From DNA to Protein a study of genomic instability candidate gen

CHAPTER ONE

INTRODUCTION

Zebrafish as a Model System

Zebrafish (Danio rerio), are a species of small minnow native to the

freshwater streams of Northern and Central India. Zebrafish belong to the

cyprinid family of teleost fish, which includes several other popular model

organisms such as medaka, pufferfish and the three-spined stickleback. These

organisms are often used in genetic and behavioral studies. Over the last

several decades, the zebrafish has gained popularity as a model organism in a

variety of backgrounds including genetics and developmental biology (Eisen

1996; Dooley and Zon 2000; Grunwald and Eisen 2002).

More recently, zebrafish have become a widely used model system for

studying human diseases, drug-targeting studies, toxicity screening, sleep and

anxiety studies, neurodegenerative and aggression disorders (Eisen 1996; Gerlai

et al. 2000; Hendricks et al. 2000; Grunwald and Eisen 2002; Gerlai 2003; Bilotta

et al. 2004; Kari et al. 2007; Feitsma and Cuppen 2008; Guo 2008; Ingham 2009;

Norton and Bally-Cuif 2010). Researchers have been using zebrafish to study

behavioral, anxiety, sleep and learning disorders because the regulatory

processes that underlie behavior in zebrafish and mammals is highly conserved.

The zebrafish has also become extremely common in cancer research, drug

discovery and aging disease studies because of the similarity of tissue histology

between fish and humans (Amatruda et al. 2002; Gerhard 2003; Rubinstein

2003; Zon and Peterson 2005). Using zebrafish gives researchers the

advantage of directly testing a mutation in a whole animal background rather

than a single cell. Additionally, using zebrafish in chemical screens allows

researchers to understand the toxicity and targets of potential therapeutics before

clinical trials are performed. The techniques developed over the last 30 years for

zebrafish researchers have made all this possible.

The basic genetics and molecular biology methods can easily be applied

in zebrafish research including PCR, cloning, gene-mapping, in situ hybridization

and mutagenesis. Histological methods have also been important in zebrafish

research since it is relatively easy to prepare paraffin sections of adult and

embryonic fish, resin mounts and sections, and even cell-specific staining in the

sections. Initial RNAi approaches in zebrafish were not very promising; however,

the development of microinjection techniques have allowed reverse genetics or

gene knockdown studies to be carried out using morpholino oligonucleotides

(MOs) (Nasevicius and Ekker 2000; Ekker and Larson 2001; Grunwald and Eisen

2002; Ingham 2009). MOs are synthetic molecules, usually 25 bases in length,

which bind complementary sequences of RNA by standard base pairing.

Structurally MOs are very similar to DNA since MOs have standard bases; the

difference is that in MOs, the bases are bound to a morpholine ring instead of

deoxyribose rings and are linked through phosphorodiamidate groups instead of

phosphates (Robu et al. 2007; Moulton 2007).

The importance of the MO chemistry is that unlike typical RNAi methods,

MOs do not degrade their target RNA molecules but rather act by steric blocking,

reducing the interaction of the target RNA with other molecules. The reduction in

gene expression using this technique is only effective for the first 4-5 days of

development, but allows for direct targeting of developmental genes (as

reviewed by Gerhard 2003). Microinjections have also aided in transplantation

and cell-labeling studies and development of transgenic zebrafish systems.

Although other methods of introducing foreign DNA into zebrafish have been

used including electroporation of fertilized eggs, particle gun bombardment,

liposome-mediated gene transfer and sperm-mediated gene transfer, the highest

survival rate and method of choice continues to be microinjection (Lele and

Krone 1996).

Teleost Evolution: Despite having more than 300 million years separating

the last common ancestor of fish and humans, the overall biology between the

two are strikingly similar (Postlethwait et al. 1999; Gerhard 2003). As a

vertebrate, zebrafish possess many anatomical and functional features that are

similar to humans (as reviewed by Gerhard 2003). Analysis of the zebrafish

genome, particularly focusing on Hox gene clusters, suggests that during the

course of vertebrate evolution, the zebrafish and human lineages have shared

two rounds of whole genome duplication prior to a third teleost-specific

duplication event (Postlethwait et al. 1994; Postlethwait et al. 1999; Postlethwait

et al. 1998). As a result of the genome duplication events, zebrafish often have

two functional homologs of the mammalian gene equivalent in which the function

of the ancestral gene is now divided between two genes. For this reason, each

zebrafish homolog has a more restricted function, often appearing as a

divergence of spatial and temporal gene expression (Carroll, Grenier and

Weatherbee 2000). Unfortunately, gene isolation and analysis in zebrafish tends

to be more difficult because of the genome duplication, but in many cases the

presence of two orthologs of a mammalian gene provides an advantage for

developmental genetic analysis by providing the opportunity to investigate gene

functions that may otherwise be obscured by other functions of those same

genes in mammalian models (Postlethwait et al. 1998; Postlethwait et al. 1999,

Carroll, Grenier and Weatherbee 2000).

Aside from the duplication event in teleosts, when compared to the human

genome sequence, the zebrafish genome demonstrates conservation of

functional domains, developmental pathways, syntenic genes, cell cycle and

tumor suppressor genes, allowing the zebrafish to be an ideal model organism

for the identification of genes and pathways involved in human diseases (Shin

and Fishman 2002; Rubinstein 2003; Lieschke and Currie 2007). Additionally the

general features of zebrafish during development through adulthood and its basic

anatomy enable it to be a useful model in other areas of research as well.

Zebrafish Development, Life Cycle, and Physical Characteristics:

Fertilization and development of zebrafish occurs externally and the optical clarity

allows for easy manipulation of individual cells or embryos for cell-fate

determination or cell-labeling studies (Gerhard 2003; Eisen and Smith 2008).

This also enables researchers to identify mutants based on phenotypes, specific

behaviors or lethality simply under a microscope. This has led to the genetic

identification of several developmental disorders and increased the usage of cell

signaling assays originally developed for use with Xenopus larvae. Other useful

advantages at the early stages of zebrafish development are that the

developmental process occurs relatively synchronously among large clutches of

eggs.

Zebrafish embryogenesis is a relatively rapid process. During the first 24

hours of development, zebrafish embryos are completely transparent making

them easy to visualize and identify the development of organs and limbs (Figure

1.1A) (Eisen 1996; Grunwald and Eisen 2002). Pigmentation can be seen by 48

hours post fertilization (hpf), which helps with genetic screens exploiting pigment

mutants such as golden and albino. By this time, zebrafish embryos display the

vertebrate-specific body plan and most organs are fully developed (Figure 1.1B).

At 72-hpf, most larvae will hatch from their chorions and swim freely (Figure

1.1C). At approximately six months of age the fish are sexually mature and

ready to breed. The adult fish reach a size of 3-4 centimeters in length; females

are often slightly larger than the males (Kimmel et al. 1995; Grunwald and Eisen

2002; Volff 2005). Adult zebrafish have a silver body with seven to nine bluish-

black horizontal stripes that are present from dorsal to caudal fins. Male

zebrafish also have a gold hue between the stripes.

Figure 1.1: Images of wildtype zebrafish embryos (A) prim-6 or approximately 25-hpf (B) long-

pec or 48-hpf (C) protruding mouth or 72-hpf (Adapted from Kimmel et al. 2005).

The breeding and aquaculture of zebrafish is relatively easy. Breeding

pairs are set up in the afternoon before the fish are fed. Depending on the

purpose of the breeding, stocks, collection or injections, the breeding tank set up

has various options. When breeding for stocks, several males and females

(group breeding) may be placed in the breeding tank and left overnight until the

following day. This allows for a heartier stock by having multiple males fertilize

eggs from various females. Another option that is particularly useful for

identifying carriers of certain genes and other mutant studies is having a single

male and a single female in a breeding tank. In other cases when embryos need

to be of specific ages for and experiment, a divider is used to separate the male

and female fish. The dividers are then pulled at specific times the following

morning and eggs are collected within a few minutes of deposition.

Streisinger and the Age of Zebrafish Genetics

George Streisinger was a molecular biologist who worked on transcription

and protein synthesis in the T4 phage (Tsugita et al. 1968). Although he was

quite established in this field, he wanted to find a vertebrate model that could be

used to continue his studies. According to several published articles, Streisinger

had a fondness for tropical fish and attempted to use several different species

from pet stores in his lab prior to settling on zebrafish (Eisen 1996; Grunwald and

Eisen 2002). In the early 1980s, breakthrough research was done when

Streisinger and associates applied mutational analysis to study embryonic

development, establishing a method to activate the development of zebrafish

without genomic contribution from sperm, giving rise to haploid embryos.

Streisinger reasoned that the zebrafish could be used in carcinogenicity and

toxicology testing of chemical and environmental agents, and could serve as a

model to establish quantitative dose-response relationships of carcinogenic

exposures in vivo (Streisinger 1984; Walker and Streisinger 1983). In order to

test for carcinogenicity, Streisinger developed a rapid genotoxicity test based on

the current experiments using mice that exploited the zebrafish pigment

mutation, golden. The golden heterozygous zebrafish were exposed to various

mutagens, including gamma rays, ethyl methanesulfonate (EMS) and ethyl-

nitrosourea (ENU) during the early stages of embryogenesis. These exposures

resulted in the appearance of mosaic patches of golden and wildtype

pigmentation in the retinal-pigmented epithelium (RPE) at 48- to 72-hpf; this

result suggested that the mutagens effectively targeted the early blastula cells

and caused genotoxic effects (Streisinger 1984). Additionally, it was concluded

that the golden cell patches arose from mutation, somatic crossing over, or loss

of the chromosome containing the golden locus (Streisinger 1984). Later, it was

found that exposure to EMS causes a large portion of chromosome breaks and

induces a higher percentage of mosaics than ENU, which causes point

mutations. These findings allowed for direct comparison of mutation frequency

and tumor development, thereby increasing the significance of using zebrafish in

these types of studies (Streisinger 1984; Currie 1996; Grunwald and Streisinger

1992). Streisinger’s lab also laid the foundation for such techniques as the ability

to breed zebrafish by in vitro fertilization and generate fully homozygous diploid

embryos using UV-inactivated sperm. Before his death in 1984, George

Streisinger showed that zebrafish could be used in classical forward genetic

screens, mutation-based genetic analysis (golden) and gynogenetic assays.

Thus, this body of work revolutionized the use of the zebrafish model system for

studying development (embryology) and genetics.

Genomic Instability Mutants in Zebrafish

Genomic instability refers to a wide range of genetic alterations from point

mutations to chromosomal anomalies leading to the disruption of the integrity of

an organism’s genetic material. The causes of genomic instability also vary from

replication errors, loss of repair mechanisms, improper chromosome segregation

and cell cycle checkpoint errors. It is now known that genomic instability is a

major contributor of diseases including cancer (Amatruda et al. 2002; Beckman

and Loeb 2005).

The ability to perform genetic screens in zebrafish using the mosaic eye

assay developed by Streisinger allows for the detection of mutant genes that lead

to increased frequencies of somatic mutations and a cancer predisposition.

Zebrafish genomic instability (gin) mutations can be induced by exposure to the

point mutagen ethyl-nitro-urea (ENU) and quickly detected by the mosaic eye

assay, which uses the golden locus on chromosome 18 to measure the somatic

loss of gene function (Moore et al. 2006). Normally at 48-hpf, wildtype and

golden heterozygous zebrafish embryos exhibit black pigmentation in the retinal-

pigmented epithelium (RPE) of the eye and the melanophores (Figure 1.2).

Embryos that are homozygous for the golden gene have a much lighter golden-

brown pigmentation; these variations can be easily distinguished under a

dissecting microscope. In order to detect the somatic mutation or gin phenotype,

golden heterozygous embryos are screened for patches of lightly pigmented RPE

cells; these patches indicate the loss-of-function of the wildtype allele in the cells

and therefore verify the mosaic.

Figure 1.2: Eye pigmentation in the zebrafish embryo. Three different 72-hpf zebrafish embryos

are shown above. (A) The embryo at the top of the photo with the gin/gin; gol/+ genotype

displays the classic multiple patch mosaic eye phenotype, due to the loss of function of the

wildtype golden allele in the lighter patches of cells. (B) The gol/gol (golden homozygous)

embryo (lower left) has the characteristic light golden-brown eye color known to this pigment

mutant. (C) A dark black color is visible in the +/+ (wildtype) or gol/+ (golden heterozygous)

embryo in the lower right portion of the photo. Photo adapted from Moore et al. 2006.

Twelve gin mutant lines were generated by ENU mutagenesis followed by

half-tetrad mapping that involved producing gynogenetic embryos from a cross

by activating eggs with irradiated sperm and applying pressure to inhibit the

second meiotic division (Moore et al. 2006). The goal of this mutant screen was

to identify new genes that may be important in maintaining genomic stability.

The mutants show a strong linkage to a centromeric marker on the same

chromosome independent of its map position. Genetic experiments with the gin

mutants indicated that they exhibited both maternal and zygotic gene expression

in the developing zebrafish embryo (Moore et al. 2006). These zebrafish

experiments proved to be the first direct genetic screen in a vertebrate for

somatic mutations based on a locus-specific assay (Moore, Gestl and Cheng

2004; Moore et al. 2006).

Consistent with the role of these genes in genomic instability, tumors

spontaneously arose in adult carriers of the 12 gin mutations. However, carriers

of the gin-10 mutation showed a strong phenotype and became the focus of a

survey to determine the frequency and variety of cancer in these fish. Although

all of the gin carriers displayed a predisposition to malignant peripheral nerve

sheath tumors, gin-10 carriers also had approximately a 10-fold increase in the

incidence in tumor formation when compared to wild type fish, and often had an

earlier appearance of tumors as well. PCR-based LOH analysis was performed

for several tumor samples, many were from gin-10 fish, and the loss of a parental

allele for a marker within the gin-10 region had been lost from the tumor tissue

itself, while it was still apparent in the normal tissue samples from the same fish.

Additional analysis on the gin-10 region in zebrafish tumors suggested that

recombination or a regional deletion was the cause of LOH (Moore et al. 2006).

Potential Candidate Genes for gin-10 Mutations

As a first step in identifying the gin-mutant genes, chromosome

assignments were made for the 12 gin mutants based on preliminary mapping

experiments. Most of the gin-mutants were confirmed using adjacent markers.

Three of the mutants, gin-5, -9, and -10 were all mapped to chromosome 18

where the golden locus is also found.

The gin-10 mutant on chromosome 18 was mapped between bases

14,738,066 and 15,810,769 (Figure 1.3). Using this mapping data and the recent

Sanger database information, a list of potential gin-10 candidate genes was

made. An overview of this region is shown in Figure 1.3. Several of these gin-10

candidates were explored as part of this dissertation project and are the focus of

Chapter Two. Although the focus for the gin-10 candidates was specifically on

mapk12, rfx4, sir2, and synbl, several other genes within the region were also

cloned and analyzed by RT-PCR. The rationale for the role of these genes in

genomic instability is discussed below.

Figure 1.3: Overview of the zebrafish chromosome 18 region were gin-10 was initially mapped in

previous studies. Although there are several additional genes within this region (based on the

Ensemble zv8 database), this diagram shows the key genes (bold) identified with potential for

genomic instability activity.

Mapk12: The evolutionary conserved mitogen activated protein kinase

(MapK) family is involved in diverse cellular processes including growth,

proliferation, differentiation, survival, inflammatory response and development.

Using sequential phosphorylation events, MapKs can transmit signals from the

cells’ environment to the nucleus and elicit an appropriate cellular response.

These phosphorylation events occur on conserved domain (Thr-XXX-Tyr), which

leads to the activation and localization of the protein. There are three major

subclasses of the MapK family: the extracellular signal related kinases (ERK), c-

jun amino-terminal kinases (JNK), and the p38 MapKs. The middle amino acid

residue of the dual-phosphorylation domain determines these subclasses;

generally ERKs have a TEY motif, JNKs have a TPY motif and the p38 MapKs

have a TGY motif (Krens et al. 2006).

Figure 1.4: Schematic of the MapK family pathways showing initiating factors and the cascade

of downstream targets (Adapted from Krens et al. 2006).

Several members of the MAPK family have been studied intensely

because of their association with a number of diseases such as cancers,

autoimmune diseases, and developmental abnormalities. For example, the Ras-

MAPK (Erk1/2) pathway has been observed as one of the most oncogenic

pathways in a variety of cancers, leading to uncontrollable cell proliferation by

activating c-myc, c-jun, elk-1, and other transcription factors that promote DNA

synthesis and cell growth. For this reason, ERK pathways are often targets for

anti-cancer therapeutics and research (Tang et al. 2005).

This family has also been shown to be important during vertebrate

development by regulating mesoderm induction and neuronal differentiation.

Examples of this have been documented in the human prostatic PC12 cell line

where a transient activation of the ERK pathway by EGF leads to proliferation

whereas a sustained activation of this pathway in response to nerve growth

factor causes differentiation (Krens et al. 2006). Activation by T- or B-cell

receptors can also cause cell differentiation into platelet precursors.

p38γ (Mapk12) protein has been shown to be highly expressed in several

human malignant cell lines, which may indicate a potential role in tumorigenesis

(Tang et al. 2005). Several studies have revealed that the ERK/MAPK pathway

activation is sufficient in transforming NIH 3T3 cells; furthermore Ras

transformation is also dependant on the JNK pathway (Tang et al. 2005).

Activation of p38 MAPKs is thought to inhibit Ras-induced cell proliferation in NIH

3T3 cells, suppress Ras transformation and induce K-Ras-dependent cell death

in human colon cancers (Tang et al. 2005). Recent studies based on the p38

family of MAPKs have shown that K-Ras is an activator of p38γ; it has the ability

to increase the expression of p38γ without phosphorylation. Consequently, this

study also provided evidence that increased expression of p38γ was a

requirement for K-Ras transformation by a new mechanism involving a complex

with other ERK proteins (Tang et al. 2005).

Figure 1.5: The proposed model showing the requirement of p38γ for K-Ras transformation and

the inhibitory response from p38α (Adapted from Tang et al 2005).

Additionally, JNK and p38 MAPKs have been shown to modulate cellular

responses to a variety of extracellular signals such as mitogens, inflammatory

cytokines and UV irradiation; often these MAPKs are involved in promoting

programmed cell death when activated (Krens et al. 2006). Interestingly, UV

irradiation specifically activates the p38 MAPK pathway in many organisms,

although it appears that each isoform is independent or has distinct biological

roles (Wang et al. 2000; Krens et al. 2006). Studies have shown that all p38

isoforms can activate the transcription factor-activating factor 2 (ATF2) in vitro,

and are involved in cell cycle arrest at particular cell cycle checkpoints. Wang et

al. (2000) demonstrated that MKK6 and all p38 isoforms are activated by γ

irradiation. Activation of this pathway is ATM-dependent and sufficient to arrest

cells in G2 following irradiation; additionally, the inhibition of MKK6 and or

specifically p38γ disrupts this checkpoint (Wang et al. 2000).

Figure 1.6: The MKK6-p38γ pathway is involved in the G2-M DNA damage checkpoint (Adapted

from Wang et al. 2000).

The mapk12 gene was initially found by screening human cDNA using a

rat erk-3 gene. Interestingly, mapk12 was isolated from a human skeletal muscle

cDNA library and was later confirmed to be identical to sapk3. From that finding

it appeared that mapk12 functioned in the differentiation of myoblasts. In

Xenopus, MAPK12 or p38γ has been shown to phosphorylate Cdc25C and is

important for G2/M progression of oocytes (Perdiguero et al. 2003). Additionally,

data from Wang et al. (2000) supports the model of an important interplay

between p38γ and the G2 cell cycle checkpoint control.

Rfx4: Regulatory factor X (RFX) proteins are evolutionarily conserved

transcription factors among S. cerevisiae, S. pombe, C. elegans, zebrafish,

mouse and humans, that possess a winged helix DNA binding motif (Morotomi-

Yano et al. 2002). These proteins appear to function in a variety of unrelated

systems including regulation of mitotic cell cycle in yeast, mammalian immune

response, brain development and brain-specific diseases and testes

development in mammals (Emery et al. 1996; Mach et al. 1996; Morotomi-Yano

et al. 2002; Zhang, Zeldin and Blackshear 2007).

RFX1, RFX2, and RFX3 were first identified in human and mouse and

classified as a family based on the 76-amino acid DNA-binding domain; RFX4

was found later fused to the estrogen receptor in two aberrant cDNA clones from

human breast tumor tissue while other RFX proteins have just been recently

identified (Emery et al. 1996; Matsushita et al. 2005). Several alternative splice

variants of RFX4 have been reported in the testis and brain and may function

during both morphogenesis and disease formation (Blackshear et al. 2003;

Matsushita et al. 2005). More recently, Matsushita et al. (2005) have shown that

there are several human isoforms of RFX4 and they are often overexpressed in

gliomas but are not detectable in other cancers including liver, colon and

stomach. The function of rfx4 in zebrafish development and disease has yet to

be explored.

Rpc2/Polr3b: RNA polymerase III (Pol III) is a 17-subunit complex that is

responsible for the transcription of various small non-coding and nuclear RNAs in

eukaryotes (Yee et al. 2007). One of the largest subunits, Rpc2, is highly

homologous to its Pol I and Pol II counterparts, and the gene happens to lie

within the gin-10 region in zebrafish. Previous studies analyzing the structure of

Pol III as well as yeast two-hybrid and biochemical investigations have identified

its unique functions including high processivity, efficient transcription termination

and recycling, RNA 3’ cleavage activity, and interaction with diverse promoters to

specific individual subunits (Yee et al. 2007). Additionally, mutational analysis in

yeast have shown that intact Pol III is essential for cell growth; reduced Pol III

function are broad including disruption of protein synthesis, incomplete ribosome

biogenesis, mRNA splicing defects and defects in membrane targeting for newly

translated proteins. Studies in human cell lines have also implicated the roles of

known oncogenes and tumor suppressors, including c-myc and Rb, in controlling

the interactions of transcription factors that bring the Pol III complex to the

promoters of its target genes (White 2005; Yee et al. 2007). Deregulated Pol III

activity has been shown to be a common feature of tumorigenic cells in culture;

however, there is little evidence that it is elevated Pol III activity. Determining the

role of rpc2 in zebrafish tumors will increase the understanding of human cancer.

Sir2: Sirtuins (sir2) are NAD-dependent deacetylases that are found in a

variety of organisms from bacteria to humans. They were originally found in

yeast and have been shown to act in transcriptional repression, recombination,

cellular division, microtubule organization, cellular responses to DNA damage

and aging (Buck, Gallo and Smith 2004; North and Verdin 2004). The sirtuin

family has a unique catalytic domain characterized by its requirement for NAD as

a cofactor (Blander and Guarente 2004; North and Verdin 2004). One of the

main functions of this protein family is the regulation of transcriptional repression

through binding a multiprotein complex. Sequence-specific DNA binding proteins

mediate the initial recruitment of sirtuin protein complexes to telomeres and other

loci for transcriptional repression (North and Verdin 2004).

There are several types of sirtuins found in yeast, mammals, and

zebrafish. Sirt1 has been implicated in the repair of DNA damage by negatively

regulating the p53 pathway; it has been shown to deacetylate p53, which is

initially acetylated at two lysine residues in response to DNA damage in order to

activate it (Sakaguchi et al. 1998; North and Verdin 2004). The human Sirt2

protein has been shown to localize in the cytoplasm and is involved in the

regulation of the microtubule network by deacetylating lysines of α-tubulin.

Recent studies have also shown that Sirt2 is upregulated prior to mitosis and is

potentially involved in cell-cycle regulation (Blander and Guarente 2004; North

and Verdin 2004). Additionally, recent proteomics research has provided a role

for Sirt2 in cancer pathogenesis since the gene is located in a region of frequent

chromosomal deletions in human gliomas, indicating that Sirt2 may act as a

tumor suppressor (North and Verdin 2004). Mammalian Sirt3 has been shown to

localize in the mitochondrial matrix mediated by an amphipathic α-helix at its

amino terminus (Buck, Gallo and Smith 2004; North and Verdin 2004). Several

studies have shown that in vitro, Sirt3 has robust histone deacetylation activity;

since it appears to be localized in the mitochondria, the relevance of its HDAC

activity is still not understood (North and Verdin 2004). Recent research has also

shown that sirtuins may play a considerable role in the genetic control of aging

and related diseases. Additionally, the mammalian Sirt4, Sirt5, Sirt6 and Sirt7

also display some histone deacetylation properties; however, that research is still

underway.

Synbl & Ric8a: Asymmetric cell division plays an important role during

embryogenesis by producing daughter cells with distinct differentiation pathways

(Cowenbergs, Spilker and Gotta 2004). Members of the RIC-8 or synembryn

family proteins are known to be key regulators of asymmetric cell division in

invertebrate and vertebrate embryogenesis. Appropriate centrosome positioning

and movement determines the alignment of the mitotic spindle and is an

essential feature of development since it determines the cleavage plane during

cytokinesis (Miller and Rand 2000; Miller et al. 2000). Many studies have been

modeled in C. elegans showing the importance of accurate spindle orientations

for an asymmetric first cell division. The movement and positioning of

centrosomes also mediates nuclear migration in a variety of cells and organisms

(Miller et al. 2000). The machinery that regulates these movements during

development are still not fully understood; however recent studies in C. elegans

and Drosophila are beginning to identify the mechanisms involved.

The ric-8 gene was first identified in a screen of C. elegans mutants that

were resistant to inhibitors of cholinesterase and defective in vesicle priming

(Miller et al. 2000; Miller and Rand 2000; Cowenbergs, Spilker and Gotta 2004).

Subsequently ric-8 has been identified as a required cytoplasmic protein for G-

protein signaling in the C. elegans nervous system and more recently as an

important part of the machinery that regulates the mitotic spindle, nuclear

migration and other centrosome-mediated events during early embryogenesis

(Miller et al. 2000; Miller 2000). RIC-8 proteins are G-protein positive regulators,

which exhibit G-protein coupled receptor-independent guanine nucleotide

exchange activity for Gα subunits. These proteins act downstream of

partitioning-defective (PAR) proteins, which contributes to spindle and

centrosome anchoring events (Wilkie and Kinch 2005).

Figure 1.7: Schematic of the Ric8 mediated asymmetric cell division in C. elegans zygotes

(Adapted from Wilkie & Kinch 2005).

Other studies in C. elegans, Drosophila and in mammals have also shown

that Ric-8 proteins are required for signaling during synaptic transmission and

thus seem to also be involved in receptor-dependent signaling (Hampoelz et al.

2005). Ric-8 mutants have been created in C. elegans, Drosophila, and in

zebrafish; all homozygous ric-8 mutants are lethal but studies have shown in all

three models that they can be rescued by injection of a transgene that covers the

ric-8 genomic locus (Hampoelz et al. 2005). In addition to the role in

development, dysregulation of asymmetric cell division has recently been shown

to play an important role in cancer.

In Xenopus, mouse and other mammals, there are two ric-8 homologs: ric-

8A (synembryn-a) and ric-8B (synembryn-b), while there is only one ric8 gene in

C. elegans that gives rise to two proteins by alternative splicing. The zebrafish

synembryn-like gene (synbl) is a ric-8b homolog and is located on chromosome

18 within the interval thought to contain the gene responsible for the ENU-

induced gin-10 genomic instability mutation. The zebrafish ric-8a gene is located

on chromosome 25 and appears to have at least three protein-coding transcripts.

The role of these different ric-8 transcripts during zebrafish development is not

clear.

Mdm1 as a gin-12 Candidate Gene

Similar to gin-10, the ENU-induced genomic instability mutation, gin-12,

has been shown to cause embryonic somatic mutations in golden heterozygous

zebrafish; this mutant was mapped to an interval on chromosome 4 in zebrafish

in previous studies (Cheng and Moore 1997; Moore, Gestle and Cheng 2004;

Moore et al. 2006). Several genes within the gin-12 region may potentially

display genomic instability activity including mdm1, which was originally cloned

from 3T3DM cells along with mdm2 and mdm3.

The well-studied Mdm2 protein has been shown to cause acentric

chromatin bodies or double minutes (DMs) and mediate the immortalization and

transformation of rat embryo fibroblasts when overexpressed in cooperation with

Ras (Freedman, Wu and Levine 1999). Additionally, Mdm2 has been shown to

negatively regulate the p53 tumor suppressor and is classified as an oncogene.

Since mdm1 was originally identified on DMs with mdm2, it is possible that it also

functions as an oncogene in the p53 pathway, giving rise to genomic instability

(Snyder et al. 1988).

Figure 1.8: Diagram of the autoregulation of p53, Mdm2 and Mdm4 following DNA damage

and/or mitogenic signals in the cell.

Unlike Mdm2, the structure and functional domains of the Mdm1 protein

have yet to be studied. Using the ExPasy Proteomics server (www.expasy.org),

phosphorylation, glycosylation, and myristoylation sites were predicted the known

zebrafish Mdm1 peptides (Figure 1.9, Appendix C). Interestingly, two potential

sumoylation (SUMO) sites were also predicted, which may play and important

role in the cellular localization and stability of the Mdm1 protein. While it does

not appear that Mdm1 shares the same functional domains as the Mdm2 protein

based on the bioinformatics predictions, it remains an interesting candidate for

the gin studies.

Figure 1.9: Predicted functional sites of the zebrafish Mdm1-001 (long) peptide. The ExPasy

Bioinformatics resource portal predicted several Protein Kinase C (PKC) phosphorylation sites,

Casein Kinase II (CK2) phosphorylation sites, five N-myristoylation (MYR) sites, two cAMP/cGMP

dependent protein kinase (cAMP) phosphorylation sites, one Tyrosine Kinase (TYR)

phosphorylation site, and one N-glycosylation site. Additionally, two SUMO sites were predicted

in the Mdm1 peptide sequence, along with a potential Nuclear Localization Signal (NLS) and a

52- amino acid serine-rich region. A larger version of this image can be found in Appendix C.

Not much is known about mdm1 in human, mouse, or zebrafish; however

there have been studies suggesting that it is involved in retinal degenerative

diseases including arrd2 in mouse (Chang et al. 2008). A nonsense mutation in

the mdm1 gene appears to cause a late-onset RPE atrophy and

hypopigmentation, similar to human AMD, in the affected mice; the complete

absence of mdm1 in mice leads to severe retinal degeneration. Further

investigation of the mdm1 gene in zebrafish as a model to knockdown and

overexpress this gene will allow for a better understanding of its function during

development and whether it has any genomic instability activity.

Summary and Specific Aims

Several zebrafish mutations have been found that display increased

frequencies of somatic mutations during embryonic development and cause an

increase in tumor formation during adulthood. These genomic instability mutants

(gin) are heritable, ENU-induced mutations that have been shown previously to

act both maternally and zygotically in the zebrafish embryo (Moore et al. 2006).

The focus of this dissertation project was to study candidate genes for the gin-10

and gin-12 genomic instability mutations.

There are three Specific Aims for this project.

1. Explore and study potential gin-10 candidate genes and determine

whether these candidates have the potential for genomic instability

activity by knocking each down in golden heterozygous embryos.

2. Distinguish the differences in gene expression and developmental

function of the paralogs synbl and ric8a, both of which have been

shown to regulate asymmetric cell division in the development of C.

elegans.

3. Characterize the gin-12 candidate gene mdm1 in early zebrafish

development using transcript expression analysis and morpholino

knockdown technology, as well as analyzing potential protein-protein

interactions.

Chapter Two focuses on gin-10 candidate genes and Specific Aims 1 & 2.

The goal of Specific Aim 1 is to explore the potential candidate genes within the

gin-10 region on chromosome 18 in zebrafish embryos and determine which of

the candidate genes may contribute to genomic instability during development.

The initial prediction was that only 1-2 genes within the gin-10 region would

model genomic instability activity and lead to future research in the

characterization of its function during development and cancer. Specific Aim 2 is

based on a candidate gene found within the gin-10 region, synembryn-like

(synbl), which has been shown to regulate asymmetric cell division in C. elegans.

A homolog was found on chromosome 25, ric8a, which recent studies have

shown that it is often abnormally regulated in certain cancer types. The purpose

of this aim is to characterize the functions of these two homologs in zebrafish by

exploring developmental expression patterns and through the use of transgenic

zebrafish models.

Chapters Three and Four explore Specific Aim 3, which is a study of the

gin-12 candidate gene mdm1. It was initially thought that the mdm1 gene might

play a role in the p53 tumor suppressor pathway by either regulating p53 directly

or as part of an overall regulatory component. Based on the amino acid

sequence of Mdm1 and it’s homology to Mdm2 and Mdm4 (known p53

regulators), it appeared that Mdm1 did not contain the p53 binding sequences

and most likely did not directly bind to the p53 protein. Thus, studies were

designed to determine whether mdm1 functioned in the p53 pathway or

displayed genomic instability activity in the zebrafish embryo.

CHAPTER TWO

AN INVESTIGATION OF THE ZEBRAFISH GIN-10 CANDIDATE GENES

Rationale and Experimental Design

An ethyl-nitrosurea (ENU) mutagenesis experiment led to the discovery of

twelve genomic instability (gin) mutants with the phenotype of increased

frequencies of embryonic somatic mutations (Moore et al. 2006). The gin-10

mutant line showed noticeably increased frequencies of tumor development and

somatic loss of heterozygosity (LOH) based on PCR assays; thus the goal of

these studies was to identify the gene or genes involved in the gin-10 line. In

previous studies, half-tetrad analysis and SSR markers were used to

approximate the gin-10 region on the zebrafish chromosome 18 (Moore et al.

2006).

Table 2.1: Potential gin-10 Candidate Genes on chromosome 18

Sanger Gene

Gene Information

si:dkey103i16.1

PTPRF interacting protein (liprin beta 1)

cry1b

cryptochrome 1b

rpc2/polr3b

RNA pol III subunit B

rfx4

regulatory factor X, 4

synbl

synembryn-like; resistance to inhibitors of cholinesterase B

si:dkey103i16.6

sir2 homolog

btbd11b

btb (POZ) domain containing 11b

mapk12 (ERK6 or p38γ)

mitogen-activated protein kinase 12a

cirh1a

cirrhosis, autosomal recessive 1a (cirhin)

rpb5

RNA pol II subunit 5

Since ENU causes point mutations that can be difficult to map, a

candidate gene approach was used to identify the potential gin-10 mutant genes.

Initially, a list of genes within the gin-10 region was created based on the SSR

mapping data, most current Sanger Ensemble database (www.ensemble.org)

and evaluated based on potential for genomic instability activity using the known

orthology information from NCBI and Sanger databases. The 10 candidate

genes identified are presented in Table 2.1

Since previous genetic studies suggested that the gin carrying mutants

must be expressed both maternally and zygotically during development, the first

step was to determine the candidate gene expression profiles during various

stages of development by RT-PCR. The synbl, mapk12, cry1b, rfx4, and sir2

candidate genes were fully cloned as cDNA from the wildtype (WT(AB)) strain of

zebrafish for future sequence comparison to gin-10 fish. Additionally, expression

profiles of the synbl, rfx4, and sir2 were explored using semi-quantitative RT-

PCR and whole mount in situ hybridization in zebrafish embryos of various ages.

To formally investigate the contribution to genomic instability of synbl, ric8a, rfx4,

and sir2, the genes were knocked down in zebrafish embryos using morpholino

oligonucleotides (MOs) in both WT(AB) and golden heterozygous fish. The use

of MOs in the zebrafish embryo provides information about the biological function

of synbl, ric8a, rfx4, and sir2. Lastly, synbl heterozygous transgenic zebrafish

were obtained from the Zebrafish International Resource Center (ZIRC), in order

to further explore the function of this candidate and the paralog, ric8a, in a

background other than WT(AB).

Cloning and Sequencing Candidate Gene Fragments from WT(AB)

Embryos

The initial phase in these studies was to determine whether the candidate

genes were expressed both maternally and zygotically in WT(AB) embryos. To

do this, zebrafish embryos were collected at various stages of development and

euthanized in order to isolate total RNA. Several pairs of primers were made to

amplify cDNA from each candidate gene (Table 2.1). The primers were designed

specifically to amplify a region of the transcript that spans more than one exon in

the genomic DNA; this allowed for the successful amplification from the RNA

without any trace of genomic DNA contamination.

Candidate gene fragments were amplified using 4-hpf and 24-hpf embryo

RNA and ligated into the pCR4 vector (Invitrogen) for sequencing. The pCR4

vector was used primarily because it allowed for a greater percentage of positive

clones since it contains the TA cloning feature and has a fast ligation time.

Additionally the sequencing promoters are much closer to the PCR insert,

allowing for better sequencing results (Figure 2.1).

Creating a Bank of cDNA Clones from WT(AB) Stocks: It had been

suggested that cDNA libraries of wild type zebrafish genes be created for this

project to provide a stable and consistent source of cDNA. Therefore, this

project began by preparing cDNA stocks from 24-, 48-, 72-hpf, and 7-dpf WT(AB)

embryos using two different commercial kits: Invitrogen’s Superscript III First

Strand Synthesis kit and New England BioLab’s Protoscript II kit. Unfortunately,

several of the candidate genes showed poor amplification from the cDNA stocks

or had numerous, non-specific amplification on the gel photos (Figure 2.2).

These results were seen using both cDNA kits with multiple primer sets for

several candidate genes; however a preliminary direct RT-PCR experiment gave

robust clean amplification of specific candidate gene fragments of interest (Figure

2.3 and Appendix 1). Thus, RT-PCR became the method of choice for cloning

the gin-10 candidate genes.

Figure 2.1: pCR4-TOPO cloning vector (Invitrogen) used to clone fragments of candidate gin-10

genes for sequencing and later probe making for other experiments.

RT-PCR and Cloning: WT(AB) embryos were collected from breeding

tanks and brought over to the lab in system water, transferred to a glass dish and

sorted into petri dishes in groups of 100 embryos or less. Embryos were allowed

to develop under normal conditions in a 27oC incubator until the desired

developmental age was reached, when they were collected into microfuge tubes

in groups of 50 and total RNA was isolated using the Trizol method (Invitrogen).

RT-PCR reactions for each candidate gene were performed several times for

each primer pair initially to determine whether the gene was expressed both

maternally (4-hpf RNA) and zygotically (after 8-hpf). All reactions were run on

the appropriate concentration agarose gel and photographed.

Figure 2.2: PCR performed using the Invitrogen Superscript III First Strand Synthesis kit and

several sets of synbl and rfx4 primer sets. (A) Most of the primer sets shown here amplified

multiple fragments when using the cDNA stocks, except for synbl F2R1. These same primer sets

amplified single, robust bands with direct RT-PCR. Additionally, synbl F4R2 did not amplify with

this method. (B) Primers for the rfx4 gene were also used to amplify fragments from cDNA, and

gave rise to multiple fragments appearing on the gel. Actin is shown as a positive control in both

gel photos.

Figure 2.3: RT-PCR results of several gin-10 candidate genes at 48- and 72-hpf.

Candidate gene fragments were amplified from WT(AB) RNA at various

ages to determine if the genes were expressed both maternally and zygotically in

the embryos. Figure 2.3 demonstrates the zygotic expression of several

candidate genes, cry1b, ps20, btbd11, rfx4, rpb5, and mapk12; this is a

representative experiment with complete RT-PCR data for all of the gin-10

candidates shown in Table 2.2. The initial objective was to narrow down the gin-

10 candidate list by removing genes that were not expressed both maternally and

zygotically. However, it appears that all of the candidates screened had both

maternal and zygotic expression. Before pursuing these candidate genes

further, it was important to search Sanger and NCBI databases for information

regarding the function and orthology of the current list of zebrafish gin-10

candidate genes. By doing so, several of the genes were immediately removed

from the list based on orthologous information in human and mouse studies. For

example, although cry1b is close to one of the gin-10 markers on chromosome

18 in zebrafish, database, and publication searches revealed that it is involved in

circadian rhythms with CLOCK genes and has been regularly studied without

inference to genomic instability activity. Other genes that were removed from the

candidate list included cirh1a, rpb5, and ps20. Full developmental expression

profiles were obtained for the rest of the candidate genes, including rpc2/polr3b,

which was also eventually removed from the candidate list since recent

publications determined that it functioned mainly in the development of the

digestive system and also did not appear to have genomic instability activity (Yee

et al 2007). Table 2.2 shows complete results for RT-PCR data, based upon

individual experiments that can be found in Appendix A.

Table 2.2: Summary of RT-PCR Maternal and Zygotic Expression Results from the gin-10

Candidate Genes on Chromosome 18 in Zebrafish.

Gene/Fragment

4-hpf

hpf

12-

hpf

18-

hpf

24-

hpf

36-

hpf

48-

hpf

72-

hpf

dpf

btbd11

F1R1

F2R2

cirh1a

F1R1

F2R2

cry1b

F1R1

F2R2

F3R3

mapk12

F1R1

F2R2

ps20

F1R1

(long)

F3R3

(short)

rfx4

F1R10

F5R9

F5R13

F6R9

F5R5

F7R6

F1R1

F4R4

rpb5

F1R1

F2R2

rpc2

F1R5

F3R1

F4R6

F5R7

F3R2

F2R3

sir2

F1R1

F2R2

F3R3

synbl

F1R1

F6R3

F3R2

F1R5

(genomic

5'UTR)

F2R1

F3R4

F4R2

F4R4

F1R3

Due to the apparent involvement in asymmetric cell division during

development of C. elegans, the focus of the gin-10 candidates shifted towards

the synbl gene and its homolog ric8a on chromosome 25. The goal was to

continue investigating the candidates that remained on the list including sir2 and

rfx4, but to concentrate particularly on the function of synbl and ric8a since

asymmetric cell division is a known characteristic of many cancer types (Aguilera

and Gomez-Gonzalez 2008).

Figure 2.4: Developmental profile of embryonic gene expression of synbl, ric8a and actin in

WT(AB) zebrafish.

A developmental RT-PCR profile was performed using WT(AB) RNA and

primers for synbl and ric8a; actin is shown as a positive control. As seen in

Figure 2.4, both synbl and ric8a appear to be expressed throughout zebrafish

embryonic development through larval stage day 7. From this data, it appears

that maternal expression of synbl is slightly more robust than its homolog, ric8a.

Noticeably, synbl expression decreases from Mid-Blastula Transition (MBT)

through 12-hpf, then remains robust and constant through 7-dpf. Expression of

ric8a appears relatively low in comparison through approximately 18-hpf, then

increases slightly at 24-hpf and remains constant through 7-dpf. It is important to

note that although actin was used as a positive control, its expression does vary

slightly during development.

Normal Expression of Candidates in Developing Zebrafish Embryos

Semi-Quantitative RT-PCR: In order to further explore the developmental

profiles of the gin-10 candidate genes, a more quantitative approach was

pursued. All of the candidates initially tested by RT-PCR showed consistent

levels of both maternal and zygotic gene expression (Table 2.2, Figure 2.4).

Rather than using qPCR, which is very expensive and unavailable in the

laboratory, it was suggested that the expression of candidate gene transcripts

was assessed by the semi-quantitative RT-PCR technique (Marone et al. 2001;

Livingston personal communication). This method uses a low-cycle RT-PCR in

combination with a partial Southern blotting procedure, allowing for better

visualization of the quantity of a transcript than standard PCR (Marone et al.

2001). For these experiments, DIG-labeled DNA probes for the candidate genes

were used on developmental blots and detected with CSPD substrate (Roche).

Figure 2.5: Semi-quantitative RT-PCR Developmental Profile of several gin-10 candidate genes,

synbl, rfx4 and sir2. ric8a is not a gin-10 candidate; however it is a paralog of synbl making it an

important gene to study and determine what expression and functional differences occur between

the two.

Previous examination of the gene expression of these particular gin-10

candidates resulted in all of the candidates displaying both maternal and zygotic

gene expression (Figure 2.4). In contrast, Figure 2.5 shows a very distinct

difference in the gene expression patterns of synbl, ric8a, rfx4, and sir2.

Particularly, there appears to be an apparent differential expression pattern

between paralogs synbl and ric8a. As demonstrated earlier, synbl has a strong

maternal expression that decreases around MBT, then steadily increases

through 24-hpf. This data suggests that synbl expression is slightly decreased at

48-hpf, peaks at 72-hpf then continues to decrease through larval stage (day 7).

Interestingly, ric8a displays very low expression through 18-hpf, where it begins

to increase until its expression peaks at 72-hpf, then decreases by 7-dpf. Since

these results initially appeared very different from the original RT-PCR

experiments, the semi-quantitative experiments were repeated several times for

verification. By using a more sensitive application of RT-PCR, a variation in gene

expression between these homologs was finally identified.

Similarly, rfx4 and sir2 exhibited robust bands when amplified by the

standard 35-cycle RT-PCR experiments at various ages. The semi-quantitative

data suggests that rfx4 has very little maternal expression while the zygotic

expression peaks by 12-hpf and remains at a constant level through 48-hpf. By

7-dpf, expression of rfx4 in WT(AB) is completely absent. Maternal expression of

sir2 appears to be very robust in comparison with original RT-PCR and some of

the other candidate genes, and then gradually decreases at MBT. From 12- to

24-hpf, there is a gradual increase in the expression level of the sir2 transcript;

by 48-hpf the transcript level appears to decrease again until it is hardly

detectable from 72-hpf through 7-dpf.

Whole-mount in situ Hybridization: This technique was used to investigate

the temporal and spatial expression of specific candidate gene transcripts at

various ages during development. WT(AB) embryos treated with 1-phenyl-2-

thiourea (PTU) to stop pigment formation and golden embryos were utilized for

these experiments. Several DIG-labeled RNA probes were made from the

candidate clone library using the Ambion MaxiScriptT3/T7 kit.

It is important to note that many ages of zebrafish embryos were used for

the in situ hybridization experiments. Due to the complexity of this procedure,

many of the embryos disintegrated during the various washing steps (See

Chapter Six for detailed methods). The results shown here are representative

examples of the data gathered.

Figure 2.6: Whole-mount in situ hybridization of a 48-hpf WT(AB) zebrafish embryo using an

DIG-labeled actin RNA probe. Both sense and antisense probes were made for each gene

tested; actin was used as a positive control since it results in the staining of the entire embryo

when using the antisense RNA probe. Other controls (not shown here) were (1) batches of

embryos that were not subject to an RNA probe but were blocked and stained, and (2) embryos

that were not subject to an RNA probe or Anti-DIG antibody but were stained. Controls were

used to verify that non-specific staining by BM Purple (Roche) would not occur in the

experimental embryos.

Figure 2.7: Whole-mount in situ hybridization of WT(AB) embryos using a DIG-labeled synbl

RNA probe (A) 8-hpf (B) 12-hpf (C) 24-hpf (D) 48-hpf (E) 72-hpf. It is important to note that

embryos were treated with PTU to prevent the development of pigment after 24-hpf.

Figure 2.8: Whole-mount in situ hybridization of WT(AB) embryos using a DIG-labeled ric8a RNA

probe (A) 8-hpf (B) 12-hpf (C) 24-hpf (D) 48-hpf (E) 72-hpf. It is important to note that embryos

were treated with PTU to prevent the development of pigment after 24-hpf.

The expression pattern of synbl in developing zebrafish embryos is shown

in Figure 2.7. These results suggest that the synbl gene is ubiquitously

expressed in the early embryo through approximately 24-hpf, then expression

becomes restricted to the brain as seen in the 48- and 72-hpf embryos. Figure

2.8 shows the in situ results for ric8a in WT(AB) embryos. Although there

appears to be some staining in 8- and 12-hpf embryos, the staining is muted and

not robust, and may be contributed to background staining. The deep purple

stain is apparent in the 24-hpf embryo, and like the synbl homolog, appears to be

ubiquitous at this time point, with some more concentrated staining in the

hindbrain and cerebellum. Likewise, between 48- and 72-hpf, staining for ric8a

appears to become restricted to the brain.

Based on teleost evolution, it was anticipated that the ric8 homologs, synbl

and ric8a, would diverge in their expression and function as a result of gene

duplication events. Thus, the differential expression patterns during zebrafish

development were expected. The data suggests that the temporal expression of

synbl and ric8a are significantly different, as it appears that ric8a is transcribed

well after MBT. Interestingly, both genes appear to be initially expressed

ubiquitously throughout the embryo, and then becoming restricted to the brain

based on in situ data. It is possible that both genes function cooperatively during

development within the central nervous system (CNS). In order to further the

understanding of these two homologs, it became necessary to identify specific

gene function during early development by knocking down each gene with

Morpholino Oligonucleotides (MOs).

Consequently, rfx4 and sir2 were still priority candidates on the gin-10 list.

The functions of these two genes in zebrafish development were still unknown.

Therefore it was important to investigate these genes further using MOs into a

WT(AB) background for functional analysis and golden heterozygotes for

genomic instability analysis.

Testing Candidate Genes for Genomic Instability

In the mosaic eye assay used to identify gin mutations, mosaicism at the

golden locus was seen in gin/gin homozygotes (Streisinger 1984; Moore et al.

2006). Candidate genes were therefore knocked down in golden heterozygous

embryos to determine if they would also cause mosaic eye pigmentation. The

preferred method of gene knockdown in zebrafish embryos is to inject the

antisense oligonucleotides known as morpholinos (MOs). MOs have become the

standard gene knockdown tool in embryonic animal systems. MOs are not

degraded by nucleases in animals, serum, or cells since cellular proteins do not

recognize the structure (backbone). Furthermore, injections of MOs do not

activate Toll-like receptors or activate innate immune responses, or modify the

methylation of DNA. Therefore, the effects of MOs during embryogenesis can be

seen up to several days following the injection.

Morpholino Design: Morpholinos have a higher binding affinity than

equivalent DNA-based antisense oligos, which allows them to target a specific

gene more effectively. Part of the reason for the success of MOs is because

they act by steric blocking as opposed to an RNase H-mediated mRNA

degradation mechanism. In eukaryotes, pre-mRNA is transcribed from the DNA

in the nucleus, introns are spliced out, and then mature mRNA is exported into

the cytoplasm where translation of the peptide product occurs (Figure 2.10A).

MOs can be made to modify or block the splicing process, or block translation

depending on the sequence of the oligo.

The most common MOs used are those that block translation by binding to

the 5’UTR of mRNA (just upstream of the AUG), which allows the MO to interfere

with the progression of the ribosomal initiation complex from the 5’ cap to the

start codon (Figure 2.10B). By preventing translation initiation, the MO effectively

knocks down gene expression; this allows for the function of the gene during

development to be identified.

Figure 2.9: Structure of a Morpholino Oligonucleotide shown in heteroduplex with RNA.

Morpholinos are synthetic molecules, approximately 25 bases in length that bind to

complementary RNA sequences by standard base pairing. Although morpholinos contain the

standard nucleic acid bases (adenine, cytosine, guanine and thymine), they are structurally

different from DNA or RNA since the bases are bound to morpholine rings rather than

deoxyribose or ribose rings. Additionally, phosphorodiamidate groups rather than phosphates

link the morpholine rings, which eliminates ionization; when injected into cells or whole embryos,

morpholinos are uncharged molecules (Adapted from Moulton 2007).

Figure 2.10: Schematic showing normal gene expression in eukaryotes and types of morpholino

targeting. (A) Normal gene expression where mRNA is transcribed and processed in the nucleus

and exported into the cytosol to be translated by ribosomes into protein precursors. (B)

Translation-blocking morpholinos work in the cytosol and target the 5’cap and start codon of the

processed mRNA, which blocks the binding of the translation initiation complex. (C) Splice-

blocking morpholinos target the pre-mRNA in the nucleus and prevent the formation of the splice

lariat or blocks the binding of splice complexes to splice sites. In some cases, these mRNAs are

translated into mis-spliced proteins (shown above) or occasionally be degraded (Adapted from

Moulton 2007).

Additionally, MOs can be used to hinder the pre-mRNA processing steps

by preventing snRNP complexes from binding to target sequences, by blocking

the nucleophilic adenine and preventing the formation of the splice lariat, or by

interfering with the binding of splice regulatory proteins (Figure 2.10C)

(Summerton and Weller 1997; Moulton 2007). While splice-blocking MOs can be

efficient in knocking down gene expression, several other effects can occur

including modified splicing, intron inclusions, and activation of cryptic splice sites.

However, assaying for splice blocking MOs can be conveniently done by RT-

PCR and visualized by agarose gel electrophoresis.

Use of the p53 ATG Morpholino: The downside to using MOs in

developing embryos is that the MO can produce non-specific or off-target effects.

It has been shown that up to 18% of MOs appear to have non-targeted

phenotypes in the central nervous system (CNS) and somite tissues of zebrafish

embryos; this is due to the activation of p53-mediated apoptosis (Robu et al.

2007). These non-target effects can be suppressed by co-injection of the

standard p53 MO along with the experimental MO (Robu et al. 2007).

Optimization of Injection Experiments using the golden Morpholino:

Knockdown experiments began by injecting golden heterozygous embryos

(between 1-8 cell stage) with various concentrations of the golden morpholino

reconstituted in Danieau buffer with 0.1% phenol red dye. The lethality of the

injections and the concentration needed to knockdown the single wildtype

pigment allele was determined, then used as a baseline in the following injections

into WT(AB) embryos to produce the golden phenotype. Additionally, golden

heterozygous and WT(AB) embryos were also injected with Danieau buffer and

0.1% phenol red dye in order to determine if any effects occurred. These

embryos were raised in the nursery and bred, showing that there were no

negative effects of the buffer solution on lethality or fertility of the fish. The data

obtained from these initial experiments was used to standardize the

concentrations used in future morpholino injection experiments and also served

as a positive control since it clearly demonstrated that the morpholino was

effectively getting into the cytosol of the zebrafish embryonic cells.

Experimental Morpholino Injections: Zebrafish embryos were injected with

MOs specific to the target candidate genes and screened for mosaic eyes at 48-

72 hpi. Based on the genetic evidence that gin-10 has maternal activity,

translation-blocking MOs were initially used for knockdown experiments in an

effort to knockout and/or significantly reduce maternal transcripts within the

embryo prior to MBT. The goal of the morpholino injections was to further

evaluate the gin-10 candidate genes and potentially narrow down the list to one

or two genes, Effects of the morpholinos on the golden heterozygous embryos

were verified by injections into WT(AB) embryos.

Table 2.3: Morpholino oligonucleotide sequences against specific gin-10 candidate genes.

Target

Gene

Morpholino

Type

Sequence

Purpose

golden

Translation

Blocker

5’-GCTGGAGAAACACGTCTGTCCTCAT-3’

Positive

Control

p53

Translation

Blocker

5’-GCGCCATTGCTTTGCAAGAATTTG-3’

Standard

rfx4

Translation

Blocker

5’-GCTCTTCCAGCAGCCCACAATGCAT-3’

Experimental

sir2

Translation

Blocker

5’-CTCTGCTCAACCTCGCCTTGCTCAT-3’

Experimental

synbl

Translation

Blocker

5’-CACTCAAACTCATCTCTGAATGATG-3’

Experimental

Splice

Blocker

5’-ACTGTCACTCTCACCTTAT-3’

Experimental

ric8a

Translation

Blocker

5’-TTAAGTCCATTTTCATCGCTGTTCC-3’

Experimental

Morpholinos were designed and ordered from Gene Tools, LLC.

Both WT(AB) and golden heterozygous embryos were injected with

various concentrations of morpholinos that were custom designed to target

specific gin-10 candidate genes (Table 2.2). Initially, embryo survival following

morpholino injections was very low (at 24-hpi) due to neural death and necrosis.

After completing the optimization of golden morpholino injections and co-injecting

embryos with the p53 morpholino, embryo survival increased while neural death

and necrosis was visibly less noticeable, which allowed for the mutant phenotype

to be screened. Controls for each injection session included setting aside

approximately 50-100 not-injected embryos to verify normal development within

the clutch, and injections of the p53 morpholino into embryos as well as dye-only

injections to verify that the phenol red solution that the morpholino was diluted

did not cause any change to normal embryonic development.

Figure 2.11: Representative results of Morpholino Injections into WT(AB) and golden

heterozygous zebrafish embryos. (A) 48-hpf WT(AB) not injected control embryo (B) 48-hpf

golden heterozygous embryo injected with sir2/p53 morpholino (C) 48-hpf WT(AB) embryo

injected with the sir2/p53 morpholino (D) 48-hpf golden heterozygous embryo injected with

rfx4/p53 morpholino (E) 48-hpf WT(AB) embryo injected with rfx4/p53 morpholino (F) 24-hpf

golden heterozygous embryo injected with rfx4/p53 morpholino (G) 48-hpf WT(AB) embryo

injected with ric8a/p53 morpholino (H) 48-hpf WT(AB) embryo injected with synbl(splice)/p53

morpholino (I) 48-hpf WT(AB) embryo injected with ric8a/synbl(splice)/p53 morpholino.

Wildtype embryos injected with sir2/p53 morpholinos displayed a

decrease in metabolic processes, delayed growth, severe pericardial edema and

a 90-100% lethality rate by 48-hpi (Figure 2.11B-C). The same phenotypes were

seen when injected into golden heterozygous embryos. Since lethality was an

issue at 48-hpi, it was not possible to use the mosaic eye assay to test for

genomic instability by observing the RPE. Therefore, a lower dosage was used

in golden heterozygous embryos, which enabled the use of the mosaic eye

assay, as well as having embryos with less severe metabolic defects, although

the defects were still observable. By doing so, the survival of the injected

embryos was increased to approximately 4-dpi. Unfortunately, no mosaics were

found in the zebrafish embryos injected with the sir2 morpholino.

Based on the cloning and morpholino experiments, it appears that the

sirtuin gene within the gin-10 region of zebrafish may be most closely related to

the human sirt3 gene. In humans, SIRT3 is a soluble protein located within the

mitochondrial matrix, and has been implicated in regulating metabolic processes

including adaptive thermogenesis based on overexpression and fasting studies in

mammalian cells (Blander and Guarente 2004; North and Verdin 2004).

Although there are a few studies that suggest SIRT3 is also found within the

nucleus and has some histone deacetylase properties, it is much more likely,

based on the well-known published data and morpholino knockdown results in

zebrafish, that the zebrafish gin-10 candidate, sir2, is involved specifically in

metabolic regulation and not genomic instability activity (Buck, Gallo and Smith

2004).

Embryos injected with the rfx4/p53 morpholinos exhibited delayed growth

through 24-hpi and some CNS abnormalities. Interestingly, these embryos either

had very long, curly tails or short, stubby tails (Figure 2.11D-F). No mosaics

were observed in any of the rfx4 morpholino injected embryos, suggesting that

this gene is not involved in the genomic instability activity seen in the gin-10

mutants.

The synbl gene was the first gin-10 candidate on the priority list due to the

known involvement in asymmetric cell division in C. elegans development (Wilkie

and Kinch 2005). Since there is a paralog on chromosome 25, it was important

to determine whether the knockdown of either of these genes causes mosaic

eyes, and whether they appear to function similarly in the developing zebrafish

embryo. Morpholinos were designed to target both synbl and ric8a, and were

used separately and in combination in WT(AB) and golden heterozygous

embryos.

Embryos injected with ric8a/p53 morpholinos appeared to have defects in

tail development, which may have been caused by non-specific off target effects

of the morpholino. More importantly, there was noticeable defects in the brain,

which most likely led to the observed cerebral edema in injected embryos 24-hpi

and older (Figure 2.11G). Embryos injected with either the synbl/p53

(Translation-blocking) morpholino or synbl splice/p53 (splice-blocking)

morpholino had severe developmental delay through 24-hpi, but interestingly

also had similar defects in brain and CNS development that resulted in embryos

with abnormally developed heads as compared to WT(AB) not-injected embryos

(Figure 2.11H). None of these morpholinos, when injected into golden

heterozygous embryos, resulted in mosaic eyes. When injected in combination

(ric8a/synbl/p53), embryo survival past 48-hpi significantly decreased while the

brain and head development still appeared to be considerably underdeveloped

(Figure 2.11I). This data suggests that synbl and ric8a work cooperatively during

zebrafish embryonic development.

Several publications have suggested that synbl mutants and knockdowns

have tiny, round, smooth melanophores that migrate abnormally during

development, which was the expected phenotype of the synbl morpholino

injections into WT(AB) embryos (Nagayoshi et al. 2008, Amsterdam et al. 2004).

After many injection sessions using various concentrations of synbl and ric8a

morpholinos, this expected phenotype was never observed. Rather, the

morphants were typically small and underdeveloped, with abnormal brain

development while the melanophores appeared to migrate properly with the

wildtype rough-edge appearance. The synbl MOs used for these experiments

were custom designed based on the VEGA annotated sequence information and

appeared to be different than those used in the published methods from

Nagayoshi et al. 2008. Additionally, the 5’ UTR and portions of exon 1 were

difficult to clone and were eventually cloned using genomic DNA from WT(AB)

embryos (Appendix A); morpholinos were designed using both the sequencing

information that was gained as a result of the cloning and the published VEGA

sequence. The differences in the phenotype of the injected embryos may have

been due to the effectiveness of the synbl MOs to appropriately target the synbl

transcripts. In order to determine the role of synbl and its paralog ric8a,

transgenic zebrafish were utilized.

Table 2.4: Summary of Morpholino Results for the gin-10 Candidate Gene Knockdowns

Morpholino

# Injected

Embryos

Survival

at 48-hpi

% Embryos displaying

Abnormal Phenotype

Golden (Control)

407

33.2

45.9

Rfx4

350

66.9

61.5

Rfx4 + p53

1137

80.1

75.4

Sir2 + p53

406

73.6

82.9

Synbl (ATG)

2422

83.4

85.2

Synbl (Splice)

325

60.9

61.1

Synbl (Splice) +

p53

558

82.8

76.6

Ric8a

1649

66.8

89.7

Ric8a + p53

1080

80.9

90.6

Ric8a + Synbl

(Splice) + p53

1235

78.1

93.6

Transgenic synembryn-like (Tg-synbl) Zebrafish from ZIRC

The Zebrafish International Resource Center (ZIRC) in Oregon listed

transgenic zebrafish embryos that were heterozygous for synbl (Tg-synbl), which

were purchased in order to evaluate the effects of the ric8a and synbl morpholino

knockdowns in a background other than wildtype and golden heterozygous.

Since homozygous mutants are lethal within a few hours after fertilization,

morpholino injections into the heterozygotes should provide better insight to the

function of each gene during development. The purchased fish were placed in

the nursery area in 10L tanks until they reached breeding age. Only two males

survived, therefore they were outcrossed with WT(AB) females in order to

maintain the stock.

Morpholino Injections into Tg-synbl Fish: The two male Tg-synbl fish were

bred to various Tg-synbl females for the morpholino injection experiments; this

self-cross should have resulted in wildtype and Tg-synbl heterozygous embryos

since the homozygous mutants are lethal. Immediately, it was recognized that

the 25% lethality by 24-hpf from these crosses did not occur, which was

interesting and unexpected. Morpholino injections using the synbl splice-blocker,

the ric8a morpholino and a combination of both were used in the resultant Tg-

synbl self-cross embryos. The results obtained appeared to be consistent with

the data from prior injections into WT(AB) and golden heterozygous embryos

(Figure 2.12).

Figure 2.12: Morpholino injections into the embryo progeny of the transgenic fish self-cross

breeding purchased from ZIRC. (A) Control, not injected 48-hpf embryo (B) 48-hpf embryo

injected with synbl(splice)/p53 morpholino (C) 48-hpf embryo injected with ric8a/synbl(splice)/p53

morpholino (D) 48-hpf embryo injected with ric8a/p53 morpholino.

PCR Analysis of the Transgenic Insert: Based on the self-cross and

morpholino results in the Tg-synbl fish, ZIRC was contacted for the genotyping

protocol for this particular transgenic line. Primer sequences were obtained from

ZIRC and ordered from IDT (primers were called Tg-synbl F1 and Tg-synbl R1).

Genomic DNA was prepared from fin clippings of all the existing ZIRC fish and

several of the young fish from a self-cross mating (not-injected embryos that

were tanked in the nursery system). According to ZIRC, the primers were

created in such a way that one primer anneals to the viral sequence and the

other anneals to the flanking genomic region, resulting in a 238bp band on an

agarose gel (Figure 2.13A).

As seen in Figure 2.13B, the PCR results suggest that approximately 40%

of the fish from ZIRC had the transgenic insert, revealing that the initial cross was

Tg-synbl/+ x WT(AB); based on the original ZIRC information, it was thought that

all of the purchased fish contained the transgenic insertion (Figure 2.14).

Figure 2.13: Identifying the Transgenic ZIRC fish (A) Schematic showing the transgenic viral

insert (red dotted line) within the gene (black line) at an intron-exon boundary. Primers designed

to specifically genotype the Tg-synbl fish obtained from ZIRC (blue arrows) will result in a PCR

band of 238bp on an agarose gel. (B) PCR results from the genotyping assay of the 20

transgenic fish received from ZIRC. These results suggest that approximately 40% of the fish

carried the transgenic insert, while the rest are wildtype.

More importantly, neither of the males carried the transgenic insert and the initial

outcross to WT(AB) simply produced wildtype fish; the embryos resulting from

self-crosses that were also genotyped did not carry the transgenic insert either.

Preservation of the transgenic line was important, so those females carrying the

transgenic insert were out crossed to robust looking WT(AB) males.

Figure 2.14: In vitro fertilization was performed at ZIRC using frozen sperm from a Tg-synbl

heterozygous male to fertilize a WT(AB) female. This resulted in half of the F1 progeny to be

heterozygous for the transgenic insert while half of the progeny was fully WT(AB).

Verification of the Transgenic Insertion Point: Since the results from the

morpholino injections did not vary from prior injection experiments, the PCR

product from the transgenic analysis was cloned into the pCR4 vector

(Invitrogen) and sent out to Macrogen Inc (S. Korea) for sequencing. Using the

NIH BLAST program, the sequencing results revealed that the transgenic insert

was on chromosome 18 but not actually interfering with the synbl gene; the

results suggested that the insert was actually in the second intron of the cry1b

gene, a former gin-10 candidate gene on the adjacent contig to the synbl gene

(Figure 2.15).

Figure 2.15: Representative results from BLAST analysis of the cloned viral insert and flanking

genomic region of the PCR genotyped ZIRC fish. These results suggest that the insert is actually

within the cry1b gene on an adjacent contig to the synbl gene.

The full viral sequence was obtained from ZIRC and new primers were

designed to amplify the ends of the insert and synbl genomic sequence. Primers

were also designed for cry1b genomic sequence based on the cloning and

sequencing results. Genomic PCR of the viral primers in combination with the

synbl primers never yielded a band on the agarose gels; only one combination of

the viral primer with a cry1b primer yielded a band (Table 2.3). The resultant

PCR product was also cloned and sequenced for verification; the results still

implied that the viral insert was in the second intron of the cry1b gene.

Table 2.5: Results of the Genomic PCR using Primers to Identify the Location of the Transgenic

Insert in WT(AB) and Confirmed Transgenic Fish.

Primer Pair

Results

WT(AB)

Male #1

Confirmed

Transgenic Female

F1/R1

F1/R5

F1/F3

F1/R3

F1/R5

F1/R1

F11/R1

F3/R1

F5/R1

R3/R1

F1/R5

F3/R3

Viral Primers Synbl Primers Cry1b Primers

This was an unfortunate finding in that it did not allow for an expanded

insight into the functional differences between the synbl paralogs, other than they

do not appear to function within the cry1b gene network (based on the

morpholino injections into the transgenic embryos in comparison with the

injections into WT(AB) and golden heterozygous embryos).

Summary of Experimental Results:

• Candidate genes were analyzed for maternal and zygotic activity in WT(AB)

zebrafish embryos by RT-PCR. Since the entire set of candidate genes

initially listed within the gin-10 region showed both maternal and zygotic

expression, the candidate gene list was narrowed down based on the

known orthology information on the Sanger and NCBI databases. This left

synbl, rfx4, and sir2 as the main candidates on the gin-10 list.

• The ric8a gene is a paralog of synbl located on chromosome 25. This gene

was being studied simultaneously with the gin-10 work, in order to

determine if there were any differences in expression and function of the

homologs during early zebrafish development.

• Expression analysis of the four genes (synbl, ric8a, rfx4, and sir2) was

performed using the semi-quantitative RT-PCR protocol and whole-mount

in situ hybridization. It was interesting to find that synbl and ric8a did

appear to have some divergence in the temporal expression, whereas synbl

is expressed throughout early development and ric8a appears to show

activity after 12-hpf. Additionally, both genes appear to be expressed

specifically in the brain and CNS after 24-hpf.

• Morpholino oligonucleotides were used to knockdown individual candidate

gene expression in both WT(AB) embryos and golden heterozygous

embryos in order to determine whether genomic instability activity occurs by

using the mosaic eye assay. Unfortunately, none of the morpholino

knockdowns led to mosaic eyes in the golden heterozygous embryos,

although some general information on the function of the candidate genes

was determined.

• Transgenic fish were obtained to further study the synbl homologs. Since

the injection phenotypes did not vary from the previous experiments and the

self-crosses did not lead to a 25% dead-loss of homozygous mutants, the

integrity of the transgenic insert was questioned. By PCR and cloning

analysis, it was concluded that the insert was only in approximately 40% of

the fish purchased, and the insert was actually affecting the cry1b gene

rather than the synbl gene.

• The conclusions and implications of these results are discussed in Chapter

Five.

CHAPTER THREE

METHODS FOR ANALYZING MDM1 PROTEIN EXPRESSION IN ZEBRAFISH

EMBRYOS

Rationale and Experimental Design

The most interesting gene within the gin-12 candidate region on the

zebrafish chromosome 4 was mdm1. This gene was originally cloned from a

transformed murine cell line, and found to be overexpressed in these cells along

with the known oncogene mdm2 (Snyder et al. 1988). The aim was to determine

whether the Mdm1 protein is involved in the regulation of the tumor suppressor

p53 following DNA damage, or interact with other proteins in that pathway. In

order to explore this hypothesis, it was necessary to first establish protein

expression analysis of Mdm1 and other p53 regulators in zebrafish embryos.

Protein expression analysis and general proteomics experiments are not

common in the current realm of zebrafish research. Therefore, the availability of

antibodies reactive in zebrafish is limited. In order to perform western blots and

possible protein-protein interaction studies, custom anti-Mdm1 antibodies were

produced for these experiments. The custom anti-Mdm1 antibodies were

assayed for specificity and quality before more detailed analysis was carried out.

This chapter describes the experiments designed to optimize protein detection in

zebrafish. These experiments utilize custom and commercially available

antibodies to assess the expression of Mdm1, Mdm2, Mdm4 and p53 protein in

zebrafish as well as in vitro transcription and translation (TNT) expression of

Mdm1, Mdm2, Mdm4 and p53 protein that are used as controls in these and

future experiments.

Optimization of Western Blotting Techniques in Zebrafish Embryos

Preparing Embryo Lysate for Western Blot Analysis: As a first step

towards analyzing protein expression in zebrafish, it was necessary to optimize

the preparation of good quality embryo protein lysates There are very few

published protocols for lysate preparation from zebrafish, since protein work is

relatively rare in this model organism. An issue with generating protein lysates

from embryos is that the chorion must be removed before the lysate can be

produced. Zebrafish embryonic chorions can be removed either manually or

dissolved with a brief pronase treatment. For these experiments, manual

dechorionation was used to prevent exposure of the embryo lysate to

proteinases. It was important that after dechorionation that the yolk also be

removed from the embryos because it contains an abundance of protein that

could impact the sensitivity of antibody detection of the target proteins. Initial

experiments utilized two syringes with 25-G needles for the removal of the yolk

from the embryo. While observing the embryo under a dissecting microscope, it

was held with one syringe while the other was used to peel the yolk away from

the body. Although this technique worked well for removing the yolk, it was time-

consuming and often led to embryo damage and poor lysate quality. Therefore

the process was changed to a technique utilizing a two-buffer system (Link,

Shevchenko and Heisenberg 2006) that proved to be optimal in removing the

yolk without causing damage to the embryo itself. This technique involved

washing embryos in a high salt deyolking buffer, which dissolved the yolks while

shaking the embryos in a microfuge tube (See Chapter Six for full protocol).

After optimizing the deyolking step, it was essential to develop a

successful protocol for preparing embryo lysates. Three different protocols were

obtained and tested for quality and standardization of lysate preparation – Myers

lab protocol, the Look lab protocol and the Zebrafish book protocol (Karlovich et

al. 1993; Maslow personal communication; O’Shea and Westerfield 1993). The

first two protocols exploited the use of RIPA buffer with various protease

inhibitors (Roche). The resulting Ponceau staining of the lysate on a western blot

showed clear protein bands, but was frequently light and inconsistent in clarity

from one sample to the next (Appendix C). The Zebrafish book protocol, which

utilized a Sample buffer containing SDS and β-ME along with sonication,

appeared to result in the most consistent lysate samples from different embryo

samples based on clarity upon transfer and intensity of the bands, when loading

5µl on an SDS-PAGE gel and staining the blot with Ponceau-S (See Chapter

Six).

Standardizing the Western Blotting Technique: Before using any custom

antibodies, commercially available antibodies for zebrafish Actin and p53 were

purchased and used to optimize the western blotting procedure. Actin was used

because of its abundance and ubiquitous expression and p53 was used since

this was potential interacting protein with Mdm1. During these studies, embryo

lysates were resolved on 10% SDS-PAGE gels and blotted as detailed in

Chapter Six. Blots were incubated in either anti-Actin or anti-p53 antibodies at

4oC from one hour to overnight using concentrations recommended by the

suppliers. Primary antibody detection was carried out with the CSPD

chemiluminescent reagent. These experiments were repeated several times for

consistency with each antibody. Representative results for the standardization

experiments can be seen in Figure 3.1.

Figure 3.1: Western blots detecting β-Actin and p53 protein in WT(AB) zebrafish embryo lysate.

(A) Two sets of lysate samples for 48-hpf embryos and 72-hpf embryos were run on a 10% SDS-

PAGE gel. A polyclonal rabbit anti-Actin antibody was purchased from Cell Signaling Technology

and used at a dilution of 1:1000, and detected with CSPD Chemiluminescence substrate on X-

Ray film. The β-Actin protein appeared at a molecular weight of 45kDa. (B) 24-hpf and 48-hpf

zebrafish embryo lysates were run on a 10% SDS-PAGE gel. A polyclonal rabbit anti-Zebrafish

Anti-p53 antibody that exclusively reacts to the zebrafish C-terminal protein sequence was

purchased from Anaspec and used at a dilution of 1:1000 for these experiments. An

immunoreactive band at approximately 53kDa that corresponds to the p53 protein is visible in the

embryo lysate.

Western Blots using Custom Mdm1 Antibodies: Custom Mdm1 antibodies

were purchased from 21st Century Biochemicals to be used for these and future

protein experiments. Two epitope sites were chosen in the Mdm1 peptide

Figure 3.2: Epitope sites for the custom zebrafish anti-Mdm1 antibodies ordered from 21st

Century Biochemicals. Both monoclonal antibodies, Mdm1-44 and Mdm1-2, were designed to be

specific to the zebrafish Mdm1 protein with the ability to detect both known protein sequences

based on the Sanger and VEGA database information. BLAST analysis and HPLC Mass

Spectometry by the supplier verified sequences of the peptides. The finished products were

supplied as affinity purified antibodies, which are extremely fragile and specific care was taken to

ensure the stability of each sample.

sequence that would detect both known peptide sequences (Figure 3.2). After

achieving consistent results with the western blots using the zebrafish anti-Actin

and anti-p53 antibodies, the new custom zebrafish anti-Mdm1 antibodies were

used at several dilutions on embryo lysate in order to obtain a consistent protocol

when using these antibodies. Based on the peptide information for Mdm1 in the

Sanger and VEGA databases, it was expected that the Mdm1 protein from

zebrafish lysate would appear at approximately 74kDa on a western blot.

Interestingly, there was consistent detection of bands from various zebrafish

lysates at approximately 55kDa. In order to determine if the custom antibodies

were correctly detecting the Mdm1 protein, the zfMdm1 was produced using TNT

so that it could be used as a control on the western blots.

In vitro Transcription and Translation of Mdm1, Mdm2, Mdm4 and p53

The full-coding sequences of mdm1, mdm2, mdm4 and p53 were

amplified by RT-PCR from WT(AB) zebrafish embryos using several sets of

primers for each gene. Upon successful amplification, the PCR products were

gel purified and inserted into the pCR4 vector (Invitrogen), and sequenced to

assess the integrity of the gene and determine the orientation (Figure 3.4A). This

vector had been used previously to transcribe mRNA for in situ hybridization

probes (Chapter Two), and therefore is a suitable vector for the in vitro

transcription of inserted genes. The first TNT experiment utilized the Promega

Coupled in vitro Transcription and Translation (TnT) kit to express the full-length

p53 clones in the pCR4. However, these experiments did not yield any protein

product on Western blots stained with commercially available anti-p53

antibodies. These results were puzzling since the p53 clones had been

sequenced to verify that they contained the full-length cDNA insert. To assess

these results, full-length zebrafish mdm2 clones in the pCR4 vector were also

expressed using the Promega TNT system. These clones also failed to produce

protein from the TnT reaction.

To troubleshoot these results, the sequencing data was revisited and it

was noted that there was no Kozak sequence prior to the ATG transcription start

sites in any of the clones (Figure 3.3). Therefore, the forward primers used to

originally amplify the full-length sequence were resynthesized to incorporate the

Kozak sequence prior to the ATG start site the four genes (Figure 3.4). The

sequences were again amplified by RT-PCR from zebrafish embryo, cloned into

the pCR4 vector and sequenced. This process produced several clones that

contained the correct full-length insert with the Kozak sequence and were

inserted in the T7 direction. These clones were used for TNT reactions.

Unfortunately, the final products of these TnT reactions did not yield any protein

on an SDS-PAGE gel.

Figure 3.3: Sequence of the mdm1 gene with the Kozak sequence cloned prior to the

transcription start site and mdm1 sequence cloned without the Kozak sequence. Shown here is

the first 831 bases of the modified mdm1 gene sequence in the pCR4 vector with the Primer,

Kozak sequence, and Transcription start site annotated (Sequence 1) and the first 628 bases of

the original cloned mdm1 sequence (Sequence 2).

Figure 3.4: The addition of the Kozak sequence to mdm1, mdm2, mdm4, and p53 zebrafish

sequences. (A) WT(AB) Zebrafish embryo RNA was used to RT-PCR the mdm1 sequence with a

modified forward primer to include the Kozak sequence directly upstream of the ATG-

Transcriptional Start site. The PCR products were run on an agarose gel, purified and cloned into

the pCR4 vector. Three RT-PCR reactions were set up to test the annealing temperatures (58oC,

60oC, and 62oC) for the new primer set. (B) Kozak cloning was performed for p53, mdm2, and

mdm4 in the same manner as shown for mdm1. Amplified gene sequences from zebrafish

embryo RNA were cloned into the pCR4 vector and verified by sequencing.

Upon further discussion with the technical support at Promega, it was

determined that the genes of interest should be subcloned into the Promega

vector, pCMVTnT, which was made specifically for this TNT kit (Figure 3.5B).

Full length mdm1, mdm2, mdm4 and p53 cDNA clones were excised from the

pCR4 vector by an EcoR1 digest and subcloned into the pCMVTnT vector using

T4 DNA ligase (New England BioLabs). The new pCMVTnT clones were

sequenced using both the T7 EEV promoter and internal primers to confirm that

the full sequence was present and correctly oriented within the vector.

Figure 3.5: Cloning vectors used for in vitro Transcription and Translation experiments. (A)

Invitrogen’s pCR4 TOPO-TA cloning vector which was initially used for the TnT experiments but

did not appear to function properly with the TnT kit (Promega). The clones in this vector were

shuttled to the (B) Promega pCMVTnT vector, which was created by Promega to be used

specifically with the TnT kit. This vector has a T7EEV promoter and 5’-β-Globin Leader directly

upstream of the Multiple Cloning Site. It also includes an SV40 late polyadenylation signal, which

ultimately allowed the TnT kit to work seamlessly with this vector. Figure B was created with the

Geneious v5.4 software program.

The confirmed pCMVTnT clones of mdm1, mdm2, mdm4, and p53 were

used in conjunction with the Promega Coupled in vitro Transcription and

Translation kit with the addition of biotinylated tRNA (lysine), so that all of the

translated products could be detected on a single western blot using the

Streptavidin-AP conjugated antibody (supplied in the Promega kit) and detected

with NBT-BCIP. As shown in Figure 3.6, protein did result from these reactions.

Two mdm1 clones in the pCMVTnT (Promega) vector were used for these TnT

reactions along with a single mdm2 and mdm4 pCMVTnT (Promega) clone. As

the results indicate, all of the reactions produced the expected size in vitro

translated protein product. The Mdm2 translated protein appears to have the

best expression with the colorimetric detection, which may be due to the number

of biotinylated lysines incorporated during the translation process. Confirming

that the TnT reaction worked correctly and produced the correct size protein was

an important and necessary step. These TnT products were then used to test

the custom Mdm1 antibodies and commercially available antibodies for Mdm2

and Mdm4 to correctly detect these proteins in zebrafish lysates.

Figure 3.6: In vitro Transcription and Translation results using the Promega vector, pCMVTnT,

coupled with the Promega TnT kit and biotin-labeled tRNA (lysine). Two mdm1 (kozak) samples

were used along with an mdm2 (kozak) and mdm4 (kozak) samples. The TnT reactions were

performed according to the Promega protocol (Chapter Six) with added biotin-labeled tRNA

(lysine). The TnT products were diluted 1:2 in 2X Sample buffer and run on a 10% SDS-PAGE

gel and transferred to nitrocellulose paper. Using Streptavidin-AP primary antibody, the TnT

products were colorimetrically detected with NBT-BCIP and shown above. Both Mdm1 products

appear around 74 kDa, while the Mdm2 and Mdm4 TnT products appear at 53 kDa and 55 kDa

respectively.

Using the in vitro Translated Mdm1 Product to Test the Custom Antibodies

Since the in vitro synthesized Mdm1 appeared to migrate at the expected

molecular mass (Figure 3.6), it was used as a control on the embryo lysate

western blots to determine the specificity of the custom Mdm1 antibodies; the

Promega TnT mastermix was also loaded on these gels as a control.

Representative results shown in Figure 3.7. The data indicate that the Mdm1-44

and Mdm1-2 antibodies do not detect the synthesized protein product but rather

detect proteins within the TnT mastermix. Furthermore, the custom Mdm1

antibodies do not appear to detect the appropriate size protein in embryo lysate

(expected size of Mdm1 is approximately 74 kDa).

Figure 3.7: Testing the Custom Monoclonal Mdm1 Antibodies by Western Blots. Identical gels

were loaded with TnT mastermix, verified Mdm1 TnT reaction product, and 24-72-hpf embryo

lysates (in duplicate). (A) Gel blot incubated in the Mdm1-2 monoclonal antibody overnight and

detected with CSPD. It is clear that this antibody is not detecting the synthesized protein product

in the TnT lane, but is detecting a 55 kDa protein in the embryo lysate. (B) Gel blot incubated in

the Mdm1-44 monoclonal antibody overnight and detected with CSPD. Similar results are seen

here, where the custom antibody is not detecting the synthesized Mdm1 protein or the correct

size protein in embryo lysate. These experiments were repeated several times for confirmation.

Since the Mdm1 antibodies did not appear to detect the Mdm1 protein in

embryo lysate or the synthesized product from the TnT reaction, the sensitivity

and specificity of the reagent was in question. The custom antibodies were

pooled together and used at a dilution of 1:500 (2.22 µg/µl) on a western blot

containing the TnT product and zebrafish embryo lysate (Figure 3.8A). Again, 74

kDa proteins were not detected in the lanes containing embryo lysate, although

protein was detected in the lysate around 55 kDa. Figure 3.8B shows two

western blots that were loaded with TnT mastermix and synthesized Mdm1,

Mdm2, and Mdm4. The first gel was incubated with custom Mdm1 antibody

mixture (1:500 dilution) while the second gel was incubated in a 1:1000 dilution of

the commercially available zebrafish anti-Mdm2 antibody (Anaspec). The results

suggest that the custom Mdm1 antibodies are solely detecting the TnT

mastermix (Figure 3.8B, Gel #1) whereas the anti-Mdm2 antibody also detects

some of the TnT mastermix proteins but clearly detects the zebrafish Mdm2

synthesized protein at approximately 53 kDa (Figure 3.8B, Gel #2). Based on

these results, it was apparent that using the custom Mdm1 antibodies for future

experiments such as Co-IPs would not be possible, since they were not detecting

endogenous or synthesized Mdm1 protein. In an attempt to identify proteins

associated with Mdm1 in zebrafish, embryos injected with a tagged version of

Mdm1 could be used for Co-IPs, which may provide further insight about

interacting protein partners and regulatory pathways during zebrafish

development.

Figure 3.8: Analysis of the Specificity of the Custom Mdm1 Antibodies. (A) The Mdm1-44 and

Mdm1-2 antibodies were both used to detect the Mdm1 in vitro translated protein and

endogenous Mdm1 protein from zebrafish embryo lysates at several developmental time points.

Each sample was loaded in duplicate to ensure quality of the results. The custom Mdm1

antibodies did not appear to detect the protein product in the TnT sample at 74-75 kDa, but did

appear to detect a band at approximately 55 kDa in the lysate samples (24-hpf lysates have a

very light band). (B) The in vitro transcribed products of Mdm1, Mdm2 and Mdm4 as well as TnT

mastermix were run on two separate gels. The custom Mdm1 antibodies were used on Gel #1,

which should have detected a 74-75 kDa band in the lane containing the Mdm1 TnT product.

Rather, the same results were apparent in all four lanes suggesting that the custom Mdm1

antibodies do not detect the zebrafish Mdm1 protein. The anti-Mdm2 antibody supplied by

Anaspec, was used on the Gel #2, which also showed some non-specific staining in the four

sample lanes, but a clear band was present at approximately 53 kDa in the Mdm2 TnT lane,

verifying that this antibody is detecting the correct protein.

FLAG and His-Tagging Mdm1

Based on the results from the custom Mdm1 antibody experiments, it was

necessary to use another method to detect the Mdm1 protein from zebrafish

embryos for possible protein-protein interaction studies. The FLAG peptide is a

widely utilized tag that can be used for affinity chromatography for the isolation of

protein complexes (Hopp, Gallis and Prickett 1988; Einhauer and Jungbaur

2001). In addition to the FLAG tag, polyhistidine or His-tags are another useful

tool that can be exploited and easily applied in protein purification and binding

assays (Hochuli et al. 1988; Hengen 1995). Thus, experiments were designed to

use tagged Mdm1 proteins that could be injected into embryos for subsequent

purification to identify possible binding partners of Mdm1, and also potentially

determine its functional and biological role in zebrafish. Since the Kozak

sequence and some restriction sites had already been added to the 5’- end of the

mdm1 gene with custom primers, it was decided that the tags would be

individually added to the 3’- end of the gene so that it would not interfere with the

previously adapted sequence. The methodology used to add these tags is

described below.

FLAG-Tagging mdm1 Clones in the pCR4 and pCMVTnT Vectors: The

pCR4 vector (Invitrogen) was used as a shuttle vector for these experiments

since it was known that the inserted gene sequences can be easily excised and

ligated into the pCMVTnT vector once they were verified by sequencing. Figure

3.9 shows the two-step primer process to insert the FLAG tag at the 3’-sequence

of the mdm1 gene by PCR. Since the mdm1 R9 primer was used in previous

experiments to amplify the full-length coding sequence of mdm1, it was modified

into two separate new primers that included a partial DNA sequence of the FLAG

tag (Figure 3.9B) and the complete DNA sequence of the FLAG tag (Figure

3.9C), primers mdm1 R12a and 12b respectively. The mdm1 coding sequence

was amplified by RT-PCR from embryo RNA using the mdm1 Kozak F2 primer in

combination with the mdm1 FLAG R12a primer to add the first portion of the

FLAG tag to the mdm1 gene sequence. The PCR product was run on an

agarose gel for size verification (Figure 3.10A), purified, then ligated into the

pCR4 vector and plated on ampicillin and carbenicillin plates in transformed in

Top 10 cells (Invitrogen). Plasmid DNA was isolated from individual colonies and

the second round of FLAG tagging was performed using this DNA and the

corresponding mdm1 primers (Figure 3.10B). Products from the second round of

FLAG amplification were ligated into the pCR4 vector (shuttle vector for cloning),

then excised out using an EcoR1 restriction digest and subcloned into the

pCMVTnT vector (Promega).

Figure 3.9: Experimental Design of the FLAG tag in the mdm1 sequence. (A) Peptide and DNA

sequence of the FLAG tag. A two-step PCR procedure was used to add portions of the tag to the

3’ end of the mdm1 coding sequence. (B) The first round of PCR inserts the 5’-

GACGACGATAAG-3’ just before the stop codon. This was done by modifying the R9 primer

(new R12a primer) to add the FLAG tag bases. (C) The second round of PCR completed the

FLAG tag by adding the 5’-GACTACAAGGAT-3’ to the tag sequence already in the coding

region. The new R12b primer was used for this.

Figure 3.10: PCR Results of FLAG Tagging the mdm1 Coding Sequence. (A) Round 1 PCR

using the mdm1 Kozak F2 primer with the FLAG Tag primer 12a. (B) Round 2 PCR results of

adding the FLAG tag using the R12b primer. Asterisks denote the portion of the gel that was

purified and ligated into the vector for cloning.

Analyzing the FLAG-Tag Clones in the pCMVTnT Vector: In order to

determine if the FLAG tag was successfully added to the 3’ end of the mdm1

coding sequence, the pCMVTnT clones were subjected to digests using the

Hpy99I restriction enzyme (New England BioLabs). Based on the vector

sequence information, plasmids containing the FLAG tag would show four bands

on an agarose gel since the enzyme cuts the vector itself in three locations, while

also recognizing a restriction site in the FLAG tag itself. Using this method

required very little time and effort, while quickly allowing for analysis of each

clone generated by the PCR and ligation process. It is important to note that in

addition to these digests, a representative sample of the FLAG tag clones were

also sent off for sequencing to verify the digest results. Not only would the

restriction digests verify the insertion of the FLAG tag, it would also identify the

direction of the mdm1 insert within the pCMVTnT vector. Hundreds of clones

were analyzed using the Hpy99I restriction enzyme and run on agarose gels.

Representative results of these digests are shown in Figure 3.12. Based on the

sequence map and the schematic of the pCMVTnT vector (Figure 3.11), if the

FLAG tag was correctly inserted into the mdm1 sequence, four bands were

expected on the agarose gels, sizes corresponding to the orientation of the insert

within the vector.

Figure 3.11: Restriction Analysis of the FLAG Tag Clones by Hpy99I. In addition to several

Hpy99I restriction sites within the pCMVTnT vector, a recognition site was also within the FLAG

tag, and used to identify clones containing the mdm1-FLAG insertion. (A) Shows the expected

fragment sizes following the digest if the mdm1-FLAG sequence was inserted in direction 1. This

was the desired result, since the insertion would be in the correct orientation for the TnT reaction.

(B) Shows the expected fragment sizes following the digest if the mdm1-FLAG sequence were

inserted in direction 2.

Figure 3.12: Results of the Hpy99I digests. All lanes contain three fragments following the

restriction digests, suggesting that the FLAG tag was never successfully inserted into the mdm1

coding sequence.

As shown in Figure 3.12, only three bands corresponding to the vector

appeared on the gels following the restriction digests using the Hpy99I enzyme.

To support these results, several of the clones were sent to Macrogen for

sequencing confirmation of the FLAG tag; the mdm1 insert was present without

the FLAG tag in each of the clones analyzed. Based on this evidence, it was

decided that the FLAG tag was too difficult to insert within the mdm1 coding

sequence, and therefore the efforts turned to using a 6X-His tag instead.

6X-His Tagging mdm1 Clones in the pCR4 Vector: Similar to the FLAG

tagging efforts, the 6X-His tag sequence was added to the mdm1 R9 primer, but

as a single-step amplification. Because the addition of the His tag significantly

increased the melting temperature of the primer, two different reverse primers

(R13 and R14) containing the His tag, were designed to be used with the Kozak

F2 forward primer. Figure 3.13 shows the location of the His tag within the

coding sequence of the mdm1 gene.

Figure 3.13: Schematic of adding the 6X-His Tag to the mdm1 Coding Sequence. (A) DNA

sequence of the 6X-His Tag, which was added to the R9 primer (new R13 and R14 primers). (B)

Location of the tag, just before the STOP codon in the mdm1 coding sequence.

Since the His tag was added in a single step by RT-PCR using zebrafish

embryo RNA, the PCR products were run on an agarose gel, purified and ligated

into the pCR4 vector (Figure 3.14). Due to the difficulties with the FLAG tagging

effort, the products were analyzed within the pCR4 vector, rather than shuttling

the gene inserts into the pCMVTnT vector (Figure 3.15). Similar to the restriction

analysis of the FLAG tag, the 6X-His tag insertion was also analyzed within the

pCR4 vector by restriction digests of hundreds of clones using the Ase1 enzyme

along with sequencing several samples of the clones (New England BioLabs).

Based on the restriction site analysis of the vector and mdm1-His insert, and the

orientation of the insert within the vector, upon digestion with Ase1, three bands

would appear on an agarose gel if the clones were positive for the 6X-His tag

(Figure 3.15).

Figure 3.14: RT-PCR results of adding the 6X-His Tag to mdm1. Each of the fragments were gel

purified and ligated into the pCR4 vector (Invitrogen) for cloning and analysis.

Figure 3.15: Schematic of the Ase1 digests to confirm the insertion of the 6X-His tag in mdm1.

Similar to the restriction analysis of the FLAG tags, the expected fragment sizes were dependent

on the orientation of the insert within the vector. In this case, orientation was not important, as

positive clones would be shuttled into the pCMVTnT vector for TnT reactions. (A) Shows the

expected fragment sizes if the mdm1-His were inserted in the T3 direction. (B) Shows the

expected fragment sizes if the mdm1-His were inserted in the T7 direction. Restriction site

locations in the vector were based on the information provided by Invitrogen.

After restriction mapping hundreds of pCR4 clones, it was evident that

none of the clones contained the 6X-His tag, since only two bands were apparent

on all of the agarose gels following the restriction digests. Representative results

of these digests are shown in Figure 3.16. Based on the restriction mapping and

sequencing evidence, it was determined that none of the clones contained either

the FLAG or His tag in the mdm1 coding sequence, and thus could not be used

for future experiments such as injections into embryos for Co-IPs.

Figure 3.16: Restriction Analysis Results of the 6X-His tag mdm1 Clones in the pCR4 Vector.

All clones that were examined by Ase1 restriction digests appeared to only have two fragments

on the agarose gels, suggesting that the His tag was not added to the 3’ mdm1 coding sequence.

Summary of Experimental Results

• The in vitro Transcription and Translation requires delicate handling and

specific vector qualities to synthesize proteins of interest. This can also be

a powerful tool in determining the quality and specificity of commercially

available and custom antibodies. In these studies, the synthesized proteins

were used as a control to determine the ability of the custom Mdm1

zebrafish antibodies to detect the correct protein on a western blot. Without

this control, the custom antibodies may have been inappropriately used in

other protein experiments, such as Co-IPs, leading to costly results.

• The use of tags is not novel, especially in zebrafish research. It was,

however, unfortunate that with all of the effort put forth, neither FLAG nor

6X-His tags were inserted into the coding sequence of the mdm1 gene.

The combination of restriction mapping and sequencing clearly identified

mdm1 clones without tags. A 5’ tag was never attempted since the Kozak

and in some cases, an Xho1 site for cloning purposes were already added

in this region.

• The mdm1 gene remains an important focus in the lab. Based on the

results of the westerns with the custom antibodies, and the inefficiency of

Tag cloning, it was decided that the gene function and expression should

be examined before looking at the protein expression.

CHAPTER FOUR

INVESTIGATING THE ROLE OF MDM1 IN ZEBRAFISH DEVELOPMENT

Experimental Design and Rationale

Similar to the gin-10 candidate studies, the gin-12 mutation was identified

on zebrafish chromosome 4 by ENU mutagenesis studies and half-tetrad

mapping (Moore et al. 2006). Within this region that maintains synteny with

human chromosome 12, several genes were identified as candidate genes for

gin-12, including mdm1. The aim of this chapter is to determine the expression

patterns of the mdm1 gene and identify its role in zebrafish development.

To formally investigate this candidate gene, the ability to study transcript

expression within the embryo was explored by in situ hybridization techniques, as

well as the ability to knockdown gene function by Morpholino oligonucleotides

(MOs). Embryos injected with biotin-labeled mdm1-Ex2 MO were stained to

identify the exact regions of the MO targets, leading to the identification of mdm1

gene expression and developmental function in zebrafish embryos. The use of

MOs to knockdown mdm1 gene expression was verified by RNA rescue, as well

as RT-PCR and in vitro protein studies. Additionally some of the knockdown

experiments were designed to further explore the potential genomic instability

activity of the mdm1 gene (Figure 4.1).

Figure 4.1: Schematic of the Experimental Design to Establish the Role of mdm1 in Zebrafish

Development.

Temporal and Spatial Expression Analysis of mdm1 in Zebrafish Embryos

Background: In addition to the full-length cloning of mdm1 that was

described in Chapter Three, RT-PCR was also used to amplify several smaller

fragments of this gene, which were cloned into the pCR4 vector. The amplified

cDNA products were sequenced for confirmation and orientation within the

plasmid, and kept in a library. These small mdm1 gene clones were later used

as templates for RNA and DNA probes in various experiments.

Developmental Expression Analysis of the mdm1 Gene: Preliminary RT-

PCR amplification of mdm1 had suggested that expression of mdm1 transcripts

was variable during zebrafish development (data not shown). In order to further

explore the temporal expression patterns of mdm1, semi-quantitative RT-PCR

was performed. This method utilizes the low-cycle RT-PCR reaction from

WT(AB) zebrafish RNA and a modified Southern blotting technique to detect

expression of the mdm1 cDNA with a DIG-labeled DNA probe (Figure 4.2).

Figure 4.2: Results of the mdm1 semi-quantitative expression analysis from developing zebrafish

embryo RNA. (A) Blot results detected with CSPD following hybridization with an mdm1 DIG-

labeled DNA probe. Expression of the transcript appears to vary significantly from maternal

expression (4-hpf) through larval stage (7-dpf). (B) ImageJ software (NIH) was used to

graphically represent the blot data from this analysis. Units indicate detection above that seen in

the negative control lane.

As shown in Figure 4.2, mdm1 transcript levels appear to vary during

zebrafish development. At 4-hpf, the maternal expression of mdm1 was very

robust suggesting that it plays a critical role in the earliest stages of

embryogenesis. The apparent decrease in mdm1 expression at 8-hpf occurs

shortly after the Mid-Blastula Transition (MBT), suggesting that the maternal

transcript of this gene has been depleted and the zygotic gene is not yet active.

The zygotic mdm1 gene is activated after this transition, which is seen in the

increased expression level at 12-hpf. This is also the time that the optic vesicle

and Kupffer’s vesicle develops in zebrafish embryos. Expression of this gene

appears to decline around 18-hpf before increasing in quantity again by 24-hpf.

At 72-hpf, mdm1 was expressed at the highest level. 72-hpf is the period of the

completion of rapid morphogenesis of the embryo into the early larval stage. By

7-dpf observed a low but consistent level of mdm1 gene expression observed in

the larval zebrafish.

Since the temporal expression of mdm1 was carried out through the semi-

quantitative analysis of the cDNA, it was important to further investigate the

expression of the mdm1 gene by identifying specific tissue expression of mdm1

mRNA in zebrafish embryos of different ages (spatial expression analysis).

Although little is known about the mdm1 gene in zebrafish or mammals, recent

studies suggest that mdm1 may be involved in the development of the retina

and/or optic nerves and may also play a role in retinal degenerative diseases

(Chang et al. 2008). To investigate the role of mdm1 in these processes,

experiments were carried out using WT(AB) embryos treated with phenylthiourea

(PTU) to inhibit pigment formation, and golden embryos. Several DIG-labeled

mdm1 RNA probes were generated (approximately 600bp-1200bp) with the

Ambion MaxiScript T3/T7 kit. As shown in Figure 4.3, the mdm1 transcript is

expressed ubiquitously around 12-hpf and appears to be concentrated near the

anterior portion of the embryo from 24-hpf through 72-hpf. In the 24-hpf embryo,

staining appears to be the darkest in the diencephalic roof, near the epiphysis

(midline, dorsal region of the diencephalon), the dorsal tectum, telencephalon,

and within the otic primordium. By 48-hpf, there has been a rapid

morphogenesis of the embryo, and it appears that the expression of the mdm1

gene is prominent in the midbrain (dorsal region), the midbrain tegmentum

(ventral region), and optic tectum. Staining appears to be consistent in the CNS

at 72-hpf, when the embryo enters the larval stage. Each set of in situ

hybridization experiments also included control embryos for no RNA probe (no

hybridization), no RNA probe/no antibody, and no antibody. These control

embryos remained translucent indicating that the staining observed in the

experimental embryos was due to mdm1 transcript targeting.

The combined results of these experiments indicate that the mdm1 gene

is expressed at different levels throughout the developmental process.

Expression begins as a maternal transcript that is ubiquitously expressed

throughout the early embryo followed by zygotic gene expression that is

ultimately confined to the anterior portion of the embryo. Specifically, based on

the location of the mdm1 transcript it appears that mdm1 may play a role in the

development of the otic primordium, optic cup, and other components of the brain

and CNS as shown in the figures above. Thus, it is important to confirm these

analyses with experiments to study and identify the function of the mdm1 gene in

vivo.

Figure 4.3: Results from the Whole-Mount in situ Hybridization Expression Analysis of mdm1

Transcripts in Zebrafish Embryos. (A) Ubiquitous expression of mdm1 in 12-hpf embryos (B)

Expression appears to be localized in the CNS of 24-hpf embryos, including midbrain, hindbrain

and otic primordium (C) Localization of mdm1 transcripts in the anterior region of the zebrafish

embryo continues at 48-hpf and also includes staining for mdm1 in the optic cup and optic tectum

(D) Robust staining for the mdm1 transcripts remain in the CNS as well as within the eyes and

cerebellum in 72-hpf embryos.

Using Morpholinos to Determine the Role of mdm1 During Zebrafish

Development

The use of synthetic antisense morpholino oligonucleotides (MOs) is the

standard method for creating gene knockdowns in zebrafish, Xenopus, and other

model systems. Successful binding of a morpholino to a gene of interest

requires identifying a region that is within 20 bp of the initiating AUG that allows

the MO to effectively block translation of mature mRNA molecules in vivo. It is

also possible to decrease expression of a gene by designing MOs that target

splice junctions of pre-mRNA. For developmental biology, the translation-

blocking MOs are especially useful for knocking down maternal mRNA, while

splice-blocking MOs are more effective on actively transcribed zygotic RNA.

Two transcripts produced from the mdm1 gene have been identified in

zebrafish embryos and adults. The long mdm1 transcript (VEGA mdm1-001) is

2193bp long and includes 14 exons. The short mdm1 transcript (VEGA mdm1-

002) is 2160bp long and is a splice variant that does not include exon 8.

Additionally, there is a third transcript that exists, but does not code for protein,

and therefore was not explored further.

Table 4.1: Morpholinos Used for the Gene Function Experiments

Target

Sequence

Comments

mdm1-ATG

5’-TGATTGCCTTGAAACGGACAGGCAT-3’

Translation Blocking

MO – targets the

maternal products

mdm1-Ex1

5’-TTTACAAAGCTTACCTTGAAACGGA-3’

Splice-blocker that

targets Exon 1

mdm1-Ex2

5’-CGCTGATTCCCTATTAGGAATATAT-3’

Splice-blocker that

targets Exon 2.

Labeled with Biotin

mdm1-Ex8

5’-AGTAACAGGTGAAATGTTACCTCAT-3’

Splice-blocker that

targets Exon 8/9 to

knockdown the long

transcript only

p53

5’-AGTAACAGGTGAAATGTTACCTCAT-3’

Standard supplied by

Gene-Tools, LLC.

Negative Control

5’-CCTCTTACCTCAgTTACAATTTATA-3’

Standard supplied by

Gene-Tools, LLC.

A translation-blocking MO was designed (mdm1-ATG) that would bind to

maternal mRNA from both mdm1 transcripts. Several MOs were also generated

to target specific splice junctions. These MOs allowed for the knockdown of both

zygotically active mdm1 transcripts or the knockdown of a single transcript

depending on the target (Table 4.1, Figure 4.4). For all injection experiments,

not-injected embryos, dye-only injected embryos and embryos injected with the

standard negative morpholino were also raised as controls.

Figure 4.4: Schematic of the mdm1 Coding Sequence (Long Transcript from VEGA), showing the

locations of exons, 5’-UTR, 3’UTR, Start and Stop Codons and approximate Morpholino Target

Sites. The mdm1-Ex1 MO is not shown here since it was designed to target Exon 1 in the 5’UTR

sequence that was originally not included in the gene sequence.

Effects of the mdm1 ATG-Blocking Morpholino in WT(AB) Embryos: To

investigate the role of mdm1 in zebrafish development, the processed maternal

transcript was first knocked down by a translation blocking MO, mdm1-ATG, in 1-

to 8-cell stage embryos. Since both mdm1 protein-coding transcripts have

identical 5’-coding sequence, using this MO allowed for the knockdown of both

transcripts to occur simultaneously. This MO appeared to be most effective

when 8-10nL of solution was injected at concentrations of 200-350µM. As

described in Chapter Two, the mdm1-ATG MO was also supplemented with

50µM p53 MO to reduce necrosis in the injected embryos. Following the mdm1-

ATG MO injections, a range of phenotypes was observed in 24-hpi embryos

(Figure 4.5). Headless embryos that died shortly after 24-hpi were the most

extreme of the observed phenotypes, while most of the injected embryos

appeared to have defects in head and eye development, typically resulting in an

embryo with a smaller than average head and tiny, underdeveloped eyes.

Defects in tail development were also noted in approximately 50% of the

embryos.

Injected embryos that survived to 72-hpi were also photographed and any

abnormal phenotypes were recorded. The most obvious and consistent

phenotype was that of embryos with eye deformities that ranged from small

underdeveloped retina to embryos with completely missing eyes (Figure 4.6).

This was noted in a large percentage of the injected embryos, as well as tail and

some head irregularities (Table 4.2). This information was critical, since it

appeared to be consistent with the previous in situ data, suggesting that the

mdm1 gene is expressed during the development of the primordial eye and CNS,

and specifically within those tissues in the 24- to 72-hpf embryo.

Figure 4.5: Photos of 48-hpi WT(AB) zebrafish embryos that were injected with the mdm1-ATG

Morpholino supplemented with 50µM p53 Standard Morpholino. (A) Embryo injected with 200µM

mdm1-ATG MO exhibits a small head/small eyes phenotype following injection. (B) Embryo

injected with 250µM mdm1-ATG MO displays an abnormal tail and small head, and does not

appear to have any eye development. (C) Embryo injected with 300µM of mdm1-ATG MO

appears to have a small head with only one developing eye; this embryo also displays defects in

tail development.

Figure 4.6: Results of 250µM mdm1-ATG Morpholino Injections supplemented with the p53

Morpholino into WT(AB) Embryos shown at 72-hpi. (A) Embryo has obvious MO effects in eye

and tail development causing an abnormally stubby appearing tail and underdeveloped eyes. (B)

Embryo shown has no eye development in addition to a severe shortening of the tail. (C) Embryo

has a small, abnormal sized head and severely underdeveloped eyes; tail is of normal length

although it is slightly twisted.

Effects of the mdm1 Splice-Blocking Morpholinos in WT(AB) Embryos:

Since the Mdm1 custom antibodies analyzed in Chapter Three did not properly

detect the in vitro synthesized Mdm1 protein or Mdm1 in the embryo lysate,

verification of the knockdown by western blot analysis was not an available

option. Fortunately, RT-PCR can be used to determine the effectiveness of

splice-blocking MOs (Morcos 2007). Therefore, if the phenotypes observed

following knockdown by a splice-blocking MO are identical to the results

observed following the translation blocking MO, and the RT-PCR results suggest

that the mdm1 transcripts were knocked down, it can be concluded that the

translation-blocking MO results are valid. Three different splice-blocking MOs

(Table 1) were custom ordered from Gene-Tools to inhibit splicing at different

intron-exon boundaries (mechanisms of MO action are explained in Chapter

Two). Figure 4.4 shows the location of MO target sites within the mdm1 gene.

Splice-blocking MOs were first designed to target both of the mdm1

transcripts. The mdm1-Ex1 MO was injected into WT(AB) embryos at a range of

concentrations from 200-550µM with and without p53 MO supplementation.

Regrettably, none of the injected embryos presented with any effects from the

injections, implying that this particular MO did not work. Experiments continued

by injecting the embryos with the mdm1-Ex2 MO that blocks splicing of Exon 2 of

the mdm1 mRNA. Eight-10nL of this MO at a concentration of 250-300µM was

injected into embryos as described above (Figure 4.7). Knockdown of the mdm1

mRNA was successful with the mdm1-Ex2 MO resulting in the same range of

phenotypes in the morphants as observed from the translation-blocking MO

experiments. These results not only confirmed the translation-blocking MO

results, but also resulted in approximately 50% of the mdm1-Ex2 MO injected

embryos exhibiting spontaneous rescue between 24- to 48-hpi. This result may

be due to the significant increase in mdm1 transcripts after 24hpf that may titrate

down the amount of the MO, resulting in a decrease the effectiveness of the

splice-blocker after 24-hpi.

Figure 4.7: Results from Injection Experiments of the mdm1-Ex2 MO into Zebrafish Embryos.

All embryos shown were injected with the mdm1-Ex2 MO, and are approximately 48-hpi. (A)

Injected with 200µM MO and has obvious defects in eye and tail development. (B) Embryo

injected with 250µM MO and also shows severe defects in eye and tail development. (C) Embryo

injected with 300µM MO and appears to have a delay in development of the tail and eyes. All

embryos shown also have pericardial edema, which is most likely a result of the injection process.

To determine whether there were any differences in resulting phenotypes

when knocking down both mdm1 transcripts as opposed to knocking down just

the long transcript, the mdm1-Ex8 MO was specifically designed to target the

mdm1-001 (VEGA) transcript and test this hypothesis. When injected into

embryos, even at extremely low doses (50µM), this particular MO was completely

lethal within a few hours following the injection. This was an unexpected result

since none of the other mdm1 MOs showed this level of lethality. Therefore the

sequence of the mdm1-Ex8 MO was re-evaluated to determine if it may be

targeting other sequences. BLAST analysis of the mdm1-Ex8 MO sequence

revealed that it actually targeted multiple transcripts on a majority of the zebrafish

chromosomes and was most likely knocking down multiple genes in the embryos.

The only difference between the two protein-coding mdm1 transcripts is exon 8,

which is very small, consisting of 11 amino acids. In an attempt to design

another MO that would target just the longer transcript, it appeared that any MO

designed around the 5’ and 3’ splice sites of exon 8 would result in a MO that

targets multiple genes, based on BLAST analysis.

Since both mdm1-ATG MO and mdm1-Ex2 MO had similar results when

injected into WT(AB) embryos, they were co-injected in combination with and

without the p53 MO (mdm1-ATG/mdm1-Ex2 combination). Interestingly, even at

very low concentrations (50-100µM), the combination of these MOs was

extremely lethal, and resulted in 0% survival after 24-hpi. A summary of the MO

injection experiments is presented in Table 4.2. It is important to note that the

total number of injected embryos does not include the embryos used for controls

including the dye-only, buffer-only and standard negative control injected

embryos. These embryos did not exhibit any developmental abnormalities

following control injections. Additionally, several clutches of embryos injected

with dye or buffer only were raised and bred, indicating that the injections did not

interfere with development and fertility.

Table 4.2: Overview of Morpholino Injection Results.

mdm1-ATG MO

mdm1-Ex1

mdm1-Ex2

mdm1-Ex8

+ p53

- p53

+ p53

- p53

+ p53

- p53

+ p53

- p53

Survival at

24-hpi

78%

52%

20%

14%

62%

32%

Survival at

48-hpi

89%

80%

98%

96%

95%

88%

No Head

Development

13%

16%

Small Heads

Only

15%

21%

11%

28%

Small

Heads/Small

Eyes

62%

56%

64%

51%

Small

Head/No Eyes

10%

Tail Defects

43%

56%

43%

56%

Abnormalities

Observed

98%

96%

Total Injected

1926

882

305

298

2792

455

764

510

Results include all of the experimental injections for the above-mentioned mdm1 MOs with and

without the supplementation of the p53 Standard MO (the p53 supplementation was 50µM

concentration of the p53 MO added to the experimental mdm1 MO solution). The mdm1-ATG

MO was injected into 1-2 cell zebrafish embryos at a range of concentrations between 200-

350µM in a 0.1% Phenol Red solution. The mdm1-Ex1 MO was injected into 1-2 cell zebrafish

embryos at a range of concentrations between 200-550µM in a 0.1% Phenol Red solution.

Following these injections, it was apparent that the mdm1-Ex1 MO was not correctly targeting the

mdm1 mRNA, as determined by observing 96-98% of injected zebrafish embryos with no defects

or abnormalities following the injections. The mdm1-Ex2 MO was injected into 1-2 cell zebrafish

embryos at a range of concentrations between 250-300µM in a 0.1% Phenol Red solution. The

mdm1-Ex8 MO was injected into 1-2 cell zebrafish embryos at 50µM (initial injections for this MO

were concentrations between 200-400µM, but the concentration was reduced based on the

lethality of this MO). Due to the lethality of this MO at low concentrations, the MO sequence was

analyzed using BLAST, which revealed that it targets multiple transcripts on several

chromosomes. Based on the data and observations noted above, it appeared that both the

mdm1-ATG MO and the mdm1-Ex2 MO were causing similar defects during the development of

zebrafish, which focused on the development of the head and eyes. Interestingly, approximately

50% of the injected embryos were observed to have defects in tail development.

Whole-mount Antibody Staining of mdm1-Ex2 Injected Embryos: In

addition to confirming the morpholino injections by RT-PCR analysis, embryos

injected with the mdm1-Ex2 MO were also subject to an in situ antibody staining

procedure that was used to detect the biotin label of that particular morpholino.

This technique allowed for the visualization of morpholino target tissues within

the whole injected embryo, and also enabled the verification of earlier whole-

mount in situ hybridization data. The 3’- biotin label on the mdm1-Ex2 MO is a

394kDa affinity tag that can be detected with avidin or streptavidin conjugated

with alkaline phosphatase.

Figure 4.8: Structure of the Biotin label added to the mdm1-Ex2 MO (from Gene Tools, LLC.).

Embryos were injected with the mdm1-Ex2 MO with p53 MO

supplementation as previously described. Injected embryos were kept at 26-

27oC in petri dishes containing aquarium water with Methylene Blue, and

collected at several developmental stages including 18-somite (18-hpi) through

prim-6 (25-hpi). Following photodocumentation of the injected embryos, they

were euthanized in Tricaine-S on ice and prepared accordingly for the antibody

staining protocol described in Chapter Six. Colorimetric detection with

Streptavidin-AP and NBT-BCIP was terminated after several minutes of

treatment. The embryos were placed in glycerol to clear and were subsequently

photographed.

Since the morpholino specifically targets mdm1 RNA, it was expected that

the results from the antibody staining procedure would be similar to the in situ

hybridization data from previous experiments where the mdm1 transcript

expression within the embryo tissue was identified. As Figure 4.9 shows, there is

a striking similarity between the antibody staining results and the initial in situ

hybridization data. In both experiments mdm1 RNA appeared to be expressed

significantly in the anterior portion of the embryo, specifically in the CNS and

primordial eye tissues. The morpholino injection and rescue results also

suggested that mdm1 may also be involved in the development of the tail, based

on the observed phenotypes. In a 19-hpi embryo (Figure 4.9A), the morpholino

appears to localize to regions of the CNS, specifically the diencephalon,

midbrain, telencephalon and otic primordium. At approximately 22-hpi (Figure

4.9B), staining continues to appear robust within the diencephalon, midbrain and

telecephalon. Additionally, the morpholino appears to localize within the optic

cup, epiphysis and cerebellum. Similar staining was also observed in the 25-hpi

embryos (Figure 4.9C and 4.9D), although the most intense staining occurred

within the optic cup, lens, epiphysis and hindbrain. It is also important to note

that all of the embryos had noticeable staining in the posterior portion of the tail,

that may correspond to the tail defect phenotype that was observed during the

injection experiments.

Figure 4.9: Whole-mount Antibody Staining of Morpholino Injected Zebrafish Embryos (A) 19-hpi

injected embryo shows very distinct staining in the diencephalon, telencephalon and midbrain (B)

Robust staining in the 22-hpi injected embryo indicates localization of the MO in the cerebellum,

optic cup and epiphysis (C) Staining is also noted in the lens and hindbrain in the 25-hpi injected

embryo (D) Lateral view of the 25-hpi injected embryo also shows staining in the optic cup and

lens.

RNA Rescue of the Morphants: Although the results from the translation

and splice blocking morpholinos were exciting, it was important to confirm that

the phenotypes observed were due to the knockdown of the mdm1 gene and not

due to non-specific effects of the morpholino injections. One method to validate

the results is to utilize RNA rescue experiments. The mMessage mMachine T7

kit (Ambion) was used to synthesize in vitro capped mRNA of the full-length

mdm1 gene with the Kozak sequence. A pCR4 clone containing the confirmed

sequence was used for the transcription reaction (cloning and sequencing of the

100

gene is described in Chapter Three). Following the transcription reaction, the

RNA was diluted 1:5 in 0.1M KCl and stored in 5µl aliquots at -80oC. Prior to

storage, the concentration of one aliquot was determined by spectrophotometry.

In order to control the RNA rescue experiments, the embryos were

injected twice; first they were injected with the morpholino and then they were

injected with the specific RNA (Ekker & Larson 2001; Nasevicius & Ekker 2000).

Up to 100 embryos at a time were placed under the dissecting microscope and

injected with the appropriate concentration of either the mdm1-ATG or mdm1-

Ex2 MOs with p53 MO supplementation. Once the embryos were injected with

the morpholino solution, half of the embryos were removed and placed in a petri

dish containing aquarium water with Methylene Blue and stored at approximately

27oC in an incubator. The remaining embryos were then injected with 8-10nL of

a 50-75pg dose of the capped mRNA. This procedure was repeated until all

embryos were injected, and continued over a several week period. Additionally,

not-injected embryos, dye-only injected embryos and embryos injected with the

standard negative morpholino were also raised during each weekly injection

session as controls. The embryos were then compared to each other to identify

whether the morpholino effects were rescued.

Figure 4.10 shows a WT(AB) not-injected embryo, an mdm1-Ex2

morphant, and the RNA rescued embryo. Based on the observed phenotypes of

the RNA rescued embryos, 98% of the RNA injected embryos appeared to have

wildtype characteristics, including appropriate developing tails, heads and eyes,

as compared to the morphants. Since the tail phenotype was also rescued in

101

these experiments, it can be concluded that the mdm1 gene may be involved in

the development of the tail as well as the eye in the early zebrafish embryo.

Figure 4.10: Zebrafish embryo morpholino rescue experiments (A) Not-injected 48-hpf golden

heterozygous embryo control (B) mdm1 MO knockdown in a golden heterozygous embryo (C) 48-

hpi golden heterozygous embryo first injected with the mdm1-Ex2 MO, then injected with

synthesized mdm1 mRNA for a full rescue of the normal phenotype.

Confirming the Injection Experiments and Validating the mdm1 Gene

Analysis

It was important to verify that the mdm1 MOs were actually targeting the

mdm1 mRNA within the zebrafish embryo. Two approaches were taken to

explore the specificity of the knockdown experiments. First, embryos injected

with the mdm1-Ex2 MO were collected and analyzed for mdm1 transcript

expression by RT-PCR. Second, an in vitro experiment was designed to test

whether the mdm1-ATG MO successfully targeted the processed mdm1

transcript. This allowed for the visualization of decreasing Mdm1 protein

expression with the addition of morpholino to the reaction mix.

102

RT-PCR Analysis of Injected Embryos: All embryos from the knockdown,

rescue, and overexpression experiments were collected into RNA Later (Qiagen),

along with non-injected embryos, and used for analysis by RT-PCR. Total RNA

was isolated from each set of embryos, and primers for mdm1 were used to

amplify the full-length coding sequence from each set of RNA. The prediction

was that embryos injected with mdm1 MOs will show decreased levels of mRNA

from the gene. These experiments to confirm the specificity of the effect of the

MO in the injected embryos were performed in triplicate, using embryo RNAs

from 14 injection sets. Representative data from the RT-PCR experiments is

shown in Figure 4.11.

The RT-PCR results suggest that the mdm1-Ex2 MO knockdowns were

efficient in targeting the mdm1 transcripts in the zebrafish embryos. The

observed level of the mdm1 transcripts in Lane 4 is significantly reduced from the

normal expression shown in Lane 3 from the non-injected WT(AB) embryos.

Although there is a weak band in Lane 4, which shows amplification from the

morphant RNA, it is reasonable to conclude that the phenotypes observed

following the MO injection was due to the mdm1-Ex2 MO. Additionally, the

results in Lane 5 suggest that the RNA rescue phenotypic effects observed

following the injections were also a result of successful knockdown and rescue of

the mdm1 gene in the zebrafish embryos, since the resultant cDNA band on the

gel is consistent with the band in Lane 3. Interestingly, the embryos only injected

with the mdm1 synthesized capped mRNA, had approximately the same RT-

103

PCR expression results of mdm1 as the non-injected and RNA rescued embryos.

This result was observed numerous times.

Figure 4.11: RT-PCR Confirmation of the Injection Experiments in Zebrafish Embryos.

Applying the in vitro Transcription and Translation System to Confirm the

Translation-blocking Effects of the mdm1-ATG Morpholino: By knocking down

mdm1 mRNA, the MDM protein is presumed to also be depleted in the injected

embryos, which would give the resulting phenotypes that were observed in the

morphants (Figures 4.5 - 4.7). Without antibodies specific to the zebrafish Mdm1

protein, verifying MDM protein appeared to be a difficult undertaking in vivo. For

this reason, the question was asked if it was possible to study the effect of the

mdm1-ATG MO on protein levels in vitro. Previous experiments that utilized the

Promega TnT kit resulted in the expression of the synthesized Mdm1 protein on

104

a western blot, which utilized biotinylated tRNA (lysine) and was then detected

with Streptavidin-AP and stained with NBT-BCIP (Chapter Three). Using various

dilutions of the mdm1-ATG MO in nuclease-free water, 0.5µl of each dilution was

added to a 10µl reaction containing the TnT mastermix, amino acid mixture,

biotinylated tRNA (lysine), and the full-length mdm1 clone in the pCMVTnT

Vector (Promega). The reaction products were then diluted 1:10 in Sample

Buffer, run on an SDS-PAGE gel, and transferred to a nitrocellulose membrane.

The TnT products were detected as previously described (Chapter Three) and

the results are shown in Figure 4.12.

As Figure 4.12A indicates, the increased concentration of the mdm1-ATG

MO added to the TnT reaction appears to decrease the expression of the

synthesized Mdm1 protein on the western blot. These results were very exciting

since they illustrate a novel way to confirm the MO targeting in vitro without a

specific antibody. Importantly, the approximate concentrations of the mdm1-ATG

MO injected into the live zebrafish embryos correspond with the total

concentration of MO in lanes 10-12 in Figure 4.12A. Since the MO added to the

TnT reaction significantly decreased the expression of the Mdm1 protein in vitro,

it is likely that the resulting phenotypes from the mdm1-ATG MO injections are

due to the decrease in maternal mdm1.

105

Figure 4.12: In vitro Synthesis of the Mdm1 Protein with Various Concentrations of the mdm1-

ATG MO to Test for Specificity. (A) The western blot of the Mdm1 TnT products with an

increasing amount of MO added to each reaction. (B) The concentration added to each TnT

reaction.

Testing mdm1 for Genomic Instability and Overexpression Experiments

Testing for Genomic Instability: Based on previous studies and mapping

experiments, the mdm1 gene is located within the gin-12 candidate region on

chromosome 4. While the interest of the experiments described throughout this

chapter were designed to determine the developmental role of the mdm1 gene, it

was also important to continue investigating mdm1 as a potential gin-12

candidate by exploiting the knockdown coupled with the mosaic eye assay. For

these experiments, golden heterozygous embryos were injected with various

concentrations of the mdm1-ATG MO with and without the supplementation of

the p53 MO. Embryo survival at 24-hpi was maximized by the addition of the p53

106

MO, as described in Chapter Two, specifically when embryos were injected with

8-10nL of 250µM MO solution in Danieau Buffer with 0.1% Phenol Red.

Unfortunately, mosaic eyes did not result in any of the golden heterozygous

injections, suggesting that the mdm1 gene was no longer a gin-12 candidate.

Overexpression Analysis of the mdm1 Gene in Zebrafish Embryos: Since

the mdm1 gene was originally identified in transformed 3T3DM cells that

expressed the gene, it was of interest to determine the effect of overexpression

of mdm1 mRNA in embryos. These experiments were carried out by injecting

both WT(AB) and golden heterozygous embryos so that the mosaic eye assay

could be used to observe potential genomic instability. The same in vitro

synthesized capped full-length mdm1 mRNA was utilized for this experiment that

was used in the previous RNA rescue experiments. Approximately 100pg of

mdm1 mRNA was injected into 1-8 cell golden heterozygous and WT(AB)

embryos, which were then kept in aquarium water at 26-27oC in an incubator and

monitored daily to remove any debris or decayed embryos and replenish the

water. At 48-hpi, the golden heterozygous embryos were placed in 2X Tricaine

solution (anesthetized) and screened for mosaicism. The injected WT(AB)

embryos were screened for any abnormalities, then subsequently photographed

and collected into RNA Later (Qiagen).

107

Figure 4.13: Injection of synthesized full-length mdm1 mRNA into golden heterozygous

zebrafish embryos. (A) 48-hpf Not-Injected embryo (B) Embryo injected with RNA.

Figure 4.13 shows that the embryos injected with an abundance of mdm1

mRNA exhibit a fully wildtype phenotype. There appears to be a slight delay in

pigment cell migration that is often a result from the injection itself and has been

observed in all of the previous injection experiments. In the golden

heterozygotes, some areas of the RPE appear lighter in color; however these are

not the expected golden-colored patches observed in true mosaics, but rather

layers of cells that have not developed pigment. By 72-hpi, these embryos had

completely black-pigmented eyes, confirming that the mdm1 gene did not have

genomic instability activity in developing embryos by overexpression.

Summary of Experimental Results:

• The semi-quantitative RT-PCR results suggest that the expression of the

mdm1 gene in developing zebrafish embryos is quite varied. Interestingly,

although this procedure was repeated several times using various embryo

RNA samples, the RT-PCR results indicated that the expression of mdm1

108

transcripts is quite low in 18-hpf embryos. It has been shown with other

experiments that there is in fact, mdm1 expression in 18-hpf embryos and

that it may potentially be involved in CNS and eye development at this

stage.

• Whole-mount in situ hybridization was used to identify the spatial

expression of the mdm1 transcript within specific embryo tissues. These

results suggested that the mdm1 transcripts were initially expressed

ubiquitously throughout the embryo, and then became confined to the

anterior or cephalic region. This indicated that the mdm1 gene may play a

role in CNS development.

• Several morpholino oligonucleotides were designed to specifically target

one or both mdm1 transcripts within the zebrafish embryo. Unfortunately,

two of the splice-blocking morpholinos did not appear to effectively target

the mdm1 gene as they either gave a “no effect” or “completely lethal”

result. However, defects in eye, head and tail development were observed

in embryos injected with either the mdm1-ATG MO or the mdm1-Ex2 MO,

suggesting that these morpholinos are both correctly targeting the mdm1

gene, and that it is in fact involved in the development of the CNS,

particularly the eye.

• RNA rescue experiments were performed in order to confirm the observed

morpholino phenotypes that occurred from the mdm1 morpholino injections.

Embryos that were co-injected with morpholino and the in vitro transcribed

109

capped mdm1 mRNA displayed a full rescue phenotype of eye, head and

tail development.

• Staining for the biotin label in whole embryos also confirmed much of the

results from previous experiments that the mdm1 gene is expressed

significantly in the anterior portion of the embryo, particularly the CNS and

eye. These results also verified the in situ data of RNA expression.

Interestingly, it also suggested that the mdm1 RNA is expressed in

primordial eye and CNS at approximately 19-hpi, which is a slightly different

result than the earlier semi-quantitative RT-PCR findings. Overall, these

results confirm the expression of the mdm1 RNA and the targeting ability of

the morpholino within the embryo. Additionally, these results imply that the

mdm1 gene is a key component in the development of the eye in zebrafish

embryos.

• The injection experiments were validated by RT-PCR analysis using total

RNA from non-injected, mdm1-Ex2 injected, RNA Rescued, and RNA

overexpressed embryos. The results confirmed that the mdm1 gene is:

o Expressed in non-injected embryos

o Was knocked down in the morpholino injected embryos

o Can be rescued in the morpholino-injected embryos with the mdm1

RNA. Additionally, the mdm1 transcript expression did not appear

to change in the embryos injected with the in vitro transcribed

mRNA.

110

• Novel use of the in vitro Transcription and Translation kit to detect the

effectiveness of the mdm1-ATG MO targeting proved to be successful.

This allowed for the verification of the translation-blocking MO without

having a specific antibody reactive to zebrafish Mdm1. Since there are

relatively few zebrafish antibodies available, this technique is extremely

useful for rapid confirmation of decreased protein expression levels in vitro.

• Morpholinos were also used to test for genomic instability activity of mdm1

in golden heterozygous embryos, as mdm1 was currently a candidate gene

on the gin-12 list. In embryos with at least one developing eye, mosaics

were not observed, resulting in mdm1 no longer being considered a gin-12

candidate gene. Overexpression of the mRNA was also performed in

golden heterozygous embryos to determine if this caused genomic

instability activity in the embryo. Again, mosaics were not observed in any

of the injected embryos. Thus, it appears that mdm1 has no potential

genomic instability activity.

111

CHAPTER FIVE

CONCLUSIONS, IMPLICATIONS, AND FUTURE DIRECTIONS

Conclusions and Implications

The overall purpose of this research was to identify and characterize a

unique set of genomic instability mutants in the freshwater zebrafish. These gin

mutants exhibit increased incidence of somatic mutation during early

embryogenesis. Identifying the genes responsible for these modifications in the

zebrafish model will eventually lead to a better understanding of the development

and disease in humans.

The data presented throughout this dissertation demonstrate an in depth

investigation into the developmental expression and function of several genes,

located within known regions of genomic instability activity in the zebrafish model

organism. This project had three distinct objectives.

1. Chapter Two focused on the first objective, which was to explore

candidate genes within the gin-10 region of chromosome 18. Specifically,

the RNA expression and developmental function of synbl, rfx4, and sir2

genes were examined.

112

2. Chapter Three, dealt with the optimization of Mdm1 protein expression

experiments and determining the specificity of custom zebrafish Mdm1

antibodies.

3. Chapter Four focused specifically on the expression and function of the

mdm1 gene, a candidate for the gin-12 studies, during the early zebrafish

developmental stages.

All of the gin-10 Candidate Genes showed both Maternal and Zygotic

Gene Expression: Preliminary genetic studies determined that the gin mutants

required both maternal and zygotic expression in developing zebrafish. Using

RT-PCR, ten candidate genes were analyzed in WT(AB) embryos for maternal

and zygotic gene expression using 4-hpf and 24-hpf embryo RNA. Interestingly,

all of the candidate genes tested appeared to have both maternal and zygotic

transcript expression (Table 2.1). Predicted Sanger and NCBI orthology

information for the gin-10 candidate genes was used to select the top three

candidate genes, synbl, rfx4, and sir2 for future experimental focus. Transcript

expression analyses of these genes utilized the semi-quantitative RT-PCR and

whole-mount in situ hybridization techniques (Figures 2.5, 2.7, and 2.8). The rfx4

and sir2 transcripts appeared to have very distinct patterns of expression

throughout the zebrafish embryogenesis according to the semi-quantitative RT-

PCR data.

While exploring the synbl candidate gene, a paralog, ric8a, was found

during the database searches. Since it appeared that these paralogs were

113

involved in asymmetric cell division in other organisms, it was important to look

for any variations in the temporal and spatial expression of these genes during

development to determine whether they function cooperatively or have diverged

in function during teleost evolution. There was a very interesting difference in the

pattern of expression between the synbl paralogs during early zebrafish

development. The maternal expression of synbl is robust at 4-hpf, then a distinct

transitional period occurs when active zygotic transcription of the synbl gene

appears around 12-hpf. Expression levels appear to peak at 24-hpf but are still

detectable through 7-dpf. The ric8a gene does not appear to have maternal

expression or be actively transcribed in the embryo until approximately 18-hpf

and peaks at 72-hpf. It is clear that there is some divergence in the expression

of these genes during zebrafish development.

The Developmental Functions of Several gin-10 Candidate Genes were

Identified by Morpholino Injections: Translation-blocking MOs were designed to

target each of the gin-10 candidate genes and the ric8a gene (Table 2.2).

Additionally, a splice-blocking MO was designed for the synbl gene, which was

used alone and in combination with the ric8a MO and synbl-ATG MO. Injections

were performed in both WT(AB) and golden heterozygous embryos, which

allowed for the use of the mosaic eye assay to determine if the gene knockdown

led to mosaic eyes. The results of the MO injections suggested that none of the

targeted genes had genomic instability activity, since mosaic eyes were never

observed at 48- through 72-hpi (Figure 2.11). However, these experiments were

114

important to understand the general role of each of these genes in zebrafish

development. For example, based on the injection experiments and the resulting

phenotypes of the morphants, it appears that the sir2 gene is involved in

metabolic activity and had a 90-100% lethality rate in the injected embryos by 48-

hpi. These results further suggest that the zebrafish sir2 gene is more closely

related to the human sirt3 gene, which has been implicated in the regulation of

metabolic processes. Based on the morpholino and cloning data, it has been

concluded that the zebrafish sir2 gene is no longer a gin-10 candidate, but may

become useful for studying metabolic regulation and adaptive thermogenesis in

zebrafish.

When using the MOs to determine differences in the function of the synbl

and ric8a genes, it was determined that although some off-target effects occurred

in the development of the tail, both MOs caused cerebral edema and the general

abnormal development in the brain, as compared to the non-injected embryos.

Some published data had suggested that the synbl gene is involved in the

migration of pigment cells of the developing zebrafish. Interestingly, all of the

synbl, ric8a and combination injected embryos had normal looking melanocytes

that appeared to migrate properly. Based on these results, it is unclear whether

the morpholinos used (synbl-ATG, synbl-splice, and ric8a-ATG) were correctly

targeting the appropriate transcripts. However, it is important to note that similar

phenotypes were observed from each of these MO injections and when injected

in combination, which suggests the reported data is consistent with synbl and

ric8a function in the developing embryo. When applying the mosaic eye assay to

115

all of the knockdowns performed, it appears that none of the genes tested had

genomic instability activity.

Re-Mapping the Viral Insert of the ZIRC Transgenic Fish Line Led to an

Unexpected Result: To continue the research on the synbl paralogs, zebrafish

embryos containing a transgenic viral insert were purchased from the Zebrafish

International Research Center in Oregon. Self-crosses of the Tg-synbl fish were

performed and the resulting embryos were injected with synbl-ATG, ric8a-ATG or

synbl-splice/ric8a-ATG MOs to determine the effect of these knockdowns in a

synbl null background. The phenotypic results of these morphants did not vary

from the initial synbl injection experiments, which was an unexpected result

(Figure 2.12). Furthermore, the self-crosses did not result in a 25% dead-loss of

homozygous mutants, and therefore the transgenic insert was remapped in order

to determine the exact chromosomal location. Genomic PCR and cloning

analyses repeatedly suggested that the transgenic insert was not affecting the

synbl gene, but rather it was located within the second intron of the cry1b gene

(Figure 2.15, Table 2.3). The cry1b gene is also on chromosome 18 and is

located within the gin-10 region; however, it is on the adjacent contig to the

location of the synbl gene, and appears to be involved in circadian rhythms

according to recent published reports. Without having the appropriate transgenic

fish for these studies, moving forward to determine the developmental

differences of these genes in vivo was not plausible at the time.

116

Protein Expression Studies were Optimized in the Zebrafish Model for

Future Experiments to Study Mdm1 Protein Interactions: Studying protein

expression by western blotting and other techniques is a trivial component in

many research laboratories. However, it is highly underused in the zebrafish

community due to the unavailability of zebrafish specific antibodies and

protocols. For this reason, custom Mdm1 antibodies were designed and ordered

to look at the expression of this protein during zebrafish development, and use it

for protein-protein interaction studies to determine whether the Mdm1 protein is

involved in the regulation of the p53 tumor suppressor or another component of

its pathway. As Chapter Three describes, much of the initial effort was placed on

optimizing the lysate preparation and western blotting techniques using various

stage zebrafish embryos (Figure 3.1). Once a protocol was defined, it was

imperative to ensure the quality and specificity of the custom antibodies. The

use of the in vitro transcription and translation kit was utilized for this purpose, to

synthesize the Mdm1 protein and several control proteins that were run on the

western blots with embryo lysate to test for the specificity of the custom

antibodies (Figure 3.5). Repeatedly, the results indicated that the custom Mdm1

antibodies did not detect the Mdm1 protein (74 kDa), but rather a protein at

approximately 55 kDa; the custom antibodies also did not detect the synthesized

Mdm1 protein on any of the blots (Figures 3.6 and 3.7).

To further study the Mdm1 protein interactions in vivo without an antibody,

injecting embryos with FLAG or 6X-His tagged Mdm1 protein would potentially

allow for the co-immunoprecipitation of the tagged Mdm1 and interacting proteins

117

for analysis. It was unfortunate that during the time spent on cloning these tags

into the full-length mdm1 sequence, none of the clones analyzed by restriction

digests or sequencing contained the tags (Figures 3.11 and 3.15). Rather than

struggle with the protein expression studies, analyzing the developmental

function and expression of the gene was pursued. Since little was known about

the mdm1 in zebrafish or other organisms, using knockdown technologies and

developing novel techniques for these studies was crucial in identifying it’s

biological role during development.

The mdm1 Gene is Involved in the Development of the Zebrafish Eye and

CNS: Using many of the techniques described in Chapter Two, the

developmental expression patterns (both temporal and spatial) of the mdm1

gene were explored specifically in Chapter Four. Expression analysis results

consistently showed that the transcript expression of mdm1 was extremely varied

during development, implying that the transcription of this gene was highly

regulated during specific developmental time points (Figure 4.2). The in situ

hybridization data also suggested that the gene expression of mdm1 was

localized to the head of the embryo from approximately 24-hpf through 72-hpf

(Figure 4.3).

Morpholinos were also used for the targeted knockdown of the mdm1

transcripts in vivo (Table 4.1). Two of the mdm1 MOs, a translation-blocker and

a biotin-labeled splice-blocker, reliably gave the same phenotypic results in the

morphants. This data showed that the mdm1 gene is involved in the

118

development of the eye and CNS, since the knockdowns caused abnormal

development of the eyes and portions of the brain (Figures 4.5, 4.6, and 4.7). In

higher concentrations of the MOs, embryos would develop without eyes;

however, the survival of these was limited. To prove that the effects of these

MOs were not due to off-target events, RNA rescue, biotin staining for the splice-

blocking MO, and RT-PCR experiments were performed (Figures 4.9, 4.10, and

4.11). The combined results of these experiments were exciting in that they all

appeared to confirm the appropriate targeting of the mdm1 splice-blocking MO

and even the earlier in situ expression analysis. Importantly, these results imply

that the mdm1 gene is a key component in zebrafish eye development.

It was also important to show that the mdm1-ATG MO was also correctly

targeting the mdm1 gene. However, the only known way to determine this is by

using western blot analysis to look at the decrease of protein expression

following the injection. Using the tools available to the lab, a novel approach was

developed that allowed for the visualization of the MO effects on the synthesized

Mdm1 protein. By adding different concentrations of the mdm1-ATG MO to TnT

reaction samples, along with the biotin labeled tRNA (lysine), the final products

were run on a gel and detected with streptavidin. The results demonstrated the

successful targeting of the MO to the Mdm1 protein by decreasing protein

expression, which corresponded to the addition of increasing concentrations of

MO (Figure 4.12). The success of this technique to show direct targeting of the

MO in vitro should prove to be a rapid and useful tool for the zebrafish research

community. In addition to this in vitro approach to testing for morpholino

119

specificity, it may be possible to also show a rescue in this system by imitating an

in vivo rescue experiment by adding synthesized mRNA to the reactions.

Although this method has not yet been tested, it may prove to be an invaluable

step in assessing the specificity and quality of morpholinos prior to embryo

injections.

Finally, the underlying goal was to determine whether the mdm1 gene had

any genomic instability activity in the developing zebrafish embryo, as it was

identified within the gin-12 region of chromosome 4. Knockdown and RNA

overexpression experiments of the mdm1 gene in golden heterozygous embryos

did not give rise to mosaics. Additionally, the phenotype of the mdm1 RNA

injected embryos was normal as compared to non-injected embryos (Figure

4.13). From these results, it can be concluded that the mdm1 gene is no longer

a gin-12 candidate.

Future Directions

The gin-10 Project: Since the genes that were focused on in the current

project did not lead to any genomic instability activity in the zebrafish embryos

with the knockdowns, it is important to reevaluate the genes within the mapped

gin-10 region. The most recent annotated zebrafish chromosome 18 information

had suggested that some of the candidates listed in Table 2.1 are actually

located outside the gin-10 region, while new genes have recently been identified.

Some preliminary RT-PCR for those genes has been done (Appendix A)

although much more work is needed for the continuation of this project. While

120

knocking down individual candidate genes is a necessary component of this

project, it may not be an effective way to identify the cause of genomic instability

activity since it can be expensive to design the MOs and is very time consuming.

An alternative approach may be to use microarrays and comparative genomics

with cDNA from the known gin-10 carriers and either WT(AB) or golden fish to

determine the location of the loss of expression on chromosome 18 that leads to

the mosaic phenotype. It would also be interesting to look at the upstream and

promoter regions of the genes within the mapped gin-10 area, to identify

important transcription factor binding sites and other regulatory elements that

may be involved in the cause of genomic instability in these gin mutants.

Another direction that can be taken, although difficult in the zebrafish, is to

fluorescently label specific genes within the gin-10 region for fluorescent in situ

hybridization (FISH) analysis of chromosome spreads. It is possible that the

genomic instability is caused by a loss of function of more than one gene in this

region, which may not be found by individual gene knockdowns or even

microarrays.

The Mdm1 Protein Expression and Interaction Studies: Given that the

custom Mdm1 antibodies failed to detect the appropriate size protein in the

embryo lysate and the TnT product, it is reasonable to continue by designing

other peptides that target other regions of the Mdm1 protein. By looking at the

Mdm1 peptide sequences, it may be possible to find better epitope sites for

western blotting and co-IP techniques with zebrafish embryo lysates. While

121

having an effective antibody is important for these studies, it would be interesting

to determine what protein the custom Mdm1 antibodies were detecting at 55 kDa

protein on the western blots. Running an SDS-PAGE with zebrafish embryo

lysate and cutting out the 55 kDa proteins for mass spectrometry analysis or

using the antibody for an immunoprecipitation (IP) combined with mass

spectrometry may allow for this determination.

Another experimental approach that was partly explored in these studies

was adding a tag to the mdm1 sequence for protein interaction studies. Since

the 5’-region of the mdm1 sequence was modified by the addition of the Kozak

sequence and restriction sites, the research efforts focused on adding a FLAG or

6X-His tag to the 3’ sequence before the stop codon. Unfortunately, the tags

were not successfully added to the mdm1 sequence, although it did not appear

that there should be any sequence issues to complicate cloning based on the

sequence information. Another method would be to clone the full-length mdm1

gene without the stop codon (insert a restriction site immediately before the stop

codon), and insert this sequence into a vector containing the tag of choice. Upon

verification of the correct reading frame, this can be synthesized in vitro and the

tagged protein can be injected into zebrafish embryos and used for co-IPs to look

at potential Mdm1 protein interactions. Although this would not allow for

exploring the endogenous Mdm1 protein interactions, it could give rise to

pertinent preliminary protein data and lead to the identification of the pathways

that Mdm1 is involved in. Additionally, other tags could be incorporated into the

mdm1 sequence for the same purposes.

122

Further Analysis of the Zebrafish mdm1 Gene Function: While the results

described in Chapter Four have implicated the zebrafish mdm1 gene as having a

role in eye development, and recently published data have suggested its

involvement in age-related retinal degeneration in mouse models, the next step

should be to determine the mechanisms by which the mdm1 gene functions

within the development of the eye. One approach would be to evaluate the gene

and protein expression of Mdm1 in various known zebrafish eye mutants, such

as cyclops, chokh and pax6, which are used in studying eye development and

diseases. It was clear from our studies that mdm1 plays a role in the proper

development of the eye during early zebrafish development. The wnt11 and shh

gene families are well studied and known to be involved in the development of

the eye in zebrafish. It may be possible that the Mdm1 protein functions within

these pathways or is part of a protein complex that gives way to some of the

known zebrafish eye mutant phenotypes.

Another experimental approach would be to work across species using the

arrd (retinal degeneration) mice and try to rescue the phenotype with the

zebrafish Mdm1. By doing so, it would show that the zebrafish model is useful

for evaluating human/mammalian diseases and would also provide insight into

the cause of this age-related eye disorder. Evaluating the functional role of this

gene in zebrafish and the mechanism underlying eye development may reveal

critical biological pathways that operate in maintaining the normal functioning of

the eye, or future therapeutic targets for those affected by retinal degenerative

disorders.

123

CHAPTER SIX

MATERIALS AND METHODS

Materials:

Tricaine-S (Tricaine Methanesulfonate) was obtained from Western

Chemicals Inc. (Ferndale, WA) and was reconstituted in deionized water (1.5g/L)

to make the 10X stock. CSPD (chemiluminescence substrate) and NitroBlock II

were obtained from Roche and prepared as directed from the manufacturer. 1-

phenyl-2-thiourea (PTU) was obtained from Sigma and stored as a 10mM

solution in deionized water.

Buffers and Solutions:

Embryo bleach solution is 65µl/100ml bleach in aquarium water.

Methylene blue solution is 0.0015M Methylene Blue in aquarium water. 10X

Tricaine is 1.5g of Tricaine powder dissolved in 1 liter of distilled water. 10X

Loading dye is 0.41% Bromophenol Blue, 50% Glycerol. Semi-quantitative RT-

PCR wash buffer is 0.1M Maleic Acid, 0.15M NaCl, 0.3% Tween-20 pH 7.5.

Semi-quantitative RT-PCR Detection buffer is 1M Tris pH 9.5, 1M NaCl.

Denaturing solution is 1.5M NaCl, 0.5M NaOH. Neutralizing solution is 1M Tris

pH 8.0, 1.5M NaCl. Low-stringency buffer is 2X SSC, 0.1% SDS. High-

124

stringency buffer is 0.5X SSC, 0.1% SDS. In situ Full Hybridization mix is 50%

Formamide, 5X SSC, 0.1% Tween-20, 50µg/ml Heparin, 500µg/ml tRNA,

adjusted to pH 6.0 with citric acid. In situ 1X blocking buffer is 1X PBT, 2%

Sheep Serum, 2 mg/ml BSA. AP buffer is 1M Tris pH 9.5, 1M MgCl2, 5M NaCl,

0.1% Tween-20. Stop solution is 1X PBT pH 5.5, 1mM EDTA. 1X Danieau

solution is 58mM NaCl, 0.7mM KCl, 0.4mM MgSO4, 0.6mM Ca(NO3)2, 5.0mM

HEPES pH 7.6. Embryo lysis buffer is 10mM Tris pH 8.1, 1mM EDTA, 0.3%

Tween-20, 0.3% NP-40. Ringer’s solution is 116mM NaCl, 2.9mM KCl, 1.8mM

CaCl2, 5mM HEPES pH 7.2. Deyolking buffer is 55mM NaCl, 1.8mM KCl,

1.25mM NaHCO2 pH7.5. Wash buffer for embryo lysate preparation is 110mM

NaCl, 35mM KCl, 27mM CaCl2, 10mM Tris pH8.5. SDS Sample buffer is 63mM

Tris pH 6.8, 10% Glycerol, 2% SDS, 5% β-ME. 2X sample buffer for western

blots is 125mM Tris, 4mM EDTA, 0.005% Bromophenol Blue, 4% SDS, 25%

Glycerol. Tobin buffer is 192mM Glycine, 25mM Tris, 0.038% SDS, 20% MeOH.

1X PBS is 137mM NaCl, 27mM KCl, 4.3mM Na2HPO4, 1.47mM KH2PO4. TBST

is 50mM Tris pH 7.5, 150mM NaCl 0.2% Tween-20. 1X assay buffer is 20mM

Tris, 1mM MgCl2. CSPD detection solution is 0.1M Diethanolamine, 1mM MgCl2,

0.25mM CSPD, 5% NitroBlock II.

Zebrafish Husbandry

Fish are raised in a modular system consisting of 1-, 3-, and 10-liter

overflow style tanks. Water conditions are maintained by the addition of sodium

bicarbonate and instant ocean solutions to pure reverse osmosis (RO) water.

125

System water is also run through a variety of filters including biofilters, carbon,

and ultraviolet (UV) light. The general conditions of the fish system water are pH

between 6.8-7.2, conductivity of approximately 350mS and a temperature of 25-

27oC. Weekly tests are also performed on the system water to check for

ammonia, nitrite and nitrate levels, and alkalinity. The facility itself has a timed

overhead lighting system that allows for a 14-hour light, 10-hour dark cycle.

Maintenance of the fish system is relatively easy and inexpensive when

compared to other facilities housing mice, rats, rabbits and other model

organisms. Due to the modularity of the system, it is very easy to change and

clean tanks and maintain consistent water conditions.

Figure 6.1: Modular Fish System

126

Typically, larvae will be raised in small dishes in a 27oC incubator until

approximately 7 days post fertilization (dpf), when they can be placed in a larger

tank and fed an algae diet. Over several weeks, the fish diet is gradually

changed over to adult food, which includes live brine shrimp, flake food, and

bloodworms, and the fish will be at breeding age within approximately six months

Primers:

Numerous oligonucleotide primers were generated (IDT) in order to obtain

the results described in this report. The primers that were used are listed below

by primer name, grouped by gene of interest.

Table 6.1: List of Primers for gin-10, synbl and mdm1 projects.

Primer

Name

Sequence

Project

BTB F1

5'- ACAGGAACCAGCAGTGCTCAAGTA -3'

gin-10

BTB F2

5'- TCTTCACCGCCTCCAACAGGTTTA -3'

gin-10

BTB F3

5'- ACAACAGCCTAGACACCGTCAACA -3'

gin-10

BTB R1

5'- TGAGCATTCGGAAACCCAGATCCT -3'

gin-10

BTB R2

5'- AGTTCGGGTGCCTCCAG AATTTA -3'

gin-10

BTB R3

5'- TACTTGAGCACTGCTGGTTCCTGT -3'

gin-10

cirh1a F1

5'- AATGGCAGCTGCGAAGATGAATGG -3'

gin-10

cirh1a F2

5'- ATGATGTTCGAGCTGTGGCTGAGA -3'

gin-10

cirh1a R1

5'- ATCTCAGCCACAGCTCGAACATCA -3'

gin-10

cirh1a R2

5'- TACTATCAGGCAGTGGCTGGCTTT -3'

gin-10

cry1b F1

5'- TGGAGATGCCAGCAGAGACAATCA -3'

gin-10

cry1b F2

5'- ATGTGGCTTTCGTGCAGCTCATTC -3'

gin-10

cry1b F3

5'- TTGCTTCCATGTTGTTGACGTGCG -3'

gin-10

cry1b R1

5'- AATGAGCTGCACGAAAGCCACATC -3'

gin-10

cry1b R2

5'- ACCACATGACACTGCAAATGCTGG -3'

gin-10

cry1b R3

5'- TCTGGCGAAGCGTTTGATTTAGCC -3'

gin-10

Hi2039A

5'- CCATCATCGCAGGCTAACTAAG -3'

synbl

Hi2039A

5'- ACCCGTGTATCCAATAAACCC -3'

synbl

127

liprin F1

5'- TCAGTCCTGGAACACATGCAGTCA -3'

gin-10

liprin F2

5'- ATTGAGGAGTTGCGGCAGTCACTA -3'

gin-10

liprin F3

5'- TGGGACTGTATGTGACTTTGGCGA -3'

gin-10

liprin R1

5'- TCGCCAAAGTCACATACAGTCCCA -3'

gin-10

liprin R3

5'- TGATGCCCTCAGAGAAAGCTCCAA -3'

gin-10

mapk12 F1

5' - CAGGCTGCTCAAACACATGAAGCA - 3'

gin-10

mapk12 F2

5' - TAGACGATCAGCCTGTTGTGTGCT - 3'

gin-10

mapk12

5' - TCCAGGTGGTCGTGTCCTTTGAAT - 3'

gin-10

mapk12

5' - GAATGCCTGCTGTGCCAGCTTTAT - 3'

gin-10

mdm 1 R8

5'- CCTCAGACGCCATGGAGCATGAAG -3'

mdm1

mdm1 F1

5' - TGCGCTCTGACCAGTCAGGAATTA - 3'

mdm1

mdm1 F10

5'- ACGCCTCACCA GGTAACAG -3'

mdm1

mdm1 F2

5' - ACACATCACGATCGAACCACTCCA -3'

mdm1

mdm1 F3

5'- AGAAGAAGAAGGAGCGACCACACA -3'

mdm1

mdm1 F8

5'- CTTCATGCTCCATGGCGTCTGAGG -3'

mdm1

mdm1 F9

5'- CCCTAGTCACTACCGGCGTA -3'

mdm1

mdm1 koz

5'- CTCGAGCTGAAAGCCGCCATGCCTGTCCGTTTCAAG -3'

mdm1

mdm1 koz

5'- CTGAAAGCCGCCATGCCTGTCCGTTTCAAG -3'

mdm1

mdm1 R1

5' - AAGGGTGGGATGGCTCTGTTGTTA -3'

mdm1

mdm1 R10

5'- ACGTTTCCTCGTGAGATCAGGCTT -3'

mdm1

mdm1 R2

5'- AATTCCGCGTTTCTCATTGTGCCC -3'

mdm1

mdm1 R3

5'- AATATGGCTTGTGCTTCTGCAGGC -3'

mdm1

mdm1 R9

5'- TTGCATAGTTGAAGTGTGCTTTGACAGATTAGC -3'

mdm1

mdm1-

rHis1

5'-

TTAATGATGATGATGATGATGTTGAGGAAGCACAAACAGCTC

-3'

mdm1

mdm1-

rHis2

5'-

TTAATGATGATGATGATGATGTGAGGAAGCACAAACAGCTC -

mdm1

mdm1a F4

5'- TCTCCTAATCTGTTGACAAACCAAAGCAAG -3'

mdm1

mdm1a F5

5'- GCCTGCAGAAGCACAAGCCATATT -3'

mdm1

mdm1a F6

5'- TGCGCTCTGACCAGTCAGGAATTA -3'

mdm1

mdm1a F7

5'- CGGAGCTGTTTGTGCTTCCTCAAT -3'

mdm1

mdm1a R4

5'- ACCAACCCAATCTCACGGCAATTC -3'

mdm1

mdm1a R5

5'- AATATGGCTTGTGCTTCTGCAGGC -3'

mdm1

mdm1a R6

5'- AAGGTGGATGTCTGAGCTGCTTCT -3'

mdm1

mdm1a R7

5'- ACGTTTCCTCGTGAGATCAGGCTT -3'

mdm1

mdm1a R8

5'- CCATTTGTTGTCTATGCACAATACTGGTC -3'

mdm1

mdm2 F1

5'- CAACGGTCACCAGCAGATAA -3'

mdm1

mdm2 F3

5'- GATTCGCGCAACGGTCACCAGCAGATAACTACCAA -3'

mdm1

mdm2 F4

5'- TCTCGGTGCTGTTCTTGGAGTGAA -3'

mdm1

128

mdm2 F5

5'- TCGCACGTGCTGGAAGTGAACAAT -3'

mdm1

mdm2 F6

5'- ACGGTCACCAGCAGATAACTACCA -3'

mdm1

mdm2 koz

5'- CTACCAAAGCCGCCATGGCAACAGAGA -3'

mdm1

mdm2 koz

5'- CTCGAGCTACCAAAGCCGCCATGGCAACAGAGA -3'

mdm1

mdm2 R1

5'- AGTTTGGGAGTTGCCTTGTG -3'

mdm1

mdm2 R2

5'- TCAGTGCCTCAGCTCATGTAAGTC -3'

mdm1

mdm2 R3

5'- ACAACCACAAGGCAACTCCCAAACTTC -3'

mdm1

mdm2 R4

5'- TCCCGGGACAGAATCCACATCATT -3'

mdm1

mdm2 R5

5'- AGACAGTCTTCGCAGCCGAAAT -3'

mdm1

mdm2 R6

5'- ATCAATACAGCACCTCACAAGC -3'

mdm1

mdm2 R7

5'- CAGTGCCTCAGCTCATGTAAGTCA -3'

mdm1

mdm2 R7b

5'- GGTATATTTCAGTGCCTCAGCTCATGTAAGTC -3'

mdm1

mdm4 F1

5'- CACACACACGTCTCCCTTTC -3'

mdm1

mdm4 F2

5'- ACGTGACGACATACACCACGTCAA -3'

mdm1

mdm4 F4

5'- CTTTCAGGGTTGCCTTGGTGGTTT -3'

mdm1

mdm4 koz

5'- GAGTGAAAAGCCGCCATGACCTCATTGGCA -3'

mdm1

mdm4 koz

5'- CTCGAGGAGTGAAAAGCCGCCATGACCTCATTGGCA -3'

mdm1

mdm4 R1

5'- CTAGCCTCGCATTGATGGTT -3'

mdm1

mdm4 R2

5'- TCTTCCCACTCAGGCAGTAGCAAA -3'

mdm1

mdm4 R4

5'- CTGGGCAAGGAGCGTGAAATTTGT -3'

mdm1

mdm4 R5

5'- GATTTGTTGCTCATGCAATGAAGGT -3'

mdm1

mdm4 R5b

5'- AGATTTGTTGCTCATGCAATGAAGGTTTTGATGA -3'

mdm1

p53 F1

5'- GGCAATCCGAAAGTCGATAA -3'

mdm1

p53 F1b

5'- TCCGGGCAATCCGAAAGTCGATAA -3'

mdm1

p53 koz F2

5'- GCAAAGCAGCCGCCATGGCGCAAAACGA -3'

mdm1

p53 R1

5'- GCTCTTTTTGGACTGCCTTTT -3'

mdm1

p53 R2

5'- GCACAGTTGTCCATTCAGCACCAA -3'

mdm1

p53 R3

5'- ATCGGCTTGCAACCAATTCCGATG -3'

mdm1

p53 R4

5'- ACAAAGGTCCCAGTGGAGTGAACA -3'

mdm1

p53 R5

5'- GCATCCCATCACCTTAATCAGAGTCGCT -3'

mdm1

p53 R5b

5'- TAGCATCCCATCACCTTAATCAGAGTCGCTTCTTCC -3'

mdm1

pME F1

5'- CATTGTGCTGGCGCGGATTCTTTA -3'

mdm1

pME F2

5'- TTGCAGCCGAATACAGTGATCCGT -3'

mdm1

pME F3

5'- GGATCCGGTGGTGCAAATCAAAGA -3'

mdm1

pME F4

5'- TGTTCTGCGCCGTTACAGATCCAA -3'

mdm1

pME R1

5'- AGCGAGGAAGCGGAAGAGGAATTA -3'

mdm1

pME R3

5'- TTATAGTCCTGTCGGGTTTCGCCA -3'

mdm1

PS20 F1

5'- AAGAAGGACGGTCAGGCTGAATGA -3'

gin-10

PS20 F2

5'- ACCGAGGACAGTGCATCAAACAGA -3'

gin-10

129

PS20 R1

5'- TGGGTTGATGATGTGGCACTCGTA -3'

gin-10

PS20 R2

5'- AATAGTTGGCGGTTCATTCCGCTG -3'

gin-10

PS20b F3

5'- TACGAGTGCCACATCATCAACCCA -3'

gin-10

PS20b R3

5'- AATGCGGTCACCAACAGAAACTGG -3'

gin-10

ptn F1

5'- TGCATTCAGTCTGTCGCTCTCACA -3'

in situ

control

ptn F2

5'- ACCACAGGAATGAAGACTCGCACT -3'

in situ

control

ptn R1

5'- AGTGCGAGTCTTCATTCCTGTGGT -3'

in situ

control

ptn R2

5'- ACAGCAGCCTCTGGTGGATTTACA -3'

in situ

control

rfx4 F5

5' - TGATGCTGATGAGCTCCACTCCAA - 3'

gin-10

rfx4 F6

5' - TGGCTCGTTTCATCTGATTCACCT - 3'

gin-10

rfx4 F7

5' - ACTGGATACTGTCATACGCGCCAA - 3'

gin-10

rfx4 R10

5'- AGGTGATGTCTGCACTGTGGATCA -3'

gin-10

rfx4 R11

5'- CAGCTGCTTGGCAAACTTCCTGAT -3'

gin-10

rfx4 R13

5'- TGCTCGTATTCAGAGTTCCTGCGT -3'

gin-10

rfx4 R5

5' - AGTTTACCCACATCGGGATGAGCA - 3'

gin-10

rfx4 R6

5' - TGATCCACAGTGCAGACATCACCT - 3'

gin-10

rfx4 R9

5'- TGCTCATCCCGATGTGGGTAAACT -3'

gin-10

ric8a F1

5'- GTTTGATTCCAACCGACGCCATGT -3'

synbl

ric8a F2

5'- ACAGAGAGGCCAAACCACACATCA -3'

synbl

ric8a F3

5'- TTCTGGATCGCCCAGATAAAGCCA -3'

synbl

ric8a R1

5'- TCTCCATGAACTCCACCAGCACTT -3'

synbl

ric8a R2

5'- ATTGGACTCTCCGAACTGTTGGCT -3'

synbl

ric8a R3

5'- TGATCAGGAACGTAAGGCGCAGAT -3'

synbl

Rp11 F2

5'- TCTGGACAGGCTGAAGGTGTTTGA -3'

in situ

control

Rp11 F3

5'- TTTCCGCTATTGTGGCCAAGCAAG -3'

in situ

control

Rp11 R2

5'- GTGGATGCAGCCTTTATGACGCAA -3'

in situ

control

Rp11 R3

5'- ACAATCTTGAGAGCAGCTGGGACA -3'

in situ

control

RPB5 F1

5'- ACGAGGCGACAAATTGAGTGCAAC -3'

gin-10

RPB5 F2

5'- ATGGTGGCAGTGATGTCCAGGTTA -3'

gin-10

RPB5 R1

5'- GGCATGAAAGCCAAGGGACCAAAT -3'

gin-10

RPB5 R2

5'- TTGGTTCTCCATTCACCAGGTCCA -3'

gin-10

rpc2 F1

5'- GCGCAAACTTCCGTTCTGAGTGTT -3'

gin-10

rpc2 F2

5'- TTACTTTAGCAGGACTTGCCCGGA -3'

gin-10

rpc2 F3

5'- TGATGCCGACCCAATGTGGTATCT -3'

gin-10

rpc2 F4

5'- TCGCTGTAATGGTTCGCAGAGTGA -3'

gin-10

rpc2 F5

5'- TGAATAAGTCCATGCCCACCGTCA -3'

gin-10

rpc2 R1

5'- TCACTCTGCGAACCATTACAGCGA -3'

gin-10

130

rpc2 R3

5'- ACTTCTCTGTATTGTGGCTGGCCT -3'

gin-10

rpc2 R4

5'- TCCTGAATATCCCAGCAAGCCACA -3'

gin-10

rpc2 R5

5'- GCTTCGCAGCATTATTGGCATCCT -3'

gin-10

rpc2 R6

5'- ACATCGGTCAGTCAGGTTGGTTGA -3'

gin-10

rpc2 R7

5'- CAGTTTGAGGCGAGGGATGATGTT -3'

gin-10

Zrpc2 F2

5'- TCAACCAACCTGACTGACCGATGT -3'

gin-10

Zrpc2 F6

5'- AGGCCAGCCACAATACAGAGAAGT -3'

gin-10

Zrpc2 R2

5'- TGATCTGGTCCTGCCTCATGTGTT -3'

gin-10

Sir2 F1

5'- TGTGGTAGTTGCTGGAGCAGGAAT -3'

gin-10

Sir2 F2

5'- TTTGACGGTCTTCCCTTCATCCCT -3'

gin-10

Sir2 F3

5'- CAGCAGTCCTCATTGGCTGTTGAT -3'

gin-10

sir2 F4

5'- GTGTCGTGATGAGCAAGGCGAGGTTGAGCAG -3'

gin-10

Sir2 R1

5'- CAAGCTGGCAAAGGGCTCAATCTT -3'

gin-10

Sir2 R2

5'- TGCTGATTCCTGCTCCAGCAACTA -3'

gin-10

Sir2 R3

5'- ATGGCTGTTCATCAGGGTCTGGAT -3'

gin-10

Sir2 R4

5'- AGTCTTCCTGCAGACAGCCTCAAT -3'

gin-10

sir2 R5

5'- TCTCCGCTGTTCTCACCACTGCTGCTGATGTATG -3'

gin-10

synbl F1

5'- AAGAATTTGGCCTCCGTCCAGGTA -3'

synbl

synbl F10

5'- AGCAGAGTCCCTTGCTGTCTTCAA -3'

synbl

synbl F11

5'- TCTCTGCCTGCCTGTCAGACATTT -3'

synbl

synbl F12

5'- CGGATGCAACTGCAAGAGGGTTTA -3'

synbl

synbl F13

5'- GAAGCGAGAAGCGAACTGATTGGT -3'

synbl

synbl F14

5'- AGCTGTTCCATCTGTTCCTGACCT -3'

synbl

synbl F2

5'- TTACTTTAGCAGGACTTGCCCGGA -3'

synbl

synbl F3

5'- CATTTGGTCAGCGCGTCACATCAT -3'

synbl

synbl F4

5'- TGACACCTGTGCTGAGTCTGTTGA -3'

synbl

synbl F5

5'- TGAAGCCCAGAGGCTTGTCAGTAT -3'

synbl

synbl F6

5'- GACAGTGTTTCTAAGTGGACGGCA -3'

synbl

synbl F7

5'- TGAAGCCCAGAGGCTTGTCAGTAT -3'

synbl

synbl F8

5'- GACAATTTCCCTCCATGCACCGTT -3'

synbl

synbl F9

5'- TTCCCGAACTCCCAAATCTTCCCA -3'

synbl

synbl R1

5'- GATGCCTGCAATCAGCCGAAGTTT -3'

synbl

synbl R12

5'- GTATGCACAGATTCCATGTGGACC -3'

synbl

synbl R13

5'- CTGTTCCTACAGGAATAACTTGGGT -3'

synbl

synbl R14

5'- GACTTTCCAAGGAACATGGTTTCA -3'

synbl

synbl R2

5'- CACAGAGTGCCACAAACACACGTT -3'

synbl

synbl R3

5'- ATGATGTGACGCGCTGACCAAATG -3'

synbl

synbl R4

5'- TAAAGGTGCCAGAGTCCCGTCTTT -3'

synbl

synbl R5

5'- TAGCCAACGGCTAACTCGCTAACA -3'

synbl

synbl R6

5'- ACCTGGACGGAGGCCAAATTCTTA -3'

synbl

synbl R7

5'- ACACTGTCAGTCCTTCTGCACACA -3'

synbl

131

synbl R8

5'- ATGGAACCCTGAGGCGACGTTTAT -3'

synbl

synbl R9

5'- TTGTAGCCGGTCAACCAGCAAATG -3'

synbl

synbl sh

5'- AGAGCGATTGTGTGTTCAGTTGAGTCAT -3'

synbl

synbl.sh

5'- AGGAGGAGGTGTCTGTTAGGTTGT -3'

synbl

zActin F1

5'- TTGGCATGGGACAGAAAGACTCCT -3'

control

zActin F2

5'- TCACACCTTCTACAACGAGCTGCG -3'

control

zActin F3

5'- CACCACGGCCGAAAGAGAAATTGT -3'

control

zActin F4

5'- TTGCGGTATCCATGAGACCACCTT -3'

control

zActin R1

5'- ACTCCTGCTTGCTGATCCACATCT -3'

control

zActin R2

5'- GAAGCTGTAGCCTCTCTCGGTCAG -3'

control

zActin R3

5'- AAATGCATGGCAAGGAACTCACCC -3'

control

zActin R4

5'- TACCTCCCTTTCCAGTTTCCGCAT -3'

control

Antibodies:

Polyclonal rabbit antibodies against zebrafish p53 and mdm2 were

purchased from Anaspec (Freemont, CA). Polyclonal rabbit antibody against

zebrafish β-actin was purchased from Cell Signaling Technologies. Polyclonal

Goat Anti-Rabbit Alkaline Phosphatase Conjugated antibodies were purchased

from both Anaspec and Southern Biotech (Birmingham, AL). Streptavidin

Alkaline Phosphatase Conjugated antibody was purchased from Thermo

Scientific as a ready-to-use solution. Digoxigenin Alkaline Phosphatase antibody

was purchased from Roche.

Embryo Collection for RNA:

Embryos were collected from breeding tanks and brought over to the lab

in system water. They were then transferred into a glass bowl and washed using

embryo bleach solution, rinsed several times in aquarium water then stored in

Methylene Blue solution. Developing embryos were sorted into new petri dishes

132

(100 embryos per dish) and allowed to develop at approximately 27oC in an

incubator until appropriate age was reached. Embryos were collected in aliquots

of 50 per 1.5ml tube. Aquarium water was removed using a pulled glass pipette

and 1ml of 10X Tricaine was added to each tube on ice and allowed to incubate

10-15 minutes. The tricaine was then removed using a pulled glass pipette and

300µl of RNA Later was added to each tube and the embryos were stored at -

80oC.

RT-PCR Recipe and PCR Machine Program:

After RNA was isolated from the frozen embryos, it was quantified using a

spectrophotometer (A260) and a 1µg/µl dilution was made. The RT-PCR recipe is

based on the recommended recipe from Invitrogen, but modified for a 20µl

reaction:

Table 6.2: Standard RT-PCR Recipe

Component

Volume µl

RNA (1µg)

PCR Water

7.24

2X Buffer

Forward Primer

(20µM)

0.68

Reverse Primer

(20µM)

0.68

Platinum Taq

0.4

Total

20µl

133

Table 6.3: RT-PCR Program

Temperature

Time

Reverse Transcription

30 minutes

Pre-Denaturation

2 minutes

Denature

45 seconds

Anneal

55 (varied slightly

based on primer Tm)

45 seconds

Extension

1 minute 30 seconds

Repeat 35 Cycles

Final Extension

10 minutes

Hold

Once finished, 2.3µl of 10X loading dye was added to each PCR tube and

the products were run on an agarose gel (0.7-2.5% depending on the expected

size of the fragment). All gels were photographed and recorded by date and

PCR number. Primers that gave a robust band on the initial gel were repeated

with 24-hpf and 72-hpf RNA and run on a 0.7-1% low-melting temperature

agarose gel for extraction. The bands were purified using Qiagen’s MinElute

PCR purification kit and stored at -20oC for several days or used immediately.

TOPO TA Cloning with pCR4 and Top10 Cells:

Cloning fragments of candidate genes was done using RT-PCR and

TOPO TA Cloning kit (Invitrogen). Purified PCR product was ligated into pCR4

(sequencing) vector at room temperature for 30 minutes. The ligated plasmid

was then transformed into chemically competent Top 10 E.coli cells on ice for 30

minutes; the cells were heat shocked for 30 seconds at 42oC and placed

immediately on ice. SOC medium was added to the cells and they were shaken

at 37oC for 1 hour prior to plating on either ampicillin or carbenicillin plates, which

134

were then incubated at 37oC overnight. The plates were removed from the

incubator, checked for colony growth, sealed in parafilm, and stored at 4oC.

Sequencing Preparation:

In order to sequence the candidate gene, plasmid DNA was isolated from

the E. coli cells using Qiagen’s miniprep kit and the insert was verified by an

EcoRI digest. The EcoRI digest was performed because there are restriction

sites flanking the insert region in the pCR4 vector. So far, there have not been

any issues with digest sites within the inserts themselves, and this remains the

best way to determine if there is an appropriate insert in the plasmid prior to

sequencing. Once the digests were analyzed, the plasmid DNA was quantified

by spectrophotometry and dilutions of 5ng/µl were made and sent to Macrogen

Inc.

Sequencing Analysis:

Sequences were downloaded from the Macrogen website and analyzed

using the Geneious software program (Biomatters, Ltd.). Primers used for

sequencing and any internal primers were identified, sequences assembled, and

consensus sequences verified by NIH BLAST analysis.

DNA Probe Synthesis for Semi-quantitative RT-PCR:

DNA probes for semi-quantitative expression analysis were made using

the DIG High Prime Labeling Kit from Roche. Plasmid DNA was linearized using

135

a restriction enzyme that did not cut the insert itself, usually SpeI, NotI, or PstI

based on the vector map. The digest was performed for 3-4 hours and the

product was run on a 1% low-melting temperature agarose gel for 90 minutes,

photographed, extracted, and purified using Qiagen’s MinElute Kit. The plasmid

DNA was quantified by spectrophotometry before DIG labeling.

Only 1µg of linearized plasmid (up to 16µl) was needed for the probe and

was denatured in a boiling water bath for 10 minutes prior to labeling. DIG-High

Prime was added to the denatured DNA mixed briefly and collected at the bottom

of the tube by centrifugation and incubated overnight at 37oC according to the

Roche Protocol. The reaction was stopped the next day by adding 0.2M EDTA

to the tube and heating to 65oC for 10 minutes.

Labeling efficiency of the probe was determined before using in a semi-

quantitative experiment. A series of probe dilutions was made from 1ng/µl to

0.01pg/µl in supplied dilution buffer. A 1µl spot of each dilution was placed on a

positively charged nylon membrane and allowed to air dry. The membrane was

UV cross-linked then placed into a small container of Maleic Acid Buffer and

incubated for 2 minutes. The membrane was then placed in blocking solution for

30 minutes at room temperature. Once blocked, the membrane was placed in a

1:10,000 anti-DIG antibody solution for 30 minutes at room temperature. The

membrane was then transferred to a dish containing wash buffer and incubated

with agitation twice for 15 minutes. The Wash Buffer was poured off and

Detection Buffer was added to the membrane for 5 minutes. The membrane was

then placed on a piece of clear transparency film and covered with 1 ml of CSPD

136

for 5 minutes. A second piece of transparency film was placed on top of the

membrane and it was exposed to x-ray film for several minutes. The probes that

worked at the 0.1pg/µl dilution well were used for semi-quantitative analysis.

Semi-quantitative RT-PCR:

To further assess the expression of candidate genes in wildtype embryos,

semi-quantitative RT-PCR analysis was performed. The number of PCR cycles

varied between 15 and 21 for each of the candidate genes; this was tested by

using two different time points of embryo RNA (usually 24-hpf and 72-hpf) in a

normal 20µl RT-PCR reaction with primers specific to a particular candidate gene

and varying the number of cycles for each set. Once all of the reactions were

complete, the products were loaded onto an agarose gel and run for 90 minutes.

If bands were present on the gel with Ethidium Bromide staining then the

reactions were repeated and the number of PCR cycles were decreased until

bands were faint and difficult to see on the gel. Once the appropriate number of

cycles was determined for each candidate, an RT-PCR developmental profile

was performed using primers for the candidate gene that only amplified a single

band and was approximately 500-1000 base pairs.

Gels were run for 90 minutes at 80 Volts and placed on the

transilluminator to verify faint or no bands and photographed. The wells were

removed from the gel and the top left corner was notched for directionality. Gels

were placed in Denaturing Solution on a rocker for 30 minutes, rinsed briefly with

sterile deionized water, then washed in Denaturing Solution for another 30

137

minutes. The Denaturing Solution was discarded and the gel was placed in

Neutralizing Solution for 30 minutes on a rocker, rinsed briefly with sterile

deionized water and washed with Neutralizing Solution for another 30 minutes.

While the gel was in the last wash, the blot transfer stack was prepared by

soaking a positively charged nylon membrane (Roche) and two pieces of

Whatman paper in 2X SSC, and a long piece of Whatman paper in 20X SSC.

The transfer set-up was made by placing a large pipette tip box lid in a glass

baking dish and placing a piece of glass on top of the lid to create a platform.

The long piece of Whatman paper was laid across the glass creating a bridge;

20X SSC was then poured onto the platform and halfway up the baking so that

both ends of the Whatman paper were submerged. The two smaller pieces of

Whatman paper that had been soaking were placed onto the center of the

platform and air bubbles were removed prior to the next step. Once the gel was

ready, it was placed facedown onto the Whatman paper. The membrane was

placed directly on top of the gel. A glass pipette was rolled across the membrane

to remove any air bubbles between it and the gel at this point. Two more pieces

of Whatman paper was placed on top of the membrane, and then approximately

3-4 inches of cut paper towels were stacked on top. The stack was finished by

placing another piece of glass on top of the paper towels, covering with saran

wrap to keep it in place and adding a weight (two heating block inserts). Another

piece of saran wrap was added to keep the weight centered on the stack, and it

was allowed to sit over night, approximately 20 hours.

138

Once the transfer was complete, the stack was carefully disassembled

and the membrane was labeled with pencil and washed in 5X SSC with agitation

for 5 minutes. The membrane was then air dried for several minutes and was UV

cross-linked (1200J) and stored between two pieces of clean Whatman paper

until hybridization. The hybridization temperature was determined by calculating

the Tm and Topt for the probe being used. The membrane was then soaked in 5X

SSC and placed in a glass hybridization tube with 10ml of warm hybridization

solution. The tube was tightly capped and placed in the pre-warmed

hybridization oven for 1-2 hours. When pre-hybridization was almost complete,

the probe was denatured by placing it in a 15ml conical tube and boiling the

probe for 5 minutes then placing the probe on ice for 1-2 minutes. The pre-

hybridization solution was removed from the glass tube and the denatured probe

was added (approximately 10 ml). Blots were hybridized overnight and the total

hybridization time was recorded.

Figure 6.2: Calculation of Hybridization Temperature for each Probe. Melting temperature of the

probe was calculated then a range of hybridization temperatures was determined by the formulas

above.

Following hybridization, the probe was poured back into a 15ml conical

tube and stored at -20oC for future use. The blot was briefly washed in a low-

stringency buffer at 68oC. The buffer was removed and replaced with fresh low-

139

stringency buffer and washed with agitation at 68oC for 5 minutes; this step was

repeated twice. Once the low-stringency washes were complete, the membrane

was washed twice in high-stringency buffer at 68oC with agitation. The

membrane was then placed in a Wash Buffer at room temperature for 5 minutes

with agitation.

The membrane was placed in a shallow dish containing 50ml of Roche 1X

Blocking Buffer and allowed to incubate with agitation for 1 hour. Once the

membrane was blocked, it was placed in a 1:10,000 anti-DIG antibody solution

and incubated for 1 hour at room temperature. To remove the excess antibody,

the membrane was rinsed with Wash Buffer briefly then incubated in fresh Wash

Buffer at room temperature twice for 15 minutes. The Wash Buffer was removed

and the membrane was soaked in detection buffer for 5 minutes. The membrane

was placed on a clean piece of Whatman paper to remove excess Detection

Buffer, then immediately transferred onto a piece of clear transparency film.

Approximately 1-2ml of CSPD was pipetted onto the membrane then a second

piece of clear transparency film was placed on top of the membrane spreading

out the CSPD. The membrane was incubated at 37oC for 10-15 minutes to

activate the CSPD and exposed to x-ray film.

Embryo Collection for In Situ Hybridization Experiments:

Embryos were collected from breeding tanks and brought over to the lab

in system water. They were then transferred into a glass bowl and washed using

an embryo bleach solution, rinsed several times in aquarium water then stored in

140

a 0.003% PTU solution, which inhibits pigment formation. Developing embryos

were sorted into new petri dishes (100 embryos per dish) and allowed to develop

at approximately 27oC in an incubator until appropriate age was reached.

Embryos were collected in aliquots of 50 per 1.5ml tube. Aquarium water was

removed using a pulled glass pipette and 1ml of 10X Tricaine was added to each

tube on ice and allowed to incubate 10-15 minutes. The tricaine was then

removed using a pulled glass pipette and 300µl of 4% PFA in PBS was added to

each tube and stored at 4oC overnight.

The next day, the PFA was removed and the embryos were washed

several times in 1X PBS with agitation (until all of the PFA was removed, usually

3-4 washes). Embryos were then manually dechorionated in 1X PBS on an

agar-coated plate if needed and dehydrated by washing in increasing MeOH in

PBS solution (25% MeOH, 50% MeOH, 75% MeOH and 100% MeOH).

Embryos were stored in 100% MeOH at -20oC.

RNA Probe Synthesis for In Situ Hybridization Experiments:

DIG-labeled RNA probes were used for in situ hybridization experiments in

zebrafish embryos. Plasmids containing the appropriate gene fragment

(approximately 600-800bp) were linearized by restriction digest, either NotI or

SpeI depending on the orientation of the insert in the plasmid. The digest ran

approximately 3-4 hours and the product was run on a 1% low-melting

temperature agarose gel for 90 minutes, extracted, and purified using Qiagen’s

141

MinElute kit. The linearized plasmid was quantified by spectrophotometry prior to

the transcription reaction.

Ambion’s MAXIScript Transcription reaction kit was used to prepare RNA

Probes for in situs with 1µg of the linearized plasmid DNA. The transcription

reaction using a DIG-labeled UTP and either T3 or T7 enzyme was prepared

depending on the orientation of the fragment in the plasmid. The transcription

reaction was incubated at 37oC for 1 hour then Turbo DNase (from kit) was

added to the reaction and incubated at 37oC for an additional 15 minutes. The

reaction was stopped by the addition of 0.5M EDTA followed by purification of the

reaction using NucAway Spin Columns. Probes were stored at -20oC.

Whole Mount In Situ Hybridization:

Since the purpose of this experiment was to see spatial expression of

candidate gene transcripts in whole embryos, it was essential to properly and

carefully handle the embryos that were extremely fragile at this point. Embryos

were removed from the freezer, thawed on ice, placed in 6-well culture dishes

according to age, and rehydrated in PBT with increasing MeOH solution (25%

PBT, 50% PBT, 75% PBT and 100% PBT). The embryos were washed in 100%

PBT several times with gentle swirling at room temperature then treated with

Proteinase K diluted in PBT at room temperature.

142

Table 6.4: Proteinase K treatment of zebrafish embryos for in situ hybridization experiments

(From Thisse & Thisse 2008).

Developmental Stage

Duration of Proteinase K Treatment

1 cell – 1 somite stage

30 seconds

1-8 somite stage

1 minute

9-18 somite stage

3 minutes

18 somite – 24hpf

10 minutes

24hpf – 7dpf

30 minutes

The Proteinase K was removed from the embryos and a second PFA fix at

room temperature was done. The PFA was removed and the embryos were

washed in 100% PBT several times with gentle swirling to prepare embryos for

pre-hybridization. The PBT was removed and the embryos were incubated in

Full hybridization mix at 65oC for 5 hours without rocking. Prior to hybridization,

the RNA probe was boiled for 10 minutes and placed on ice. The probe was

added to each well of embryos, they were gently swirled for several seconds and

incubated for 1-2 days in a 65oC oven.

The probe was removed and stored at -80oC for future use. Embryos

were briefly washed in Full hybridization mix, then in a hybridization mix that did

not contain tRNA or heparin. Embryos continued to be washed in an increasing

2X SSC solution (25% 2X SSC, 50% 2X SSC, 75% 2X SSC and 100% 2X SSC).

The 2X SSC solution was removed from the embryos and they were washed

twice in a 0.2X SSC with 0.01% Tween-20 solution for 30 minutes. The embryos

were then washed in increasing PBT in 0.2X SSC solutions (25% PBT, 50%

PBT, 75% PBT and 100% PBT) in preparation for blocking.

The PBT was removed from the wells and the embryos were incubated at

room temperature in 1X blocking buffer for 4 hours. A 1:10,000 anti-DIG

143

antibody solution was made in 1X blocking buffer; the blocking buffer was

removed from the embryos and the antibody was added and incubated at 4oC

with gentle rocking overnight.

The following day, the antibody solution was removed and the embryos

were washed several times with PBT at room temperature. The PBT was

removed and the embryos were allowed to briefly dry in the wells to reduce the

formation of precipitate in the wells. Embryos were washed several times in AP

buffer at room temperature. BM Purple was used to stain the embryos; the AP

buffer was removed and fresh AP buffer added to each well. The BM Purple was

briefly centrifuged and added to each well, the entire dish was covered in foil and

placed in a drawer until color developed. The color development was stopped by

removing the BM Purple solution and adding fresh Stop Solution, incubating for

15 minutes at room temperature, and repeating this process. Embryos were

stored in Stop solution at 4oC or cleared in glycerol and photographed.

Fish Breeding for Injections:

Adult fish (1-2 females with 1 male) were placed in a breeding tank

containing a divider the evening prior to injections. The next morning, the

dividers were pulled out of the tanks and the fish were allowed to breed. As soon

as the females laid eggs, the divider was replaced and the eggs were collected

into a small plastic container and transported to the microinjection room.

Dividers were subsequently re-pulled and replaced until the fish stopped

144

breeding for that day. Females were given a minimum of two weeks rest

between injection experiments while the males were bred every 1-2 weeks.

Morpholino Injections:

Custom MOs were designed to the 5’ UTR and early coding sequence of

the specific candidate genes in order to block translation. The morpholino

powder was first resuspended in 1X Danieau solution to a concentration of 1mM

stock MO. Once the stock solution was prepared, MOs were diluted in 1%

phenol red and 1X Danieau at various concentrations. Prior to injections, the

MOs were heated to 65oC for 10 minutes, vortexed, and briefly centrifuged so

that any precipitate that may have formed was at the bottom of the tube.

Approximately 10µl of MO was loaded into a 1.2mm pulled glass capillary tube

containing filament using a gel-loading tip. The needle was placed in a vertical

mount (beaker with a mound of clay along the edge) until the MO solution

migrated to the tip of the needle. It was then placed in the micromanipulator and

the tip was broken using forceps in aquarium water.

Embryos were brought over from the fishroom and were immediately

transferred into glass bowls containing 0.0015M MeBlue in aquarium water.

Embryos were briefly observed under dissecting microscope to make sure that

they were all less than 8-cell stage and approximately all the same age for

consistency. 100 embryos were transferred to an area of a large petri dish lid

containing MeBlue water placed on the dissecting scope with a black

background. Embryos were held with a pair of forceps while the needle was

145

carefully extended through the chorion and into the yolk just below the

developing cell(s). The MO solution was released using a remote foot pedal

(pressure and volume settings were adjusted based on the needle) and a red

spot was seen in the yolk once the solution was injected. Embryos were

transferred to a new petri dish containing MeBlue water and allowed to develop

in a 27oC incubator.

Figure 6.3: Morpholino Injection set up.

Screening Morphants:

Although mosaicism can be seen at 48-72hpf, the morphant embryos

were screened daily for any obvious phenotypic changes. Any dead or decayed

embryo (which is often seen since it is difficult to differentiate 1-cell embryos from

water-activated eggs at the time of injection) were removed from the dishes and

recorded. The MeBlue water was also changed daily to keep the embryos as

clean as possible.

At 48-72hpf, embryos were placed in 2X Tricaine and viewed under the

dissecting scope. Embryos were scored for mosaicism and any developmental

146

abnormalities observed; the embryos were then grouped together by phenotype

and placed into a 6-well culture dish. Groups and/or single embryos were

photographed then collected into either 4% PFA in PBS or RNA Later for future

use.

Fin Clipping Adult Zebrafish:

Adult fish were placed in a 1X Tricaine solution until anesthetized. Each

fish was then removed from the solution, briefly blotted on a paper towel and

placed on a small cutting board. Using a scalpel, a small portion of the lower tail

fin was cut and placed in a 0.5ml thin-walled tube containing 50µl of embryo lysis

buffer on ice. The fish was promptly placed in a 1L tank and swirled around until

it began swimming on its own.

Genomic DNA Preps from Fin Clips:

After the fin clips were obtained, the 0.5ml tubes were incubated at 98oC

for 10 minutes. Then 5µl of Proteinase K (10mg/ml) was added to each tube;

they were briefly vortexed, centrifuged, and then placed in a 55oC incubator

overnight. The next day, the tubes were vortexed and re-centrifuged, and an

additional 45µl of lysis buffer and 5µl of Proteinase K were added to each tube.

Tubes were spun down and incubated at 55oC overnight. The DNA was then

stored at -20oC.

147

Genomic DNA PCR:

Before PCR was performed on the genomic samples, they were diluted

1:20 with Nuclease-free water and incubated at 98oC to inactivate the Proteinase

K. The PCR recipe is based on the recommended usage from Takara, but

modified for a 20µl reaction.

Table 6.5: Genomic DNA PCR recipe used to amplify a portion of the transgenic insert and

flanking genomic DNA in ZIRC fish.

Component

Volume (µl)

Diluted DNA

2.4

PCR Water

11.8

Takara 10X Buffer

Takara dNTPs

1.6

Forward Primer

Reverse Primer

Takara Taq

0.2

Total

20µl

A standard 35-cycle PCR program was used to amplify fragments from

genomic DNA. Once finished, 2.3µl of 10X loading dye was added to each

sample, and the products were run on an agarose gel.

Embryo Collection and Lysate Preparation for Western Blot Analysis:

Embryos were collected from breeding tanks and brought over to the lab

in system water. They were then transferred into a glass bowl and washed using

an embryo bleach solution, rinsed several times in aquarium water then stored in

Methylene Blue solution. Developing embryos were sorted into new petri dishes

(100 embryos per dish) and allowed to develop at approximately 27o C in an

incubator until appropriate age was reached. Embryos were dechorionated

manually using #5 Forceps in a 10X Tricaine solution. Embryos remained in the

148

10X Tricaine on ice for 10-15 minutes, and then were transferred to a 1.5ml tube.

While on ice, the Tricaine was removed using a pulled glass pipette and 1ml of

cold Ringer’s Solution was added to the tube. The embryos were gently rocked

for 2 minutes to wash and the Ringer’s Solution was removed with the glass

pipette. 1ml of Deyolking Buffer was then added to the tube and the embryos

were shaken for 5 minutes at 1100rpm. The embryos were centrifuged at 300 x

g for 30 seconds and the supernatant was discarded. A wash buffer was then

added to the embryos, shaken for 2 minutes at 1100rpm, and centrifuged to

pellet the embryos. The supernatant was discarded and embryos were

centrifuged again in order to remove any excess liquid that remained in the tube.

Once the excess liquid was removed, 200µl of SDS Sample Buffer was added to

the tube. The embryos were homogenized with a pestle and power

homogenizer. The sample was then boiled for 5 minutes then centrifuged at high

speed for 2 minutes. The supernatant was transferred to a new tube and stored

at -20oC (Link et al. 2006, O’Shea 1993).

Western Blot Analysis of Zebrafish Embryo Lysate:

Embryo lysates were diluted in 2X Cracking Buffer for a total volume of

15µl and incubated at 95oC for 5 minutes. The samples were then loaded onto a

10% SDS-Polyacrylamide gel and run at 180 Volts for approximately 45-50

minutes. Once the gel was finished, the blocking stack was assembled in a

shallow dish containing Tobin Buffer by placing two pieces of thick Whatman

paper on the bottom of the dish then the gel was carefully placed on top of the

149

blotting paper. The nitrocellulose membrane was hydrated in a shallow dish

containing deionized water then placed on top of the gel. Finally two more

pieces of thick blotting paper was placed on top of the membrane to complete the

stack. The stack was then removed from the dish and flipped over (membrane

side down) and placed in a Hoefer Scientific Blotter; the blotter was run at

110mAmps for 1 hour. Once the gel transfer was complete, the membrane was

stained with 0.1% Ponceau S to verify the protein transfer. Stained membranes

were also scanned or photographed for documentation. The membrane was

then rinsed several times in 1X PBS then blocked in 5% Nonfat Milk in TBST

overnight at 4oC. The membrane was then brought up to room temperature in

the blocking solution prior to the addition of antibody. The primary antibody was

diluted in blocking solution (1:500 – 1:4000 depending on the antibody), and the

membrane was incubated in primary antibody for 1-2 hours at room temperature.

The membrane was then washed several times in TBST then incubated for 1

hour at room temperature in AP-Conjugated secondary antibody (1:3000 dilution)

in blocking solution. Prior to detection with CSPD, the blot was washed several

times in TBST then in 1X Assay buffer. The membrane was placed on a piece of

transparency film on a flat surface, and 1-2ml of CSPD detection solution was

pipetted on and allowed to incubate for 5 minutes. A second piece of

transparency film was then placed on top of the membrane and exposed to x-ray

film.

150

In Vitro Expression of Protein:

Recombinant protein was produced from full-length clones in the

pCMVTnT expression vector (Promega) using the TnT Coupled Reticulocyte

Lysate System as detailed by the manufacturer (Promega). Upon completion of

the 90-minute TnT reaction, a portion of the sample was combined with an equal

volume of 2X cracking buffer and boiled for 5 minutes for western blotting; the

remaining sample was stored at -80oC for future use.

Whole-mount Antibody Staining of Morpholino Injected Embryos

Following the morpholino injections, embryos were placed in cold 2X

Tricaine-S solution in petri dishes and dechorionated with forceps. The embryos

were then rinsed four times in cold 1X PBS for 15 minutes, then fixed overnight in

4% PFA in PBS at 4oC. The following day, the fixed embryos were washed

several times in 1X PBT for 5 minutes. Embryos were then blocked in 10%

Normal Goat Serum (NGS) in 1X PBT for 1 hour at room temperature with gentle

rocking. The NGS block was then removed and the embryos were incubated

overnight at 4oC with the primary antibody (Streptavidin-AP) in 10% NGS in 1X

PBT. The primary antibody solution was removed the next day, and the embryos

were washed 4 times in 1X PBT for 5 minutes each. Secondary antibody (AP-

Conjugate) in 10% NGS/1X PBT (1:1000 dilution) was then added to the

embryos and incubated at room temperature for 2 hours. The antibody solution

was removed and the embryos were rinsed several times in 1X PBS for 10

151

minutes. The antibody was detected using NBT-BCIP and the reaction was

stopped after successful color development with several washes of 1X PBS.

152

LITERATURE CITED

Aguilera, A., and Gómez-González, B. (2008). Genome instability: a mechanistic

view of its causes and consequences. Nat Rev Genet 3, 204-217.

Amatruda, J., Shepard, JL., Stern, HM., and Zon, LI (2002). Zebrafish as a

cancer model system. Cancer Cell 1, 229-231.

Amsterdam, A., Nissen, RM., Sun, Z., Swindell, EC., Farrington, S. and Hopkins,

N. (2004). Identification of 315 genes essential for early zebrafish

development. Proc Natl Acad Sci U S A 101, 12792-12797.

Beckman, R.A., and Loeb, L.A. (2005). Genetic Instability in Cancer: Theory and

Experiment. Seminars in Cancer Biology 15, 423-435.

Beckwith, L.G., Moore, J.L., Tsao-Wu, G.S., Harshbarger, J.C., and Cheng, K.C.

(2000). Ethylnitrosurea induces neoplasia in zebrafish (Danio rerio).

Laboratory Investigation 80, 379-385.

Bilotta, J., Barnett, J.A., Hancock, L. and Saszik, S. (2004). Ethanol Exposure

alters zebrafish development: A novel model of fetal alcohol syndrome.

Neurotoxicology and Teratology 26, 737-743.

Blackshear, P., Graves, JP., Stumpo, DJ., Cobos, I., Rubenstein, JL. and Zeldin,

DC. (2003). Graded phenotypic response to partial and complete

deficiency of a brain-specific transcript variant of the winged helix

transcription factor RFX4. Development 130, 4539-4552.

Blander, G., and Guarente, L. (2004). The Sir2 Family of Protein Deacetylases.

Annu Rev Biochem 73, 417-435.

Brockerhoff, S.E., Dowling, J.E., and Hurley, J.B. (1998). Zebrafish retinal

mutants. Vision Research 38, 1335-1339.

153

Buck, S., Gallo, CM., and Smith, JS. (2004). Diversity in the Sir2 family of

protein deacetylases. J Leukoc Biol 75, 939-950.

Carroll, S., Grenier JK., Weatherbee, SD. (2000). From DNA to Diversity:

Molecular genetics and the evolution of animal design, 2nd edn (Oxford,

Blackwell Publishing).

Cavenee, W.K., Dryja, T.P., Phillips, R.A., Benedict, W.F., Godbout, R., Gallie,

B.L., Murphee, A.L., Strong, L.C., and White, R.L. (1983). Expression of

recessive alleles by chromosomal mechanisms in retinoblastoma. Nature

305, 779-784.

Chakrabarti, S., Streisinger, G., Singer, F. and Walker, C. (1983). Frequency of

gamma-ray induced specific locus and recessive lethal mutations in

mature germ cells of the zebrafish, Brachydanio rerio. Genetics 103, 109-

123.

Chang, B., Mandal, Md. N. A., Chavali, V. R. M., Hawes, N. L., Khan, N. W.,

Hurd, R. E., Smith, R. S., Davisson, M. L., Kopplin, L., Klein, B. E. K.,

Klein, R., Iyengar, S. K., Heckenlively, J. R. and Ayyagari, R. (2008). Age-

Related retinal degeneration (arrd2) in a novel mouse model due to a

nonsense mutation in the mdm1 gene. Hum Mol Genet 17, 3929-3941.

Chen, L., Gilkes, D.M., Pan, Y., Lane, W.S., and Chen, J. (2005). ATM and

Chk2-dependent phosphorylation of MDMX contribute to p53 activation

after DNA damage. EMBO 24, 3411-3422.

Cheng, K., and Moore, JL. (1997). Genetic dissection of vertebrate processes in

the zebrafish: a comparison of uniparental and two-generation screens.

Biochem Cell Biol 75, 525-533.

Couwenbergs, C., Spilker, AC. and Gotta, M. (2004). Control of embryonic

spindle positioning and Galpha activity by C. elegans RIC-8. Curr Biol 14,

1871-1876.

Currie, P.D. (1996). Zebrafish genetics: Mutant cornucopia. Current Biology 6,

1548-1552.

154

David, N., Martin, CA., Segalen, M., Rosenfeld, F., Schweisguth, F. and

Bellaïche, Y. (2005). Drosophila Ric-8 regulates Galphai cortical

localization to promote Galphai-dependent planar orientation of the mitotic

spindle during asymmetric cell division. Nat Cell Biol 7, 1083-1090.

Detrich, H.W., III,, Westerfield, M., and Zon, L.I. (1999). The Zebrafish: Genetics

and Genomics, Vol 60 (San Diego, Academic Press).

Dooley, K., and Zon, L.I. (2000). Zebrafish: a model system for the study of

human disease. Current Opinion in Genetics & Development 10, 252-256.

Driever, W., Solnica-Krezel, L., Schier, A.F., Neuhauss, S.C.F., Malicki, J.,

Stemple, D.L., Stainer, D.Y.R., Zwartkruis, F., Abdelilah, S., Rangini, Z.,

Belak, J. and Boggs, C. (1996). A Genetic Screen for Mutations Affecting

Embryogenesis in Zebrafish. Development 123, 37-46.

Drummond AJ, A.B., Buxton S, Cheung M, Cooper A, Duran C, Field M, Heled J,

Kearse M, Markowitz S, Moir R, Stones-Havas S, Sturrock S, Thierer T,

Wilson A (2011). www.geneious.com

Einhauer, A., and Jungbaur, A. (2001). The FLAG peptide, a versitile fusion tag

for the purification of recombinant proteins. J Biochem Biophys Methods

49, 455-465.

Eisen, J. (1996). Zebrafish make a big splash. Cell 87, 969-977.

Eisen, J.S., and Smith, J.C. (2008). Controlling morpholino experiments: don't

stop making antisense. Development 135, 1735-1743.

Ekker, S.C., and Larson, J.D. (2001). Morphant technology in model

developmental systems. Genesis 30, 89-93.

Emery, P., Durand, B., Mach, B. and Reith, W. (1996). RFX proteins, a novel

family of DNA binding proteins conserved in the eukaryotic kingdom.

Nucleic Acids Res 24, 803-807.

Feitsma, H., and Cuppen, E. (2008). Zebrafish as a Cancer Model. Mol Cancer

Res 6, 685-694.

155

Fornzler, D., Her, H., Knapik, E.W., Clark, M., Lehrach, H., Postlethwait, J.H.,

Zon, L.I., and Beier, D.R. (1998). Gene Mapping in Zebrafish Using

Single-Strand Conformation Polymorphism Analysis. Genomics 51, 216-

222.

Fox, E.J., and Loeb, L.A. (2010). Lethal Mutagenesis: Targeting the Mutator

Phenotype in Cancer. Seminars in Cancer Biology 20, 353-359.

Freedman, D., Wu, L. and Levine, AJ. (1999). Functions of the MDM2

oncoprotein. CMLS, Cell Mol Life Sci 55, 96-107.

Gasteiger, E., Gattiker, A., Hoogland, C., Ivanyi, I., Appel, R.D., Bairoch, A.

(2003) ExPASy: the proteomics server for in-depth protein knowledge and

analysis. Nuc Acids Res. 31, 3784-3788.

Gerhard, G. (2003). Comparitive aspects of zebrafish (Danio rerio) as a model

for aging research. Experimental Gerontology 38, 1333-1341.

Gerlai, R. (2003). Zebra fish: An uncharted Behavior Genetic Model. Behavior

Genetics 33, 461-468.

Gerlai, R., Lahav, M., Guo, S. and Rosenthal, A (2000). Drinks like a fish: zebra

fish (Danio rerio) as a behavior genetic model to study alcohol effects.

Pharmacology, Biochemistry and Behavior 67, 773-782.

Gestl, E.E., Kauffman, E.J., Moore, J.L., and Cheng, K.C. (1996). New conditions

for generation of gynogenetic half-tetrad embryos in zebrafish. J Hered

88, 76-79.

Gonzalez, C. (2007). Spindle orientation, asymmetric division and tumour

suppression in Drosophila stem cells. Nat Rev Genet 8, 462-472.

Greene, J. (1999). A little fish makes a big splash. In NCRR Reporter.

Grunwald, D.J., and Eisen, J.S. (2002). Headwaters of the zebrafish --

emergence of a new model vertebrate. Nat Rev Genet 3, 717-724.

156

Grunwald, D.J., and Streisinger, G. (1992). Induction of recessive lethal and

specific locus mutations in the zebrafish with ethyl nitrosurea. Genet Res

Camb 59, 103-116.

Guo, S. (2008). Animal Models for Anxiety Disorders. In Biomarkers for

Psychiatric Disorders, C.W. Turk, ed. (San Francisco, Springer Science),

pp. 203-216.

Haffter, P., Granato, M., Brand, M., Mullins, M.C., Hammerschmidt, M., Kane,

D.A., Odenthal, J., van Eeden, F. J. M., Jiang, Y-J., Heisenberg, C-P.,

Kelsh, R. N., Furutani-Seiki, M., Vogelsang, E., Beuchle, D., Schach, U.,

Fabian, C. and Nusslein-Volhard, C. (1996). The Identification of Genes

with Unique and Essential Functions in the Development of the Zebrafish,

Danio rerio. Development 123, 1-36.

Haffter, P., Odenthal, J., Mullins, M.C., Lin, S., Farrell, M.J., Vogelsang, E., Haas,

F., Brand, M., van Eeden, F.J.M., Furutani-Seiki, M., et al. (1996).

Mutations affecting pigmentation and shape of the adult zebrafish.

Developmental genes in evolution 206, 260-276.

Hampoelz, B., Hoeller, O., Bowman, SK., Dunican, D., and Knoblich, JA. (2005).

Drosophila Ric-8 is essential for plasma-membrane localization of

heterotrimeric G proteins. Nat Cell Biol 7, 1099-1105.

Hansen, C., Bartek, J., and Jensen, S. (2008). A functional link between the

human cell cycle-regulatory phosphatase Cdc14A and the atypical

mitogen-activated kinase Erk3. Cell Cycle 7, 325-334.

Hendricks, J.C., Sehgal, A. and Pack, A.I. (2000). The need for simple animal

model to understand sleep. Progress in Neurobiology 61, 339-351.

Hengen, P. (1995). Purification of His-Tag fusion proteins from Escherichia coli.

Trends Biochem Sci 20, 285-286.

Hochuli, E., Bannwarth, W., Döbeli, H., Gentz, R., and Stüber, D. (1988). Genetic

Approach to Facilitate Purification of Recombinant Proteins with a Novel

Metal Chelate Adsorbent. Nat Biotech 6, 1321-1325.

157

Hopp, T.P., Gallis, B., and Prickett, K.S. (1988). A short polypeptide marker

sequence useful for recombinant protein identification and purification.

Bio/Technology 6, 1204-1210.

Ingham, P. (2009). The power of the zebrafish for disease analysis. Hum Mol

Genet 18, R107-112.

Ito, A., Kawaguchi, Y., Lai, C-H., Kovacs, JJ., Higashimoto, Y., Appella, E., and

Yao, T-P. (2002). MDM2±HDAC1-mediated deacetylation of p53 is

required for its degradation. EMBO 21, 6236-6245.

Iwakuma, T., and Lozano, G. (2003). MDM2, an introduction. Molecular cancer

research : MCR 1, 993-1000.

Johnston, C., Afshar, K., Snyder, JT., Tall, GG., Gonczy, P., Siderovski, DP and

Willard, FS. (2008). Structural Determinants Underlying the Temperature-

sensitive Nature of a G Mutant in Asymmetric Cell Division of

Caenorhabditis elegans. J Biol Chem 283, 21550-21558.

Kari, G., Rodeck, U., and Dicker, AP. (2007). Zebrafish: an emerging model

system for human disease and drug discovery. Clin Pharmacol Ther 82,

70-80.

Karlovich, C., John, RM., Ramirez, L., Stainier, DYR., and Myers, RM. (1998).

Characterization of the Huntington’s disease (HD) gene homolog in the

zebrafish Danio rerio. Gene 217, 117-125.

Kawakami, K. (2005). Transposon Tools and Methods in Zebrafish.

Developmental dynamics : an official publication of the American

Association of Anatomists 234, 244-254.

Kelsh, R.N., Schmid, B., and Eisen, J.S. (2000). Genetic analysis of

melanophore development in zebrafish embryos. Developmental biology

225, 277-293.

Kimmel CB, B.W., Kimmel SR, Ullmann B and Schilling TF. (1995). Stages of

embryonoc development of the zebrafish. Developmental Dynamics 203,

253-310.

158

Knudson, A.G. (1971). Mutation and cancer: statistical study of retinoblastoma.

Proceedings of the National Academy of Sciences of the United States of

America 68, 820-823.

Knudson, A.G. (1985). Hereditary cancer, oncogenes, and antioncogenes.

Cancer Research 45, 1437-1443.

Krens, S., Spaink, HP., Snaar-Jagalska, BE. (2006). Functions of the MAPK

family in vertebrate-development. FEBS Lett 580, 4984-4990.

Langenau, D.M., Feng, H., Berghmans, S., Kanki, J.P., Kutok, J.L., and Look,

A.T. (2005). Cre/lox-regulated transgenic zebrafish model with conditional

myc-induced T cell acute lymphoblastic leukemia. Proceedings of the

National Academy of Sciences of the United States of America 102, 6068-

6073.

Langheinrich U, H.E., Stott G, Vacun G. (2002). Zebrafish as a model organism

for the identification and characterization of drugs and genes affecting p53

signaling. Curr Biol 12, 2023-2028.

Lele, Z., and Krone, P.H. (1996). The zebrafish as a model system in

developmental, toxicological and transgenic research. Biotechnology

Advances 14, 57-72.

Lieschke, G.J., and Currie, P.D. (2007). Animal models of human disease:

zebrafish swim into view. Nat Rev Genet 8, 353-367.

Link, V., Shevchenko, A., and Heisenberg, C-P. (2006). Proteomics of early

zebrafish embryos. BMC Developmental Biology 6.

Livingston, B.T. (2008), K. Griffett, ed. (Tampa).

Mach, B., Steimle, V., Martinez-Soria, E., and Reith, W. (1996). Regulation of

MHC class II genes: lessons from a disease. Annu Rev Immunol 14, 301-

331.

Marone, M., Mozzetti, S., De Ritis, D., Pierelli, L., and Scambia, G. (2001).

Semiquantitative RT-PCR analysis to assess the expression levels of

multiple transcripts from the same sample. Biol Proced Online 3, 19-25.

159

Maslow, T. (2010). Preparation of Zebrafish Embryo Protein Lysates, K. Griffett,

ed.

Matsushita, H., Uenaka, A., Ono, T., Hasegawa, K., Sato, S., Koizumi, F.,

Nakagawa, K., Toda, M., Shingo, T., Ichikawa, T., Noguchi, Y., Tamiya,

T., Furuta, T., Kawase, T., Date, I., and Nakayama, E. (2005).

Identification of glioma-specific RFX4-E and -F isoforms and humoral

immune response in patients. Cancer Sci 96, 801-809.

Matsuzaki, F. (2005). Drosophila G-protein signalling: intricate roles for Ric-8?

Nat Cell Biol 7, 1047-1049.

Matthews, M., Trevarrow, B., and Matthews, J. (2002). A virtual tour of the Guide

for zebrafish users. Lab animal 31, 34-40.

Miller, K., and Rand, JB. (2000a). A Role for RIC-8 (Synembryn) and GOA-1

(Goa) in Regulating a Subset of Centrosome Movements During Early

Embryogenesis in Caenorhabditis elegans. Genetics 156, 1649-1660.

Miller, K.G., Emerson, M.D., McManus, J.R., and Rand, J.B. (2000b). RIC-8

(Synembryn): A Novel Conserved Protein that Is Required for Gqa

Signaling in the C. elegans Nervous System. Neuron 27, 289-299.

Mizgireuv, I.V., and Revskoy, S.Y. (2006). Transplantable tumor lines generated

in clonal zebrafish. Cancer research 66, 3120-3125.

Moore, J., Gestl, EE., and Cheng, KC. (2004). Mosaic eyes, genomic instability

mutants, and cancer susceptibility. Methods Cell Biol 76, 555-568.

Moore, J., Rush, LM., Breneman, C., Mohideen, MA., and Cheng, KC. (2006).

Zebrafish genomic instability mutants and cancer susceptibility. Genetics

174, 585-600.

Morcos, P.A. (2007). Achieving targeted and quantifiable alteration of mRNA

splicing with Morpholino oligos. Biochem and Biophys Res Commun 358,

521-527.

160

Moreira, J., and Deutsch, A. (2005). Pigment pattern formation in zebrafish

during late larval stages: a model based on local interactions.

Developmental dynamics : an official publication of the American

Association of Anatomists 232, 33-42.

Morotomi-Yano, K., Yano, K., Saito, H., Sun, Z., Iwama, A., and Miki, Y. (2002).

Human regulatory factor X 4 (RFX4) is a testis-specific dimeric DNA-

binding protein that cooperates with other human RFX members. J Biol

Chem 277, 836-842.

Moulton, J.D. (2007). Using morpholinos to control gene expression. In Curr

Protoc Nucleic Acid Chem.

Muggerud, A., Edgren, H., Wolf, M., Kleivi, K., Dejeux, E., Tost, J., Sørlie, T., and

Kallioniemi, O. (2009). Data integration from two microarray platforms

identifies bi-allelic genetic inactivation of RIC8A in a breast cancer cell

line. BMC Med Genomics 11.

Mullins, M., Hammerschmidt, M., Haffter, P., and Nüsslein-Volhard, C. (1994).

Large-scale mutagenesis in the zebrafish: in search of genes controlling

development in a vertebrate. Curr Biol 4, 189-202.

Nagayoshi, S., Hayashi, E., Abe, G., Osato, N., Asakawa, K., Urasaki, A.,

Horikawa, K., Ikeo, K., Takeda, H., and Kawakami, K. (2008). Insertional

mutagenesis by the Tol2 transposon-mediated enhancer trap approach

generated mutations in two developmental genes: tcf7 and synembryn-

like. Development 135, 159-169.

Nasevicius, A., and Ekker, S.C. (2000). Effective Targeted Gene "Knockdown" in

Zebrafish. Nat Genet 26, 216-220.

Nawrocki, L., BreMiller, R., Streisinger, G. and Kaplan, M. (1985). Larval and

Adult Visual Pigments of the Zebrafish, Brachydanio rerio. Vision Res 25,

1569-1576.

North, B., and Verdin, E. (2004). Sirtuins: Sir2-related NAD-dependent protein

deacetylases. Genome Biology 5, 224.

Norton, W., and Bally-Cuif, L. (2010). Adult zebrafish as a model organism for

behavioural genetics. BMC Neuroscience 11.

Students also viewed