RNA subunit of RNase P from several mammals. Genomics, 18(2):418 – 422, 1993.

[9] S. M. Altmann, M. T. Mellon, D. L. Distel, and C. H. Kim. Molecular and functional analysis of an interferon gene from the zebraﬁsh, danio rerio. Journal of Virology, 77(3):1992–2002, 2003.

[10] S. F. Altschul, W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. Basic Local Alignment Search Tool. Journal of Molecular Biology, 215(3):403–410, 1990.

[11] V. Ambros, R. C. Lee, A. Lavanway, P. T. Williams, and D. Jewell. Micror- nas and other tiny endogenous rnas in c. elegans. Current Biology, 13(10): 807–818, 2003.

[12] A. Ammayappan and V. N. Vakharia. Molecular characterization of the Great Lakes viral hemorrhagic septicemia virus (VHSV) isolate from USA. Virology Journal, 6(1):171, 2009.

[13] A. Ammayappan and V. N. Vakharia. Nonvirion protein of novirhabdovirus suppresses apoptosis at the early stage of virus infection. Journal of Virology, 85(16):8393–8402, 2011.

[14] M. G. Andersson, P. C. J. Haasnoot, N. Xu, S. Berenjian, B. Berkhout, andG.Akusj¨arvi. Suppression of RNA interference by adenovirus virus- associated RNA. Journal of virology, 79(15):9556–9565, 2005.

[15] O. E. Andrews, D. J. Cha, C. Wei, and J. G. Patton. Rnai-mediated gene silencing in zebraﬁsh triggered by convergent transcription. Scientiﬁc Reports, 4, 2014. Bibliography 169

[16] N. Aronin. Target selectivity in mrna silencing. Gene therapy, 13(6):509–516, 2006.

[17] B. Ason, D. K. Darnell, B. Wittbrodt, E. Berezikov, W. P. Kloosterman, J. Wittbrodt, P. B. Antin, and R. H. A. Plasterk. Diﬀerences in vertebrate microRNA expression. Proceedings of the National Academy of Sciences, 103 (39):14385–14389, 2006.

[18] T. O. Auer and F. Del Bene. Crispr/cas9 and talen-mediated knock-in approaches in zebraﬁsh. Methods, 2014.

[19] B. Austin. Vibrios as causal agents of zoonoses. Veterinary Microbiology, 140(3–4):310 – 317, 2010.

[20] S. Backes, R. A. Langlois, S. Schmid, A. Varble, J. V. Shim, D. Sachs, and B. R. tenOever. The mammalian response to virus infection is independent of small RNA silencing. Cell Reports, 2014.

[21] R. Badakov and A. Ja´zwińska. Efficient transfection of primary zebrafish fibroblasts by nucleofection. Cytotechnology, 51(2):105–110, 2006.

[22] M. Baer, T. W. Nilsen, C. Costigan, and S. Altman. Structure and transcription of a human gene for H1 RNA, the RNA component of human RNase P. Nucleic Acids Research, 18(1):97–103, 1990.

[23] M. B. Bahramian and H. Zarbl. Transcriptional and posttranscriptional silencing of rodent 1(i) collagen by a homologous transcriptionally self-silenced transgene. Molecular and Cellular Biology, 19(1):274–283, 1999.

[24] S. J. Baigent and J. W. McCauley. Inﬂuenza type A in humans, mammals and birds: Determinants of virus virulence, host-range and interspecies transmission. BioEssays, 25(7):657–671, 2003.

[25] D. Balciunas, K. J. Wangensteen, A. Wilber, J. Bell, A. Geurts, S. Siva- subbu, X. Wang, P. B. Hackett, D. A. Largaespada, R. S. McIvor, and S. C. Ekker. Harnessing a high cargo-capacity transposon for genetic applications in vertebrates. PLoS Genetics, 2(11), 2006. Bibliography 170

[26] I. Band´ın and C. P. Dopazo. Host range, host speciﬁcity and hypothesized host shift events among viruses of lower vertebrates. Veterinary Research, 42(1):1–15, 2011.

[27] S. Bannister, T. Wise, D. Cahill, and T. J. Doran. Comparison of chicken 7SK and U6 RNA polymerase III promoters for short hairpin RNA expression. BMC Biotechnology, 7(1):79, 2007.

[28] C. Bardeleben, R. L. Moore, and R. K. Wayne. Isolation and Molecular Evolution of the Selenocysteine tRNA (Cf TRSP) and RNase P RNA (Cf RPPH1) Genes in the Dog Family, Canidae. Molecular Biology and Evolu- tion, 22(2):347–359, 2005.

[29] C. F. Basler, X. Wang, E. M¨uhlberger, V. Volchkov, J. Paragas, H.-D. Klenk, A. Garc´ıa-Sastre, and P. Palese. The ebola virus vp35 protein functions as a type i ifn antagonist. Proceedings of the National Academy of Sciences,97 (22):12289–12294, 2000.

[30] D. C. Baulcombe. RNA as a target and an initiator of post-transcriptional gene silencing in trangenic plants. Plant molecular biology, 32(1-2):79–88, 1996.

[31] A. A. Bazzini, M. T. Lee, and A. J. Giraldez. Ribosome proﬁling shows that miR-430 reduces translation before causing mrna decay in zebraﬁsh. Science, 336(6078):233–237, 2012.

[32] V. M. Bedell, Y. Wang, J. M. Campbell, T. L. Poshusta, C. G. Starker, R. G. Krug, II, W. Tan, S. G. Penheiter, A. C. Ma, A. Y. H. Leung, S. C. Fahrenkrug, D. F. Carlson, D. F. Voytas, K. J. Clark, J. J. Essner, and S. C. Ekker. In vivo genome editing using a high-eﬃciency TALEN system. Nature, pages 1–7, 2012.

[33] Y. Bennasser, S.-Y. Le, M. Benkirane, and K.-T. Jeang. Evidence that HIV- 1 encodes an siRNA and a suppressor of RNA silencing. Immunity, 22(5): 607–619, 2005. Bibliography 171

[34] Y. Bennasser, C. Chable-Bessia, R. Triboulet, D. Gibbings, C. Gwizdek, C. Dargemont, E. J. Kremer, O. Voinnet, and M. Benkirane. Competition for XPO5 binding between Dicer mRNA, pre-miRNA and viral RNA regulates human Dicer levels. Nature Structural & Molecular Biology, 18(3):323–327, 2011.

[35] B. Berkhout and A. T. Das. HIV-1 Escape From RNAi Antivirals: Yet Another Houdini Action? Molecular Therapy — Nucleic Acids, 1(6):e26, 2012.

[36] B. Berkhout and J. J. Eekels. RNAi as antiviral therapy: The HIV-1 case. In RNA Interference from Biology to Therapeutics, pages 221–242. Springer, 2013.

[37] B. Berkhout and J. Haasnoot. The interplay between virus infection and the cellular RNA interference machinery. FEBS Letters, 580(12):2896 – 2902, 2006.

[38] E. Bernstein, A. A. Caudy, S. M. Hammond, and G. J. Hannon. Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, 409(6818):363–365, 2001.

[39] B. R. Bill, A. M. Petzold, K. J. Clark, L. A. Schimmenti, and S. C. Ekker. A primer for morpholino use in zebraﬁsh. Zebraﬁsh, 6(1):69–77, 2009.

[40] V. Bitko, A. Musiyenko, O. Shulyayeva, and S. Barik. Inhibition of respira- tory viruses by nasally administered siRNA. Nature Medicine, 11(1):50–55, 2004.

[41] P. R. Blackburn, J. M. Campbell, K. J. Clark, and S. C. Ekker. The crispr system—keeping zebraﬁsh gene targeting fresh. Zebraﬁsh, 10(1):116–118, 2013.

[42] D. Boden, O. Pusch, F. Lee, L. Tucker, P. R. Shank, and B. Ramratnam. Promoter choice aﬀects the potency of HIV1 speciﬁc RNA interference. Nu- cleic Acids Research, 31(17):5033–5038, 2003. Bibliography 172

[43] H. Bohle, N. Lorenzen, and B. D. Schyth. Species speciﬁc inhibition of viral replication using dicer substrate siRNAs (DsiRNAs) targeting the viral nucleoprotein of the ﬁsh pathogenic rhabdovirus viral hemorrhagic septicemia virus (VHSV). Antiviral Research, 90(3):187–194, 2011.

[44] M. T. Bohnsack, K. Czaplinski, and D. GORLICH.¨ Exportin 5 is a RanGTP- dependent dsRNA-binding protein that mediates nuclear export of pre- miRNAs. RNA, 10(2):185–191, 2004.

[45] S. Boonanuntanasarn, G. Yoshizaki, and T. Takeuchi. Speciﬁc gene silencing using small interfering RNAs in ﬁsh embryos. Biochemical and Biophysical Research Communications, 310(4):1089–1095, 2003.

[46] S. Boonanuntanasarn, S. Panyim, and G. Yoshizaki. Usage of putative ze- braﬁsh U6 promoters to express shRNA in Nile tilapia and shrimp cell extracts. Transgenic Research, 18:323–325, 2009.

[47] K. B¨orner, D. Niopek, G. Cotugno, M. Kaldenbach, T. Pankert, J. Willem- sen, X. Zhang, N. Sch¨urmann, S. Mockenhaupt, and A. Serva. Robust rnai enhancement via human argonaute-2 overexpression from plasmids, viral vectors and cell lines. Nucleic acids research, 2013.

[48] M. M. Bradford. A rapid and sensitive method for the quantitation of mi- crogram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72(1–2):248 – 254, 1976. ISSN 0003-2697.

[49] M. Brand, M. Granato, and C. Nüsslein-Volhard. Keeping and raising zebrafish. Zebrafish, 261:7–37, 2002.

[50] A. M. Brownawell. Exportin-5, a novel karyopherin, mediates nuclear export of double-stranded RNA binding proteins. The Journal of Cell Biology, 156 (1):53–64, 2002.

[51] T. R. Brummelkamp, R. Bernards, and R. Agami. A system for stable expression of short interfering RNAs in mammalian cells. Science, 296(5567): 550 –553, 2002. Bibliography 173

[52] J. Burchard, A. L. Jackson, V. Malkov, R. H. Needham, Y. Tan, S. R. Bartz, H. Dai, A. B. Sachs, and P. S. Linsley. microRNA-like oﬀ-target transcript regulation by siRNAs is species speciﬁc. RNA, 15(2):308–315, 2009.

[53] D. J. Burgess. Small RNAs: A new mechanism of microRNA target recognition. Nature Reviews Genetics, 13(4):233–233, 2012.

[54] D. J. Burgess. Small RNAs: Antiviral RNAi in mammals. Nature Reviews Genetics, 14(12):821–821, 2013.

[55] J. S. Burgos, J. Ripoll-Gomez, J. M. Alfaro, I. Sastre, and F. Valdivieso. Zebraﬁsh as a new model for herpes simplex virus type 1 infection. Zebraﬁsh, 5(4):323–333, 2008. ISSN 1557-8542 (Electronic); 1545-8547 (Linking).

[56] L. Cade, D. Reyon, W. Y. Hwang, S. Q. Tsai, S. Patel, C. Khayter, J. K. Joung, J. D. Sander, R. T. Peterson, and J. R. J. Yeh. Highly eﬃcient generation of heritable zebraﬁsh gene mutations using homo- and heterodimeric TALENs. Nucleic Acids Research, 40(16):8001–8010, 2012.

[57] S. Campbell, B. Collet, K. Einer-Jensen, C. J. Secombes, and M. Snow. Iden- tifying potential virulence determinants in viral haemorrhagic septicaemia virus (VHSV) for rainbow trout. Dis Aquat Org, 86:205–212, 2009.

[58] S. Campbell, A. McBeath, C. Secombes, M. Snow, and B. Collet. Interferon response following infection with genetically similar isolates of viral haemorrhagic septicaemia virus (VHSV) exhibiting contrasting virulence in rainbow trout. Fish & Shellﬁsh Immunology, 30(1):287–294, 2011.

[59] W. T. G. Campus. The Danio rerio sequencing project. Wellcome Trust Genome Campus.

[60] W. Cao, R. Hunter, D. Strnatka, C. A. McQueen, and R. P. Erickson. DNA constructs designed to produce short hairpin, interfering RNAs in transgenic mice sometimes show early lethality and an interferon response. Journal Applied Genetics, 46(2):217–225, 2005. Bibliography 174

[61] N. J. Caplen. Gene therapy progress and prospects. Downregulating gene expression: the impact of RNA interference. Gene Therapy, 11(16):1241– 1248, 2004.

[62] R. W. Carthew and E. J. Sontheimer. Origins and mechanisms of mirnas and sirnas. Cell, 136(4):642–655, 2009.

[63] D. Castanotto, K. Sakurai, R. Lingeman, H. Li, L. Shively, L. Aagaard, H. Soifer, A. Gatignol, A. Riggs, and J. J. Rossi. Combinatorial delivery of small interfering RNAs reduces RNAi eﬃcacy by selective incorporation into RISC. Nucleic Acids Research, 35(15):5154–5164, 2007.

[64] C. Catalanotto, M. Pallotta, P. ReFalo, M. S. Sachs, L. Vayssie, G. Macino, and C. Cogoni. Redundancy of the two dicer genes in transgene-induced posttranscriptional gene silencing in Neurospora crassa. Molecular and Cel- lular Biology, 24(6):2536–2545, 2004.

[65] S. Chelouﬁ, C. O. Dos Santos, M. M. W. Chong, and G. J. Hannon. A dicer-independent miRNA biogenesis pathway that requires Ago catalysis. Nature, 465(7298):584–589, 2010.

[66] J.-Y. Chen, Y.-K. You, J.-C. Chen, T.-C. Huang, and C.-M. Kuo. Organi- zation and promoter analysis of the zebraﬁsh (Danio rerio) interferon gene. DNA and Cell Biology, 24(10):641–650, 2005.

[67] R. Chenna, H. Sugawara, T. Koike, R. Lopez, T. J. Gibson, D. G. Higgins, and J. D. Thompson. Multiple sequence alignment with the clustal series of programs. Nucleic Acids Research, 31(13):3497 –3500, 2003.

[68] C. Chesnutt and L. Niswander. Plasmid-based short-hairpin RNA interference in the chicken embryo. Genesis, 39(2):73–78, 2004.

[69] S. S. Chong, P. Hu, and N. Hernandez. Reconstitution of transcription from the human U6 small nuclear RNA promoter with eight recombinant polypeptides and a partially puriﬁed RNA polymerase III complex. Journal of Biological Chemistry, 276(23):20727–20734, 2001. Bibliography 175

[70] D. Cifuentes, H. Xue, D. W. Taylor, H. Patnode, Y. Mishima, S. Chelouﬁ, E. Ma, S. Mane, G. J. Hannon, N. D. Lawson, S. A. Wolfe, and A. J. Gi- raldez. A Novel miRNA Processing Pathway Independent of Dicer Requires Argonaute2 Catalytic Activity. Science, 328(5986):1694–1698, 2010.

[71] B. D. Clarke, D. M. Cummins, K. A. McColl, A. C. Ward, and T. J. Doran. Characterization of zebraﬁsh polymerase III promoters for the expression of short-hairpin RNA interference molecules. Zebraﬁsh, 10(4):472–479, 2013.

[72] B. Collet and C. J. Secombes. Type I-interferon signalling in ﬁsh. Fish & Shellﬁsh Immunology, 12(5):389–397, 2002.

[73] C. Collins, G. Ganne, and B. Collet. Isolation and activity of the promoters for stat1 and 2 in atlantic salmon Salmo salar. Fish & Shellﬁsh Immunology, 2014.

[74] L. Cong, F. A. Ran, D. Cox, S. Lin, R. Barretto, N. Habib, P. D. Hsu, X. Wu, W. Jiang, L. A. Marraﬃni, et al. Multiplex genome engineering using crispr/cas systems. Science, 339(6121):819–823, 2013.

[75] M. J. Crim and L. K. Riley. Viral Diseases in Zebraﬁsh: What Is Known and Unknown. ILAR Journal, 53(2), 2012.

[76] G. E. Crooks, G. Hon, J.-M. Chandonia, and S. E. Brenner. WebLogo: a sequence logo generator. Genome Research, 14(6):1188–1190, 2004.

[77] B. R. Cullen. Is RNA interference involved in intrinsic antiviral immunity in mammals? Nature Immunology, 7(6):563–567, 2006.

[78] B. R. Cullen, S. Cherry, and B. R. tenOever. Is RNA interference a physiologically relevant innate antiviral immune response in mammals? Cell Host & Microbe, 14(4):374 – 378, 2013. ISSN 1931-3128.

[79] D. M. Cummins, S. G. Tyack, and T. J. Doran. Characterisation and comparison of the chicken H1 RNA polymerase III promoter for short hairpin RNA expression. Biochemical and Biophysical Research Communications, 416(1–2):194–198, 2011. Bibliography 176

[80] J. T. Cuperus, N. Fahlgren, and J. C. Carrington. Evolution and functional diversiﬁcation of miRNA genes. The Plant Cell Online, 23(2):431–442, 2011.

[81] J. M. Cutr´ın,J.G.Olveira,I.Band´ın, and C. P. Dopazo. Validation of real time RT-PCR applied to cell culture for diagnosis of any known genotype of viral haemorrhagic septicaemia virus. Journal of Virological Methods, 162 (1):155–162, 2009.

[82] F. Czauderna, A. Santel, M. Hinz, M. Fechtner, B. Durieux, G. Fisch, F. Leenders, W. Arnold, K. Giese, A. Klippel, et al. Inducible shRNA expression for application in a prostate cancer mouse model. Nucleic Acids Research, 31(21):e127–e127, 2003.

[83] R. Dahm and R. Geisler. Learning from Small Fry: The Zebraﬁsh as a Ge- netic Model Organism for Aquaculture Fish Species. Marine Biotechnology, 8(4):329–345, 2006.

[84] T. Dalmay, R. Horseﬁeld, T. H. Braunstein, and D. C. Baulcombe. SDE3 encodes an RNA helicase required for post-transcriptional gene silencing in Arabidopsis. The EMBO Journal, 20(8):2069–2077, 2001.

[85] L. T. Dang, H. Kondo, I. Hirono, and T. Aoki. Inhibition of red seabream iridovirus (RSIV) replication by small interfering RNA (siRNA) in a cell culture system. Antiviral Research, 77(2):142–149, 2008.

[86] D. A. Danzeiser, O. Urso, and G. R. Kunkel. Functional characterization of elements in a human U6 small nuclear RNA gene distal control region. Molecular and Cellular Biology, 13(8):4670–4678, 1993.

[87] R. M. Das, N. J. Van Hateren, G. R. Howell, E. R. Farrell, F. K. Bangs, V. C. Porteous, E. M. Manning, M. J. McGrew, K. Ohyama, M. A. Sacco, et al. A robust system for RNA interference in the chicken using a modiﬁed microRNA operon. Developmental Biology, 294(2):554–563, 2006.

[88] A. J. Davidson and L. I. Zon. The “deﬁnitive” (and “primitive”) guide to zebraﬁsh hematopoiesis. Oncogene, 23(43):7233–7246, 2004. Bibliography 177

[89] E. De Clercq. Interferon and its inducers—a never-ending story:“old” and “new” data in a new perspective. Journal of Infectious Diseases, 194 (Supplement 1):S19–S26, 2006.

[90] W. de Vries, J. Haasnoot, R. Fouchier, P. de Haan, and B. Berkhout. Dif- ferential RNA silencing suppression activity of NS1 proteins from diﬀerent inﬂuenza A virus strains. Journal of General Virology, 90(8), 2009.

[91] M. A. Denti, A. Rosa, O. Sthandier, F. G. De Angelis, and I. Bozzoni. A new vector, based on the polII promoter of the U1 snRNA gene, for the expression of sirnas in mammalian cells. Molecular Therapy, 10(1):191–199, 2004.

[92] S. Diederichs, S. Jung, S. M. Rothenberg, G. A. Smolen, B. G. Mlody, and D. A. Haber. Coexpression of Argonaute-2 enhances RNA interference toward perfect match binding sites. Proceedings of the National Academy of Sciences, 105(27):9284–9289, 2008.

[93] S. Dios, A. Romero, R. Chamorro, A. Figueras, and B. Novoa. Eﬀect of the temperature during antiviral immune response ontogeny in teleosts. Fish and Shellﬁsh Immunology, 29(6):1019 – 1027, 2010. ISSN 1050-4648.

[94] J. G. Doench, C. P. Petersen, and P. A. Sharp. siRNAs can function as miRNAs. Genes & Development, 17(4):438–442, 2003.

[95] M. Dong, Y.-F. Fu, T.-T. Du, C.-B. Jing, C.-T. Fu, Y. Chen, Y. Jin, M. Deng, and T. X. Liu. Heritable and Lineage-Speciﬁc Gene Knockdown in Zebraﬁsh Embryo. PLoS ONE, 4(7):e6125, 2009.

[96] Z. Dong, J. Peng, and S. Guo. Stable Gene Silencing in Zebraﬁsh with Spatiotemporally Targetable RNA Interference. Genetics, 193(4):1065–1071, 2013.

[97] B. W. Draper, P. A. Morcos, and C. B. Kimmel. Inhibition of zebraﬁsh fgf8 premRNA splicing with morpholino oligos: A quantiﬁable method for gene knockdown. genesis, 30(3):154–156, 2001. Bibliography 178

[98] C. M. Duarte, M. Holmer, Y. Olsen, D. Soto, N. Marb`a, J. Guiu, K. Black, and I. Karakassis. Will the oceans help feed humanity? BioScience, 59(11): 967–976, 2009.

[99] J. R. Ecker and R. W. Davis. Inhibition of gene expression in plant cells by expression of antisense RNA. Proceedings of the National Academy of Sciences, 83(15):5372–5376, 1986.

[100] P. S. Eder, A. Srinivasan, M. C. Fishman, and S. Altman. The RNA subunit of ribonuclease P from the zebraﬁsh, Danio rerio. Journal of Biological Chemistry, 271(35):21031–21036, 1996.

[101] J. S. Eisen and J. C. Smith. Controlling morpholino experiments: don’t stop making antisense. Development, 135(10):1735–1743, 2008.

[102] S. M. Elbashir, W. Lendeckel, and T. Tuschl. RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes & Development, 15(2):188–200, 2001.

[103] A. Ellis. Innate host defense mechanisms of ﬁsh against viruses and bacteria. Developmental & Comparative Immunology, 25(8):827–839, 2001.

[104] E. Elsayed, M. Faisal, M. Thomas, G. Whelan, W. Batts, and J. Winton. Isolation of viral haemorrhagic septicaemia virus from muskellunge, Esox masquinongy (Mitchill), in Lake St Clair, Michigan, USA reveals a new sublineage of the North American genotype. Journal of Fish Diseases,29 (10):611–619, 2006.

[105] S. C. Fahrenkrug, A. Blake, D. F. Carlson, T. Doran, A. Van Eenennaam, D. Faber, C. Galli, Q. Gao, P. B. Hackett, N. Li, E. A. Maga, W. M. Muir, J. D. Murray, D. Shi, R. Stotish, E. Sullivan, J. F. Taylor, M. Walton, M. Wheeler, B. Whitelaw, and B. P. Glenn. Precision genetics for complex objectives in animal agriculture. Journal of Animal Science, 88(7):2530– 2539, 2010. Bibliography 179

[106] C. M. Fauquet, M. A. Mayo, J. Maniloﬀ, U. Desselberger, and L. A. Ball. Virus taxonomy: VIIIth report of the International Committee on Taxonomy of Viruses. 2005.

[107] B. J. Fenner, R. Thiagarajan, H. K. Chua, and J. Kwang. Betanodavirus B2 is an RNA interference antagonist that facilitates intracellular viral RNA accumulation. Journal of Virology, 80(1):85–94, 2005.

[108] G. D. Fewell and K. Schmitt. Vector-based RNAi approaches for stable, inducible and genome-wide screens. Drug discovery today, 11(21):975–982, 2006.

[109] S. Finke, J. H. Cox, and K.-K. Conzelmann. Diﬀerential transcription attenuation of rabies virus genes by intergenic regions: Generation of recombinant viruses overexpressing the polymerase gene. Journal of Virology,74 (16):7261–7269, 2000.

[110] A. Fire, S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, and C. C. Mello. Potent and speciﬁc genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391(6669):806–811, 1998.

[111] A. Fjose and Ø. Drivenes. RNAi and microRNAs: from animal models to disease therapy. Birth Defects Research Part C: Embryo Today: Reviews, 78(2):150–171, 2006.

[112] A. Fjose and X.-F. Zhao. Inhibition of the microRNA pathway in zebraﬁsh by siRNA. Methods in Molecular Biology, 629:239–255, 2010.

[113] A. S. Flynt and E. C. Lai. RNAi in Xenopus: look before you leap. Genes & Development, 25(11):1105–1108, 2011.

[114] S. M. Freier, R. Kierzek, J. A. Jaeger, N. Sugimoto, M. H. Caruthers, T. Neil- son, and D. H. Turner. Improved free-energy parameters for predictions of rna duplex stability. Proceedings of the National Academy of Sciences,83 (24):9373–9377, 1986. Bibliography 180

[115] Z. Fu. Genetic comparisons of rhabdoviruses from animals and plants. In Z. Fu, editor, The world of rhabdoviruses, pages 1–24. Springer, 2005.

[116] K. A. Garver, A. G. Dwilow, J. Richard, T. F. Booth, D. R. Beniac, and B. W. Souter. First detection and conﬁrmation of spring viraemia of carp virus in common carp, cyprinus carpio l., from hamilton harbour, lake on- tario, canada. Journal of Fish Diseases, 30(11):665–671, 2007. ISSN 1365- 2761.

[117] A. Gatignol, S. Lain´e, and G. Clerzius. Dual role of TRBP in HIV replication and RNA interference: viral diversion of a cellular pathway or evasion from antiviral immunity? Retrovirology, 2(1):65, 2005.

[118] Q. Ge, H. Ilves, A. Dallas, P. Kumar, J. Shorenstein, S. A. Kazakov, and B. H. Johnston. Minimal-length short hairpin RNAs: the relationship of structure and RNAi activity. RNA, 16(1):106–117, 2010.

[119] J. C. Giering, D. Grimm, T. A. Storm, and M. A. Kay. Expression of shRNA from a tissue-speciﬁc pol II promoter is an eﬀective and safe RNAi therapeutic. Molecular Therapy, 16(9):1630–1636, 2008.

[120] O. Gilad, S. Yun, M. A. Adkison, K. Way, N. H. Willits, H. Bercovier, and R. P. Hedrick. Molecular comparison of isolates of an emerging ﬁsh pathogen, koi herpesvirus, and the eﬀect of water temperature on mortality of experimentally infected koi. Journal of General Virology, 84(10):2661– 2667, 2003.

[121] E. Gomez-Casado, A. Estepa, and J. M. Coll. A comparative review on European-farmed ﬁnﬁsh RNA viruses and their vaccines. Vaccine, 29(15): 2657–2671, 2011.

[122] R. I. Gregory, K.-p. Yan, G. Amuthan, T. Chendrimada, B. Doratotaj, N. Cooch, and R. Shiekhattar. The Microprocessor complex mediates the genesis of microRNAs. Nature, 432(7014):235–240, 2004. Bibliography 181

[123] F. Grey, R. Tirabassi, H. Meyers, G. Wu, S. McWeeney, L. Hook, and J. A. Nelson. A viral microRNA down-regulates multiple cell cycle genes through mRNA 5 UTRs. PLoS pathogens, 6(6):e1000967, 2010.

[124] S. Griﬃths-Jones, R. J. Grocock, S. van Dongen, A. Bateman, and A. J. Enright. miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Research, 34(suppl 1):D140–D144, 2006.

[125] S. Griﬃths-Jones, H. K. Saini, S. van Dongen, and A. J. Enright. mirbase: tools for microrna genomics. Nucleic Acids Research, 36(suppl 1):D154– D158, 2008.

[126] D. Grimm. The dose can make the poison: lessons learned from adverse in vivo toxicities caused by RNAi overexpression. Silence, 2(1):1–6, 2011.

[127] D. Grimm, K. Pandey, and M. A. Kay. Adeno-Associated Virus Vectors for Short Hairpin RNA Expression. In D. R. Engelke and J. J. Rossi, editors, RNA Interference, pages 381–405. Academic Press, 2005.

[128] D. Grimm, K. L. Streetz, C. L. Jopling, T. A. Storm, K. Pandey, C. R. Davis, P. Marion, F. Salazar, and M. A. Kay. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature, 441(7092):537– 541, 2006.

[129] D. Grimm, L. Wang, J. S. Lee, N. Schürmann, S. Gu, K. Börner, T. A. Storm, and M. A. Kay. Argonaute proteins are key determinants of RNAi efficacy, toxicity, and persistence in the adult mouse liver. The Journal of Clinical Investigation, 120(9):3106–3119, 2010.

[130] C. Gwizdek, B. Ossareh-Nazari, A. M. Brownawell, A. Doglio, E. Bertrand, I. G. Macara, and C. Dargemont. Exportin-5 mediates nuclear export of minihelix-containing RNAs. Journal of Biological Chemistry, 278(8):5505– 5508, 2003. Bibliography 182

[131] I. Ha, B. Wightman, and G. Ruvkun. A bulged lin-4/lin-14 RNA duplex is suﬃcient for caenorhabditis elegans lin-14 temporal gradient formation. Genes & Development, 10(23):3041–3050, 1996.

[132] J. Haasnoot, W. de Vries, E.-J. Geutjes, M. Prins, P. de Haan, and B. Berkhout. The Ebola virus VP35 protein is a suppressor of RNA silencing. PLoS Pathogens, 3(6):e86, 2007.

[133] K. M. Halbig, A. C. Lekven, and G. R. Kunkel. Zebraﬁsh U6 small nuclear RNA gene promoters contain a SPH element in an unusual location. Gene, 421(1-2):89–94, 2008.

[134] S. M. Hammond. Dicing and slicing: The core machinery of the RNA interference pathway. FEBS Letters, 579(26):5822–5829, 2005.

[135] S. M. Hammond, E. Bernstein, D. Beach, and G. J. Hannon. An RNA- directed nuclease mediates post-transcriptional gene silencing in Drosophila cells. Nature, 404(6775):293–296, 2000.

[136] L. He and G. J. Hannon. MicroRNAs: small RNAs with a big role in gene regulation. Nature Reviews Genetics, 5(7):522–531, 2004.

[137] S. He, E. Salas-Vidal, S. Rueb, S. G. Krens, A. H. Meijer, B. E. Snaar- Jagalska, and H. P. Spaink. Genetic and Transcriptome Characterization of Model Zebraﬁsh Cell Lines. Zebraﬁsh., 3(4):441–453, 2006.

[138] J. Heasman. Morpholino oligos: making sense of antisense? Developmental Biology, 243(2):209–214, 2002.

[139] J. D. Heidel, S. Hu, X. F. Liu, T. J. Triche, and M. E. Davis. Lack of interferon response in animals to naked siRNAs. Nature Biotechnology,22 (12):1579–1582, 2004.

[140] A. Helwak, G. Kudla, T. Dudnakova, and D. Tollervey. Mapping the human miRNA interactome by CLASH reveals frequent noncanonical binding. Cell, 153(3):654–665, 2013. Bibliography 183

[141] J. Heppell and H. L. Davis. Application of DNA vaccine technology to aquaculture. Advanced Drug Delivery Reviews, 43(1):29 – 43, 2000.

[142] P. Herbomel, B. Thisse, and C. Thisse. Ontogeny and behaviour of early macrophages in the zebraﬁsh embryo. Development, 126(17):3735–3745, 1999.

[143] N. Hernandez. Small nuclear RNA genes: a model system to study fundamental mechanisms of transcription. Journal of Biological Chemistry, 276 (29):26733–26736, 2001.

[144] W. Herr. Oct-1 and Oct-2: Diﬀerential transcriptional regulation by proteins that bind to the same DNA sequence. Cold Spring Harbor Monograph Archive, 22:1103–1135, 1992.

[145] E. Herrera-Carrillo, A. Harwig, Y. P. Liu, and B. Berkhout. Probing the shrna characteristics that hinder dicer recognition and consequently allow ago-mediated processing and agoshrna activity. RNA, 2014.

[146] T. M. Hinton and T. J. Doran. Inhibition of chicken anaemia virus replication using multiple short-hairpin RNAs. Antiviral Research, 80(2):143–149, 2008.

[147] N. Hitoshi, Y. Ken-ichi, and M. Jun-ichi. Eﬃcient selection for high- expression transfectants with a novel eukaryotic vector. Gene, 108(2):193– 199, 1991.

[148] B. Hoﬀmann, M. Beer, H. Sch¨utze, and T. Mettenleiter. Fish rhabdoviruses: molecular epidemiology and evolution. The World of Rhabdoviruses, pages 81–117, 2005.

[149] H.-H. Hoﬀmann, W. M. Schneider, and C. M. Rice. Interferons and viruses: an evolutionary arms race of molecular interactions. Trends in immunology, 36(3):124–138 1471–4906, 2015.

[150] T. Holen, M. Amarzguioui, E. Babaie, and H. Prydz. Similar behaviour of single-strand and double-strand siRNAs suggests they act through a common RNAi pathway. Nucleic Acids Research, 31(9):2401–2407, 2003. Bibliography 184

[151] K. M. Hope, R. N. Casey, G. H. Groocock, R. G. Getchell, P. R. Bowser, and J. W. Casey. Comparison of quantitative RT-PCR with cell culture to detect viral hemorrhagic septicemia virus (VHSV) IVb infections in the Great Lakes. Journal of Aquatic Animal Health, 22(1):50–61, 2010.

[152] V. Hornung, M. Guenthner-Biller, C. Bourquin, A. Ablasser, M. Schlee, S. Uematsu, A. Noronha, M. Manoharan, S. Akira, A. de Fougerolles, et al. Sequence-speciﬁc potent induction of IFN-α by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nature Medicine, 11(3):263– 270, 2005.

[153] K. Howe, M. D. Clark, C. F. Torroja, J. Torrance, C. Berthelot, M. Muffato, J. E. Collins, S. Humphray, K. McLaren, L. Matthews, S. McLaren, I. Sealy, M. Caccamo, C. Churcher, C. Scott, J. C. Barrett, R. Koch, G.-J. Rauch, S. White, W. Chow, B. Kilian, L. T. Quintais, J. A. Guerra-Assun¸cão, Y. Zhou, Y. Gu, J. Yen, J.-H. Vogel, T. Eyre, S. Redmond, R. Baner- jee, J. Chi, B. Fu, E. Langley, S. F. Maguire, G. K. Laird, D. Lloyd, E. Kenyon, S. Donaldson, H. Sehra, J. Almeida-King, J. Loveland, S. Trevan- ion, M. Jones, M. Quail, D. Willey, A. Hunt, J. Burton, S. Sims, K. McLay, B. Plumb, J. Davis, C. Clee, K. Oliver, R. Clark, C. Riddle, D. Eliott, G. Threadgold, G. Harden, D. Ware, B. Mortimer, G. Kerry, P. Heath, B. Phillimore, A. Tracey, N. Corby, M. Dunn, C. Johnson, J. Wood, S. Clark, S. Pelan, G. Griffiths, M. Smith, R. Glithero, P. Howden, N. Barker, C. Stevens, J. Harley, K. Holt, G. Panagiotidis, J. Lovell, H. Beasley, C. Hen- derson, D. Gordon, K. Auger, D. Wright, J. Collins, C. Raisen, L. Dyer, K. Leung, L. Robertson, K. Ambridge, D. Leongamornlert, S. McGuire, R. Gilderthorp, C. Griffiths, D. Manthravadi, S. Nichol, G. Barker, S. White- head, M. Kay, J. Brown, C. Murnane, E. Gray, M. Humphries, N. Sycamore, D. Barker, D. Saunders, J. Wallis, A. Babbage, S. Hammond, M. Mashreghi- Mohammadi, L. Barr, S. Martin, P. Wray, A. Ellington, N. Matthews, M. Ell- wood, R. Woodmansey, G. Clark, J. Cooper, A. Tromans, D. Grafham, C. Skuce, R. Pandian, R. Andrews, E. Harrison, A. Kimberley, J. Garnett, N. Fosker, R. Hall, P. Garner, D. Kelly, C. Bird, S. Palmer, I. Gehring, Bibliography 185

A. Berger, C. M. Dooley, Z. Ersan-Ur¨¨ un, C. Eser, H. Geiger, M. Geisler, L. Karotki, A. Kirn, J. Konantz, M. Konantz, M. Oberländer, S. Rudolph- Geiger, M. Teucke, K. Osoegawa, B. Zhu, A. Rapp, S. Widaa, C. Langford, F. Yang, N. P. Carter, J. Harrow, Z. Ning, J. Herrero, S. M. J. Searle, A. En- right, R. Geisler, R. H. A. Plasterk, C. Lee, M. Westerfield, P. J. de Jong, L. I. Zon, J. H. Postlethwait, C. Nüsslein-Volhard, T. J. P. Hubbard, H. R. Crollius, J. Rogers, and D. L. Stemple. The zebrafish reference genome sequence and its relationship to the human genome. Nature, pages 1–6, 2013.

[154] D. J. Y. Hsieh and C. F. Liao. Zebraﬁsh M2 muscarinic acetylcholine receptor: cloning, pharmacological characterization, expression patterns and roles in embryonic bradycardia. British journal of pharmacology, 137(6):782–792, 2009.

[155] M. Huang, F.-J. Jia, Y.-C. Yan, L.-H. Guo, and Y.-P. Li. Transactivated minimal E1b promoter is capable of driving the expression of short hairpin rna. Journal of Virological Methods, 134(1):48–54, 2006.

[156] G. Hutv´agner, J. McLachlan, A. E. Pasquinelli, E.´ B´alint, T. Tuschl, and P. D. Zamore. A Cellular Function for the RNA-Interference Enzyme Dicer in the Maturation of the let-7 Small Temporal RNA. Science, 293(5531): 834–838, 2001.

[157] C. Ill and H. Chiou. Gene therapy progress and prospects: recent progress in transgene and RNAi expression cassettes. Gene Therapy, 12(10):795–802, 2005.

[158] T. Isshiki, T. Nishizawa, T. Kobayashi, T. Nagano, and T. Miyazaki. An outbreak of VHSV (viral hemorrhagic septicemia virus) infection in farmed Japanese ﬂounder Paralichthys olivaceus in Japan. Diseases of Aquatic Or- ganisms, 47(2):87–99, 2001.

[159] L. E. Iverson and J. K. Rose. Localized attenuation and discontinuous synthesis during vesicular stomatitis virus transcription. Cell, 23(2):477–484, 1981. Bibliography 186

[160] T. Iwamoto. Characterization of Striped jack nervous necrosis virus subge- nomic RNA3 and biological activities of its encoded protein B2. Journal of General Virology, 86(10):2807–2816, 2005.

[161] L. R. Iwanowicz and A. E. Goodwin. A new bacilliform fathead minnow rhabdovirus that produces syncytia in tissue culture. Archives of Virology, 147(5):899–915, 2002.

[162] A. L. Jackson and P. S. Linsley. Recognizing and avoiding siRNA off-target effects for target identification and therapeutic application. Nature Reviews Drug Discovery, 9(1):57–67, 2010.

[163] I. Jensen and B. Robertsen. Eﬀect of double-stranded RNA and interferon on the antiviral activity of Atlantic salmon cells against infectious salmon anemia virus and infectious pancreatic necrosis virus. Fish & Shellﬁsh Immunology, 13(3):221–241, 2002.

[164] X. Ji. The mechanism of RNase III action: how dicer dices. pages 99–116, 2008.

[165] L. Jiang, P. Zhao, G. Wang, T. Cheng, Q. Yang, S. Jin, P. Lin, Y. Xiao, Q. Sun, and Q. Xia. Comparison of factors that may aﬀect the inhibitory eﬃ- cacy of transgenic RNAi targeting of baculoviral genes in silkworm, Bombyx mori. Antiviral Research, 97(3):255–263, 2013.

[166] B. O. Johnsen and A. J. Jenser. The gyrodactylus story in norway. Aqua- culture, 98(1–3):289 – 302, 1991.

[167] K. L. Johnson, B. D. Price, L. D. Eckerle, and L. A. Ball. Nodamura virus nonstructural protein B2 can enhance viral RNA accumulation in both mammalian and insect cells. Journal of Virology, 78(12):6698–6704, 2004.

[168] C. L. Jopling, M. Yi, A. M. Lancaster, S. M. Lemon, and P. Sarnow. Mod- ulation of hepatitis C virus RNA abundance by a liver-speciﬁc MicroRNA. Science, 309(5740):1577–1581, 2005. Bibliography 187

[169] P. V. Jørgensen. Egtved virus: temperature-dependent immune response of trout to infection with low-virulence virus. Journal of Fish Diseases, 5(1): 47–55, 1982.

[170] A. D. Judge, G. Bola, A. C. Lee, and I. MacLachlan. Design of noninﬂam- matory synthetic siRNA mediating potent gene silencing in vivo. Molecular Therapy, 13(3):494–505, 2006.

[171] M. T. Jurado, P. Garc´ıa-Valtanen, A. Estepa, and L. Perez. Antiviral activity produced by an IPNV-carrier EPC cell culture confers resistance to VHSV infection. Veterinary Microbiology, 166(3):412–418, 2013.

[172] T. Katahira and H. Nakamura. Gene silencing in chick embryos with a vector-based small interfering RNA system. Development, growth & diﬀer- entiation, 45(4):361–367, 2003.

[173] K. Kawakami. Tol2: a versatile gene transfer vector in vertebrates. Genome Biology, 8(Suppl 1):S7, 2007.

[174] K. Kawakami and A. Shima. Identification of theTol2 transposase of the medaka fish Oryzias latipes that catalyzes excision of a nonautonomous¡ Tol2 element in zebrafish Danio rerio. Gene, 240(1):239–244, 1999.

[175] K. Kawakami, A. Shima, and N. Kawakami. 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. Proceedings of the National Academy of Sciences, 97(21):11403–11408, 2000.

[176] K. M. Keene, B. D. Foy, I. Sanchez-Vargas, B. J. Beaty, C. D. Blair, and K. E. Olson. Rna interference acts as a natural antiviral response to o’nyong-nyong virus (alphavirus; togaviridae) infection of anopheles gambiae. Proceedings of the National Academy of Sciences of the United States of America, 101 (49):17240–17245, 2004.

[177] A. Kelly and A. F. Hurlstone. The use of RNAi technologies for gene knockdown in zebraﬁsh. Brieﬁngs in Functional Genomics, 2011. Bibliography 188

[178] J. R. Kennerdell and R. W. Carthew. Use of dsRNA-mediated genetic interference to demonstrate that frizzled and frizzled 2 act in the wingless pathway. Cell, 95(7):1017, 1998.

[179] A. Khvorova, A. Reynolds, and S. D. Jayasena. Functional siRNAs and miRNAs exhibit strand bias. Cell, 115(2):209–216, 2003.

[180] H. Kilpinen, S. M. Waszak, A. R. Gschwind, S. K. Raghav, R. M. Witwicki, A. Orioli, E. Migliavacca, M. Wiederkehr, M. Gutierrez-Arcelus, N. I. Panousis, et al. Coordinated eﬀects of sequence variation on DNA binding, chromatin structure, and transcription. Science, 342(6159):744–747, 2013.

[181] D.-H. Kim, M. A. Behlke, S. D. Rose, M.-S. Chang, S. Choi, and J. J. Rossi. Synthetic dsrna dicer substrates enhance rnai potency and eﬃcacy. Nature Biotechnology, 23(2):222–226, 2005.

[182] M. S. Kim and K. H. Kim. Effects of NV gene knock-out recombinant viral hemorrhagic septicemia virus (VHSV) on Mx gene expression in Epithelioma papulosum cyprini (EPC) cells and olive flounder (Paralichthys olivaceus). Fish & Shellfish Immunology, 32(3):459–463, 2012.

[183] R. K. Kim and M. Faisal. The Laurentian Great Lakes strain (MI03) of the viral haemorrhagic septicaemia virus is highly pathogenic for juvenile muskellunge, Esox masquinongy (Mitchill). Journal of Fish Diseases, 33(6): 513–527, 2010.

[184] S.-M. Kim, K.-N. Lee, S.-J. Lee, Y.-J. Ko, H.-S. Lee, C.-H. Kweon, H.- S. Kim, and J.-H. Park. Multiple shrnas driven by u6 and cmv promoter enhances eﬃciency of antiviral eﬀects against foot-and-mouth disease virus. Antiviral research, 87(3):307–317, 2010.

[185] W.-S. Kim, S.-R. Kim, D. Kim, J.-O. Kim, M. Park, S.-I. Kitamura, H.-Y. Kim, D.-H. Kim, H.-J. Han, S.-J. Jung, et al. An outbreak of VHSV (viral hemorrhagic septicemia virus) infection in farmed olive ﬂounder Paralichthys olivaceus in Korea. Aquaculture, 296(1):165–168, 2009. Bibliography 189

[186] C. B. Kimmel, W. W. Ballard, S. R. Kimmel, B. Ullmann, and T. F. Schilling. Stages of embryonic development of the zebraﬁsh. Developmental Dynamics, 203(3):253–310, 1995.

[187] P. d. Kinkelin, M. Dorson, and A. Hattenberger-Baudouy. Interferon synthesis in trout and carp after viral infection. Developmental and Comparative Immunology, (Suppl. 2):167–174, 1982.

[188] M. Kiriakidou, P. T. Nelson, A. Kouranov, P. Fitziev, C. Bouyioukos, Z. Mourelatos, and A. Hatzigeorgiou. A combined computational- experimental approach predicts human microrna targets. Genes & development, 18(10):1165–1178, 2004.

[189] S. A. Knoepfel, M. Centlivre, Y. P. Liu, F. Boutimah, and B. Berkhout. Selection of RNAi-based inhibitors for anti-HIV gene therapy. World Journal of Virology, 1(3):79, 2012.

[190] A. Kozomara and S. Griﬃths-Jones. mirbase: integrating microrna annotation and deep-sequencing data. Nucleic Acids Research, 39(suppl 1): D152–D157, 2011.

[191] T. Kudo and S. Sutuo. Usage of putative chicken U6 promoters for vector- based RNA interference. Journal of Reproduction and Development, 51(3): 411–417, 2005.

[192] R. Kuhn, F. Schwenk, M. Aguet, and K. Rajewsky. Inducible gene targeting in mice. Science, 269(5229):1427–1429, 1995.

[193] G. R. Kunkel. Identiﬁcation of a SPH element in the distal region of a human U6 small nuclear RNA gene promoter and characterization of the SPH binding factor in HeLa cell extracts. Gene expression, 6(2):59–72, 1996.

[194] K. A. La Fauce and L. Owens. RNA interference reduces Pmerg DNV expression and replication in an in vivo cricket model. Journal of Invertebrate Pathology, 100(2):111–115, 2009. Bibliography 190

[195] L. S. Lambeth, T. G. Wise, R. J. Moore, M. S. Muralitharan, and T. J. Do- ran. Comparison of bovine RNA polymerase III promoters for short hairpin RNA expression. Animal Genetics, 37(4):369–372, 2006. ISSN 1365-2052.

[196] L. S. Lambeth, Y. Zhao, L. P. Smith, L. Kgosana, and V. Nair. Target- ing Marek’s disease virus by RNA interference delivered from a herpesvirus vaccine. Vaccine, 27(2):298–306, 2009.

[197] U. Langheinrich, E. Hennen, G. Stott, and G. Vacun. Zebraﬁsh as a model organism for the identiﬁcation and characterization of drugs and genes affecting p53 signaling. Current Biology, 12(23):2023–2028, 2002.

[198] S. E. LaPatra, L. Barone, G. R. Jones, and L. I. Zon. Eﬀects of Infectious Hematopoietic Necrosis Virus and Infectious Pancreatic Necrosis Virus In- fection on Hematopoietic Precursors of the Zebraﬁsh. Blood Cells, Molecules, and Diseases, 26(5):445–452, 2000.

[199] C.-H. Lecellier, P. Dunoyer, K. Arar, J. Lehmann-Che, S. Eyquem, C. Him- ber, A. Saib, and O. Voinnet. A cellular microRNA mediates antiviral defense in human cells. Science Signaling, 308(5721):557, 2005.

[200] H. Ledford. Transgenic salmon nears approval. Nature, 497(7447):17–18, 2013.

[201] R. C. Lee, R. L. Feinbaum, and V. Ambros. The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell, 75(5):843–854, 1993.

[202] Y. Lee, C. Ahn, J. Han, H. Choi, J. Kim, J. Yim, J. Lee, P. Provost, O. R˚admark, and S. Kim. The nuclear RNase III Drosha initiates microRNA processing. Nature, 425(6956):415–419, 2003.

[203] Y. Lee, M. Kim, J. Han, K.-H. Yeom, S. Lee, S. H. Baek, and V. N. Kim. MicroRNA genes are transcribed by RNA polymerase II. EMBO Journal, 23(20):4051–60, 2004. Bibliography 191

[204] J. Leonard and D. Schaﬀer. Antiviral RNAi therapy: emerging approaches for hitting a moving target. Gene Therapy, 13(6):532–540, 2005.

[205] J. N. Leonard and D. V. Schaﬀer. Computational design of antiviral RNA interference strategies that resist human immunodeﬁciency virus escape. Jour- nal of virology, 79(3):1645–1654, 2005.

[206] A. K. Leung, A. G. Young, A. Bhutkar, G. X. Zheng, A. D. Bosson, C. B. Nielsen, and P. A. Sharp. Genome-wide identiﬁcation of Ago2 binding sites from mouse embryonic stem cells with and without mature micrornas. Nature Structural & Molecular Biology, 18(2):237–244, 2011.

[207] J. P. Levraud, P. Boudinot, I. Colin, A. Benmansour, N. Peyrieras, P. Her- bomel, and G. Lutfalla. Identiﬁcation of the zebraﬁsh IFN receptor: implications for the origin of the vertebrate IFN system. The Journal of Im- munology, 178(7):4385–4394, 2007.

[208] J.-P. Levraud, N. Palha, C. Langevin, and P. Boudinot. Through the looking glass: witnessing host–virus interplay in zebraﬁsh. Trends in microbiology, 22(9):490–497, 2014.

[209] B. P. Lewis, C. B. Burge, and D. P. Bartel. Conserved seed pairing, often ﬂanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell, 120(1):15–20, 2005.

[210] Y. Li, J. Lu, Y. Han, X. Fan, and S.-W. Ding. RNA interference functions as an antiviral immunity mechanism in mammals. Science, 342(6155):231–234, 2013.

[211] Y.-X. Li, M. J. Farrell, R. Liu, N. Mohanty, and M. L. Kirby. Double- Stranded RNA Injection Produces Null Phenotypes in Zebraﬁsh. Develop- mental Biology, 217(2):394–405, 2000.

[212] V. Libri, A. Helwak, P. Miesen, D. Santhakumar, J. G. Borger, G. Kudla, F. Grey, D. Tollervey, and A. H. Buck. Murine cytomegalovirus encodes a Bibliography 192

miR-27 inhibitor disguised as a target. Proceedings of the National Academy of Sciences, 109(1):279–284, 2012.

[213] J. A. Lindbo, L. Silva-Rosales, W. M. Proebsting, and W. G. Dougherty. Induction of a highly speciﬁc antiviral state in transgenic plants: implications for regulation of gene expression and virus resistance. The Plant Cell Online, 5(12):1749–1759, 1993.

[214] J. Liu, M. A. Carmell, F. V. Rivas, C. G. Marsden, J. M. Thomson, J.-J. Song, S. M. Hammond, L. Joshua-Tor, and G. J. Hannon. Argonaute2 Is the Catalytic Engine of Mammalian RNAi. Science, 305(5689):1437–1441, 2004.

[215] J. Liu, M. A. Valencia-Sanchez, G. J. Hannon, and R. Parker. MicroRNA- dependent localization of targeted mRNAs to mammalian P-bodies. Nature Cell Biology, 7(7):719–723, 2005.

[216] W. Liu, Y. Guan, and P. Collodi. A zebrafish cell culture assay for the identification of microRNA targets. Zebrafish., 7(4):343–348, 2010.

[217] W.-Y. Liu, Y. Wang, Y.-H. Sun, Y. Wang, Y.-P. Wang, S.-P. Chen, and Z.-Y. Zhu. Efficient RNA interference in zebrafish embryos using siRNA synthe- sized with SP6 RNA polymerase. Development, Growth and Differentiation, 47(5):323–331, 2005.

[218] Y. P. Liu, N. C. T. Schopman, and B. Berkhout. Dicer-independent processing of short hairpin RNAs. Nucleic Acids Research, 41(6):3723–3733, 2013.

[219] K. J. Livak and T. D. Schmittgen. Analysis of relative gene expression data using real-time quantitative pcr and the 2δδct method. Methods, 25(4):402 – 408, 2001. ISSN 1046-2023.

[220] M.-W. Lu, Y.-M. Chao, T.-C. Guo, N. Santi, Ø. Evensen, S. K. Kasani, J.-R. Hong, and J.-L. Wu. The interferon response is involved in nervous necrosis Bibliography 193

virus acute and persistent infection in zebraﬁsh infection model. Molecular Immunology, 45(4):1146–1152, 2008.

[221] P. Y. Lu, F. Xie, and M. C. Woodle. In vivo application of RNA interference: from functional genomics to therapeutics. Advances in Genetics, 54:117, 2005.

[222] R. Lu, M. Maduro, F. Li, H. W. Li, G. Broitman-Maduro, W. X. Li, and S. W. Ding. Animal virus replication and RNAi-mediated antiviral silencing in Caenorhabditis elegans. Nature, 436(7053):1040–1043, 2005.

[223] S. Lu and B. R. Cullen. Adenovirus VA1 noncoding RNA can inhibit small interfering RNA and microRNA biogenesis. Journal of Virology, 78(23): 12868–12876, 2004.

[224] M. Ludwig, N. Palha, C. Torhy, V. Briolat, E. Colucci-Guyon, M. Br´emont, P. Herbomel, P. Boudinot, and Jean-Pierre. Whole-Body Analysis of a Viral Infection: Vascular Endothelium is a Primary Target of Infectious Hematopoietic Necrosis Virus in Zebraﬁsh Larvae. PLoS Pathogens, 7(2), 2011.

[225] E. Lund and J. E. Dahlberg. Substrate Selectivity of Exportin 5 and Dicer in the Biogenesis of MicroRNAs. Cold Spring Harbor Symposia on Quantitative Biology, 71, 2006.

[226] E. Lund, S. G¨uttinger, A. Calado, J. E. Dahlberg, and U. Kutay. Nuclear export of microRNA precursors. Science, 303(5654):95–98, 2004.

[227] E. Lund, M. D. Sheets, S. B. Imboden, and J. E. Dahlberg. Limiting Ago protein restricts RNAi and microRNA biogenesis during early development in Xenopus laevis. Genes & Development, 25(11):1121–1131, 2011.

[228] Y. Luo, S. Weng, Q. Wang, X. Shi, C. Dong, Q. Lu, X. Yu, and J. He. Tiger frog virus can infect zebraﬁsh cells for studying up- or down-regulated genes by proteomics approach. Virus Research, 144(1–2):171 – 179, 2009. ISSN 0168-1702. Bibliography 194

[229] C. R. MacKay, J. P. Wang, and E. A. Kurt-Jones. Dicer’s role as an antiviral: still an enigma. Current opinion in immunology, 26:49–55, 2014.

[230] P. Maillard, C. Ciaudo, A. Marchais, Y. Li, F. Jay, S. Ding, and O. Voinnet. Antiviral rna interference in mammalian cells. Science, 342(6155):235–238, 2013.

[231] P. I. Mäkinen, J. K. Koponen, A.-M. Kärkkäinen, T. M. Malm, K. H. Pulkki- nen, J. Koistinaho, M. P. Turunen, and S. Ylä-Herttuala. Stable RNA interference: comparison of U6 and H1 promoters in endothelial cells and in mouse brain. The Journal of Gene Medicine, 8(4):433–441, 2006.

[232] S. Mangos, B. Vanderbeld, R. Krawetz, K. Sudol, and G. M. Kelly. Ran binding protein RanBP1 in zebraﬁsh embryonic development*. Molecular reproduction and development, 59(3):235–248, 2001.

[233] M. Marcos-L´opez, P. Gale, B. C. Oidtmann, and E. J. Peeler. Assessing the impact of climate change on disease emergence in freshwater ﬁsh in the united kingdom. Transboundary and Emerging Diseases, 57(5):293–304, 2010. ISSN 1865-1682.

[234] J. Marra. When will we tame the oceans? Nature, 436(7048):175–176, 2005.

[235] J. Martinez and T. Tuschl. RISC is a 5 phosphomonoester-producing RNA endonuclease. Genes & Development, 18(9):975–980, 2004.

[236] I. Matejusova, P. McKay, F. Bland, and M. Snow. Application of a sensitive, speciﬁc and controlled realtime PCR assay to surveillance indicates a low prevalence of viral haemorrhagic septicaemia virus (VHSV) in wild herring, Clupea harengus L., in Scottish waters. Journal of Fish Diseases, 33(10): 841–847, 2010.

[237] C. Matranga, Y. Tomari, C. Shin, D. P. Bartel, and P. D. Zamore. Passenger- Strand Cleavage Facilitates Assembly of siRNA into Ago2-Containing RNAi Enzyme Complexes. Cell, 123(4):607–620, 2005. Bibliography 195

[238] M. Matsumoto and T. Seya. TLR3: interferon induction by double-stranded RNA including poly (I: C). Advanced Drug Delivery Reviews, 60(7):805–812, 2008.

[239] I. W. Mattaj, N. A. Dathan, H. D. Parry, P. Carbon, and A. Krol. Changing the RNA polymerase speciﬁcity of U snRNA gene promoters. Cell, 55(3): 435–442, 1988.

[240] C. Mayor, M. Brudno, J. R. Schwartz, A. Poliakov, E. M. Rubin, K. A. Frazer, L. S. Pachter, and I. Dubchak. Vista: visualizing global dna sequence alignments of arbitrary length. Bioinformatics, 16(11):1046–1047, 2000.

[241] K. A. McColl, B. Clarke, and T. J. Doran. Role of genetically engineered animals in future food production. Australian Veterinary Journal, 91(3): 113–117, 2013.

[242] G. J. McIntyre and G. C. Fanning. Design and cloning strategies for con- structing shRNA expression vectors. BMC Biotechnology, 6(1):1, 2006.

[243] G. J. Mcintyre, J. L. Groneman, Y.-H. Yu, A. Tran, and T. L. Apple- gate. Multiple shRNA combinations for near-complete coverage of all HIV-1 strains. AIDS Res Ther, 8(1):1, 2011.

[244] G. J. Mcintyre, Y.-H. Yu, M. Lomas, and G. C. Fanning. The eﬀects of stem length and core placement on shrna activity. BMC molecular biology, 12(1): 34, 2011.

[245] M. F. McLoughlin and D. A. Graham. Alphavirus infections in salmonids – a review. Journal of Fish Diseases, 30(9):511–531, 2007. ISSN 1365-2761.

[246] N. D. Meeker and N. S. Trede. Immunology and zebraﬁsh: Spawning new models of human disease. Developmental and comparative immunology,32 (7):745–757, 2008.

[247] G. Meister, M. Landthaler, A. Patkaniowska, Y. Dorsett, G. Teng, and T. Tuschl. Human argonaute2 mediates RNA cleavage targeted by miR- NAs and siRNAs. Molecular Cell, 15(2):185 – 197, 2004. Bibliography 196

[248] D.-M. Meng, L. Wang, J.-R. Xu, S.-L. Yan, L. Zhou, and Q.-S. Mi. Fabp4- cre-mediated deletion of the miRNA-processing enzyme dicer causes mouse embryonic lethality. Acta diabetologica, pages 1–2, 2013.

[249] C. Meyer, F. Grey, C. N. Kreklywich, T. F. Andoh, R. S. Tirabassi, S. L. Orloﬀ, and D. N. Streblow. Cytomegalovirus microRNA expression is tissue speciﬁc and is associated with persistence. Journal of Virology, 85(1):378– 389, 2011.

[250] Y. Mishima. Widespread roles of micrornas during zebraﬁsh development and beyond. Development, growth & diﬀerentiation, 54(1):55–65, 2012.

[251] A. M. Monteys, R. M. Spengler, J. Wan, L. Tecedor, K. A. Lennox, Y. Xing, and B. L. Davidson. Structure and activity of putative intronic miRNA promoters. Rna, 16(3):495–505, RNA.

[252] F. E. Moore, D. Reyon, J. D. Sander, S. A. Martinez, J. S. Blackburn, C. Khayter, C. L. Ramirez, J. K. Joung, and D. M. Langenau. Improved somatic mutagenesis in zebraﬁsh using transcription activator-like eﬀector nucleases (TALENs). PLoS ONE, 7(5), 2012.

[253] C. Mork, P. Hershberger, R. Kocan, W. Batts, and J. Winton. Isolation and characterization of a rhabdovirus from starry ﬂounder (Platichthys stella- tus) collected from the northern portion of Puget Sound, Washington, USA. Journal of General Virology, 85(2):495–505, 2004.

[254] P. Mourrain, C. B´eclin, T. Elmayan, F. Feuerbach, C. Godon, J.-B. Morel, D. Jouette, A.-M. Lacombe, S. Nikic, N. Picault, et al. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell, 101(5):533–542, 2000.

[255] A. Mukherjee, S. Shrivastava, J. Bhanja Chowdhury, R. Ray, and R. B. Ray. Transcriptional suppression of mir-181c by hepatitis c virus enhances homeobox a1 expression. Journal of Virology, 88(14):7929–7940, 2014. Bibliography 197

[256] J. L. Muñoz-Jordán,G.G.Sánchez-Burgos, M. Laurent-Rolle, and A. Garc´ıa-Sastre. Inhibition of interferon signaling by dengue virus. Proceedings of the National Academy of Sciences, 100(24):14333– 14338, 11 2003. URL http://www.pnas.org/content/100/24/14333. abstractN2-Denguevirusisaworldwide-distributedmosquito-borneflaviviruswitha Itstranscribedpolyproteiniscleavedbyhost-andvirus-encodedpeptidasesinto10pr someofwhichareofunknownfunction.Althoughdenguevirus-infectedcellsseemtobere theviralproductsthatmediatethisresistanceareunknown.Therefore, wehaveanalyzedtheabilityofthe10denguevirus-encodedproteinstoantagonizetheIF WefoundthatexpressioninhumanA549cellsofthedenguevirusnonstructuralproteinsN NS4A,orNS4BenhancesreplicationofanIFN-sensitivevirus. Moreover,expressionofNS4Band,toalesserextent, ofNS2AandNS4Aproteinsresultsindown-regulationofIFN-\protect\ begingroup\immediate\write\@unused\def\MessageBreakΩ\let\ protect\edefYourcommandwasignored.\MessageBreakTypeI toreplaceitwithanothercommand, \MessageBreakortocontinuewithoutit.\errhelp\let\def\ MessageBreakΩ(inputenc)\def\errmessagePackageinputencError: Unicodechar\u8:notsetupforusewithLaTeX. ΩΩSeetheinputencpackagedocumentationforexplanation. ΩTypeHforimmediatehelp\endgroup-stimulatedgeneexpression. CellsexpressingNS4Borinfectedwithdenguevirusdonotexhibitnuclearsignaltransd 1ontreatmentwithIFN-\protect\begingroup\immediate\ write\@unused\def\MessageBreakΩ\let\protect\ edefYourcommandwasignored.\MessageBreakTypeI toreplaceitwithanothercommand, \MessageBreakortocontinuewithoutit.\errhelp\let\def\ MessageBreakΩ(inputenc)\def\errmessagePackageinputencError: Unicodechar\u8:notsetupforusewithLaTeX. ΩΩSeetheinputencpackagedocumentationforexplanation. ΩTypeHforimmediatehelp\endgrouporIFN-\protect\ Bibliography 198

begingroup\immediate\write\@unused\def\MessageBreakΩ\let\ protect\edefYourcommandwasignored.\MessageBreakTypeI toreplaceitwithanothercommand, \MessageBreakortocontinuewithoutit.\errhelp\let\def\ MessageBreakΩ(inputenc)\def\errmessagePackageinputencError: Unicodechar\u8:notsetupforusewithLaTeX. ΩΩSeetheinputencpackagedocumentationforexplanation. ΩTypeHforimmediatehelp\endgroup, indicatingthatNS4BmightbeinvolvedinblockingIFNsignalingduringdenguevirusinfection Thisprotein,encodedbyapositivestrandRNAvirus, isimplicatedasanIFN-signalinginhibitor.

[257] S. Murphy, A. Pierani, C. Scheidereit, M. Melli, and R. G. Roeder. Puri- ﬁed octamer binding transcription factors stimulate RNA polymerase III- mediated transcription of the 7SK RNA gene. Cell, 59(6):1071–1080, 1989.

[258] A. G. Murray and E. J. Peeler. A framework for understanding the potential for emerging diseases in aquaculture. Preventive veterinary medicine, 67(2): 223–235, 2005.

[259] E. Myslinski, A. Krol, and P. Carbon. Optimal tRNA (Ser) Sec gene activity requires an upstream SPH motif. Nucleic Acids Research, 20(2):203–209, 1992.

[260] E. Myslinski, J.-C. Am´e, A. Krol, and P. Carbon. An unusually compact external promoter for RNA polymerase III transcription of the human H1RNA gene. Nucleic Acids Research, 29(12):2502–2509, 2001.

[261] H. Nakano, S. Amemiya, K. Shiokawa, and M. Taira. RNA Interference for the Organizer-Speciﬁc Gene Xlim-1 in Xenopus Embryos. Biochemical and Biophysical Research Communications, 274(2):434–439, 2000.

[262] C. Napoli, C. Lemieux, and R. Jorgensen. Introduction of a chimeric chalcone synthase gene into petunia results in reversible co-suppression of homologous genes in trans. The Plant Cell Online, 2(4):279–289, 1990. Bibliography 199

[263] A. Nasevicius and S. C. Ekker. Eﬀective targeted gene /‘knockdown/’ in zebraﬁsh. Nature Genetics, 26(2):216–220, 2000.

[264] R. L. Naylor, R. J. Goldburg, J. H. Primavera, N. Kautsky, M. C. M. Bev- eridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney, and M. Troell. Eﬀect of aquaculture on world ﬁsh supplies. Nature, 405(6790):1017–1024, 2000.

[265] B. Novoa, A. Romero, V. Mulero, I. Rodr´ıguez, I. Fern´andez, and A. Figueras. Zebraﬁsh (Danio rerio) as a model for the study of vaccination against viral haemorrhagic septicemia virus (VHSV). Vaccine, 24(31–32): 5806–5816, 2006.

[266] B. Novoa, S. Mackenzie, and A. Figueras. Inﬂammation and innate immune response against viral infections in marine ﬁsh. Current pharmaceutical design, 16(38):4175–4184, 2010.

[267] B. Novoa, A. Figueras, J. D. Lambris, and G. Hajishengallis. Zebraﬁsh: Model for the Study of Inﬂammation and the Innate Immune Response to Infectious Diseases, volume 946, pages 253–275. Springer New York, 2012. ISBN 978-1-4614-0106-3.

[268] A. Nyk¨anen, B. Haley, and P. D. Zamore. ATP requirements and small interfering RNA structure in the RNA interference pathway. Cell, 107(3): 309–321, 2001.

[269] A. C. Oates, A. E. Bruce, and R. K. Ho. Too much interference: Injection of double-stranded RNA has nonspecific effects in the zebrafish embryo. Developmental Biology, 224(1):20–28, 2000.

[270] K. Okamura, A. Ishizuka, H. Siomi, and M. C. Siomi. Distinct roles for argonaute proteins in small RNA-directed RNA cleavage pathways. Genes & Development, 18(14):1655–1666, 2004.

[271] M. Olejniczak, P. Galka, and W. J. Krzyzosiak. Sequence-non-speciﬁc effects of rna interference triggers and microRNA regulators. Nucleic Acids Research, 38(1):1–16, 2010. Bibliography 200

[272] N. J. Olesen and H. F. Skall. Viral haemorrhagic septicemia virus. In Manual of Diagnostic Tests for Aquatic Animals, pages 279–298. World Organisation for Animal health (OiE), Paris, France, 2009.

[273] S. Omoto and Y. R. Fujii. Regulation of human immunodeﬁciency virus 1 transcription by nef microRNA. Journal of General Virology, 86(3):751–755, 2005.

[274] A. Orioli, C. Pascali, A. Pagano, M. Teichmann, and G. Dieci. RNA polymerase III transcription control elements: themes and variations. Gene, 493 (2):185–194, 2012.

[275] B. Packham and L. Vasek. New laws will curb super-trawler from ﬁshing in Australian waters for at least two years, 2012.

[276] P. J. Paddison, A. A. Caudy, E. Bernstein, G. J. Hannon, and D. S. Con- klin. Short hairpin RNAs (shRNAs) induce sequence-speciﬁc silencing in mammalian cells. Genes & Development, 16(8):948–958, 2002.

[277] N. Palha, F. Guivel-Benhassine, V. Briolat, G. Lutfalla, M. Sourisseau, F. El- lett, C.-H. Wang, G. J. Lieschke, P. Herbomel, O. Schwartz, and J.-P. Lev- raud. Real-time whole-body visualization of chikungunya virus infection and host interferon response in zebraﬁsh. PLoS Pathogens, 9(9), 2013.

[278] D. M. Parichy, M. R. Elizondo, M. G. Mills, T. N. Gordon, and R. E. Engeszer. Normal table of postembryonic zebraﬁsh development: staging by externally visible anatomy of the living ﬁsh. Developmental Dynamics, 238 (12):2975–3015, 2009.

[279] J. S. Parker, E. A. Parizotto, M. Wang, S. M. Roe, and D. Barford. Enhance- ment of the seed-target recognition step in {RNA} silencing by a piwi/mid domain protein. Molecular Cell, 33(2):204 – 214, 2009.

[280] L. Pase, J. E. Layton, W. P. Kloosterman, D. Carradice, P. M. Waterhouse, and G. J. Lieschke. miR-451 regulates zebraﬁsh erythroid maturation in vivo via its target gata2. Blood, 113(8):1794–1804, 2009. Bibliography 201

[281] D. M. Patrick, C. C. Zhang, Y. Tao, H. Yao, X. Qi, R. J. Schwartz, L. Jun- Shen Huang, and E. N. Olson. Defective erythroid diﬀerentiation in miR-451 mutant mice mediated by 14-3-3 . Genes & Development, 24(15):1614–1619, 2010.

[282] M. R. Paule and R. J. White. Transcription by RNA polymerases I and III. Nucleic Acids Research, 28(6):1283–1298, 2000.

[283] D. Pauly, J. Alder, E. Bennett, V. Christensen, P. Tyedmers, and R. Watson. The future for ﬁsheries. Science, 302(5649):1359–1361, 2003.

[284] I. M. Pedersen, G. Cheng, S. Wieland, S. Volinia, C. M. Croce, F. V. Chisari, and M. David. Interferon modulation of cellular micrornas as an antiviral mechanism. Nature, 449(7164):919–922, 2007.

[285] T. Pelisseir and M. Wassenegger. A DNA target of 30 bp is suﬃcient for RNA-directed DNA methylation. RNA, 6(01):55–65, 2000.

[286] J. T. Perez, A. M. Pham, M. H. Lorini, M. A. Chua, J. Steel, B. R tenOever, et al. MicroRNA-mediated species-speciﬁc attenuation of inﬂuenza A virus. Nature Biotechnology, 27(6):572–576, 2009.

[287] S. P. Persengiev, X. Zhu, and M. R. Green. Nonspeciﬁc, concentration- dependent stimulation and repression of mammalian gene expression by small interfering RNAs (siRNAs). RNA, 10(1):12–18, 2004.

[288] S. Pfeﬀer, M. Zavolan, F. A. Grasser, M. Chien, J. J. Russo, J. Ju, B. John, A. J. Enright, D. Marks, and C. Sander. Identiﬁcation of virus-encoded microRNAs. Science Signaling, 304(5671):734, 2004.

[289] P. E. Phelan, M. E. Pressley, P. E. Witten, M. T. Mellon, S. Blake, and C. H. Kim. Characterization of snakehead rhabdovirus infection in zebraﬁsh (Danio rerio). Journal of Virology, 79(3):1842–1852, 2005.

[290] C. C. Pritchard, H. H. Cheng, and M. Tewari. Microrna proﬁling: approaches and considerations. Nature Reviews Genetics, 13(5):358–369, 2012. Bibliography 202

[291] T. A. Rand, S. Petersen, F. Du, and X. Wang. Argonaute2 cleaves the anti-guide strand of siRNA during RISC activation. Cell, 123(4):621–630, 2005.

[292] R. E. Randall and S. Goodbourn. Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures. Journal of General Virology, 89(1):1–47, 2008.

[293] M. Rao and S. Sockanathan. Molecular mechanisms of RNAi: implications for development and disease. Birth Defects Research Part C: Embryo Today: Reviews, 75(1):28–42, 2005.

[294] F. Ratcliﬀ, B. D. Harrison, and D. C. Baulcombe. A similarity between viral defense and gene silencing in plants. Science, 276(5318):1558–1560, 1997.

[295] J. Reed and H. Muench. A simple method of estimating 50 percent endpoints. American Journal of Hygiene, 27:493–497, 1938.

[296] M. Rehmsmeier, P. Steffen, M. Höchsmann, and R. Giegerich. Fast and effective prediction of microRNA/target duplexes. RNA, 10(10):1507–1517, 2004.

[297] A. Reynolds, E. M. Anderson, A. Vermeulen, Y. Fedorov, K. Robinson, D. Leake, J. Karpilow, W. S. Marshall, and A. Khvorova. Induction of the interferon response by siRNA is cell type–and duplex length–dependent. RNA, 12(6):988–993, 2006.

[298] G. D. Rienzo, J. H. Gutzman, and H. Sive. Efficient shRNA-mediated inhibition of gene expression in zebrafish. Zebrafish, 2012.

[299] S. Ro, C. Park, D. Young, K. M. Sanders, and W. Yan. Tissue-dependent paired expression of miRNAs. Nucleic Acids Research, 35(17):5944–5953, 2007.

[300] B. Robertsen, V. Bergan, T. Røkenes, R. Larsen, and A. Albuquerque. At- lantic salmon interferon genes: cloning, sequence analysis, expression, and Bibliography 203

biological activity. Journal of Interferon and Cytokine Research, 23(10): 601–612, 2003.

[301] M. E. Robu, J. D. Larson, A. Nasevicius, S. Beiraghi, C. Brenner, S. A. Farber, and S. C. Ekker. p53 activation by knockdown technologies. PLoS Genetics, 3(5):e78, 2007.

[302] R. Roelz, I. H. Pilz, M. Mutschler, and H. L. Pahl. Of mice and men: Human RNA polymerase III promoter U6 is more eﬃcient than its murine homologue for shRNA expression from a lentiviral vector in both human and murine progenitor cells. Experimental Hematology, 38(9):792–797, 2010.

[303] N. Romano and G. Macino. Quelling: transient inactivation of gene expression in Neurospora crassa by transformation with homologous sequences. Molecular Microbiology, 6(22):3343–3353, 2006.

[304] J. J. Rossi. RNAi and the P-body connection. Nature Cell Biology, 7(7): 643–644, 2005.

[305] W. E. Rudzinski and T. M. Aminabhavi. Chitosan as a carrier for targeted delivery of small interfering RNA. International Journal of Pharmaceutics, 399(1):1–11, 2010.

[306] S. Ruiz, C. Tafalla, A. Cuesta, A. Estepa, and J. M. Coll. In vitro search for alternative promoters to the human immediate early cytomegalovirus (IE- CMV) to express the G gene of viral haemorrhagic septicemia virus (VHSV) in ﬁsh epithelial cells. Vaccine, 26(51):6620–6629, 2008.

[307] S. Ruiz, B. D. Schyth, P. Encinas, C. Tafalla, A. Estepa, N. Lorenzen, and J. M. Coll. New tools to study RNA interference to ﬁsh viruses: Fish cell lines permanently expressing siRNAs targeting the viral polymerase of viral hemorrhagic septicemia virus. Antiviral Research, 82(3):148–156, 2009.

[308] C. L. Sadowski, R. W. Henry, S. M. Lobo, and N. Hernandez. Targeting TBP to a non-TATA box cis-regulatory element: a TBP-containing complex Bibliography 204

activates transcription from snRNA promoters through the PSE. Genes & Development, 7(8):1535–1548, 1993.

[309] M. A. Samols, J. Hu, R. L. Skalsky, and R. Renne. Cloning and identiﬁ- cation of a microRNA cluster within the latency-associated region of Ka- posi’s sarcoma-associated herpesvirus. Journal of Virology, 79(14):9301– 9305, 2005.

[310] J. D. Sander and J. K. Joung. Crispr-cas systems for editing, regulating and targeting genomes. Nature Biotechnology, 32(4):347–355, 04 2014.

[311] C. Sandrock, T. Kelly, et al. Clinical review: update of avian inﬂuenza a infections in humans. Critical Care, 11(2):209, 2007.

[312] M. Sarasin-Filipowicz, J. Krol, I. Markiewicz, M. H. Heim, and W. Filipow- icz. Decreased levels of microRNA miR-122 in individuals with hepatitis C responding poorly to interferon therapy. Nature Medicine, 15(1):31–33, 2009.

[313] M. Schaub, E. Myslinski, C. Schuster, A. Krol, and P. Carbon. Staf, a promiscuous activator for enhanced transcription by RNA polymerases II and III. The EMBO journal, 16(1):173–181, 1997.

[314] M. Scherr and M. Eder. Gene silencing by small regulatory RNAs in mammalian cells. Cell Cycle, 6(4):444–449, 2007.

[315] T. D. Schmittgen and K. J. Livak. Analyzing real-time PCR data by the comparative CT method. Nature Protocols, 3(6):1101–1108, 2008.

[316] L. Schramm and N. Hernandez. Recruitment of RNA polymerase III to its target promoters. Genes & Development, 16(20):2593–2620, 2002.

[317] L. Schramm, P. S. Pendergrast, Y. Sun, and N. Hernandez. Diﬀerent human TFIIIB activities direct RNA polymerase III transcription from TATA- containing and TATA-less promoters. Genes & Development, 14(20):2650– 2663, 2000. Bibliography 205

[318] C. Schuster, E. Myslinski, A. Krol, and P. Carbon. Staf, a novel zinc ﬁnger protein that activates the RNA polymerase III promoter of the selenocysteine tRNA gene. The EMBO journal, 14(15):3777, 1995.

[319] H. Schütze, P. J. Enzmann, E. Mundt, and T. C. Mettenleiter. Identifica- tion of the non-virion (NV) protein of fish rhabdoviruses viral haemorrhagic septicaemia virus and infectious haematopoietic necrosis virus. Journal of General Virology, 77(6):1259–1263, 1996.

[320] H. Schutze, E. Mundt, and T. C. Mettenleiter. Complete Genomic Sequence of Viral Hemorrhagic Septicemia Virus, a Fish Rhabdovirus. Virus Genes, 19(1):59–65, 1999.

[321] D. S. Schwarz, G. Hutv´agner, T. Du, Z. Xu, N. Aronin, and P. D. Zamore. Asymmetry in the assembly of the RNAi enzyme complex. Cell, 115:199– 208, 2003.

[322] D. S. Schwarz, Y. Tomari, and P. D. Zamore. The rna-induced silencing complex is a mg2+-dependent endonuclease. Current Biology, 14(9):787–91, 2004.

[323] B. D. Schyth. RNAi-mediated gene silencing in ﬁshes? Journal of Fish Biology, 72(8):1890–1906, 2008.

[324] B. D. Schyth, N. Lorenzen, and F. S. Pedersen. Antiviral activity of small interfering RNAs: speciﬁcity testing using heterologous virus reveals interferon-related eﬀects overlooked by conventional mismatch controls. Vi- rology, 349(1):134–141, 2006.

[325] B. D. Schyth, N. Lorenzen, and F. S. Pedersen. A high throughput in vivo model for testing delivery and antiviral eﬀects of siRNAs in vertebrates. Molecular Therapy, 15(7):1366–1372, 2007.

[326] R. J. Seeley, A. Perlmutter, and V. A. Seeley. Inheritance and longevity of infectious pancreatic necrosis virus in the zebraﬁsh, Brachydanio rerio Bibliography 206

(Hamilton-Buchanan). Applied and Environmental Microbiology, 34(1):50– 55, 1977.

[327] G. L. Sen and H. M. Blau. Argonaute 2/RISC resides in sites of mammalian mRNA decay known as cytoplasmic bodies. Nature cell biology, 7(6):633– 636, 2005.

[328] N. Senghaas and R. W. K¨oster. Culturing and Transfecting Zebraﬁsh PAC2 Fibroblast Cells. Cold Spring Harbor Protocols, 2009(6):pdb. prot5235, 2009.

[329] G. J. Seo, R. P. Kincaid, T. Phanaksri, J. M. Burke, J. M. Pare, J. E. Cox, T.-Y. Hsiang, R. M. Krug, and C. S. Sullivan. Reciprocal inhibition between intracellular antiviral signaling and the rnai machinery in mammalian cells. Cell host & microbe, 14(4):435–445, 2013.

[330] Y. Shao, C. Y. Chan, A. Maliyekkel, C. E. Lawrence, I. B. Roninson, and Y. Ding. Eﬀect of target secondary structure on RNAi eﬃciency. RNA,13 (10):1631–1640, 2007.

[331] J. S. Shapiro, S. Schmid, L. C. Aguado, L. R. Sabin, A. Yasunaga, J. V. Shim, D. Sachs, S. Cherry, et al. Drosha as an interferon-independent antiviral factor. Proceedings of the National Academy of Sciences, 111(19):7108–7113, 2014.

[332] R. Shi and V. Chiang. Facile means for quantifying microRNA expression by real-time PCR. BioTechniques, 39(4):519–525, 2005.

[333] M. Shinya, K. Kobayashi, A. Masuda, M. Tokumoto, Y. Ozaki, K. Saito, T. Kawasaki, Y. Sado, and N. Sakai. Properties of gene knockdown system by vector-based sirna in zebraﬁsh. Development, growth & diﬀerentiation, 55(9):755–765, 2013.

[334] D. Siolas, C. Lerner, J. Burchard, W. Ge, P. S. Linsley, P. J. Paddison, G. J. Hannon, and M. A. Cleary. Synthetic shRNAs as potent RNAi triggers. Nature Biotechnology, 23(2):227–231, 2004. Bibliography 207

[335] R. W. Siu, R. Fragkoudis, P. Simmonds, C. L. Donald, M. E. Chase-Topping, G. Barry, G. Attarzadeh-Yazdi, J. Rodriguez-Andres, A. A. Nash, A. Mer- its, et al. Antiviral RNA interference responses induced by Semliki Forest virus infection of mosquito cells: characterization, origin, and frequency- dependent functions of virus-derived small interfering RNAs. Journal of Virology, 85(6):2907–2917, 2011.

[336] H. F. Skall, N. J. Olesen, and S. Mellergaard. Viral haemorrhagic septicaemia virus in marine ﬁsh and its implications for ﬁsh farming – a review. Journal of Fish Diseases, 28(9):509–529, 2005.

[337] H. F. Skall, N. J. Olesen, and S. Mellergaard. Prevalence of viral haemorrhagic septicaemia virus in Danish marine ﬁshes and its occurrence in new host species. Diseases of Aquatic Organisms, 66(2):145–152, 2005.

[338] C. A. Sledz, M. Holko, M. J. De Veer, R. H. Silverman, and B. R. Williams. Activation of the interferon system by short-interfering RNAs. Nature Cell Biology, 5(9):834–839, 2003.

[339] A. R. Soares, P. M. Pereira, B. Santos, C. Egas, A. C. Gomes, J. Arrais, J. L. Oliveira, G. R. Moura, and M. A. Santos. Parallel DNA pyrosequencing unveils new zebraﬁsh microRNAs. BMC Genomics, 10(1):195, 2009.

[340] J. Sprague, L. Bayraktaroglu, D. Clements, T. Conlin, D. Fashena, K. Frazer, M. Haendel, D. G. Howe, P. Mani, S. Ramachandran, K. Schaper, E. Segerdell, P. Song, B. Sprunger, S. Taylor, C. E. Van Slyke, and M. West- erfield. The zebrafish information network: the zebrafish model organism database. Nucleic Acids Research, 34:581–585, 2006.

[341] D. Stein, E. FOSTER, S.-B. HUANG, D. WELLER, and J. SUMMERTON. A speciﬁcity comparison of four antisense types: morpholino, 2-O-methyl RNA, DNA, and phosphorothioate DNA. Antisense and Nucleic Acid Drug Development, 7(3):151–157, 1997. Bibliography 208

[342] N. Suh, L. Baehner, F. Moltzahn, C. Melton, A. Shenoy, J. Chen, and R. Blelloch. MicroRNA function is globally suppressed in mouse oocytes and early embryos. Current Biology, 20(3):271–277, 2010.

[343] D. A. Suhy, S.-C. Kao, T. Mao, L. Whiteley, H. Denise, B. Souberbielle, A. D. Burdick, K. Hayes, J. F. Wright, H. Lavender, et al. Safe, long- term hepatic expression of anti-hcv shrna in a nonhuman primate model. Molecular Therapy, 20(9), 2012.

[344] C. Sullivan and C. H. Kim. Zebraﬁsh as a model for infectious disease and immune function. Fish and Shellﬁsh Immunology, 25(4), 2008.

[345] C. S. Sullivan, A. T. Grundhoﬀ, S. Tevethia, J. M. Pipas, and D. Ganem. SV40-encoded microRNAs regulate viral gene expression and reduce susceptibility to cytotoxic T cells. Nature, 435(7042):682–686, 2005.

[346] J. Summerton. Morpholino antisense oligomers: the case for an RNase H- independent structural type. Biochimica et Biophysica Acta (BBA) - Gene Structure and Expression, 1489(1), 1999.

[347] C. Sun, L. Hu, S. Liu, G. Hu, and S. Zhang. Antiviral activity of phosvitin from zebraﬁsh Danio rerio. Developmental and comparative immunology, 2013.

[348] Y. Suzuki, T. Ito, T. Suzuki, R. E. Holland, T. M. Chambers, M. Kiso, H. Ishida, and Y. Kawaoka. Sialic acid species as a determinant of the host range of inﬂuenza a viruses. Journal of virology, 74(24):11825–11831, 2000.

[349] H. Tabara, M. Sarkissian, W. G. Kelly, J. Fleenor, A. Grishok, L. Timmons, A. Fire, and C. C. Mello. The Gene, RNA Interference, and Transposon Silencing in C. elegans. Cell, 99(2):123–132, 1999.

[350] I. Takami, S. R. Kwon, T. Nishizawa, and M. Yoshimizu. Protection of japanese ﬂounder paralichthys olivaceus from viral hemorrhagic septicemia (vhs) by poly (i: C) immunization. Diseases of aquatic organisms, 89(2): 109–115, 2010. Bibliography 209

[351] G. Tang. siRNA and miRNA: an insight into RISCs. Trends in Biochemical Sciences, 30(2), 2005.

[352] M. Tanguy and E. A. Miska. Antiviral rna interference in animals: piecing together the evidence. Nature Structural & Molecular Biology, 20(11):1239– 1241, 2013.

[353] D. Taxman, L. Livingstone, J. Zhang, B. Conti, H. Iocca, K. Williams, J. Lich, J. Ting, and W. Reed. Criteria for eﬀective design, construction, and gene knockdown by shRNA vectors. BMC Biotechnology, 6(1):7, 2006. ISSN 1472-6750.

[354] D. J. Taxman, L. S. Lambeth, and C. A. Smith. Methods in Molecular Biology, volume 942, pages 205–232. Humana Press, 2013. ISBN 978-1- 62703-118-9.

[355] N. G. H. Taylor, P. F. Dixon, K. R. Jeﬀery, E. J. Peeler, K. L. Denham, and K. Way. Koi herpesvirus: distribution and prospects for control in England and Wales. Journal of Fish Diseases, 33(3), 2010.

[356] M. Teichmann, Z. Wang, and R. G. Roeder. A stable complex of a novel transcription factor IIB-related factor, human TFIIIB50, and associated proteins mediate selective transcription by RNA polymerase III of genes with upstream promoter elements. Proceedings of the National Academy of Sci- ences, 97(26), 2000.

[357] M.-I. Thoulouze, E. Bouguyon, C. Carpentier, and M. Br´emont. Essential Role of the NV Protein of Novirhabdovirus for Pathogenicity in Rainbow Trout. Journal of Virology, 78(8):4098–4107, 2004.

[358] S. Tong, Y. Li, P. Rivailler, C. Conrardy, D. A. A. Castillo, L.-M. Chen, S. Recuenco, J. A. Ellison, C. T. Davis, I. A. York, et al. A distinct lineage of inﬂuenza a virus from bats. Proceedings of the National Academy of Sciences, 109(11), 2012. Bibliography 210

[359] S. Tsuchiya, M. Oku, Y. Imanaka, R. Kunimoto, Y. Okuno, K. Terasawa, F. Sato, G. Tsujimoto, and K. Shimizu. MicroRNA-338-3p and microRNA- 451 contribute to the formation of basolateral polarity in epithelial cells. Nucleic Acids Research, 37, 2009.

[360] J. L. Umbach and B. R. Cullen. The role of RNAi and microRNAs in animal virus replication and antiviral immunity. Genes & Development,23 (10), 2009.

[361] A. Urasaki, G. Morvan, and K. Kawakami. Functional dissection of the tol2 transposable element identiﬁed the minimal cis-sequence and a highly repetitive sequence in the subterminal region essential for transposition. Genetics, 174(2):639–649, 2006.

[362] D. Vallone, C. Santoriello, S. B. Gondi, and N. S. Foulkes. Basic Protocols for Zebraﬁsh Cell Lines. Circadian Rhythms: Methods and Protocols, 362, 2007.

[363] A. van den Berg, J. Mols, and J. Han. RISC-target interaction: cleavage and translational suppression. Biochimica et biophysica acta, 1779(11):668, 2008.

[364] M. van Gestel, S. van Erp, L. Sanders, M. Brans, M. Luijendijk, M. Merkestein, R. Pasterkamp, and R. Adan. shrna-induced saturation of the microrna pathway in the rat brain. Gene therapy, 2014.

[365] J. T. van Mierlo, A. W. Bronkhorst, G. J. Overheul, S. A. Sadanandan, J.- O. Ekstr¨om, M. Heestermans, D. Hultmark, C. Antoniewski, and R. P. van Rij. Convergent evolution of argonaute-2 slicer antagonism in two distinct insect rna viruses. PLoS Pathogens, 8(8), 2012.

[366] M. Varela, P. Diaz-Rosales, P. Pereiro, G. Forn-Cun´ı, M. M. Costa, S. Dios, A. Romero, A. Figueras, and B. Novoa. Interferon-induced genes of the expanded iﬁt family show conserved antiviral activities in non-mammalian species. PLoS ONE, 9(6), 2014. Bibliography 211

[367] T. A. Vickers, W. F. Lima, J. G. Nichols, and S. T. Crooke. Reduced levels of Ago2 expression result in increased siRNA competition in mammalian cells. Nucleic Acids Research, 35(19), 2007.

[368] O. Voinnet. Induction and suppression of RNA silencing: insights from viral infections. Nature Reviews Genetics, 6(3):206–220, 2005.

[369] L. Wang, W. L., H.-X. Zhang, J.-H. Zhang, W.-H. Chen, X.-F. Ruan, and C. Xia. In vitro eﬀects of recombinant zebraﬁsh IFN on spring viremia of carp virus and infectious hematopoietic necrosis virus. Journal of Interferon and Cytokine Research, 26(4), 2006.

[370] L. Wang, J.-y. Zhou, J.-h. Yao, D.-r. Lu, X.-j. Qiao, and W. Jia. U6 promoter-driven siRNA injection has nonspecific effects in zebrafish. Bio- chemical and Biophysical Research Communications, 391(3):1363–1368, 2010.

[371] A. Wargelius, S. Ellingsen, and A. Fjose. Double–Stranded RNA Induces Speciﬁc Developmental Defects in Zebraﬁsh Embryos. Biochemical and Bio- physical Research Communications, 263(1), 1999.

[372] T. Watanabe, H. Hatakeyama, C. Matsuda-Yasui, Y. Sato, M. Sudoh, A. Takagi, Y. Hirata, T. Ohtsuki, M. Arai, K. Inoue, et al. In vivo therapeutic potential of Dicer-hunting siRNAs targeting infectious hepatitis C virus. Scientiﬁc Reports, 4, 2014.

[373] P. M. Waterhouse, M. W. Graham, and M.-B. Wang. Virus resistance and gene silencing in plants can be induced by simultaneous expression of sense and antisense RNA. Proceedings of the National Academy of Sciences,95 (23), 1998.

[374] J.-X. Wei, J. Yang, J.-F. Sun, L.-T. Jia, Y. Zhang, H.-Z. Zhang, X. Li, Y.-L. Meng, L.-B. Yao, and A.-G. Yang. Both Strands of siRNA Have Potential to Guide Posttranscriptional Gene Silencing in Mammalian Cells. PLoS ONE, 4(4), 2009. Bibliography 212

[375] E. M. Westerhout and B. Berkhout. A systematic analysis of the eﬀect of target RNA structure on RNA interference. Nucleic Acids Research, 35(13): 4322–4330, 2007.

[376] E. M. Westerhout, M. Vink, P. J. Haasnoot, A. T. Das, and B. Berkhout. A conditionally replicating hiv-based vector that stably expresses an antiviral shrna against hiv-1 replication. Molecular Therapy, 14(2):268–275, 2006.

[377] R. J. White. Transcription by rna polymerase iii: more complex than we thought. Nature Reviews Genetics, 12(7):459–463, 2011.

[378] E. Wienholds, M. J. Koudijs, F. J. M. van Eeden, E. Cuppen, and R. H. A. Plasterk. The microRNA-producing enzyme Dicer1 is essential for zebraﬁsh development. Nature Genetics, 35(3):217–218, 2003.

[379] J. L. Wiesen and T. B. Tomasi. Dicer is regulated by cellular stresses and interferons. Molecular immunology, 46(6):1222–1228, 2009.

[380] B. Wightman, I. Ha, and G. Ruvkun. Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell, 75(5), 1993.

[381] C. E. Willett, A. Cortes, A. Zuasti, and A. G. Zapata. Early hematopoiesis and developing lymphoid organs in the zebraﬁsh. Developmental Dynamics, 214(4), 1999.

[382] J. Winton, W. Batts, P. DeKinkelin, M. LeBerre, M. Bremont, and N. Fijan. Current lineages of the epithelioma papulosum cyprini (EPC) cell line are contaminated with fathead minnow, Pimephales promelas, cells. Journal of Fish Diseases, 33(8):701–704, 2010.

[383] T. G. Wise, D. J. Schafer, L. S. Lambeth, S. G. Tyack, M. P. Bruce, R. J. Moore, and T. J. Doran. Characterization and comparison of chicken U6 promoters for the expression of short hairpin RNAs. Animal Biotechnology, 18(3), 2007. Bibliography 213

[384] K. Wolf, M. Gravell, and R. G. Malsberger. Lymphocystis virus: isolation and propagation in centrarchid ﬁsh cell lines. Science, 151(3713), 1966.

[385] X. Xia, H. Zhou, and Z. Xu. Multiple shRNAs expressed by an inducible pol II promoter can knock down the expression of multiple target genes. Biotechniques, 41(1):64, 2006.

[386] X. G. Xia, H. Zhou, H. Ding, E. B. Aﬀar, Y. Shi, and Z. Xu. An enhanced u6 promoter for synthesis of short hairpin RNA. Nucleic Acids Research,31 (17), 2003.

[387] J. Xie, L. L¨u, M. Deng, S. Weng, J. Zhu, Y. Wu, L. Gan, S.-M. Chan, and J. He. Inhibition of reporter gene and Iridovirus-tiger frog virus in ﬁsh cell by RNA interference. Virology, 338(1), 2005.

[388] X. Xu, L. Zhang, S. Weng, Z. Huang, J. Lu, D. Lan, X. Zhong, X. Yu, A. Xu, and J. He. A zebraﬁsh (Danio rerio) model of infectious spleen and kidney necrosis virus (ISKNV) infection. Virology, 376(1), 2008.

[389] J.-S. Yang, T. Maurin, N. Robine, K. D. Rasmussen, K. L. Jeﬀrey, R. Chand- wani, E. P. Papapetrou, M. Sadelain, D. O’Carroll, and E. C. Lai. Con- served vertebrate mir-451 provides a platform for Dicer-independent, Ago2- mediated microRNA biogenesis. Proceedings of the National Academy of Sciences, 107(34), 2010.

[390] R. YI. Overexpression of Exportin 5 enhances RNA interference mediated by short hairpin RNAs and microRNAs. RNA, 11(2), 2005.

[391] R. Yi, Y. Qin, I. G. Macara, and B. R. Cullen. Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes & Devel- opment, 17(24):3011–3016, 2003.

[392] S.-Y. Ying and S.-L. Lin. Current perspectives in intronic micro RNAs (miRNAs). Journal of biomedical science, 13(1), 2006. Bibliography 214

[393] J.-Y. Yu, S. L. DeRuiter, and D. L. Turner. RNA interference by expression of short-interfering RNAs and hairpin RNAs in mammalian cells. Proceedings of the National Academy of Sciences, 99(9), 2002.

[394] R. A. Zambon, V. N. Vakharia, and L. P. Wu. RNAi is an antiviral immune response against a dsRNA virus in Drosophila melanogaster. Cellular Microbiology, 8(5), 2006.

[395] P. D. Zamore and B. Haley. Ribo-gnome: the big world of small RNAs. Science Signaling, 309(5740):1519, 2005.

[396] P. D. Zamore, T. Tuschl, P. A. Sharp, and D. P. Bartel. Rnai: double- stranded rna directs the atp-dependent cleavage of mrna at 21 to 23 nucleotide intervals. Cell, 101(1):25–33, 2000.

[397] A. Zapata, B. Diez, T. Cejalvo, C. Gutierrez-de Frias, and A. Cortes. On- togeny of the immune system of ﬁsh. Fish & Shellﬁsh Immunology, 20(2): 126–136, 2006.

[398] H. M. Zbikowska. Fish can be ﬁrst – advances in ﬁsh transgenesis for commercial applications. Transgenic Research, 12:379–389, 2003.

[399] Y. Zeng and B. R. Cullen. Sequence requirements for microRNA processing and function in human cells. RNA, 9(1), 2003.

[400] Y. Zeng, X. Cai, and B. R. Cullen. Use of RNA polymerase II to transcribe artiﬁcial microRNAs. Methods in Enzymology, 392, 2005.

[401] K. Zenke and K. H. Kim. Novel fugu U6 promoter driven shRNA expression vector for eﬃcient vector based RNAi in ﬁsh cell lines. Biochemical and Biophysical Research Communications, 371(3), 2008.

[402] Y. B. Zhang and J. F. Gui. Molecular regulation of interferon antiviral response in ﬁsh. Developmental and comparative immunology, 38(2):193– 202, 2012. Bibliography 215

[403] X.-F. Zhao, A. Fjose, N. Larsen, J. V. Helvik, and Ø. Drivenes. Treat- ment with small interfering RNA affects the microRNA pathway and causes unspecific defects in zebrafish embryos. FEBS Journal, 275(9), 2008.

[404] Z. Zhao, Y. Cao, M. Li, and A. Meng. Double-stranded RNA injection produces nonspeciﬁc defects in zebraﬁsh. Developmental Biology, 229(1): 215–223, 2001.

[405] M. Zuker. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Research, 31(13):3406–3415, 2003.

[406] L. O. Zwillenberg, M. H. Jensen, and H. L. Zwillenberg. Electron microscopy of the virus of viral haemorrhagic septicaemia of rainbow trout (Egtved virus). Archiv f Virusforschung, 17(1):1–19, 1965.

Appendix A

Synthetic Oligonuclotides used in this study

Name Sequence PCR zU6-1 F GAT TCC AAA TGG TGC ATG AG zU6-1 R AGA TCA GCA GTC AGG CTC AG zU6-2 F CTC GGT CAT TCA GTA AGA CG zU6-2 R GAA CCA AGA GCT GGA GGG AG zU6-3 F GTT TGA TGG GCC GTT GCT G zU6-3 R GAA CTA GGA GCC TGG AGG AC zU6-4 F GCT CTG CTC TAG TGA GAG CAG zU6-4 R GAG CTG GGA GTC TGG AGG AC zH1 F TCC TGC TAG GAC AGC GAG TG zH1 R CTG TTC ATG AGC GCT ACG

qPCR β-actin SYBR F CGA GCA GGA GAT GGG AAC C β-actin SYBR R CAA CGG AAA CGC TCA TTG C IFN F GAA TGG CTT GGC CGA TAC AGG ATA IFN R GAT TGG CTT GGC CGA TAC AGG ATA

217 Appendix A. Synthetic oligonuclotides used in this study 218

VHSV G Gene R GGG TTG GGT CGC CAT GTT TC VHSV G Gene F CCT CTG TCC GAC CTT GGT AG Ago2 F CGC ATG TGT TAC GTG GCA TC Ago2 R AAA GTG CTG GAC CAT GTG CT Xpo5 F CAG CCA GTT TCT GAG GTC GT Xpo5 R GTA CCT AGT GCG AGT ACG GC U6 qPCR snRNA R AGA TGG AAC GCT TCA CGA AT U6 qPCR snRNA F TTG CTT CGG CAG CAC ATA H1 qPCR snRNA F GAG GTC ACC CAC TGA GTG CT H1 qPCR snRNA R GTT GAT CCG CTG AAT GTG TG MCP 138 F CGG CGT CAA GGA CTT TAT G MCP 407 R GCG TTG TCT TCG GGA ATA C MCP 952 F ATG CCC ACC TTC ACC TAC MCP 1515 R CTG TTC GGT GGA CAA TGC MCP 2857 F GTA CAG AGC CAC CTC TTG MCP 3597 R CGT CTT GAG GAG CTG TTC DNP-1851 F CAC CTT CGG CCT GGA CTA TGT G DNP-2272 R TGC TGG CAG CGG GTA TCT TC DNP-3161 F CCA GCA CCT TCG GCA ACA TC DNP-3526 R TCG CCG ACC AGT ACC TGA AC DNP-4374 F CGA CTT TGA GGA GAC GTT TG DNP-3121 F ATG GTG TAC GCT TGC TTC DNP-3572 R AGG CCA GAG TAC TGG ATG DNP-4717 R TCC AGT CCA CCT TCT GTA TG

shRNA Construction zU6-1shEGFP TCT AGA TTC CAA AAA AGC TGA CCC TGA AGT TCA TCT CTC TTG AAG ATG AAC GTC AGG GTC AGC AGA TCA GCA GTC AGG CTC AG zU6-2shEGFP TCT AGA TTC CAA AAA AGC TGA CCC TGA Appendix A. Synthetic oligonuclotides used in this study 219

AGT TCA TCT CTC TTG AAG ATG AAC GTC AGG GTC AGC GAA CCA AGA GCT GGA GGG AG zU6-3shEGFP TCT AGA TTC CAA AAA AGC TGA CCC TGA AGT TCA TCT CTC TTG AAG ATG AAC GTC AGG GTC AGC GAA CTA GGA GCC TGG AGG AC zU6-4 shEGFP TCT AGA TTC CAA AAA AGC TGA CCC TGA AGT TCA TCT CTC TTG AAG ATG AAC GTC AGG GTC AGC GAG CTG GGA GTC TGG AGG AC zH1 shEGFP TCT AGA TTC CAA AAA AGC TGA CCC TGA AGT TCA TCT CTC TTG AAG ATG AAC GTC AGG GTC AGC GGG TTA TGA CGT AGT CG TD391 TCT AGA TTC CTT TTT TCC GAA ATG TTC GTC ACC AAA TTC TCT TGA AAT TTG GTG ACG AAC ATT TCG GGA ACC AAG AGC TGG AGG GAG TD392 TCT AGA TTC CTT TTT TGA ACA AAT TGG CCA AGA GAC TTC TCT TGA ATC AGA GAA CCG GTT AAA CAA GGA ACC AAG AGC TGG AGG GAG TD393 TCT AGA TTC CTT TTT TGG CAA TTC AAG AGG TCA GTG ATC TCT TGA ATC ACT GAC CTC TTG AAT TGC CGA ACC AAG AGC TGG AGG GAG zH1-shVHSVG906 CTC GAG TTC CAA AAA AAC AAC CAT CTC ATC ACC AAC ATT CTC TTG AAA TGT TGG TGA TGA GAT GGT TGG GGT TA TGA CGT AGT CTG CG Appendix A. Synthetic oligonuclotides used in this study 220

pEGFP Intron F GTT ATC CCC TGA TTC TGT GG pEGFP Intron R AAA ACT AGT ATA TTA ACG CTT ACA ATT TAC GCC zH1-shVHSVG906 F GTC GAC CAT CAA CTA CCC AAT CAC TGC Xpo5 SP6 F TCT AGA ATT TAG GTG ACA CTA TAG ATG GCG ATG GA Xpo5 SP6 R GAA TTC TCA GGG TTC AAA GAT GGT GGC SpeI U62 F AAA AAA CTA GTC TCG GTC ATT CAG TAA GA SpeI H1 F AAA AAC TAG TCA TCA ACT ACC CAA TCA CTG C M13 F CGC CAG GGT TTT CCC AGT CAC GAC M13 R TCA CAC AGG AAA CAG CTA TGA C H1 shScramble TCT AGA TTC CAA AAA AAC CAA CCA CTA CTT AAC CA U6 shScramble TCT AGA TTC CTT TTT ACC AAC CAC H1-Oct F AAA ATC TAG ACT TCC CTT AAT GTT CGA CCA C H1-STAF AAA TCT AGA GAT TTG TGT ACA ATG CGT TAA ATA AC U62 OCT d AAA TCT AGA GCT TTA CAG TTT GAA AAA TAC CAC AG VHSV F624 ACC TCG CCC TGT CAA ACT CAT VHSV R624 ATT GCC TTG ACC ACC CTG TGA VHSV F1018 CTC ATT TCT CCT CTC AAA GTT TCG VHSV R1018 CCG TCT GTG TTG TTG TCT ACC shVHSVG2 F BamHI GAA TGG ATC CTC CCT CCA GCT CTT GGT TC shVHSVG2 R KpnI GAT AGG TAC CCG TTG GGA GCT CTC CCA TA U62shVHSVG2 XhoI AAA ACT CGA GAT GTT CTG CGC ATG CCT GTG Appendix A. Synthetic oligonuclotides used in this study 221

U62 shVHSVG2 R TGC TTC CGG CTC GTA TGTTG

RNA Probes LL91 GAU GAA CUU CAG GGU CAG C mir-16 AUC GUC GUC CAU UUA UAA CCG CGG ACA GAG GG

Appendix B

Plasmids used or generated in this study

223 Plasmids used or generated in this study 224 Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin (pGEM) (pGEM) (pGEM) (pGEM) (pGEM) (pGEM) (pGEM) (Promega) :GFP:GFP-U62:shG0906 EF1a GFP Tol 2 and shRNA Expression Ampicillin EF1a GFP Tol 2 and shRNA Expression Ampicillin α α NamepEGFP-N1 CMV-EGFP Expression Description Kanamycin Selection pGEM-T A-T Cloning Vector pUC118pZFU6 1-shEGFP shEGFP Expression Cloning Vector Ampicillin pZFU6 2-shEGFP shEGFP Expression pZFU6 3-shEGFP shEGFP Expression pZFU6 4v1-shEGFP shEGFP Expression pZFU6 4v2-shEGFP shEGFP Expression pZFH1-shEGFP shEGFP Expression pFugu-shEGFP shEGFP Expression pT2-002 CAAGS Transposase Expression Ampicillin pTPase mRNA SP6 Transposase Ampicillin pMiniTolpMTol BC1pMTol BC2 Mini Tol 2 Cloning Mini Vector Tol 2 Cloning Vector + SV40 EF1a Poly GFP A Tol 2 Ampicillin Ampicillin Ampicillin pTol2-EF1 pTol2-EF1 Plasmids used or generated in this study 225 :GFP-H1:shG0906:GFP-U62:shScramble:GFP-U62:shG1207 EF1a EF1a GFP GFP Tol Tol 2:GFP-H1:shG1207 2 and and shRNA shRNA Expression Expression:GFP-H1:shScramble EF1a GFP Ampicillin Ampicillin Tol 2 and:GFP-U62-Half:shG0906 shRNA Expression EF1a GFP EF1a Tol 2 GFP:GFP-U62-Oct1:shG0906 EF1a and Tol Ampicillin GFP 2 shRNA Tol and Expression 2 shRNA and Expression EF1a shRNA:GFP-U62-SPH:shG0906 GFP Expression Tol 2 Ampicillin and shRNA Ampicillin Expression EF1a:GFP-U62-TATA:shG0906 Ampicillin GFP Tol EF1a 2 GFP and Tol shRNA Ampicillin 2 Expression:GFP-H1-Oct1:shG0906 and shRNA Expression Ampicillin :GFP-H1-Sph:shG0906 EF1a GFP Ampicillin Tol 2 and shRNA:GFP-H1-PSE:shG0906 Expression EF1a GFP Tol 2:GFP-H1-TATA:shG0906 and Ampicillin shRNA EF1a Expression GFP Tol EF1a 2 GFP and Tol shRNA 2 Expression and Ampicillin shRNA Expression Ampicillin Ampicillin α α α α α α α α α α α α α pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1 pTol2-EF1

Appendix C

Accession number of VHS virus glycoprotien and polymerase used in sequence conservation analysis

Glycoprotien Z93431.1 Z93430.1 Z93429.1 Z93428.1 Z93427.1 Z93426.1 Z93425.1 Z93424.1 Z93423.1 Z93422.1 Z93421.1 Z93420.1 Z93419.1 Z93418.1 Z93417.1 Z93416.1 Z93415.1 Z93413.1 Z93411.1 Z93410.1 Z93409.1 Z93408.1 Z93407.1 Z93406.1 Z93405.1 Z93404.1 Z93414.2 AM086383.1 AM086382.1 AM086381.1 AM086380.1 AM086379.1 AM086378.1 AM086377.1 AM086376.1 AM086375.1 AM086374.1 AM086373.1 AM086372.1 AM086371.1 AM086370.1 AM086369.1 AM086368.1 AM086367.1 AM086366.1 AM086365.1 AM086364.1 AM086363.1 AM086362.1 AM086361.1 AM086360.1 AM086359.1 AM086358.1 AM086357.1 AM086356.1 AM086355.1 AM086354.1 Z93412.2 F N665788.1 AB179621.1

227 Appendix C. Accession number of VHS virus glycoprotien and polymerase used in sequence conservation analysis 228

FJ362515.1 AJ233396.1 X73873.1 NC000855.1 Y18263.1 JF792424.1 FJ460590.1 FJ460591.1 AB839745.1 AB839746.1 AB839747.1 AB839748.1 AF143863.1 AF143862.1 KF477302.1 KC778774.1 AB672614.1 FJ362510.1 GQ385941.1 X89213.1 AB490792.1 EU481506.1