1 GENE TRAP RESULTS IN A SIGNIFICANT KNOCK DOWN OF PAX8 EXPRESSION IN ZEBRAFISH, BUT DOES IT AFFECT OTIC VESICLE DEVELOPMENT? By NICHOLE SUZANNE GEBHART A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007
2 2007 Nichole Suzanne Gebhart
3 To my wonderful Mother, Sister, and Family, who have always believed in me and supported me in everything I do; and to my Brother, who always expected the best from me.
4 ACKNOWLEDGMENTS I would like to thank my chair, Dr. Fumihito Ono, and my supervisory committee, Dr. Barbara Battelle and Dr. Peter McGuire, for their time, dedication, mentorship, and friendship. I would also like to thank Dr. Koichi Kawakami for providing the Tol2 Gene Trap vector, T2KSAG, to our lab for our studies. I would like to thank the faculty and staff of the Interdisciplinary Program in Biomedical Sciences at the University of Florida for the education I have received during my graduate studies. Last, but not least, I would like to thank the faculty, staff, and friends of the Whitney Laboratory for Marine Bioscience for their time, mentorship, friendship, and sincere dedication to me and all of their students, pushing us to our fullest potential as young scientists.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...........................................................................................................4 LIST OF FIGURES................................................................................................................ ....7 ABSTRACT....................................................................................................................... ........9 Chapter 1 INTRODUCTION .............................................................................................................11 Zebrafish ( Danio rerio ) as a Model Organism for Development.........................................13 Genetics Tools in Zebrafish Studies...................................................................................14 Morpholino Oligonucleotides Knock Down Protein Expression..................................14 Mutagenesis Methods Knock Out Gene Function........................................................15 Gene Trap Can Knock Down Gene Expression and Label Specific Cell Populations...16 The Pax Gene Family .........................................................................................................18 Zebrafish Ear Development and the Role of Pax8...............................................................21 2 MATERIALS AND METHODS ........................................................................................26 Fish Strains and Husbandry................................................................................................26 Construction and Injection of Tol2 Plasmid........................................................................26 Genomic Characterization of Tol2DsRed Insertion.............................................................27 Identification of Founder Fish.....................................................................................27 Identification of DNA Insertion Site............................................................................27 Identification of Fusion Transcripts of RFP with Endogenous Transcripts...................28 Identification of Homozygous Adults from a Heterozygous Intercross...............................29 Imaging........................................................................................................................ ......30 Timing of Expression..................................................................................................30 Confocal Imaging........................................................................................................30 Differential Interference Co ntrast (DIC) Imaging........................................................31 Quantitation of pax8 Expression.........................................................................................31 3 RESULTS...................................................................................................................... ....38 Gene Trap Fish are Identified by Expression of RFP..........................................................38 Tol2DsRed Inserted into the First Intron of the pax8 Gene.................................................38 Tol2DsRed Insertion Results in Fusion Transcripts of DsRed-Express to pax8 ...................39 RFP Expression in the Tol2DsRed Gene Trap Line of Fish................................................40 Pax8 mRNA Expression is Dramatically Knocked Down in Homozygous Embryos..........43 Knock Down of pax8 Expression May Result in a Subtle Mutant Phenotype Early in Development..................................................................................................................44
6 4 DISCUSSION................................................................................................................... .61 Tol2 Gene Trap Captures Endogenous pax8 Transcripts and Represents Pax8 Expression..................................................................................................................... .61 Gene Trap May Result in Abnormal Ear Development in Homozygous Embryos...............65 Implications of the pax8 gene trap fish line........................................................................67 Conclusion..................................................................................................................... ....68 LIST OF REFERENCES..........................................................................................................70 BIOGRAPHICAL SKETCH.....................................................................................................76
7 LIST OF FIGURES Figure page 1-1 Tol2 transposition...........................................................................................................24 2-1 Tol2DsRed gene trap construct........................................................................................34 2-2 Inverse PCR (IPCR) to amplify regions of trapped gene..................................................35 2-3 PCR to distinguish heterozygous and homozygous adults of the pax8 gene trap line........36 2-4 Measurement of otic vesicle using DIC...........................................................................37 2-5 Primer design for pax8 quantitative PCR (qPCR)............................................................37 3-1 Founder of the Tol2DsRed gene trap line........................................................................47 3-2 Insertion of Tol2DsRed...................................................................................................48 3-3 Amplification of pax8 transcripts fused to DsRed-Express..............................................49 3-4 RFP fluorescence visible by 14.5 hpf in heterozygous pax8 gene trap embryos...............50 3-5 RFP expression in heterozygotes at 21 hpf......................................................................51 3-6 RFP expression in heterozygotes at 29 hpf......................................................................52 3-7 RFP expression in heterozygotes at 2 dpf........................................................................53 3-8 RFP expression in heterozygotes at 3 dpf........................................................................54 3-9 RFP and GFP expression in heterozygous embryos in an Isl1-GFP background at 4 dpf............................................................................................................................ .....54 3-10 RFP expression in homozygotes at 21 hpf......................................................................55 3-11 RFP expression in homozygote at 29 hpf........................................................................56 3-13 Pax8 quantitative PCR (qPCR).......................................................................................58 3-14 Images of otic vesicle at 24 hpf......................................................................................59 3-15 Otic vesicle width at 26 hpf............................................................................................60 4-1 Comparison of RFP fluorescence and pax8 mRNA expression........................................69
8 LIST OF TERMS/SYMBOLS/ABBREVIATIONS cDNA Complementary DNA CNS Central nervous system D. rerio Danio rerio (zebrafish genus and species name) DHPLC Denaturing high performance liquid chromatography DNA Deoxyribonucleic acid dpf Days post fertilization EGFP Enhanced green fluorescent protein Ensembl Ensembl is a joint project between EMBL Â– European Bioinformatics Institute (EBI) and the Welcome Trust Sanger Institute (WTSI) to develop a software system that produces and maintains automatic annotation on selected eukaryotic genomes. Contains information about predicted genes, predicted introns and exons, and sequence location. www.ensembl.org ES cells Embryonic stem cells FGF Fibroblast growth factor proteins GFP Green fluorescent protein hb Hindbrain HOX Homeobox proteins hpf Hours post fertilization IPCR Inverse PCR Isl1-GFP Line of fish with the Islet-1 motoneuron-specific promoter driving expression of GFP. Motoneurons that innervate the dorsal musculature express GFP in the Isl1-GFP line of fish. kb Kilobase mhb Midbrain-hindbrain boundary mRNA Messenger RNA MO Morpholino oligonucleotide; oligonucleotide attached to a morpholino ring designed to bind to splice junctions of mRNA to block translation of the protein product in zebrafish op Otic placode PAX Paired box proteins PCR Polymerase chain reaction qPCR Quantitative PCR RACE Rapid amplification of cDNA ends RFP Red fluorescent protein RNA Ribose nucleic acid; sc Spinal cord SNP Single nucleotide polymorphism TILLING Target induced local lesions in genomics
9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science GENE TRAP RESULTS IN A SIGNIFICANT KNOCK DOWN OF PAX8 EXPRESSION IN ZEBRAFISH, BUT DOES IT AFFECT OTIC VESICLE DEVELOPMENT? By Nichole Suzanne Gebhart December 2007 Chair: Fumihito Ono Major: Medical Sciences Developmental studies are facilitated by methods of gene disruption like mutagenesis and gene knockdown. In zebrafish, a transposon-based system of gene knockdown and cell labeling, the Tol2 gene trap, has been used to create insertional mutants to identify and study gene function. Although this method is effective in creating lines of zebrafish with fluorescently labeled cells, to date no lines have been identified that exhibit a mutant phenotype. The lack of a gene trap line displaying a mutant phenotype raises the question whether the Tol2 gene trap is an effective insertional mutagenesis tool to interfere with gene function and use in developmental studies. Therefore, the goal of the studies described in this thesis was to use the Tol2 gene trap to establish a gene trap line of fish with a knock down gene expression and to identify a mutant phenotype in the gene trap line of fish. The Tol2 gene trap used for these studies contained the DsRed-Express Red Fluorescent Protein (RFP) as a reporter gene; red fluorescence was used to select for embryos carrying the Tol2DsRed insertion. Two gene trap fish were identified with the same RFP expression pattern. These fish exhibited a specific spatial and temporal expression pattern of the RFP during embryonic development. The unique RFP expression suggested the insertion of Tol2DsRed occurred in a developmentally regulated gene. The genomic sequence
10 surrounding the Tol2 insertion was determined, and the trapped gene was identified as the pax8 gene. Pax proteins are highly conserved transcription factors that play crucial roles in development. Members of the Pax protein family exhibit unique, yet overlapping, expression during development. The strict regulation of each Pax protein dictates the transcriptional network of each cell to influence specification, differentiation, growth, and migration of that cell. Embryos of the pax8 gene trap line produce fusion transcripts of pax8 mRNA and DsRedExpress, resulting in similar expression of RFP and pax8 mRNA. This study will show quantitatively that the Tol2 insertion causes a significant knockdown of pax8 expression in homozygous embryos during development. Despite the significant knockdown, the pax8 gene trap fish survive to adulthood and are viable. However, preliminary studies suggest that at one day post fertilization, these fish exhibit otic vesicles that are reduced in size compared to wild type. If confirmed by more detailed studies, this would be the first report that demonstrates the Tol2 gene trap can result in a mutant phenotype, similar to phenotypes produced using other mutagenesis and gene knockdown methods.
11 CHAPTER 1 INTRODUCTION Homeotic transcription factors regulate development through the control of gene expression. The highly conserved genes that code for these transcription factors are called master regulatory genes, and they include the Hox and Pax genes. Hox genes, first characterized in Drosophila are the earliest of the master genes to be expressed and control the segmentation of the developing embryo and the establishment of body axes in symmetrical organisms (Lewis, 1978). Pax genes are expressed slightly after the commencement of Hox gene expression and, along with Hox genes, play roles in brain regionalization and organ formation (Stoykova and Gruss, 1994; reviewed by Dahl et al., 1997 and Mansouri et al., 1998). Hox and Pax proteins control development by the binding of their conserved domains to specific DNA sequences to activate or inhibit gene expression (Bopp et al., 1986; Treisman et al., 1989). Without these domains, the Hox and Pax proteins are unable to recognize, bind, and regulate DNA transcription. The genes expressed in each cell, under the control of Hox and Pax, ultimately determine the cell fate in which they are expressed. Type, location, differentiation, and survival of the cells are all determined by gene expression (Gaunt et al., 1986; Chalepakis et al., 1991). If gene expression is not regulated properly, development of the organism will not progress normally. Improper development can cause structural and functional abnormalities that potentially result in disease. Therefore, gene expression is strictly controlled spatially and temporally by homeotic transcription factors. Although master control genes have been studied extensively, many questions remain. For example, are there additional downstream targets of these transcriptional regulators that have not been identified? What happens when the functions of these genes are compromised and, as a
12 result, are unable to regulate transcription of the downstream targets? How does this compromise normal development? To identify and further understand the roles of genes and transcriptional networks in vertebrate development, methods such as mutagenesis and gene knockdown are used to interfere with gene expression and, ultimately, the function of the gene. The purpose of interfering with expression is to facilitate studies of gene and protein function that are involved in vertebrate development, with the expectation that disruption of normal gene and protein expression will cause developmental abnormalities. Gene trap is an insertional mutagenesis method that is used to identify and potentially knockdown the expression of developmentally regulated genes in model organisms. For example, this method can be used to generate stable lines of transgenic zebrafish with the potential of knocking down gene expression and, consequently, disrupting protein function. The zebrafish Tol2 gene trap system was engineered as a mutagenesis tool with two purposes: to establish lines of zebrafish with fluorescently labeled cells and to establish lines that contain insertional mutations that interfere with gene function (Kawakami et al., 2004). The purpose of labeling cells with fluorescence using a reporter gene is to render specific populations of cells visible in live embryos during cell differentiation and development. Labeled cells, coupled with mutagenesis, produce an embryo that can be used to study the consequences of the missing gene on development in real time in a live zebrafish. The gene trap technique was optimized in zebrafish using the Tol2 transposon (Kawakami et. al., 1998, 1999, 2000, and 2004). The Tol2 gene trap was shown to effectively label tissues with GFP fluorescence in specific cells that express the trapped gene and capture the endogenous transcript of the trapped gene; it was suggested that the insertion resulted in a knock down of
13 gene expression. However, no quantitative evidence was given to support the claim that endogenous gene expression was knocked down as a result of the Tol2 insertion. Further, since the original Tol2 gene trap method was established, no line of fish has been reported to exhibit a mutant phenotype, bringing into question whether this method can serve as a tool to study development through disruption of gene function. We have established a line of zebrafish using the Tol2 gene trap containing the DsRedExpress reporter gene. These fish exhibit a specific spatial and temporal expression pattern of the RFP during embryonic development, suggesting the insertion occurred in a developmentally regulated gene. In the studies described in this thesis, I characterized the gene trap line and outline four main points: 1) I determined the genomic sequence surrounding the Tol2 insertion and identified the trapped gene as the pax8 gene. 2) I showed that embryos of the pax8 Tol2 gene trap line express RFP in a similar manner as previous reports of pax8 mRNA expression. 3) I provided quantitative data to reveal that the Tol2 insertion results in a significant knockdown of pax8 expression in homozygous embryos during development. 4) This is the first report that addresses the issue as to whether the use of Tol2 gene trap can result in a mutant phenotype similar to phenotypes produced using other mutagenesis and gene knockdown methods. Zebrafish ( Danio rerio ) as a Model Organism for Development Developmental processes are conserved among vertebrates, allowing the use of model organisms to further our understanding of the molecular, cellular, and genetic aspects of development. Zebrafish have several advantages as a model organism for embryonic and developmental studies compared to higher vertebrates. Fertilization is external, and the embryos
14 are large enough to enable genetic manipulation of the organism at the beginning stages of development. The clutch size from one breeding pair is large (up to 200 embryos), facilitating large-scale screening for mutagenesis studies. Development takes place ex utero rendering the embryos accessible and visible during the beginning stages of embryogenesis. In other vertebrate models, such as mouse and rat, development is slower and occurs in utero making studies during early embryonic stages difficult. Zebrafish development is rapid, and organ systems are established within a few days. The central nervous system (CNS) and other organ systems of zebrafish are relatively simple compared to higher vertebrates with regards to size and cell number, reducing the complexity associated with studying mammalian models. Additionally, zebrafish embryos are transparent, and cell labeling techniques allow for developing structures to be viewed in live embryos in real time. The embryonic tissues in other vertebrate models such as mouse and rat are opaque, hindering the ability to visualize cells and developing structures in live embryos. In comparison with higher vertebrate models, the simpler organ systems, accessibility of the embryo, fast development, and transparency of embryonic tissue make the zebrafish an ideal organism for developmental studies. Genetics Tools in Zebrafish Studies Morpholino Oligonucleotides Knock Down Protein Expression Gene disruption is an effective method to identify and study the function of genes in development. In zebrafish, morpholino oligonucleotides (MOs) target and knock down the expression of specific genes (Nasevicius and Ekker, 2000). MOs bind to specific regions of mRNA to block mRNA spicing or translation of mRNA into the protein product. However, the action of the MO is transient and ineffective after a few days. Therefore, this method is not ideal for studying gene function for an extended period of time or at a later stage in development. Also, MOs can cause nonspecific effects including cell death, developmental defects, and neural
15 degeneration (reviewed by Heasman, 2002). These nonspecific effects complicate the interpretation of MO-induced phenotypes, making it unclear whether the phenotype is due to the lack of functional protein product or a byproduct of the MO. Other disadvantages are nonspecific binding of MOs to mRNAs that block translation of multiple protein products, the difficulty of delivering the same amount of MO to each embryo (making the phenotype or severity of the phenotype heterogeneous among a group of injected embryos), and the potential to introduce foreign material into the embryo during injection. Mutagenesis Methods Knoc k Out Gene Function Although no technique exists for zebrafish to target and knock out specific genes, random mutagenesis using the chemical mutagen N -ethylN -nitrosourea (ENU) efficiently produces null mutations (Grunwald & Streissinger, 1992). However, it is time consuming and laborious to determine the specific gene that is mutated. TILLING (targeting induced local lesions in genomes) is also a widely used mutagenesis tool in zebrafish. Unlike chemical mutagenesis, TILLING utilizes a DNA methyltransferase isolated from Arabidopsis to provide a large number of mutants (McCallum et al., 2000). While TILLING is highly efficient at generating mutants, it requires the use of DHPLC (denaturing high performance liquid chromatography) to identify the location of the mutations, a technique that is laborious and demands highly specialized equipment. Genes are more easily identified using methods of insertional mutagenesis in which a known sequence of exogenous DNA integrates into the genome of the organism. In this method, mutations are introduced into the genomic DNA using a retroviral vector or transposon. Insertional mutagenesis is also a random approach, but the vector or transposon sequence that inserts into the DNA is known, thus facilitating the identification of the disrupted gene containing the mutation. A disadvantage of using these methods is that the mutation must cause
16 an observable phenotype in the embryo in order for a mutant to be identified. Genetic mutations do not always cause obvious phenotypes, and mutants with subtle phenotypes are often overlooked. To aid in the identification of mutants, insertional mutagenesis techniques utilizing reporter genes have been established. Gene Trap Can Knock Down Gene Expression and Label Specific Cell Populations Gene trap is an insertional mutagenesis method that was originally pioneered in Drosophila as a P element-based enhancer trap (OÂ’Kane and Gehring, 1987; Bellen et al., 1989; Bier et al., 1989; Wilson et al., 1989). The P element-based enhancer trap vector contains a weak promoter gene fused to the reporter gene lacZ LacZ is expressed when the vector inserts near an endogenous enhancer. Flies that express lacZ are identified as insertional mutants. Gene trap methods were also employed in mice before they were used in zebrafish. In mice, a gene trap vector containing a splice acceptor and reporter gene lacking a promoter is introduced into embryonic stem (ES) cells. When the vector inserts within a gene the reporter gene is expressed, and ES cells with trapped genes are identified by expression of the reporter gene (Gossler et al., 1989; Friedrich and Soriano, 1991; Skarnes et al., 1992). A major disadvantage of gene trap in mice is that insertional mutants are identified as ES cells in vitro Because these are cells, the expression of the reporter gene in the cell provides no information as to the tissue specific expression of the trapped gene. Although gene trap methods have existed in zebrafish for some years, these methods did not facilitate genome-wide screening for insertional mutants (Bayer and Campos-Ortega, 1992; Lin et al., 1994). It wasnÂ’t until the discovery of functional transposons in zebrafish that efficient gene trap methods began to emerge (Ivics et al., 1997; Raz et al., 1998; Fadool et al., 1998; Kawakami et al,. 1998). Tol2 was the first vertebrate transposon discovered to be active in the zebrafish germ line (Kawakami et al., 2000). The Tol2 gene trap was engineered by a geneticist as a method to
17 randomly label specific cells and tissues in live zebrafish embryos. This technique utilizes a transposable DNA element, the Tol2 transposon, to insert exogenous DNA into the genome (Fig. 1-1). The sequence for transposase, the enzyme responsible for transposition of the DNA, is removed so that transposition of the Tol2 DNA occurs only when transposase mRNA is coinjected into the embryo. The transposase mRNA degrades soon after injection, rendering its enzymatic action transient. Therefore, transposition of the Tol2 DNA in the genomic DNA is more likely to occur once to trap a single gene. To engineer the Tol2 gene trap Kawakami and others inserted the EGFP sequence into the Tol2 DNA sequence. The EGFP sequence was placed in the reverse orientation in relation to transposase to prevent expression of EGFP by endogenous promoter activity located near the 5Â’ end of the Tol2 element (Fig. 1-2) (Kawakami and Shima, 1999; Kawakami et al., 2004). A gene is trapped when the Tol2 DNA containing the EGFP sequence inserts downstream of a promoter in a gene coding region, and expression of EGFP is driven by the endogenous promoter of the trapped gene. A splice acceptor is located at the beginning of the EGFP sequence so that during mRNA splicing, the EGFP transcript is fused to the exons of the endogenous transcript. GFP is expressed because the EGFP mRNA is fused to the mRNA of the endogenous gene product. A polyA signal is located at the 3Â’ end of the EGFP sequence so that transcription, driven by the promoter of the trapped gene, terminates after the EGFP sequence. The end result is a knockdown of endogenous gene expression downstream of the Tol2 insertion. A start codon sequence is positioned at the beginning of the EGFP sequence so that the EGFP protein will be expressed even if the Tol2 DNA inserts into a gene sequence upstream of the endogenous start codon. A stop codon is located at the end of the EGFP sequence so that translation of the fused EGFP-endogenous transcript ends at that point.
18 Together, the polyA mRNA tail and stop codon at the end of the EGFP sequence function to interfere with the endogenous gene and protein expression. GFP fluorescence is used to identify embryos with a trapped gene in the Tol2 system, even when the embryos lack an obvious phenotype. The ability to identify insertional mutants that do not have an obvious phenotype is a major advantage of the Tol2 gene trap system compared to the other methods of insertional mutagenesis. Similar to other insertional mutagenesis methods, in gene trap a known DNA sequence inserts into the genome, facilitating the identification of the gene sequence surrounding the insertion. Unlike random chemical mutagenesis, primers specific for the 3Â’ and 5Â’ Tol2 sequence are used to amplify the regions of DNA flanking the insertion that contain the endogenous gene sequence. The Tol2 DNA integrates stably into the genome of the founder fish, and the trapped gene is passed on to the next generation. Experiments are then conducted on the progeny and subsequent generations. Through crossing heterozygous adults of the same gene trap line, a mutant containing two copies of the insertion can be produced, thus having no wild type allele of the trapped gene. Although a specific gene is not targeted as in MO-induced knock down of gene expression, gene trap is advantageous because experiments are performed on embryos that are not manipulated early in development, while MO injection requires penetration of the embryo immediately after fertilization. Furthermore, while MO action on gene expression is transient, gene trap knockdown persists through adulthood. The advantages of using gene trap to establish mutant fish lines make it a powerful tool for examining gene function during development. The Pax Gene Family This study focuses on Pax8, a member of the highly conserved paired-box (Pax) DNAbinding, transcriptional proteins. Pax proteins control the expression of genes by binding of their conserved domains to DNA to activate or inhibit transcription. The highly conserved paired
19 domain, first identified in Drosophila as the segmentation gene paired recognizes and binds to highly specific DNA sequences (Fig. 1-3) (Treisman et al., 1991; Xu et al., 1995). Further interaction with the DNA takes place via a partial or whole homeodomain, depending on the Pax protein subclass (Underhill and Gros, 1997). Transcriptional activity of the Pax proteins is also regulated by activating and inhibitory domains located at the C-terminus (Dorfler and Busslinger, 1996). Protein-protein interactions also influence transcription via an octapeptide domain (Lechner & Dressler, 1996). Pax gene regulatory domains have been highly conserved throughout evolution, with homologs of Pax genes present in organisms as diverse as human and ascidian (Holland et al., 1995). The high conservation of Pax protein transcriptional domains is indicative of how vital these genes are in development. Pax gene expression is also tightly controlled and is evident by the unique spatial and temporal expression patterns of each gene in the Pax gene family (Krauss et al., 1991). Studies of Pax gene expression in the embryonic mouse brain showed that the genes exhibited unique, yet overlapping, expression during development (Stoykova and Gruss, 1994). The particular expression patterns correspond with anatomical boundaries of the developing brain, suggesting that the Pax proteins play a role in brain regionalization and establishment of functional boundaries. Specifically, the Pax2/5/8 genes are expressed during neural differentiation. The timing of expression suggests these genes are involved in the generation of specific populations of neurons, migration of these neurons, and the specific connections that are made in the central nervous system. Regulation of Pax proteins and their targets is crucial for normal development of other organ systems as well. They function to influence cell specification, differentiation, growth, survival and migration (Reviews see Gruss & Walther, 1992; Stuart et al., 1994; Mansour et al.,
20 1996; Robson et al., 2006). For example, mutations in the pax3 and pax6 genes result in phenotypes that are due to abnormal cell migration (Edelman and Jones, 1995). Also, loss of function mutations in many of these genes play roles in disease and disorders (reviewed by Dahl et al., 1997). Pax8 is known to be involved in human congenital hypothyroidism (Macchia et al., 1998; Tell et al., 1999) and Pax6 in Aniridia (Glaser et al., 1992). The Pax genes are also expressed in the adult and are associated with specific cancers, such as Pax2 in kidney cancers (Dressler, 1992; Gnarra and Dressler, 1995; Igarashi et al., 2001; Daniel et al., 2001) and Pax8 in thyroid cancer (Kroll et al., 2000; Lui et al., 2005). Pax8 was first isolated in the mouse and reported to be expressed in the developing excretory system and thyroid gland (Plachov et al., 1990). Several years later it was shown that pax8 was expressed in the developing murine brain (Stoykova and Gruss, 1994). Pax8-/null mice die shortly after birth because follicular cells of the thyroid gland fail to develop and are unable to produce thyroid hormone that is crucial for development (Mansouri et al., 1998). Besides the effect on thyroid development, the original knockout study reported no other abnormalities. When thyroxine1 treatment was given postnatally to the Pax8-/null mutants, these mice can survive to adulthood. A closer look at the adults revealed Pax8-/mice suffer from hearing loss despite thyroxine substitution, indicating the Pax8 mutation leads to auditory defects (Christ et al., 2004). The hearing defects in the Pax8 null mice suggested Pax8 plays an important role in vertebrate ear development. More recent studies also reveal a crucial role for Pax2 and Pax8 in kidney development, which was overlooked in the original knockout studies (Narlis et al,. 2007). The Pax8-/mouse illustrates how the effects of a loss of function mutation in one organ system can mask the precise role of the gene in another developing system. 1 Thyroid hormone.
21 Therefore, it is possible that there are other consequences in development, particularly in the CNS, when Pax8 is not functioning properly. Zebrafish Ear Development and the Role of Pax8 Zebrafish ear development begins with an ectodermal thickening into the otic placode by 16 hpf (Hadden & Lewis, 1996). The thickening is triggered by signals from the hindbrain, the fibroblast growth factors Fgf3 and Fgf8, that are necessary and sufficient for otic induction (Phillips et al., 2004). Members of the Pax2/5/8 family were first implicated in vertebrate ear formation when the expression of these genes were observed in ectodermal cells of the otic placode during development (Pfeffer et al., 1998). In zebrafish, pax8 expression begins at the 1somite stage (10 1/3 hpf) and is one of the earliest known markers of otic cells in vertebrates (Pfeffer et al., 1998). The early expression of pax8 in the pre-placodal ectoderm suggests it is one of the first genes to respond to otic signals from the hindbrain, and that it potentially plays a role in determining otic cell fate. Previous studies of pax8 knockdown in zebrafish showed varying phenotypes in ear development. When Pax8 expression is knocked down using MOs targeting pax8 splice variants, morphological development of the otic placode is delayed, and the resulting effect on the ear is mild (Hans et al., 2004). Using additional sequence information of the 5Â’UTR, Mackereth and others (2005) were able to knock down pax8 function more efficiently using MOs targeted to one or all of the three main isoforms of pax8 When multiple pax8 isoforms were targeted, they produced a more profound phenotype in otic development. Several questions were unable to be addressed in the pax8 morphant studies. First, they showed a reduction in the number of hair cells in the pax8 morphant. Hair cells are very sensitive to movement and sound; when there are fewer hair cells, the response to these stimuli is expected to be impaired (Hudspeth et al., 1992). Because the knockdown of pax8 does not persist
22 throughout development the pax8 morphants, the effects on behavior could not be assessed. The question remains, is there a behavioral phenotype associated with the abnormally small otic vesicles? Also, the effects on more mature ear structures were not characterized because the embryos were not viable later in development. Other unanswered questions involve other systems in which pax8 is expressed during development; for example, what are the effects of the knockdown on kidney and thyroid development? How does the knockdown affect specification of hindbrain tissues? The gene trap line can be used to further investigate the effects of the pax8 knockdown on development. There were also problems with the pax8 morphant studies that complicated the interpretation of the morpholino-induced phenotype. There was variability in the severity of the otic phenotype, with 10% of pax8 morphants exhibiting no phenotype. Some pax8 morphants had a severe otic phenotype, with defects in the semicircular canals and fused or absent cristae (Mackereth et al., 2005). The embryos with the most severe phenotype degenerated and died as development progressed, creating even greater problems with the interpretation of the results. Although the pax8 MO studies provided new insights into early development of the zebrafish ear, the complications associated with using MOs to block expression in developmental studies raised as many questions as were answered. Although the Pax8-/null mouse has also been used to study the role of Pax8 in certain areas of development, no pax8 mutant has been reported in zebrafish, hindering developmental studies using this organism. The only studies of Pax8 in zebrafish have been accomplished by disrupting Pax8 protein function using MOs. The interpretation of the results in the pax8 MO studies was complicated. Additionally, the effects of the pax8 knockdown in the developing CNS, kidney, and thyroid follicles were not thoroughly studied (Hans et al., 2004; Mackereth et
23 al., 2005; Wendl et al., 2002). Thus, if the pax8 gene trap line of zebrafish results in a pax8 mutant, this model system can provide much information to further our understanding of the role of Pax8 in embryonic development. With the advancement of mutagenesis techniques and availability of newer vertebrate models, zebrafish and gene trap have emerged as a potentially powerful tool to study gene and protein function during development. I established a gene trap line of zebrafish using an insertional mutagenesis technique involving the Tol2 transposon and DsRed-Express reporter gene. This line of fish contains an insertion of Tol2DsRed in the pax8 gene and RFP is anticipated to represent Pax8 expression. Pax8 is a member of the Paired-box ( Pax ) gene family. Pax proteins are transcriptional regulators that play key roles in development. Specifically, Pax8 is suspected to play a key role in ear development due to the effects of otic vesicle formation in pax8 MO studies. The pax8 gene trap line of fish will be used to show whether the Tol2 gene trap has the potential to significantly knockdown gene expression and produce a mutant phenotype. A successful gene trap of pax8 is expected to disrupt otic vesicle development, and can validate the use of the Tol2 gene trap in mutagenesis and developmental studies.
24 Figure 1-1. Tol2 transposition. Schematic of Tol2 transposition from gene trap construct to genomic DNA. Transposase recognizes the 5Â’ and 3Â’ ends of the Tol2 DNA sequence, cuts the Tol2 sequence from the construct, produces a break in the genomic DNA, and ligates the Tol2 DNA to the genomic DNA.
25 Figure 1-2. The Tol2 gene trap construct. The Tol2 gene trap construct, T2KSAG, was obtained from Koichi Kawakami at the National Institute of Genetics (Shizuoka, Japan). The transposase coding of the Tol2 DNA was removed. Inserted within the Tol2 DNA sequence is a splice acceptor upstream of the EGFP sequence (with a start and stop codon) and downstream a polyA signal. The EGFP sequence was inserted in a reverse orientation to prevent expression due to endogenous promoter activity. Figure 1-3. Pax8 protein conserved domains. The paired domain, located at the N-terminus, and homeodomain recognize and bind to DNA sequences. Members of the Pax gene family contain either a partial or whole homeodomain. The transactivation and inhibitory domains are located at the C-terminus and regulate transcriptional activity. The octapeptide is involved in protein-protein interactions.
26 CHAPTER 2 MATERIALS AND METHODS All chemicals were obtained from Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Pittsburg, PA) and all primers were obtained from IDT (Coralville, IA) unless otherwise noted. Fish Strains and Husbandry Brachydanio rerio were obtained at ages 61-120 days (Aquatic Eco-Systems; Apopka, FL). These fish were used to produce embryos for injection of the DNA construct and mRNA. All fish were kept on a 14 hours light and 10 hours dark schedule. The water temperature was maintained between 27-29 degrees Celsius. All fish were handled according to guidelines set forth by the Institutional Animal Care and Use Committee (IACUC). Construction and Injection of Tol2 Plasmid The Tol2 construct T2KSAG containing the EGFP reporter gene was obtained from Koichi Kawakami at the National Institute of Genetics (Shizuoka, Japan). T2KSAG consists of the Tol2 transposable DNA element with the transposase coding region removed. The EGFP sequence between the BamHI and NotI sites in pT2KSAG was removed and replaced with the RFP DsRed-Express-DR sequence (Clontech; Mountain View, CA). The plasmid was purified using the Qiagen Miniprep Kit (Qiagen; Valencia, CA) and termed Tol2DsRed (Fig 2-1). Transposase mRNA was prepared from pCS-TP (Kawakami et al., 2004) using the mMachine SP6 Kit (Ambion; Austin, TX) and MegaClear Kit (Ambion). The Tol2DsRed plasmid and transposase mRNA were diluted to a final concentration of 50 ng/ml in filter sterilized milliQ (mQ) H2O. The transposase mRNA and Tol2DsRed plasmid were injected into fertilized zebrafish eggs as previously described (Ono et al., 2001). Surviving embryos were raised to adulthood.
27 Genomic Characterization of Tol2DsRed Insertion Identification of Founder Fish Injected embryos that survived were raised to adulthood and mated to wild type fish. The embryos from these matings were screened for RFP fluorescence at 72-120 hpf using the Zeiss Stemi SVII stereomicroscope (Carl Zeiss Microimaging Inc; Thornwood, NY) and UV lamp. Fish that expressed RFP were isolated and raised to adulthood. These fish were termed founder fish of the Tol2DsRed gene trap line and mated to wild types. Embryos from this mating that were positive for RFP fluorescence were used to determine the Tol2DsRed insertion site and RFP fusion transcripts. Identification of DNA Insertion Site Inverse PCR (IPCR) was performed on genomic DNA to amplify the gene sequence surrounding the Tol2DsRed insertion (Fig. 2-2). First, DNA was isolated from 10 embryos at 2dpf. Embryos were anesthetized with 1X tricaine, decapitated, and chopped into small pieces. The tissue was digested in 500 l extraction buffer (445 l Elution Buffer (Qiagen), 50 l 2% TritonX, and 5 l PCR grade proteinase K (Roche; Indianapolis, IN) at 50 C for 1 hour. Genomic DNA was extracted using standard phenol/chloroform extraction. DNA was precipitated using standard ethanol precipitation and resuspended in 30 l Elution Buffer (Qiagen) overnight. One g DNA was incubated separately with the restriction enzymes AluI, HaeIII, and MboI (New England BioLabs; Ipswich, MA), then the digested DNA was diluted to 1 ng/ l and self ligated using Quick T4 DNA Ligase (New England Biolabs; Ipswich, MA) per the manufacturerÂ’s instructions to circularize the individual fragments. The circularized DNA was purified and concentrated to 5 l using the Gene Clean kit (Qbiogene; Irvine, CA). The concentrated DNA was diluted 1:10 for a final volume of 50 l using autoclaved, mQ H2O.
28 IPCR was performed on 5 l of the diluted, ligated DNA using nested primers specific for the 5Â’ and 3Â’ ends of the Tol2 sequence and LAtaq polymerase with provided reagents (Takara; Shiga, Japan). PCR was performed as previously described with modifications (Kawakami et al., 2004). The nested primers specific for the 5Â’ end of Tol2 to amplify the sequence surrounding the insertion site were: round 1, Tol2-5Â’/f1 (5Â’-AGT ACT TTT TAC TCC TTA CA-3Â’) and Tol25Â’/r1 (5Â’-GAT TTT TAA TTG TAC TCA AG-3Â’); round 2, Tol2-5Â’/f2 (5Â’-TAC AGT CAA AAA GTA CT-3Â’) and Tol2-5Â’/r2 (5Â’-AAG TAA AGT AAA AAT CC-3Â’). The nested primers specific for the 3Â’ end of Tol2 to amplify the sequence surrounding the insertion site were: round 1, Tol2-3Â’/f1 (5Â’-TTT ACT CAA GTA AGA TTC TAG-3Â’) and Tol2-3Â’/r1 (5Â’-CTC CAT TAA AAT TGT ACT TGA-3Â’); round 2, Tol2-3Â’/f2 (5Â’-ACT TGT ACT TTC ACT TGA GTA-3Â’) and Tol2-3Â’/r2 (5Â’-GCA AGA AAG AAA ACT AGA GA-3Â’). The PCR products were run on a 1% agarose gel, and the most prominent band or single band was extracted using the Qiaquick Gel Extraction Kit (Qiagen). Sequencing reactions were performed on the gel extracted PCR products using the ABI Prism Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems; Foster City, CA). Reactions were purified using Performa DTR Gel Filtration Cartridges (Edge BioSystems; Gaithersburg, MD). Electrophoresis and data analysis of the purified sequencing reactions were performed using the ABI Prism 310 Genetic Analyzer (Applied Biosystems). These sequences, excluding the Tol2 portion of the sequence, were used for the Sanger D. rerio Ensembl genome blast (http://www.ensembl.org/Multi/blastview?species=Danio_rerio). Identification of Fusion Transcripts of RFP with Endogenous Transcripts RNA was isolated from 10 RFP expressing embryos at 4 dpf in order to amplify transcripts of DsRed-Express fused to endogenous transcripts of the trapped gene. First, the fish were
29 anesthetized as previously described, decapitated, and chopped into small pieces. The tissue was homogenized using QIAshredder columns (Qiagen). The RNA was extracted and isolated from the homogenized tissue using the RNeasy Mini Kit (Qiagen). cDNA was produced from the RNA using the BD SMART RACE cDNA Amplification Kit (Clontech). A 5Â’ rapid amplification of cDNA ends (RACE) was performed on the cDNA using nested primers specific for the 5Â’ end of the DsRed-Express-DR sequence. The nested primer for round 1 was RFP2: 5Â’CAC GCC GAT GAA CTT CAC CTT GTA GAT GAA GGA GCC G-3Â’, and round 2 was RFP4: 5Â’-CAG GGA GGA GTC CTG GGT CAC GGT CAC-3Â’. The PCR products were run on an agarose gel. Bands were extracted using the Qiaquick Gel Extraction Kit (Qiagen). Sequencing reactions were performed on the gel extracted 5Â’RACE products as described above. The DsRed-Express sequence was removed from the amplified sequences and the remaining sequences used for the Sanger D. rerio Ensembl genome blast to compare exons of pax8 with amplified sequences. Identification of Homozygous Adults from a Heterozygous Intercross PCR was used to distinguish homozygous adults from heterozygous adults produced from a heterozygous intercross of the pax8 gene trap line. Pure populations of wild type, heterozygous, and homozygous embryos were obtained for isolation of genomic DNA. Two homozygous adults of the pax8 gene trap line were mated to produce a pure population of homozygous embryos. A homozygous male and wild type female were mated to produce a pure population of heterozygous embryos. Two wild type adults were mated to produce a pure population of wild type embryos. Tissue was obtained from the caudal fin of adult fish anesthetized in ice cold 1% Tricaine. These fish were previously screened and found positive for RFP fluorescence as embryos. The tissue was digested using proteinase K (Qiagen) and DNA extracted using the DNeasy blood and tissue kit (Qiagen). Primers were designed in the first
30 intron of pax8 on both sides of the Tol2DsRed insertion to amplify the wild type allele; a forward primer upstream from the insertion location, pax8 F5: 5Â’-TAT TGT GTG TTT CTA AAG TCA ACC C-3Â’, and a reverse primer downstream of the insertion location, pax8 R6: 5Â’TAG ATC ACG GTG TAA ACA TCG ACA A-3Â’. PCR using these primers generates approximately a 100 bp product from the wild type pax8 allele, whereas a very large product or no product is generated from the pax8 allele containing the Tol2DsRed insertion (Fig. 2-3). If no product was produced, PCR using alpha-tubulin specific primers was performed to verify the integrity of the DNA, alpha tubulin sense: 5Â’-CGT GGT CAC TAC ACT ATT GGC AAG G-3Â’, and alpha tubulin antisense: 5Â’-AGG TAG CCA GAG GGA AGT GGA TAC G-3Â’. The identity of a homozygous fish was further verified by the production of 100% RFP positive offspring when mated to wild type fish. Imaging Timing of Expression To follow the RFP expression in heterozygous embryos, male homozygous fish were mated to wild type females to generate a pure population of heterozygous embryos. The embryos were placed in 2 or 3% methylcellulose on a Petri dish and oriented. Live embryos were observed every two hours, beginning at 8 hpf until 20 hpf using a stereomicroscope and UV lamp. The heterozygous embryos, along with homozygous embryos, were also observed at later stages up to 7 dpf. Confocal Imaging Heterozygous adults of the pax8 gene trap line were mated to heterozygous Isl1-GFP fish. Embryos were screened for RFP and GFP fluorescence using a stereomicroscope and UV lamp. Embryos exhibiting dual fluorescence were mounted on a glass coverslip (Fisher) or FluoroDish (World Precision Instruments, Inc.; Sarasota, FL) using 1.5% low-melting
31 temperature agarose. Images were obtained using a Leica scanning confocal microscope LSC Sp2 (Leica Microsystems GmbH; Wetzlar, Germany). Differential Interference Contrast (DIC) Imaging Live embryos were mounted on a FluoroDish using 3% methylcellulose and overlaid with H2O. DIC images were obtained at 1dpf using the Olympus upright microscope BX51W1 (Olympus America Inc; Center Valley, PA). Measurements of otic vesicle width were taken using the Image Pro-Plus Software (Media Cybernetics, Inc; Bethesda, MD). The width of the otic vesicle was determined by measuring from the most rostral and caudal points of the otic vesicle (Fig. 2-4). The mean width +/the standard error of the mean (S.E.M.) was calculated; wild type n=7, homozygous n=6. Quantitation of pax8 Expression For quantitative PCR (qPCR), primers were designed to amplify a unique target sequence within the pax8 mRNA. A sequence containing regions present in all pax8 isoforms was used for primer design. Regions of high homology, repetitive sequences, and single nucleotide polymorphisms (SNPs) were omitted from the target sequence as follows: regions of high homology within the pax8 mRNA were identified through a BLAST search of the pax8 mRNA in the D rerio genome (http://www.ncbi.nlm.nih.gov/BLAST/); repeat sequences were identified using the RepeatMasker program (www.repeatmasker.org); and SNPs were identified using the SNP BLAST of the D. rerio genome (http://www.ncbi.nlm.nih.gov/BLAST/). The remaining pax8 target sequence was submitted to Applied Biosystems, and primers were designed to amplify the junction between exons 5 and 6 (Fig. 2-5). A detector sequence was designed that bound to the sequence spanning the exon junction of 5 and 6. The primers and detector sequences were as follows: primer PAX8-e56F, 5Â’-CTC CCG CTG GAC ACT AAA GG -3Â’,
32 and primer PAX8-e56R, 5Â’-GCG GAG ATT CAG GAG GAG TGA-3Â’, and detector PAX8e56M1, 5Â’-ACA CAC ACT GAT CCC C-3Â’. RNA was isolated from embryos at 1 and 2 dpf from wild type, heterozygous, and homozygous embryos of the pax8 gene trap line as described previously. cDNA was prepared using the High Capacity cDNA Archive kit (Applied Biosystems). Primers for -tubulin were used as the endogenous control: primer A-TUBs, 5Â’-AGA CCT GGA GCC CAC TGT-3Â’, primer A-TUBas, 5Â’-TGG AAC AGC TGA CGG TAT GTC-3Â’, and detector A-TUB, 5Â’-CAT TGA TGA GGT GCG CAC TG-3Â’. q-PCR was performed on cDNA using the TaqMan Gene Expression System and ABI Prism 7000 Sequence Detection System (Applied Biosystems). All reactions were performed in triplicate. The Comparative CT method ( CT) of relative quantitation of gene expression was used to analyze the results. CT values were generated by the ABI Prism 7000 system software and corresponds to the PCR cycle number in which fluorescence from the detector was greater than the background fluorescence. The CT values were used to calculate a fold change in the expression of pax8 in heterozygous and homozygous embryos of the pax8 gene trap line compared to wild type. First, the Ct was determined: Ct = Ct (target) Â– Ct (endogenous control) Where the target was the pax8 exon junction 5-6 and the endogenous control was alphatubulin. Next, the Ct was determined: Ct = Ct (calibrator) Ct (experimental), where the calibrator was wild type and experimental was heterozygote and homozygote. The standard deviation (stdev) of Ct was determined in order to calculate the fold change from the Ct value: standard deviation (stdev) of Ct = (stdevtarget 2 + stdevendogenous control 2)1/2 and stdev of Ct = stdev Ct. The low and high fold change was calculated using the Ct and stdev of Ct: low fold change = 2[-( Ct + stdev. Ct)]; high fold change = 2[-( Ct Â– stdev Ct)]. Finally, the fold change was calculated
33 from the high and low fold change: fold change = average (high fold: low fold).
34 Figure 2-1 Tol2DsRed gene trap construct. Tol2DsRed construct with a splice acceptor sequence, start and stop codons within the DsRed-Express-DR sequence, and SV40 polyA sequence.
35 Figure 2-2. Inverse PCR (IPCR) to amplify regions of trapped gene. After genomic DNA is isolated from RFP positive embryos, it is digested in three separate reactions with three different restriction enzymes: AluI, HaeIII, and MboI. After digestion, the DNA fragments are self-ligated. IPCR is performed on the circular DNA using primers specific for the 5Â’ and 3Â’ ends of the Tol2 DNA. The primers amplify in opposite directions so that the end product contains the endogenous gene sequence surrounding the Tol2 sequence.
36 Figure 2-3. PCR to distinguish heterozygous and homozygous adults of the pax8 gene trap line. The pax8 specific primers used for genotyping are located upstream and downstream of the Tol2DsRed insertion site within intron 1 the pax8 gene. PCR amplifying a wild type allele will produce a product of approximately 100bp, while PCR amplifying a pax8 allele with the Tol2DsRed insertion will either not produce a product or a product of several kb.
37 Figure 2-4. Measurement of otic vesicle using DIC. DIC image of zebrafish head showing how the measurements of the otic vesicles were obtained. The most rostral and caudal locations of the otic vesicles were determined, and the distance between these two locations was measured. Figure 2-5. Primer design for pax8 quantitative PCR (qPCR). A set of primers were designed for the Taqman qPCR downstream of the Tol2DsRed insertion in the pax8 gene. The primers were designed to amplify the exon junctions of 5 and 6. A detector sequence spanning the exon junction was also designed. Binding of the detector sequence ultimately quantitates the expression levels of pax8. The detector binds to the sequence spanning the exon junction and will only bind to cDNA generated from mRNA transcripts. The detector will not bind to genomic DNA because of the presence of an intronic sequence in between exons 5 and 6.
38 CHAPTER 3 RESULTS Gene Trap Fish are Identified by Expression of RFP The EGFP sequence was removed from the Tol2 vector T2KSAG and replaced with the DsRed-Express-DR RFP sequence. The purpose of using a RFP for our gene trap experiments was to generate fish that express a fluorescent protein with a different excitation and emission spectrum than the original Tol2 vector. The RFP expressing gene trap fish can be crossed with fish expressing GFP in known regions, thereby providing a reference point for orientation of certain structures in the embryo like the brain and spinal cord. The Tol2 vector containing DsRed-Express was injected into newly fertilized embryos, and the surviving embryos were raised to adulthood. The injected adults were then either incrossed with other injected fish or outcrossed with wild type fish. The embryos produced from these breedings were screened for RFP fluorescence at 72-120 hpf. We identified two embryos from one mating pair with identical RFP expression patterns (Fig 3-1). The RFP expressing embryos were raised to adulthood and termed founders of this gene trap line. The founders were mated to wild types to produce the heterozygotes of the gene trap line used to identify the trapped gene and captured transcripts. Tol2DsRed Inserted into the First Intron of the pax8 Gene To identify the trapped gene, the DNA surrounding the Tol2DsRed insertion was amplified. Nested primers specific for the 3Â’ and 5Â’ ends of the Tol2 sequence were used for IPCR to amplify the circularized DNA sequence that flanked the Tol2 insertion. IPCR amplified unique sequences that aligned with the predicted pax8 gene. These sequences aligned with intron 1 of pax8 upstream of the sequences that correspond to conserved transcriptional and regulatory domains of the Pax8 protein (Fig 3-2a). The amplified sequences were not found to be
39 significantly homologous to other endogenous DNA sequences. An 8 bp duplication of the pax8 sequence was identified at both the 5Â’ and 3Â’ ends of the Tol2 sequence, which is characteristic of Tol2 transposition (Fig 3-2b). The identification of the duplication sequence indicated pax8 was in fact the gene trapped by the Tol2 vector. Because insertion of Tol2DsRed occurred at the 5Â’ end of the pax8 gene, a knockdown of pax8 mRNA and Pax8 protein expression is expected. Tol2DsRed Insertion Results in Fusion Transcripts of DsRed-Express to pax8 Insertion of Tol2DsRed in the first intron of pax8 is predicted to result in fusion mRNA transcripts of pax8 and DsRed-Express. To determine if the insertion of the Tol2DsRed vector in the pax8 gene sequence was able to capture the endogenous pax8 transcripts and cause the observed RFP expression pattern, the fusion transcripts of DsRed-Express to the endogenous mRNA were amplified in a 5Â’ RACE. Nested primers specific for DsRed-Express amplified two fusion transcripts. The transcripts that were fused to DsRed-Express aligned to the predicted exons 1a and 1c of pax8 (Fig. 3-3). Pax8 was previously cloned in zebrafish, and alternative splicing of the mRNA was shown to produce ten splice variants with two putative start codons (Fig. 3-3) (Mackereth et al., 2005). One putative start codon is located in exon 1c and another in exon 2. The Tol2DsRed sequence inserted into the first intron; therefore, translation of the protein product is expected to begin with the pax8 start codon in splice variants containing exon 1c, while translation of the protein product is expected to begin with the DsRed-Express start codon in transcripts that do not contain exon 1c. If the putative start codon located in exon 1c is real, this would result in a frameshift of the RFP and would not produce a functional fluorescent protein. Regardless of whether translation results a functional RFP, translation is expected to end with the DsRedExpress sequence and therefore terminate before exon 2 of the pax8 transcript. Thus, although RFP fluorescence may not completely reflect total Pax8 expression because all pax8 transcripts
40 will not be represented by RFP expression, the end result should be a knockdown of total Pax8 protein expression. RFP Expression in the Tol2DsRed Gene Trap Line of Fish The Tol2 construct inserted into the first intron of the pax8 gene and was shown to be fused to the endogenous pax8 transcript. Because RFP expression is controlled by the pax8 promoter in the pax8 gene trap line of fish, RFP fluorescence is expected to reflect Pax8 protein expression. To determine whether the expression pattern of RFP is similar to the previously reported pax8 mRNA expression pattern, fluorescence was observed in heterozygous embryos using fluorescent and scanning confocal microscopy. RFP fluorescence was observed in live heterozygous embryos using a stereomicroscope beginning at 8 hpf. Screening for fluorescence in the embryos was began at this time due to previous reports of pax8 mRNA expression. Pfeffer and others used in situ hybridization to detect pax8 mRNA and showed that it was expressed as early as 8 hpf (Pfeffer et al., 1998). Fluorescence was not visible until around 14 hpf and it appeared in the otic region (Fig. 3-4). By 21 hpf, RFP fluorescence was still visible surrounding the otic placode2; in addition it was detected at the mhb and in three bilateral pairs of hindbrain neurons (Fig. 3-5). These pairs are most likely located in rhombomeres 5, 6, and 7, based on previous reports of pax8 mRNA expression in these neurons (Pfeffer et al., 1998). No fluorescence was observed in the spinal cord at this time. The in situ studies of pax8 expression suggested the mRNA was no longer present in the otic placode at the 20-somite stage, or 19 hpf. Yet at 21 hpf, fluorescence was still 2 Precursor to otic vesicle and ear.
41 visible3. At 29 hpf, fluorescence was visible, but weaker, in the otic placode and surrounding region (Fig. 3-6). At this time, fluorescence was stronger at the mhb and in the three pairs of neurons in the hindbrain. RFP fluorescence was no longer present in the otic region at 2 dpf while a low level of fluorescence was seen in the spinal cord (Fig. 3-7). RFP continued to be expressed at the mhb and in a large population of neurons in the hindbrain. Also at 2 dpf, RFP fluorescence was observed in suspected thyroid follicle cells and the pronephros. By 3 dpf, RFP fluorescence was stronger in the spinal cord, as well as the mhb and hindbrain (Fig 3-8). RFP expression continues at the mhb, hindbrain, and spinal cord through 7 dpf. The embryos were not observed after 7 dpf, but RFP was most likely still being expressed. In summary, heterozygotes of the pax8 gene trap line show RFP fluorescence in the same regions as pax8 mRNA expression, but the timing of fluorescence and mRNA expression differs. The timing of pax8 mRNA and RFP protein expression was expected to differ because mRNA is transcribed before protein translation. Because pax8 mRNA and RFP was expressed in the same tissues, this confirmed RFP was being expressed in regions of pax8 expression. RFP expression in the spinal cord appeared to be restricted to a specific population of spinal neurons. The identify of RFP expressing neurons in the spinal cord can be determined by crossing the gene trap fish to fish with populations of known spinal neurons labeled. The type of spinal neurons can be distinguished based on their location in the spinal column: motor neurons are located in a ventral position, sensory neurons are located in a dorsal position, and interneurons are located in between the sensory and motor neurons in the spinal column. Heterozygous fish of the pax8 gene trap line were crossed to Isl1-GFP heterozygous fish to 3 Pax8 mRNA expression was previously reported in the optic vesicle (Pfeffer et al., 1998). Due to the reflection of the light at the eye, it was unclear when and if RFP was expressed in this region.
42 determine the type of neurons expressing pax8 in the spinal cord. In the Isl1-GFP fish line, expression of GFP is driven by the islet-1 promoter (Higashijima et al., 2000). The islet-1 gene is expressed in all postmitotic motoneurons (Ericson et al., 1992). Only a portion of the islet-1 promoter is used to drive GFP expression in the Isl1-GFP fish line; as a consequence, GFP is expressed in secondary motoneurons that innervate the dorsal muscular and sporadically in a few interneurons of the spinal cord (Fumihito Ono, personal communication). Leica scanning confocal microscopy was used to obtain images of RFP and GFP fluorescence in heterozygous embryos of the pax8 gene trap line with an Isl1-GFP background at 4 dpf (Fig. 3-9). The GFP expressing motor neuron cell bodies are located in the ventral region of the spinal cord. The RFP expressing neuron cell bodies are located just dorsal to the motoneuron cell bodies in the central region of the spinal column, with their axons descending into the region of the motoneurons. The location of RFP fluorescence in pax8 expressing neurons of the spinal cord with respect to GFP expression in motoneurons suggests the RFP expressing neurons are interneurons. Heterozygotes of the pax8 gene trap line were incrossed to generate embryos of three genotypes: wild type, heterozygous, and homozygous. The embryos from the incross that expressed RFP were compared, and no obvious differences in the development of RFP expressing tissues were observed. These fish were raised to adulthood and fish homozygous for the Tol2DsRed insertion in pax8 were identified using PCR. Primers specific for the pax8 gene were designed to amplify the region spanning the insertion site (fig. 2-3). Homozygous embryos were identified by the lack of a 100 bp band and presence of a band several kb in size (or no band at all).4 To compare RFP expression in homozygous and heterozygous embryos, male and 4 If no band was produced, PCR to amplify alpha-tubulin as a positive control was used to verify the integrity of the DNA.
43 female homozygous adults were mated to produce homozygous progeny, and male homozygous adults were mated to wild type female adults to produce heterozygous embryos. Pax8 mRNA Expression is Dramatically Knocked Down in Homozygous Embryos Fish homozygous for the Tol2DsRed insertion are expected to have a substantial knockdown of pax8 because of the location of the insertion in intron 1 of the pax8 gene. Quantitative PCR using primers targeting the junction between exons 5 and 6 was performed to measure the level of pax8 expression. These locations were chosen for two reasons: 1) by amplifying the exon junction, the possibility of amplifying genomic DNA is eliminated; and 2) these exons are present in all reported splice variants of pax8 and amplification of these regions ensures the presence of all splice variants will be detected (Mackereth et al., 2005). qPCR results show that at both 1 and 2 dpf, pax8 expression in heterozygous embryos is knocked down, with a fold change from wild type expression of 0.581 and 0.732 respectively (Fig. 3-13)5. More significantly, in homozygous embryos, pax8 expression was drastically knocked down at 1 and 2 dpf, with a fold change from wild type expression of 0.024 and 0.03 respectively. It is assumed that a reduction in mRNA expression represents a reduction in protein expression as well. Thus, a significant reduction of Pax8 protein in the homozygous fish will compromise the function of Pax8 as a transcription factor and the transcriptional networks in the cells in which it is expressed. If Pax8 plays a significant role in embryonic development, the homozygous fish are expected to exhibit an abnormal anatomical, physiological, or behavioral phenotype. 5 These values represent the results of one qPCR reaction. Other reactions were performed and gave similar results.
44 Knock Down of pax8 Expression May Result in a Subtle Mutant Phenotype Early in Development qPCR showed that pax8 was almost completely knocked out in homozygous embryos at 1 and 2 dpf. However, RFP expression in homozygous embryos was similar to heterozygotes with no obvious abnormalities or differences, with the exception of fluorescence intensity. If the knockdown of pax8 altered the development of the structures pax8 is normally expressed, alteration in the morphology of the RFP expressing structures is expected. It is also expected to alter cell fate because members of the pax gene family are known to play a role in cell specification, growth, survival, and migration. If the fate of the pax8 expressing cells is affected when pax8 is knocked down, it is expected to result in either an increase or decrease in the number of cells expressing RFP. Stronger fluorescence is expected in homozygous embryos because these fish are expressing two copies of DsRed-Express, while the heterozygous embryos are expressing only one copy. At 21 hpf, RFP expression was similar in the region surrounding the otic placode, at the midbrain hindbrain boundary, and in three pairs of hindbrain neurons (Fig. 3-10). Fluorescence was stronger in these regions, as expected, but the morphology of the fluorescent structures appeared the same. At 29 hpf, RFP was expressed in the same regions in the heterozygous embryo, with the expression in the hindbrain appearing to extend caudally to pairs of neurons in the hindbrain (Fig. 3-11). RFP may also be expressed more caudally in the hindbrain of heterozygous embryos at this time but is too low to visibly detect. At 2 dpf, RFP expression remained similar to expression in the heterozygous embryo with no anatomical abnormalities (Fig. 3-12). RFP was expressed at the mhb, hindbrain, and spinal cord, but no longer in the otic region. At this time, expression of RFP in the thyroid follicle cells was evident and more apparent in the pronephric ducts compared to heterozygous embryos. No differences in
45 RFP expressing cells and structures were observed in the homozygous fish of the pax8 gene trap line using stereomicroscopy. Previous studies of a pax8 knockdown in zebrafish using morpholinos reported a defect in otic development. Because our gene trap line resulted in a significant knockdown of pax8 we expected to see a similar otic vesicle phenotype in homozygous embryos of the pax8 gene trap line as in pax8 -MO studies. The pax8 knockdown studies showed a delay in otic placode formation and an otic vesicle reduced in size at 30 hpf (Hans et al., 2004; Mackereth et al., 2005). At approximately 24 hpf, we observed otic vesicles reduced in size in homozygous embryos compared to age-matched wild type (Fig. 3-14). The severity of the phenotype, in regards to the size of the otic vesicle, varied among the homozygous embryos with the most severe phenotype depicted (Fig. 3-14). Measurements of the otic vesicles from various embryos at 26 hpf indicate the reduced size is present among the homozygous embryos as a whole, with some variability. The mean width of the otic vesicle in wild type was 89.6 +/2.4 m, while slightly shorter in homozygous embryos at 83.4 +/0.6 m (wild type n=7; homozygote n=6) (Fig. 3-15). Further studies are needed to determine whether the otic phenotype causes a behavioral phenotype in older embryos and adults. The results show that the Tol2DsRed inserted into the first intron of the pax8 gene and produced fusion transcripts of DsRed and pax8 mRNA. The fusion transcripts led to expression of RFP in the same regions as pax8 mRNA expression. Homozygous embryos of the pax8 gene trap line have a significant knockdown of pax8 mRNA expression. The knockdown is expected to prevent translation of the Pax8 protein. However, RFP expression in homozygous embryos appeared normal compared to heterozygous embryos. Preliminary observations suggest the otic vesicle is slightly reduced in embryos 1 dpf. This agrees with previous studies of pax8
46 knockdown using MOs. However, further studies need to be performed in order to make a definitive conclusion on the presence of a mutant phenotype in the gene trap fish.
47 Figure 3-1. Founder of the Tol2DsRed gene trap line. Image of embryo at 2dpf positive for RFP fluorescence, indicating a Â‘trappedÂ’ gene.
48 A B Figure 3-2. Insertion of Tol2DsRed. A) Insertion site of Tol2DsRed in the pax8 gene. Tol2DsRed inserted into the first intron of the predicted pax8 gene. The insertion occurred upstream of conserved protein domain sequences that are needed for proper function of Pax8 as a transcription factor. B) Genomic sequence of pax8 flanking the Tol2DsRed insertion. An 8 bp duplication of the pax8 sequence surrounding the Tol2 insertion site was identified. This duplication is characteristic of Tol2 transposition.
49 Figure 3-3. Amplification of pax8 transcripts fused to DsRed-Express. 5Â’ RACE using DsRed-Express specific primers amplified two splice variants of pax8 fused to DsRed-Express. One splice variant contained exon 1a while the other variant contained exons 1a and 1c. The start codons of each splice variant are underlined.
50 A B Figure 3-4. RFP fluorescence visible by 14.5 hpf in heterozygous pax8 gene trap embryos. Fluorescent stereomicroscope image of embryos heterozygous for the Tol2DsRed insertion in pax8 (lateral view). A) At 12.5 hpf, no fluorescence was visible. B) Fluorescence was first observed in the otic placode region around 14.5 hpf. op=otic placode.
51 Figure 3-5. RFP expression in heterozygotes at 21 hpf. Fluorescent stereomicroscope image of embryo heterozygous for the Tol2DsRed insertion in pax8 at 21 hpf (lateral view). Fluorescence is visible at the mhb, op, and in three pairs of hindbrain neurons. Scalebar=200 m. hb=hindbrain, mhb=midbrain-hindbrain boundary; op=otic placode.
52 Figure 3-6. RFP expression in heterozygotes at 29 hpf. Fluorescent stereomicroscope image of embryo heterozygous for the Tol2DsRed insertion in pax8 at 29 hpf (lateral view). Fluorescence is stronger at the mhb and in the three pairs of hindbrain neurons and weaker in the area surrounding the op. Scalebar=200 m. hb=hindbrain; mhb=midbrain-hindbrain boundary; op=otic placode.
53 Figure 3-7. RFP expression in heterozygotes at 2 dpf. Fluorescent stereomicroscope images of embryos heterozygous for the Tol2DsRed insertion in pax8 at 2 dpf (lateral view). RFP is expressed at the mhb, hindbrain, spinal cord, and thyroid follicle cells. Scalebar=200 m. hb=hindbrain; mhb=midbrain-hindbrain boundary; pn=pronephros; sc=spinal cord; th=thyroid follicles.
54 A B Figure 3-8. RFP expression in heterozygotes at 3 dpf. Scanning confocal image of embryos at 3 dpf (lateral view). A) At 3dpf, RFP fluorescence is still visible at the mhb, hb, sc, th, and nd of the pronephros. B) Image of wild type to show endogenous fluorescence in the eye and yolk sac, but not the mhb, hb, sc, th, and nd. Hb=hindbrain; mhb=midbrain-hindbrain boundary; nd=nephric duct of the pronephros; sc=spinal cord; th=thyroid follicles. Figure 3-9. RFP and GFP expression in heterozygous embryos in an Isl1-GFP background at 4 dpf. Scanning confocal image of embryo heterozygous for the Tol2DsRed insertion in the pax8 gene in an Isl1-GFP background at 4 dpf. GFP is expressed in spinal motoneurons that innervate the dorsal musculature. The position of RFP expressing pax8 spinal cord neurons in relation to the GFP expressing motoneurons suggests pax8 is expressed in a population of spinal cord interneurons. hb=hindbrain; mhb=midbrain-hindbrain boundary; sc=spinal cord.
55 Figure 3-10. RFP expression in homozygotes at 21 hpf. Fluorescent stereomicroscope image of embryo homozygous for the Tol2DsRed insertion in pax8 at 21 hpf (lateral view). Fluorescence is visible at the mhb, otic placode, and in three pairs of hindbrain neurons. However, there is no obvious difference in expression compared to heterozygous. Scalebar=200 m. hb=hindbrain, mhb=midbrain-hindbrain boundary; op=otic placode.
56 Figure 3-11. RFP expression in homozygote at 29 hpf. Fluorescent stereomicroscope image of homozygous embryo at 29 hpf. Fluorescence is visible at the mhb, surrounding the otic placode, and in pairs of hindbrain neurons. Expression of RFP appears to be expanded caudally to neuron pairs in the hindbrain compared to heterozygous expression. Scalebar=200 m. hb=hindbrain; mhb=midbrain-hindbrain boundary; op=otic placode.
57 Figure 3-12. RFP expression in homozygote at 2 dpf. Fluorescent stereomicroscope image of a homozygous embryo 2 dpf. RFP expression is strong at the mhb, hindbrain, spinal cord, and thyroid follicular cells. Expression is strong enough to see fluorescence in the nephric ducts of the pronephros. Scalebar=200 m. hb=hindbrain; mhb=midbrain-hindbrain boundary; pn=pronephros; sc=spinal cord; th=thyroid follicles.
58 Figure 3-13. Pax8 quantitative PCR (qPCR). qPCR shows a reduction of pax8 mRNA expression in heterozygotes and a dramatic reduction in homozygous embryos for the Tol2DsRed insertion in the pax8 gene at 1 and 2 dpf.
59 Figure 3-14. Images of otic vesicle at 24 hpf. DIC images of embryos at 24 hpf. The otic vesicle is smaller in homozygous embryos compared to heterozygous and wild type. The size of the otic vesicle varied among the homozygous embryos, and the most extreme phenotype is depicted in this image. Scalebar=100 m.
60 Figure 3-15. Otic vesicle width at 26 hpf. Otolith size is reduced in embryos with a significant knockdown of pax8 expression at 26 hpf. Measurements taken of the otolith from DIC images. Values expressed are the average +/S.E.M. homozygous n=6; wild type n=7.
61 CHAPTER 4 DISCUSSION The Tol2DsRed inserted into the first intron of the predicted pax8 gene sequence. The insertion produced fusion transcripts of DsRed and pax8 mRNA. The fusion of mRNA transcripts led to expression of RFP in the same regions as pax8 mRNA expression, suggesting RFP was being expressed in the same cells as pax8 Although homozygous embryos of the pax8 gene trap line have a significant knockdown of pax8 mRNA expression, RFP expression in the homozygous embryos appeared normal compared to heterozygous embryos. Observations of otic vesicle development suggest the otic vesicles are slightly reduced in homozygous embryos 1 dpf compared to wild type embryos. Previous studies of pax8 knockdown using MOs show a more pronounced otic phenotype; thus, a reduced otic vesicle in homozygous embryos of the pax8 gene trap line would further support the result of the pax8 knockdown studies. However, further studies are needed in order to conclude whether the size of the otic vesicle is reduced, and to what extent the size is reduced, in homozygous embryos of the gene trap line. Further studies are also needed to determine whether the phenotype persists through development and if the knockdown causes abnormal development of more mature ear Tol2 Gene Trap Captures Endogenous pax8 Transcripts and Represents Pax8 Expression Cloning of pax8 in zebrafish revealed alternative splicing of exons 1a, 1b, and 1c (Mackereth et al., 2005). Out of 54 clones, the different splicing events produced three possible combinations of exon 1: 1) exon 1a; 2) exons 1a and 1c; and 3) exons 1a, 1b, and 1c. 5Â’ RACE amplified only two combinations of exon 1 fused to DsRed-Express: one consisting solely of exon 1a, and one consisting of exon 1a and 1c. A fusion transcript containing exon 1a, 1b, and 1c was not amplified. Cloning of pax8 by Mackereth et al predicted a frequency of 3.7% for transcripts containing exons 1a, 1b, and 1c, while a frequency of 74.2% and 22.1% were
62 predicted for transcripts containing exon 1a and exon 1a and 1c respectively. The low abundance of the splice variant (containing exons 1a, 1b, and 1c) may explain why that transcript was not amplified in this 5Â’ RACE. It may be possible to amplify this variant by using more template for the 5Â’ RACE reaction. Although the Tol2DsRed construct contains a splice acceptor, the DsRed-Express sequence may not be spliced into all variants of the pax8 mRNA for reasons that are unclear. Minute amounts of pax8 transcript were detected using qPCR, suggesting exons downstream of the insertion were spliced into a few transcripts. Whether the DsRed-Express sequence failed to be spliced into those final transcripts or the polyA signal did not function properly is not known. Regardless, total pax8 mRNA was significantly knocked down-almost a knock out. The presence of a few transcripts and protein molecules is not expected to compensate for such a dramatic decrease in expression. Therefore, this gene trap line is expected to be similar to that of a null mutant. As mentioned previously, two start codons are predicted for the different splice variants of pax8 (Mackereth et al., 2005). One start codon is located in exon 1c, while the other is located in exon 2. The Tol2DsRed inserted in intron 1; therefore, translation of all variants that do not contain exon 1c will begin with DsRed-Express. Of the splice variants, 74.2% are predicted to contain only exon 1a. Therefore, the majority of pax8 protein isoforms should be represented by RFP fluorescence. However, the remaining splice variants are predicted to contain exon 1c. A putative start codon is located in exon 1c. Translation at this point would cause a frameshift of the DsRed-Express reading frame and result in a nonfunctional DsRed-Express protein. Assuming the putative start codon was correctly identified, 25.8% of the transcripts will not be represented by RFP fluorescence. The RFP expression in the pax8 gene trap line will not
63 accurately represent total Pax8 protein expression if the transcripts containing exon 1c are expressed at different times and different locations than the other transcripts that result in a functional RFP protein. However, because the majority of transcripts are predicted to contain exon 1a, RFP expression will be highly representative of Pax8 expression. RFP expression is driven by the pax8 promoter, and therefore will be expressed in cells and tissues that normally express Pax8. Because Pax8 is a transcription factor, it is expected to localize to the nucleus. However, the subcellular location of RFP is not expected to represent that of the Pax8 protein because DsRed-Express does not have a cellular localization signal sequence. The RFP is expected to fill the cytoplasm of the cell, rendering the cytoarchitecture of the cell visible. RFP fluorescence was observed at later time points compared to pax8 mRNA expression (Fig. 4-1). The difference was expected because mRNA and protein was compared, and protein is translated from mRNA; thus, mRNA must be present before the protein product. An excellent example of the difference is the expression of pax8 mRNA and visualization of RFP in the otic placode. A previous study of pax8 expression first detected pax8 mRNA at 8 hpf in two regions: the area posterior to what is presumed to become the midbrain-hindbrain boundary region and the neural keel6 (Pfeffer et al., 1998). In heterozygous fish of the pax8 gene trap line, fluorescence was not clearly visible in the otic placode until 14.5 hpf. It is possible that there were low levels of RFP fluorescence before 14.5 hpf in the region posterior to the midbrainhindbrain boundary and neural keel. However, endogenous fluorescence of the yolk sac may have masked the fluorescence until RFP levels were high enough to distinguish from the 6 The neural keel develops from the neural plate and into the neural rod. In teleosts, this occurs during primary neurulation (source: http://db.yeastgenome.org/cgibin/GO/goTerm.pl?goid=14025)
64 endogenous signal. In a previous study, pax8 mRNA was detected in the otic placode at the 1somite stage, approximately 10 1/3 hpf, about 4 hours prior to the appearance of fluorescence. mRNA continued to be detected in the otic placode through the 5-somite stage, approximately 11 2/3 hpf. After this stage, expression in the otic placode was down-regulated and weakly expressed by 14 hpf. These pax8 mRNA levels fall before RFP expression is detected in the pax8 gene trap line. By 19 hpf, pax8 mRNA was no longer detected in the otic placode. RFP expression in the otic placode ceased much later than the disappearance of pax8 mRNA, sometime between 29 and 48 hpf. There are other discrepancies in the timing, but not the location, of pax8 mRNA and RFP expression in the gene trap fish. Both pax8 mRNA and RFP are expressed in the midbrain-hindbrain boundary (mhb), hindbrain, and spinal cord. There may be several reasons for the differences in previous reports of pax8 mRNA expression and when RFP was observed in the gene trap fish. The estimated maturation rate of DsRed-Express is less than one hour (Bevis and Glick, 2002). Therefore, the time between the pax8 mRNA and DsRed-Express expression is expected to be less than an hour and relatively close in time. Yet fluorescence was not observed in the otic placode until over 4 hours after mRNA was first detected. mRNA is transcribed and edited before the protein can be translated from the mRNA; therefore the protein is expected to appear after the mRNA. Additionally, there are many factors that can affect the rate of protein translation and delay the appearance of detectable levels of the protein product (You & Yin, 2000). This can explain why fluorescence was not observed in the otic region, mhb, and hindbrain until several hours after mRNA was detected. However, this does not explain the prolonged delay in fluorescence in the spinal cord. Most likely there are low levels of RFP present in the spinal cord before fluorescence is visible.
65 A more sensitive assay to detect RFP, such as immunostaining using an anti-RFP antibody, would most likely show that the protein is present before fluorescence was observed. Gene Trap May Result in Abnormal Ea r Development in Homozygous Embryos The developmental abnormalities of the ear in pax8 morphants were anticipated in the pax8 gene trap line due to the significant knockdown of pax8 mRNA in these embryos. The gene trap fish are expected to more accurately depict the consequences of impaired Pax8 function compared to the MO studies for several reasons. First, the knockdown of pax8 expression is so great in the gene trap line that the result is nearly a knock out. Gene trap prevents the full-length pax8 transcript from being produced, eliminating uncertainty as to whether the protein product is knocked down as well. On the other hand, MOÂ’s bind to transcript that has already been synthesized, meaning the Pax8 protein is already present as well. Further, no Pax8 antibody exists for use in zebrafish, making it almost impossible to measure Pax protein levels. Therefore, the degree of protein knockdown is uncertain in pax8 morphants. The pax8 MOs used in the study had the potential to bind to other members of the pax gene family, specifically pax2/5 that are believed to play redundant roles with pax8 (Mackereth et al., 2005) If pax2 and pax5 compensate for the lack of pax8 binding of the MOs to all three genes would cause a phenotype that would not normally be produced if pax8 is selectively targeted. In the pax8 morphant, nonspecific necrosis was observed when higher MO concentrations were used, illustrating how the interpretation of morphant phenotypes is complicated (Bruce Riley, personal communication). The phenotype of the gene trap fish is not complicated by the effects of MOs because these fish were not injected with MOs. Preliminary observations suggest that when Pax8 function is compromised during development in the gene trap line, ear formation is affected and the otic vesicle becomes slightly reduced in size. The observations reported in this thesis, in comparison to the pax8 MO results,
66 suggest a milder phenotypic result of a pax8 knockdown. The more severe phenotype in the MO experiments may be due to the pax8 MO binding to pax2 and pax5 transcripts in addition to pax8 blocking translation of all three protein products. A knockdown of all three Pax proteins (Pax2/5/8) would prevent compensatory action by Pax2 and Pax5, thus leading to a more severe phenotype. In the gene trap line, only the Pax8 protein levels should be affected. It is possible that Pax2 and Pax5 are able to compensate for the loss of Pax8, thus reducing the severity of the otic phenotype observed. Several questions remain: what role does Pax8 play in otic development? A reduced size of the otic vesicle indicates a reduction in otic tissue. In the pax8 morphant, no increase in cell death in the preotic region was detected. Mackereth et al. proposed that Pax8 makes the preplacodal cells competent to respond to otic hindbrain signals; thus a lack of Pax8 results in the inability to induce an otic fate, and possibly less proliferation of otic placode cells occurs as well (Mackereth et al., 2005). There was also a reduced number of hair cells in the pax8 morphant. This raises the question: is the number of hair cells reduced in the pax8 gene trap fish? Because hair cells are very sensitive to movement and sound, it would be expected that when there are fewer hair cells, the responses to these stimuli are impaired. The question remains if there is a behavioral phenotype associated with the abnormally small otic vesicles. If there is a phenotype, is it due to a reduction in hair cell number or another defect? Further, how does the pax8 knockdown affect other ear structures? The ear of the pax8 gene trap fish can be examined at later stages of development to make clear the affect on the mature adult ear. Also, experiments can be performed in adult fish of the gene trap line, leading to new information regarding the consequence, or lack thereof, of a pax8 knockdown. More in depth studies on the pax8 gene trap line may hold the answers to these and other questions.
67 Implications of the pax8 gene trap fish line Heterozygous embryos of the pax8 gene trap line will be useful tools for developmental studies. First, these embryos can be used to investigate the effect of toxins on embryonic development. Comparisons of RFP fluorescence between groups treated with a toxin or suspected toxin and untreated groups may reveal toxic effects on the morphology of developing structures that fluoresce, as well as effects on gene expression in relation to timing of RFP fluorescence or lack of RFP fluorescence. The pax8 gene trap line can also be used in combination with other lines of mutant fish to show the effect of other mutations on development. In particular, if cell specification is affected in a mutant fish line, this may be observed by a shift in the location of RFP fluorescence when the mutant is in a genetic background of the pax8 gene trap line. Also, the ability to render multiple populations of neurons visible, as in the pax8 gene trap fish in the Isl1-GFP background, may provide a useful tool for studies of spinal cord development, injury, and regeneration. Further studies are needed to determine the type of interneurons, the connections that are made, and how this neuron population functions in locomotory behavior. Thousands of zebrafish genes involved in organogenesis and disease have been revealed using mutagenesis screens. (Driever et al., 1996; Haffter et al., 1996; Amsterdam et al., 1999; Wienholds et al., 2002; Amsterdam et al., 2004). The significant knockdown of Pax8 expression may compromise the development and/or function of the systems in which it is expressed, like the CNS, ear, eye, kidney, and thyroid follicles. If a mutant phenotype is discovered in these structures, the pax8 gene trap fish may also serve as a model for disease. Further studies using the homozygous embryos may reveal developmental defects in these regions.
68 Conclusion This is the first report that quantitatively shows the Tol2 gene trap can cause a significant knockdown in expression of a gene. We showed that insertion of the Tol2DsRed in the first intron of the pax8 gene leads to a significant knockdown of pax8 mRNA expression downstream of the insertion. This knockdown is expected to occur in Pax8 protein expression because there is very little pax8 mRNA present for translation into protein. Preliminary observations of homozygous embryos suggest the knockdown of pax8 results in a smaller otic vesicle. These results would agree with previous studies utilizing pax8 MOs to knockdown expression. Further studies need to be completed in order to make a definitive conclusion regarding the otic phenotype in homozygous embryos of the gene trap line. If future studies support these preliminary findings, this would be the first report of a mutant phenotype in a gene trap line of fish.
69 Figure 4-1. Comparison of RFP fluorescence and pax8 mRNA expression. Comparison of RFP fluorescence and pax8 mRNA expression as previously reported by Pfeffer et al., 1998. RFP fluorescence appears after reports of pax8 mRNA expression, as expected. However, the time between pax8 mRNA and visible RFP protein expression differs depending on the structure. The onset of visible fluorescence in the spinal cord is unclear, but occurs sometime between 28 hpf and 2 dpf.
70 LIST OF REFERENCES Amsterdam, A., Burgess, S., Golling, G., Chen, W., Sun, Z., Townsend, K., Farrington, S., Haldi, M. and Hopkins, N. (1999). A large-scale insertional mutagenesis screen in zebrafish. Genes Dev 13, 2713-24. Amsterdam, A. and Hopkins, N. (2006). Mutagenesis strategies in zebrafish for identifying genes involved in development and disease. Trends Genet 22, 473-8. Bayer, T. A. and Campos-Ortega, J. A. (1992). A transgene containing lacZ is expressed in primary sensory neurons in zebrafish. Development 115, 421-6. Bellen, H. J., O'Kane, C. J., Wilson, C., Grossniklaus, U., Pearson, R. K. and Gehring, W. J. (1989). P-element-mediated enhancer detection: a versatile method to study development in Drosophila. Genes Dev 3, 1288-300. Bier, E., Vaessin, H., Shepherd, S., Lee, K., McCall, K., Barbel, S., Ackerman, L., Carretto, R., Uemura, T., Grell, E. et al. (1989). Searching for pattern and mutation in the Drosophila genome with a P-lacZ vector. Genes Dev 3, 1273-87. Bopp, D., Burri, M., Baumgartner, S., Frigerio, G. and Noll, M. (1986). Conservation of a large protein domain in the segmentation gene paired and in functionally related genes of Drosophila. Cell 47, 1033-40. Chalepakis, G., Fritsch, R., Fickenscher, H., Deutsch, U., Goulding, M. and Gruss, P. (1991). The molecular basis of the undulated/Pax-1 mutation. Cell 66, 873-84. Christ, S., Biebel, U. W., Hoidis, S., Friedrichsen, S., Bauer, K. and Smolders, J. W. (2004). Hearing loss in athyroid pax8 knockout mice and effects of thyroxine substitution. Audiol Neurootol 9, 88-106. Dahl, E., Koseki, H. and Balling, R. (1997). Pax genes and organogenesis. Bioessays 19, 755-65. Daniel, L., Lechevallier, E., Giorgi, R., Sichez, H., Zattara-Cannoni, H., Figarella-Branger, D. and Coulange, C. (2001). Pax-2 expression in adult renal tumors. Hum Pathol 32, 282-7. Dorfler, P. and Busslinger, M. (1996). C-terminal activating and inhibitory domains determine the transactivation potential of BSAP (Pax-5), Pax-2 and Pax-8. Embo J 15, 197182. Dressler, G. R. and Douglass, E. C. (1992). Pax-2 is a DNA-binding protein expressed in embryonic kidney and Wilms tumor. Proc Natl Acad Sci U S A 89, 1179-83.
71 Driever, W., Solnica-Krezel, L., Schier, A. F., Neuhauss, S. C., Malicki, J., Stemple, D. L., Stainier, D. Y., Zwartkruis, F., Abdelilah, S., Rangini, Z. et al. (1996). A genetic screen for mutations affecting embryogenesis in zebrafish. Development 123, 3746. Edelman, G. M. and Jones, F. S. (1995). Developmental control of N-CAM expression by Hox and Pax gene products. Philos Trans R Soc Lond B Biol Sci 349, 305-12. Ericson, J., Thor, S., Edlund, T., Jessell, T. M. and Yamada, T. (1992). Early stages of motor neuron differentiation revealed by expression of homeobox gene Islet-1. Science 256, 1555-60. Fadool, J. M., Hartl, D. L. and Dowling, J. E. (1998). Transposition of the mariner element from Drosophila mauritiana in zebrafish. Proc Natl Acad Sci U S A 95, 5182-6. Friedrich, G. and Soriano, P. (1991). Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice. Genes Dev 5, 1513-23. Gaunt, S. J., Miller, J. R., Powell, D. J. and Duboule, D. (1986). Homoeobox gene expression in mouse embryos varies with position by the primitive streak stage. Nature 324, 662-4. Glaser, T., Walton, D. S. and Maas, R. L. (1992). Genomic structure, evolutionary conservation and aniridia mutations in the human PAX6 gene. Nat Genet 2, 232-9. Gnarra, J. R. and Dressler, G. R. (1995). Expression of Pax-2 in human renal cell carcinoma and growth inhibition by antisense oligonucleotides. Cancer Res 55, 4092-8. Gossler, A., Joyner, A. L., Rossant, J. and Skarnes, W. C. (1989). Mouse embryonic stem cells and reporter constructs to detect developmentally regulated genes. Science 244, 463-5. Grunwald, D. J. and Streisinger, G. (1992). Induction of recessive lethal and specific locus mutations in the zebrafish with ethyl nitrosourea. Genet Res 59, 103-16. Gruss, P. and Walther, C. (1992). Pax in development. Cell 69, 719-22. Haddon, C. and Lewis, J. (1996). Early ear development in the embryo of the zebrafish, Danio rerio. J Comp Neurol 365, 113-28. Hans, S., Liu, D. and Westerfield, M. (2004). Pax8 and Pax2a function synergistically in otic specification, downstream of the Foxi1 and Dlx3b transcription factors. Development 131, 5091-102. Heasman, J. (2002). Morpholino oligos: making sense of antisense? Dev Biol 243, 209-14.
72 Higashijima, S., Hotta, Y. and Okamoto, H. (2000). Visualization of cranial motor neurons in live transgenic zebrafish expressing green fluorescent protein under the control of the islet-1 promoter/enhancer. J Neurosci 20, 206-18. Holland, N. D., Holland, L. Z. and Kozmik, Z. (1995). An amphioxus Pax gene, AmphiPax-1, expressed in embryonic endoderm, but not in mesoderm: implications for the evolution of class I paired box genes. Mol Mar Biol Biotechnol 4, 206-14. Hudspeth, A. J. (1992). Hair-bundle mechanics and a model for mechanoelectrical transduction by hair cells. Soc Gen Physiol Ser 47, 357-70. Igarashi, T., Ueda, T., Suzuki, H., Tobe, T., Komiya, A., Ichikawa, T. and Ito, H. (2001). Aberrant expression of Pax-2 mRNA in renal cell carcinoma tissue and parenchyma of the affected kidney. Int J Urol 8, 60-4. Ivics, Z., Hackett, P. B., Plasterk, R. H. and Izsvak, Z. (1997). Molecular reconstruction of Sleeping Beauty, a Tc1-like transposon from fish, and its transposition in human cells. Cell 91, 501-10. Kawakami, K., Koga, A., Hori, H. and Shima, A. (1998). Excision of the tol2 transposable element of the medaka fish, Oryzias latipes, in zebrafish, Danio rerio. Gene 225, 17-22. Kawakami, K. and Shima, A. (1999). Identification of the Tol2 transposase of the medaka fish Oryzias latipes that catalyzes excision of a nonautonomous Tol2 element in zebrafish Danio rerio. Gene 240, 239-44. Kawakami, K., Shima, A. and Kawakami, N. (2000). Identification of a functional transposase of the Tol2 element, an Ac-like element from the Japanese medaka fish, and its transposition in the zebrafish germ lineage. Proc Natl Acad Sci U S A 97, 114038. Kawakami, K., Takeda, H., Kawakami, N., Kobayashi, M., Matsuda, N. and Mishina, M. (2004). A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev Cell 7, 133-44. Krauss, S., Johansen, T., Korzh, V. and Fjose, A. (1991). Expression pattern of zebrafish pax genes suggests a role in early brain regionalization. Nature 353, 267-70. Kroll, T. G., Sarraf, P., Pecciarini, L., Chen, C. J., Mueller, E., Spiegelman, B. M. and Fletcher, J. A. (2000). PAX8-PPARgamma1 fusion oncogene in human thyroid carcinoma [corrected]. Science 289, 1357-60. Lechner, M. S. and Dressler, G. R. (1996). Mapping of Pax-2 transcription activation domains. J Biol Chem 271, 21088-93.
73 Lewis, E. B. (1978). A gene complex controlling segmentation in Drosophila. Nature 276, 56570. Lin, S., Yang, S. and Hopkins, N. (1994). lacZ expression in germline transgenic zebrafish can be detected in living embryos. Dev Biol 161, 77-83. Lui, W. O., Foukakis, T., Liden, J., Thoppe, S. R., Dwight, T., Hoog, A., Zedenius, J., Wallin, G., Reimers, M. and Larsson, C. (2005). Expression profiling reveals a distinct transcription signature in follicular thyroid carcinomas with a PAX8PPAR(gamma) fusion oncogene. Oncogene 24, 1467-76. Mackereth, M. D., Kwak, S. J., Fritz, A. and Riley, B. B. (2005). Zebrafish pax8 is required for otic placode induction and plays a redundant role with Pax2 genes in the maintenance of the otic placode. Development 132, 371-82. Mansour, S. J., Candia, J. M., Matsuura, J. E., Manning, M. C. and Ahn, N. G. (1996). Interdependent domains controlling the enzymatic activity of mitogen-activated protein kinase kinase 1. Biochemistry 35, 15529-36. Mansouri, A., Chowdhury, K. and Gruss, P. (1998). Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19, 87-90. McCallum, C. M., Comai, L., Greene, E. A. and Henikoff, S. (2000). Targeted screening for induced mutations. Nat Biotechnol 18, 455-7. Narlis, M., Grote, D., Gaitan, Y., Boualia, S. K. and Bouchard, M. (2007). Pax2 and pax8 regulate branching morphogenesis and nephron differentiation in the developing kidney. J Am Soc Nephrol 18, 1121-9. Nasevicius, A. and Ekker, S. C. (2000). Effective targeted gene 'knockdown' in zebrafish. Nat Genet 26, 216-20. O'Kane, C. J. and Gehring, W. J. (1987). Detection in situ of genomic regulatory elements in Drosophila. Proc Natl Acad Sci U S A 84, 9123-7. Ono, F., Higashijima, S., Shcherbatko, A., Fetcho, J. R. and Brehm, P. (2001). Paralytic zebrafish lacking acetylcholine receptors fail to localize rapsyn clusters to the synapse. J Neurosci 21, 5439-48. Pfeffer, P. L., Gerster, T., Lun, K., Brand, M. and Busslinger, M. (1998). Characterization of three novel members of the zebrafish Pax2/5/8 family: dependency of Pax5 and Pax8 expression on the Pax2.1 (noi) function. Development 125, 3063-74. Phillips, B. T., Storch, E. M., Lekven, A. C. and Riley, B. B. (2004). A direct role for Fgf but not Wnt in otic placode induction. Development 131, 923-31.
74 Plachov, D., Chowdhury, K., Walther, C., Simon, D., Guenet, J. L. and Gruss, P. (1990). Pax8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110, 643-51. Raz, E., van Luenen, H. G., Schaerringer, B., Plasterk, R. H. and Driever, W. (1998). Transposition of the nematode Caenorhabditis elegans Tc3 element in the zebrafish Danio rerio. Curr Biol 8, 82-8. Robson, E. J., He, S. J. and Eccles, M. R. (2006). A PANorama of PAX genes in cancer and development. Nat Rev Cancer 6, 52-62. Skarnes, W. C., Auerbach, B. A. and Joyner, A. L. (1992). A gene trap approach in mouse embryonic stem cells: the lacZ reported is activated by splicing, reflects endogenous gene expression, and is mutagenic in mice. Genes Dev 6, 903-18. Stoykova, A. and Gruss, P. (1994). Roles of Pax-genes in developing and adult brain as suggested by expression patterns. J Neurosci 14, 1395-412. Stuart, E. T., Kioussi, C. and Gruss, P. (1994). Mammalian Pax genes. Annu Rev Genet 28, 21936. Tell, G., Pellizzari, L., Esposito, G., Pucillo, C., Macchia, P. E., Di Lauro, R. and Damante, G. (1999). Structural defects of a Pax8 mutant that give rise to congenital hypothyroidism. Biochem J 341 ( Pt 1), 89-93. Treisman, J., Gonczy, P., Vashishtha, M., Harris, E. and Desplan, C. (1989). A single amino acid can determine the DNA binding specificity of homeodomain proteins. Cell 59, 553-62. Treisman, J., Harris, E. and Desplan, C. (1991). The paired box encodes a second DNA-binding domain in the paired homeo domain protein. Genes Dev 5, 594-604. Underhill, D. A. and Gros, P. (1997). The paired-domain regulates DNA binding by the homeodomain within the intact Pax-3 protein. J Biol Chem 272, 14175-82. Wendl, T., Lun, K., Mione, M., Favor, J., Brand, M., Wilson, S. W. and Rohr, K. B. (2002). Pax2.1 is required for the development of thyroid follicles in zebrafish. Development 129, 3751-60. Wienholds, E., Schulte-Merker, S., Walderich, B. and Plasterk, R. H. (2002). Target-selected inactivation of the zebrafish rag1 gene. Science 297, 99-102. Wilson, C., Pearson, R. K., Bellen, H. J., O'Kane, C. J., Grossniklaus, U. and Gehring, W. J. (1989). P-element-mediated enhancer detection: an efficient method for isolating and characterizing developmentally regulated genes in Drosophila. Genes Dev 3, 1301-13.
75 Xu, W., Rould, M. A., Jun, S., Desplan, C. and Pabo, C. O. (1995). Crystal structure of a paired domain-DNA complex at 2.5 A resolution reveals structural basis for Pax developmental mutations. Cell 80, 639-50. You, L. and Yin, J. (2000). Patterns of regulation from mRNA and protein time series. Metab Eng 2, 210-7.
76 BIOGRAPHICAL SKETCH Nichole Suzanne Gebhart was born in 1980 in Dayton, Ohio. She is the second of three children and grew up in Germantown, Ohio. She attended Valley View High School in Germantown where she was very active, participating in Varsity volleyball and softball, the National Honor Society, and the Student Council. Nichole graduated ninth in her class in 1998 and went on to Denison University in Granville, Ohio. At Denison, she was a member of the Varsity softball team and Kappa Kappa Gamma sorority. She spent her senior year commuting from Granville to Columbus, Ohio to conduct research at the Center for Molecular Neurobiology at the Ohio State University with Dr. John Oberdick. She reported the results of her research in her senior thesis titled, Â“Behavioral Effects of Protein Interactions in Purkinje Cells of the Cerebellum.Â” Nichole graduated from Denison University in 2002 with a B.A. in biology and a minor in art history. After graduation, Nichole worked as a full time research assistant with Dr. Oberdick at the Ohio State University. She returned to school at the University of Florida College of Medicine as a student of the Interdisciplinary Program in Biomedical Sciences (IDP) two years later. In May of 2005, she joined the lab of Dr. Fumihito Ono at The Whitney Laboratory for Marine Bioscience in St. Augustine, Florida to conduct research for her dissertation project. Despite a family tragedy in August of 2005, Nichole continued to pursue her degree and completed the required coursework in the advanced IDP concentration of neuroscience. She passed her Ph.D. candidacy exam in December 2006. In May 2007, her advisor, Dr. Ono, left the Whitney Lab and University of Florida for a position with the National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health. Nichole chose to stay at the Whitney Lab and University of Florida to continue working toward her Ph.D. However, in August 2007, she switched from the Ph.D. track to M.S. track in the IDP. At that time, she had more than
77 enough data to write her thesis. Nichole successfully defended her Masters thesis in October 2007. Nichole graduated in December 2007 from the University of Florida College of Medicine with a Master of Science Degree in medical sciences.