University of Southern California Dissertations and Theses

Opposing function of chd7 and phf6 in zebrafish craniofacial development

(USC Thesis Other)

Opposing function of chd7 and phf6 in zebrafish craniofacial development

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
I

Opposing function of chd7 and phf6 in
zebrafish craniofacial development

By
Yuhan Sun
1

Mentor: Ruchi Bajpai
1,2

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfilment of the Requirements for the Degree
MASTER OF SCIENCE
(Biochemistry and Molecular Biology)
August 2016

1 Department of Biochemistry & Molecular Biology, Keck School of Medicine, University of
Southern California, 90033 CA, USA
2: Center for Craniofacial Molecular Biology, Ostrow School of Dentistry, USC

II

ACKNOWLEDGMENTS
I would like to express my special appreciation and thanks to my advisor, Dr. Ruchi Bajpai, for
your constant encouragement and guidance in researching and writing this thesis. Thank you for
giving me this opportunity to participate in this amazing lab. I would also like to thank my
other thesis committee members, Dr. Zoltan Tokes and Dr. Ching-Ling Lien, for your guiding
comments and suggestions.

I would like to thank all the members of Ruchi’s lab for being a wonderful supportive team, and
for their helpfulness: Susan Smith, Kaivalya Shevade, Jennifer Oki, Annie Lynch, Uma
Sundaram, Erin Moran, Casey Griffin, Candida Toribio, Peter Guan, and George Tseng.

I would also like to express thanks to members of the Crump lab for their help and support.
Especially, thanks to Megan Matsutani for sharing with me many transgenic fluorescent fish.
Thanks to Lindsey and Joanna for sharing the protocol for CRISPR/Cas9-based trangenesis and
Pingfei for helping with troubleshooting PCR problems.

Thank you everyone for your help and support!

III

ABSTRACT
Background: In our studies of neural crest cells (NCCs) and craniofacial development, our lab
has identified a physical interaction between two human proteins: CHD7, a chromatin
remodeling protein, and PHF6, a dual PHD finger protein of unknown function. Mutations in
these genes result in distinct syndromes in humans with contrasting craniofacial defects.
Patients with CHARGE syndrome, with CHD7 mutations, have dysmorphic faces with reduced
neural crest derived structural and sensory tissues. On the other hand, patients with Borjeson
Forssman Lehmann syndrome (BFLS), caused by plant homeodomain Finger Protein 6 (PHF6)
mutations, have a thick calvarium and broad jaw.

Purpose: Using zebrafish as a model system, in this project we generated novel tools for
understanding BFLS and CHARGE syndromes and defined the functions of chd7 and phf6 in
craniofacial development.

Methods: A clustered regularly interspaced short palindromic repeats (CRISPR)
/CRISPR-associated (Cas) system was used to knock out these two genes in zebrafish embryos.
PCR was used to identify the mutants. We also observed effects of mutation or knockdown on
activation of known (sox10) and novel (RARG , CCND1) neural crest enhancers using confocal

IV

microscopy. In-situ hybridization will be performed in future to check the expression of some
markers of NCCs and craniofacial tissue.

Results: Chd7 mutant fish have phenotypes including developmental delay, reduced amounts of
facial tissue, small heads, small eyes and dysmorphic craniofacial cartilage, similar to defects
seen in CHARGE patients. In contrast, phf6 mutants have a significantly smaller or absent brain
with a normal face.

Conclusions: Chd7 and phf6 have opposing functions in craniofacial development in zebrafish.
Mutations in these genes cause phenotypes similar but more severe to CHARGE and BFLS
patients, respectively.

Key words ： CRISPR/Cas9, Morpholino, neural crest, zebrafish

V

Table of Contents

ACKNOWLEDGMENTS .................................................................................................. II
ABSTRACT ...................................................................................................................... III
Table of Contents ................................................................................................................ V
Introduction ......................................................................................................................... 1
Zebrafish – The Animal Model……………………………………………………………..1
Reverse Genetic Techniques – The Study Approach……………………………………….4
ZFN (Zinc-Finger Nucleases)………………………………………………………………4
TALEN (Transcription Activator-Like Effector Nucleases)………………………………..5
CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)…………………...7
MO (Morpholino)…………………………………………………………………………...9
Neural Crest Cells………………………………………………………………………….11
Our previous studies in other animal models………………………………………………15
Materials and Methods ..................................................................................................... 20
Creation of Cas9 gRNA DNA template:…………………………………………………..20
chd7 morpholino…………………………………………………………………………..23

VI

Microinjection……………………………………………………………………………..24
Imaging……………………………………………………………………………………25
Genotyping………………………………………………………………………………...25
Kits………………………………………………………………………………………...26
List of Zebrafish lines used in this study:…………………………………………………26
Results ................................................................................................................................. 27
Zebrafish is a good model for CHARGE Syndrome………………………………………27
chd7 MO knockdown and rescue………………………………………………………….27
The opposing functions of chd7 and phf6 in neural crest cell (NCC) migration………….30
phf6 fish have a small brain (microcephaly)………………………………………………33
chd7 fish have an asymmetric face………………………………………………………..37
Discussion ........................................................................................................................... 42
Further Study ..................................................................................................................... 46
Works Cited ....................................................................................................................... 47
University of Southern California Master ’s Thesis Introduction
1

Introduction
Zebrafish – The Animal Model
Zebrafish (Danio Rerio) belong to the minnow family of the order Cypriniformes
[1]
. These
tropical fresh water fish are widely used in developmental biology studies and disease modeling.
Zebrafish develop fast, taking only about 24 hours to become free-swimming larvae with
well-formed organs and tissues. Being vertebrates, they also share many similarities with
humans; remarkably, the zebrafish and human genomes are 87% similar. For this reason, the
zebrafish have become a well-known animal model in the study of biology
[2]
.
Zebrafish offer many advantages for developmental studies. Compared to mice, zebrafish
produce 10 -30 times more offspring from each parental cross, and can be mated once a week.
Therefore, it is possible to derive many different test groups from one family, which
diminishes the differences between each individual. Furthermore, the developing embryos are
transparent, which makes them easier to study
[3]
. Unlike drosophila, zebrafish are vertebrates
and therefore have a closer relationship to humans. Moreover, zebrafish have a strong
regenerative capacity, and can even regenerate part of the heart
[4][5]
.
As a model to study the facial cartilage, larval zebrafish also have some advantages. The
most obvious one is that the generation of facial tissue can be continuously imaged over time
University of Southern California Master ’s Thesis Introduction
2

in living embryos. Since fish and humans are both vertebrates, the cartilage structures also
share some similarities. Moreover, the formation period is very fast in zebrafish, lasting only 4
days.
The zebrafish embryo develops rapidly at 28.5 º C, with precursors to all major organs
appearing within 36 hours of fertilization. The embryo begins as a yolk with a big single cell on
top (in figure 1, one-cell stage
[6]
), and during this stage we injected the CRISPR/Cas9 or
morphlino. After about 30 minutes, the embryo divides into two cells, and continues dividing.
At 3 hours, there are approximately 1,000 cells, and there are 4 germ cells at this stage, which
implies that any genetic change occurring in any one of these germ cells within the first three
hours will be inherited by a large number of progeny. The cells then migrate down the sides of
the yolk to form the epiboly (at about 8 hours). At 10 hours, the cell progresses to the bud stage,
and the somites grow with an average speed of two somites per hour
[7][8]
. Around 13 hours, the
neural tube begins to form by epithelial infolding, and can be seen and imaged in real time
[9]
. At
10 somites (around 14 hours), neural crest cells emerge from the neural tube. The number,
migration and patterning of early neural crest cells as well as their maturation and
differentiation can be followed with multiple neural crest-specific enhancers tagged to
fluorescent proteins. These include sox10, which is expressed in early migrating neural crest
from 12 to 24 hours, after which it is downregulated and turned on again in all developing
University of Southern California Master ’s Thesis Introduction
3

cranial cartilage. Another is RAGR, which is a human enhancer that is expressed in early
migrating neural crest with high expression levels in the first two branchial arches. RARG is
expressed in many craniofacial tissues but is excluded from the developing cartilage. At 24
hours, the fish has 30 somites, and at 2 days it is out of the oolemma. The yolk shrinks over
time because the fish uses the nutrition as it matures during the first few days. After 5 days, the
mouth opens and the fish can eat some food, such as paramecia. At this stage, the ventral view of
the fish clearly shows the developing craniofacial cartilage and heart. Any defects in these
structures can be easily assessed. After three months, the adult fish reaches reproductive
maturity.

Figure 1: The developmental stages of zebrafish (by Kimmel et al.)
University of Southern California Master ’s Thesis Introduction
4

Reverse Genetic Techniques – The Study Approach
Today, reverse genetic techniques have become so powerful that one can make a targeted
mutation in any gene of interest. The three main reverse genetic techniques are zinc finger
nuclease (ZFN), transcription activator-like effector nuclease (TALEN) and clustered regularly
interspaced short palindromic repeats (CRISPR)/CRISPR-associated (Cas) systems, which can
induce double-stranded breaks (DSBs) within the target site. In animal cells, DSBs can be
repaired by error-free homologous recombination (HR) or error-prone non-homologous end
joining (NHEJ). The latter will introduce small insertions or deletions (indels) near the target
site, and thus will destroy the original open reading frame and the whole gene.
ZFN (Zinc-Finger Nucleases)
In 2008, a research group at the University of Massachusetts Medical School successfully
used artificial restriction enzymes known as zinc-finger nucleases (ZFNs) to mutate a specific
target site in zebrafish. Thus, ZFNs became the first reverse genetic technology to be used in
zebrafish
[10]
.
ZFNs originate from a transcription factor family which forms alpha-beta-beta secondary
University of Southern California Master ’s Thesis Introduction
5

structure, and are generated by fusing a zinc finger DNA-binding domain to a DNA-cleavage
domain. A 16-amino-acid residue determines the specificity of the particular ZFN. One ZFN
protein recognizes a specific sequence of three bases
[11][12]
. Therefore, Zinc finger domains can
be engineered to target a specific DNA sequence, and induce a DSB at this site (Figure 2). By
taking advantage of endogenous DNA repair machinery, these reagents can be used to precisely
alter the genomes of higher organisms.
However, there are many disadvantage of ZFNs. Even the affinity of each ZFN is high,
when they are combined, the final affinity may be low. There are also many off-target effects.
Gradually, ZFNs have been replaced by other reverse genetic technologies.

Figure 2: The structure and mechanism of ZFN (by Sigma-Aldrich website
[13]
)

TALEN (Transcription Activator-Like Effector Nucleases)
In 2011, Professor Bo Zhang’s lab at Peking University used TALEN technology to mutate
a target gene in zebrafish
[14]
. TALEN( TAL effector nucleases) are restriction enzymes that can
University of Southern California Master ’s Thesis Introduction
6

be engineered to cut specific sequences of DNA. Compared to ZFNs, TALENs have higher
efficiency and specificity as they can target a longer sequence. TALENs are also easier to
synthesize and less expensive.
TALENs are made by fusing a TAL effector DNA-binding domain to a DNA cleavage
domain (Figure 3). TAL effectors are proteins that are secreted by Xanthomonas bacteria via
their type III secretion system, which is typically used to infect plants
[15]
. TAL effectors
recognize specific bases, which can be used in constructing a tandem array of TALs that will
bind a desired sequence ending in a nuclease, together referred to as a TALEN. Each TAL
effector protein unit is 33 to 35 amino acids long, with only amino acids 12 and 13 differing.
These two amino acids are used for recognizing a specific DNA base. By utilizing this
mechanism, scientists can design a series of TALENs to recognize a particular target sequence
as desired. The DNA cleavage domain added to a TALEN artificially introduces a DSB, which
may cause indels in this area
[16]
.
Despite the advantages of this system, making a TALEN is still difficult, and the molecular
mass of the molecule is larger than that of a CRISPR Cas9 protein.

University of Southern California Master ’s Thesis Introduction
7

Figure 3: The structure and mechanism of a TALEN (by GeneCopoeia website
[17]
)

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)
CRISPR is one of the most important discoveries so far in 21
st
century biology. CRISPR
recognizes and cuts a target gene in a manner analogous to RNA interference in eukaryotic
organisms
[18]
. Since the recognizing part of the CRISPR/Cas9 system is an RNA fragment, it is
much easier to form than TALENs or ZFNs, which require different proteins for different target
sites.
The CRISPR/Cas system is a prokaryotic immune system that confers resistance to foreign
genetic elements such as plasmids and phages, and provides a form of acquired immunity
(Figure 4c)
[19]
. When exotic DNA invades a prokaryon, the CRISPR/Cas9 system will silence it
to protect itself. The current CRISPR classification scheme groups Cas operons into three major
divisions: I, II and III. Cas9 belongs to the well-studied type II Cas system, and is thought to be
the only operon that uses CRISPR RNA to silence the exotic DNA
[20]
. In nature, the type II
CRISPR/Cas system captures the exotic DNA at the CRISPR location, transcribes the DNA and
makes CRISPR RNA (crRNA). crRNA binds to trans-activating crRNA (tracrRNA) and cuts the
exotic DNA via the Cas protein (Figure 4a). Researchers found that the guide RNA (gRNA) that
combines crRNA and tracrRNA has the same function (Figure 4b)
[21]
. By delivering the Cas9
University of Southern California Master ’s Thesis Introduction
8

protein and appropriate guide RNAs into a cell, the organism's genome can be cut at any desired
location. Therefore, in the laboratory, the gRNAs are designed to mutate the target genes with
the CRISPR protein.
Theoretically, CRISPR can target any sequence which fits in a 5’-GG-N(18-22)-NGG-3’
structure. This type of sequence occurs randomly, approximately once in every 128bp. CRISPR
can therefore be used in many animals, including mice, zebrafish and human cells
[22]
. If the
desired target site is highly unusual and does not contain this kind of sequence, a previous study
has shown that the GG at 5’ terminal, which is restricted by the T7 promoter used to transcribe
gRNA, can still work when one G is replaced with A. The resulting
5’-(G/A)(G/A)-N(18-22)-NGG-3’ sequence is more frequently found, about one in 32 base
pairs. Therefore, sites that can be targeted via CRISPR/Cas9’s are easy to find.

University of Southern California Master ’s Thesis Introduction
9

C

Figure 4: The mechanism of CRISPR Cas9 knockout. a. The structure of natural CRISPR
Cas9 protein; b. The structure of artificial CRISPR Cas9 protein; c, CRISPR/Cas9 system

MO (Morpholino)
“Morpholino” (MO) is shorthand for phosphorodiamidate morpholino oligomer (PMO), a
type of oligomer molecule used in molecular biology to knock down gene expression. Gene
knockdown is achieved by preventing cells from making a targeted protein, and it is a powerful
method for learning about the function of a particular protein. These MOs can be used for
studies of several model organisms, including mice, zebrafish, frogs, and sea urchins
[23]
.
The molecular structure of MOs contains a backbone of methylenemorpholine rings and
phosphorodiamidate linkages. Different MO molecules recognize and bind to different RNA
University of Southern California Master ’s Thesis Introduction
10

bases (Figure 5a). Therefore, researchers designing MO sequences to block the access of other
molecules target small specific sequences of the base-pairing surfaces of RNA
[24]
.
MO knocks down a gene in two ways: blocking translation and modifying pre-mRNA
splicing. When blocking translation, the MO is bound to the 5'-untranslated region of mRNA,
and prevents the translation initiation complex from binding or moving onto the mRNA (Figure
5b)
[25]
. For interfering with pre-mRNA, MOs are designed to prevent splice-directing small
nuclear ribonucleoproteins (snRNP) complexes from binding to the borders of introns on
pre-mRNA (Figure 5d-e), or to interfere with the binding of splice regulatory proteins
[26]
.

University of Southern California Master ’s Thesis Introduction
11

Figure 5: The molecular structure and knockdown mechanism of MOs. a. Molecular
structure of morpholino. b. Using a morpholino to block translations. c. Normal mRNA
splicing. d, e. Two different strategies used to modify pre-mRNA splicing.

Neural Crest Cells
In vertebrates, there is a transient cell population known as the neural crest that exists for a
short period of time. Neural crest cells then differentiate and migrate to different parts of the
body, in turn giving rise to diverse cell lineages—including melanocytes, craniofacial cartilage
and bone, smooth muscle, peripheral and enteric neurons and glia
[27]
.
The formation of neural crest cells happens after gastrulation (Figure 6). The neural crest
cells exist at the border of the neural plate and the non-neural ectoderm. During neurulation, the
borders of the neural plate roll over at the dorsal midline to form the neural tube. Subsequently,
neural crest cells between the dorsal neural tube and overlying ectoderm delaminate from the
neuroepithelium and migrate to the periphery, where they differentiate into varied cell types
[28]
.
In zebrafish, the neural crest cells form the facial tissue, the facial cartilage structure, some
heart tissue, etc.
University of Southern California Master ’s Thesis Introduction
12

Figure 6: The formation of neural crest during the process of neurulation. First, the neural
plate folds to form the neural groove. Then, after the neural tube is closed, neural crest
delaminates from the region between the dorsal neural tube and overlying ectoderm and
migrates out.

In zebrafish, the neural crest cells (NCCs) migrate out and form four branchial arches
(Figure 7)
[29][30]
. When the fish is mature, there are seven pharyngeal arches. The arches also
correspond to the cartilage structure, which means that different portions of the NCCs form
University of Southern California Master ’s Thesis Introduction
13

different tissues.

a
b
Figure 7: Schematic drawing of zebrafish neural crest development. a, Schematic drawing
of the pharyngeal region of a 24-hpf zebrafish embryo. The pharyngeal arches are comprised
of mesoderm (green), neural crest cells (gray), endodermal pouches (red), medial pharyngeal
endoderm (orange), and lateral ectoderm (blue)
[29]
. b. Schematic representation of a grown
zebrafish possessing 6 endodermal pouches and 7 pharyngeal arches. The color-coded streams
correspond to colored pharyngeal arch cartilages depicted in the 5-days-postfertilization (dpf)
schematic. bb, basibranchial; bh, basihyal; cb, ceratobranchial; ch, ceratohyal; hs,
hyosymplectic; M, Meckel's cartilage; p, pharyngeal arch; pq, palatoquadrate; r,
rhombomere
[30]
.
University of Southern California Master ’s Thesis Introduction
14

Figure 8: Putative neural crest gene-regulatory network functioning at the neural plate
border in vertebrates. Red arrows represent proven direct regulatory interactions. Black
arrows show genetic interactions based on loss-of-function and gain-of-function studies. Gray
lines denote repression.

Underlying the development of neural crest is a gene regulatory network (Figure 8)
[31]
.
Previous studies have shown that Rho GTPases and cadherins function in delamination by
regulating cell morphology and adhesive properties; Sox9 and Sox10 regulate neural crest
differentiation by activating many cell-type-specific effectors downstream
[32]
. We can use
sox10 as a post-migratory neural crest marker, Foxd3 as a pre-migratory neural crest marker and
University of Southern California Master ’s Thesis Introduction
15

wnt1 as a marker that can indicate both the neural crest and the neural tube.
Our previous studies in other animal models

Figure 9: The opposing phenotypes of CHARGE syndrome and BFLS patients.

Our lab has identified a physical interaction between two human proteins: CHD7, a
chromatin remodeling protein, and PHF6, a dual PHD finger protein of unknown function.
Mutations in these genes result in distinct syndromes in humans with contrasting craniofacial
University of Southern California Master ’s Thesis Introduction
16

defects. Patients with CHARGE syndrome, who have CHD7 mutations, have a dysmorphic face
with reduced amounts of neural crest-derived structural and sensory tissues. On the other hand,
patients with Borjeson Forssman Lehmann syndrome (BFLS), caused by plant homeodomain
Finger Protein 6 (PHF6) mutation, have a thick calvarium and broad jaw (Figure 9). These
phenotypes are observed in neural crest-derived cells. Therefore, we hypothesized that these
two genes may have some function during the formation of neural crest cells.
We hypothesized that PHF6 is a negative regulator of neural crest cell development and
differentiation. While CHD7 is known to promote these processes, PHF6 may be a negative
regulator of CHD7 function. To test this hypothesis, we decided to generate chd7 and phf6
mutant. Thus, we tried to complement in vitro human embryonic stem cell and Xenopus animal
model studies with genetic studies in zebrafish to assess the evolutionary conservation of this
process.
In a previous in vitro study of human embryonic stem cells (ES cells), normal ES were
found to generate migrating neural crest cells when differentiated into neuroectodermal cells.
Strong PHF6 knockdown cells (~ 90% KD) are selected against and weak knockdown cells
(30% KD) generated neural crest cells sooner than wild type cells (day 5 compared to day 7-9 in
WT). In contrast, partial CHD7 knockdown (~50%KD) ES cells recapitulate the
haploinsufficiency in CHARGE Syndrome and generate fewer neural crest cells compared to
University of Southern California Master ’s Thesis Introduction
17

the control group. Further, strong knockdown of CHD7 (75%KD) results in complete absence of
neural crest cell formation. Interestingly, these defects resulting from CHD7 knockdown can be
rescued by a partial decrease in PHF6 levels. This means, in vitro, the functional relationship
between CHD7 and PHF6 is antagonistic. Haploinsufficiency of one of these proteins disrupts
the balance of positive and negative regulation of NCC formation, which can be restored by
down-regulating the other protein.
A study of Xenopus in vivo, mopholino-mediated knockdown of chd7 and phf6 confirmed
the findings from human stem cell studies. The results indicated that chd7 knockdown animals
have fewer neural crest cells than controls, whereas phf6 knockdown animals have earlier
neural crest cell migration and a larger number of neural crest cells (EM and RB unpublished
results). Further, the chd7 morphant phenotype can be rescued by a dose-dependent
overexpression of a dominant negative phf6 mRNA in Xenopus embryos, supporting an
opposing functional relationship between chd7 and phf6 (RB unpublished results).
I performed a MO knockdown of chd7 in zebrafish, to test that whether zebrafish is a
good model for studying CHARHE syndrome. While a chd7 knockdown phenotype using a
morpholino has been described in zebrafish, these experiments did not show conclusive rescue
of rescue of several CHARGE-like, morphant phenotypes with wild type chd7 mRNA and in a
second study rescue experiments were not attempted. I tested the specificity of
University of Southern California Master ’s Thesis Introduction
18

morpholino-mediated knockdown of the chd7 gene in zebrafish. The morpholino is designed to
block the splicing site after the eighth exon. The knockdown fish were identified by reverse
transcriptase PCR (RT-PCR).
The hypothesis that, chd7 and phf6 have an opposing function has not been tested in a
genetic model system. Therefore, my contribution to this project was to establish a genetic
model for each of these two syndromes by knocking out these two genes in zebrafish. In this
project, I used CRISPR/Cas9 to mutate the chd7 and phf6 genes in zebrafish embryos. We
co-injected Cas9 mRNA with gRNAs, and we also tried to inject Cas9 protein with gRNAs and
observe the resulting phenotype in the injected embryos. My goal was to induce deletion of a
functional domain in the phf6 gene and make a point mutation in a catalytic domain in the chd7
gene. Several fluorescent reporters were used for this analysis. Specifically, we used the sox10
promoter-driven dsred transgenic line (sox10::dsred), foxd3-eGFP transgenic line and two
enhancer-reported transgenic lines (hwnt1enh::GFP and hRARGenh::eGFP) newly generated
in our lab. These reporters were used to observe the expression of neural crest cells and
craniofacial tissue.
Here, I characterize the process by which we have successfully made targeted mutations
and knockdown fish for phf6 and chd7. I have partially characterized the mutant fish phenotypes
and detailed genotyping analysis is ongoing. Chd7 mutant fish have phenotypes including
University of Southern California Master ’s Thesis Introduction
19

developmental delay, decreased facial tissue, small head, small eyes and a defect of the
craniofacial cartilage tissue, whereas phf6 mutants have significantly a smaller brain with a
mostly normal face. Therefore, chd7 and phf6 have opposing functions in craniofacial and
neuro-development in zebrafish. These phenotypes are reminiscent of features seen in patients
with mutations in CHD7 and PHF6, respectively.
University of Southern California Master ’s Thesis Introduction
20

5’ - GNNNNNNNNNNNNNNNNNN(N) - NGG - 3 ’
Target sequence
PAM area
Materials and Methods
Creation of Cas9 gRNA DNA template:
First of all, a unique, approximately 19-base-long, Cas9/gRNA target must be identified on
the zebrafish genome that ends in NGG.

Plus-strand Cas9/gRNA target:
Since only one DNA stand is needed for recognition by CRISPR/Cas9 the negative strand
can also be targeted.
Alternative Cas9/gRNA target:
To improve the efficiency and specificity of targeting, the following criteria should be met:
i. A,T,C and G bases are well-distributed;
ii. stretches of more than four of the same bases are avoided;
iii. high specificity near the PAM area should be ensured to reduce the possibility of off
target binding;
iv. Cas9 target sites near the 5’ prime terminal should be avoided for knockout
purposes (since the gene may have an alternate transcription start site that may
result in a functional protein). Also avoid designing targets in the last exon (which
5’ – CCN - NNNNNNNNNNNNNNNNNN(N)C - 3’
Target sequence PAM area
University of Southern California Master ’s Thesis Introduction
21

may consist of 3 rime UTR); ideally targets should aim to destroy important
domains.
The sequence of a gRNA DNA template should be: [T7 promoter]-[Target Sequence]-[start
of gRNA sequence], which is: 5’- aattaatacgactcactata-[20 bp Target Sequence]-gttttagagctaga
aatagc-3’. My goal was to generate zebrafish chd7 and phf6 mutants to model CHARGE and
BFLS, respectively. Both chd7 and phf6 are single-copy genes in zebrafish. I tried to design two
gRNAs to mutate the chd7 gene, and four gRNAs to make a deletion in the second PHD domain
of the phf6 gene.
I designed two CRISPR cutting sites on the chd7 DEXDc domain, which is essential for
ATP-dependent helicase activity, to induce indels (small fragment insertion or deletion) at this
area and therefore destroy functional chd7 protein. Phf6, on the other hand, is a dual PHD
finger protein without any known enzymatic activity. There are two naturally occurring
alternately spliced isoforms of phf6, and the shorter one (phf6b) lacks the second domain
(PHD-F2). Recent studies in our lab have shown that the shorter phf6b functions as a
dominant negative. Thus, I designed four gRNAs to try to delete PHD-F2. The gRNAs I
designed are shown in Figure 10.
University of Southern California Master ’s Thesis Introduction
22

Figure 10: The designed DNA templates of chd7 and phf6 gRNA; red arrows point to the
gRNA target sites.

For convenient detection of the mutant alleles, I also identified a restriction enzyme cutting
site on the chd7 gRNA targeting site. If the gene is mutated in this area, resulting in the loss of
the enzyme, the gene will not be cut and there will be a bigger band in the electrophoresis gel,
which makes the genotyping easier. The restriction enzymes I used are Hpy188 III(TCAGGA)
and Mwo I (CTGGCC), both purchased from NEB (Table 1).
University of Southern California Master ’s Thesis Introduction
23

The CRISPR/Cas9 vector is used to obtain the DNA template of Cas9 mRNA. Then, in
vitro transcription is performed to generate Cas9 mRNA and gRNAs.

Table 1: The method I used to test the chd7 mutations and phf6 deletions.

chd7 morpholino
The chd7 Morpholino (MO) was designed by Chetana Sachidanandan’s lab (Figure 11), and
purchased from Gene Tools, LLC. The MO sequence that we used for chd7 knockdown was
University of Southern California Master ’s Thesis Introduction
24

designed across exon 8 and intron 8-9 (indicated in the following figure)
MO: 5'-ATGGAGGGTCAATTCTAACCTCAGT 3'.

Figure 11: The chd7 MO design. a, The location of chd7 MO on chd7 protein. b, The MO’s
location and expected PCR products from wild-type and morphant transcripts.
Microinjection
Embryos were injected at the one-cell stage (Figure 12). In this experiment, I injected 2 nl
of injection mixture per embryo. The mixture contained phenol red as a dye to indicate
successful injection into the cell. The concentration of Cas9 mRNA was about 150 ng/ul. The
gRNA concentration does not seem to be highly important, since reported concentrations range
greatly, from 1 ng/ul to almost 100 ng/ul. Higher concentrations of gRNA do not seem to cause
any additional defects or mortality. In my experiment, I injected 100 ng/ul. In some experiments,
I injected Cas protein (by PNA bio) rather than Cas9 mRNA. The Cas9 protein’s concentration
University of Southern California Master ’s Thesis Introduction
25

was about 150ng/ul. For the MO experiment, the final concentration used was 0.15 mM. In the
chd7 MO rescue test, 1.5ng mRNA were injected together with 1.2 ng chd7 MO.

Figure 12: Using the microinjection needle to inject the reagent into embryo at the
one-cell stage
Imaging
Images of the embryos were taken with a confocal microscope (Broadband Confocal Leica
TCS SP5 II). All photos were analyzed and assembled on Leica LAS AF Software.
Genotyping
Tail clippings or collected embryos are placed into 50mM NaOH to destroy the tissue, at
95 ºC for 20 minutes. When the tissues were lysis, 3 ul Tris-HCl (pH 8.0) was added to
neutralize the pH. PCR was then conducted to identify mutants. GoTaq® DNA Polymerase (by
Promega) was used to duplicate DNA.
cell
University of Southern California Master ’s Thesis Introduction
26

Kits
1. MinElute®Gel Extraction Kit (by QIAGEN) was used for gel extracting.
2. MAXIscript® T7 Kit and MEGAscript® T7 Kit (by Invitrogen) was used for in vitro
transcription of Cas9 mRNA, gRNAs, and chd7 mRNA (used to rescue the chd7 MO
phenotype).
3. MicroPoly(A) Purist
TM
Kit (by Thermo Fisher Scientific) was used for purifying
mRNA.
4. SuperScript® III First-Strand Synthesis System for RT-PCR (by Invitrogen) was used
to synthesize cDNA from purified total mRNA.
List of Zebrafish lines used in this study:
1. Tubingen wild type;
2. AB outcross wild type
3. Sox10::dsRed
el10

4. Sox10::GFP
5. Foxd3:: GFP
6. hRARG::GFP (RB, unpublished)
7. hWNT1::GFP (RB, unpublished)
University of Southern California Master ’s Thesis Results
27

Results
Zebrafish is a good model for CHARGE Syndrome
chd7 MO knockdown and rescue
Chd7 is highly expressed in the developing neural tube and migrating neural crest. The
cartilage structure was labeled by sox10-dsRed in Figure 13. I used chd7 MO to knockdown
chd7 function in zebrafish. As the amount of injected chd7 MO is increased, the cartilage
defect observed in the injected fish becomes more severe (Figure 13), which means the chd7
MO has a dose dependent effect.

Figure 13: Phenotypic spectrum of defects in zebrafish models of CHARGE syndrome
(knockdown of chd7, 8dpf). The sox10 is labeled by dsRed fluorescence. In 8 days, sox10
marked the cartilage structure. In this experiment, chd7 were knocked down by chd7
morpholino. The results show that chd7 MO has a dose dependent effect.
University of Southern California Master ’s Thesis Results
28

The cartilage structure of the mature zebrafish is shown in Figure 14
[33]
. Compared to the
chd7 MO fish shown in Figure 13, the knockdown fish lost neural-crest derived cartilage first.
Many of them still have the mesoderm-derived cartilage, and pectoral fin cartilage is never
affected. Therefore, chd7 has some function on neural crest cell formation, and thus interferes
with the cartilage taking shape.

Figure 14: The cartilage structure of mature zebrafish. Diagrams show the cranial neural
crest-derived cartilage (blue), bone and teeth (red), mesoderm-derived cartilage (green),
pectoral fin cartilage (black), and eyes (yellow).

I also performed a rescue experiment using MO resistant normal human CHD7 mRNA
and human CHD7 mRNA which was mutated at the ATPase binding site of the CHD7 protein.
I injected the mRNA into sox10-GFP fish and recorded the developmental delay of the F
0
fish
(founder zero the injected fish, Figure 15, 16). Human CHD7 mRNA can partially rescue the
phenotype caused by chd7 MO, while human CHD7 mutated mRNA did not show such a
rescue. In chd7 MO plus human CHD7 ATPase mutant mRNA group, the CHD7 ATPase
University of Southern California Master ’s Thesis Results
29

positive cells are selective-against, which implicated that human CHD7 mutant mRNA makes
the cell even worse. In conclusion, ATPase biding site is important for CHD7 function.

Figure 15: Human CHD7 mRNA rescue experiment. A. The phenotype of represented fish
from each of the following: wild type fish, chd7 MO fish, chd7 MO plus dsRed mRNA fish,
chd7 MO plus human CHD7 mRNA fish and chd7 MO plus human CHD7 mutant fish. b,
Examples of fish with developmental delay. Many CHARGE syndrome defects like
dysmorphic faces, reduced cartilage, eye defects and heart defects were also noted in these
morphant fish.

University of Southern California Master ’s Thesis Results
30

Figure 16: chd7 MO rescue experiment. a, Developmental delay exhibited at 24hpf by chd7
MO fish. Human CHD7 mRNA can partially rescue the chd7 MO knockdown function. At 24
hpf, wild type fish have 30 somites. Fish with 19-29 somites were considered to have a mildly
abnormal phenotype. Fish with 10-18 somites were considered to have a severely abnormal
phenotype. Fish with less than 9 somites were considered to have extreme developmental delay.
*: p value less than 0.05. One spot represents 5 fish. b. The RT-PCR results confirm successful
injection of chd7 MO and human CHD7 mRNA into the embryos. Note the morphant PCR
product identified in chd7 MO injected but not in control fish (lane1) and the human CHD7
specific product in both wild type and ATPase mutant CHD7 co-injected fish (lanes
4-6). Therefore, the rescue is not due to lack of chd7 MO. We compared the
human CHD7 mRNA bands in different times, the mutated CHD7 mRNA are selected-against,
and lost completely at 24 hours (lane 6).

Importantly, the rescue was confirmed to be a result of overexpression of
morpholino-resistant human mRNA for wild type CHD7 as the rescued fish still expressed the
morphant product. It is possible that the rescue is partial due to an insufficient amount of mRNA
or the inability of human CHD7 to perfectly compensate for all zebrafish chd7 functions.
The opposing functions of chd7 and phf6 in neural crest cell (NCC)
migration.
In a previous study, our lab showed that, in human embryonic stem (ES) cells, CHD7
promotes NCC formation. On the other hand, PHF6 inhibits this process. In this project, I
designed chd7 and phf6 gRNA, and then injected them together with Cas9 mRNA into
foxd3-GFP fish embryos. Although foxd3 is expressed in pre-migratory NCCs
[34]
, in this
University of Southern California Master ’s Thesis Results
31

fluorescent reporter line, the perdurance of GFP allows detection of early migrating NCCs as
well. Confocal imaging of injected embryos showed that in phf6-injected fish, foxd3
expression was not reduced and migrating NCCs could be visualized. However, in some
chd7-injected fish, there were significantly fewer migrating and pre-migrating NCCs, both in
the pharyngeal arches (white arrow) and the supraorbital tracks (Figure 17). This suggests that
in zebrafish, chd7 and phf6 have opposing functions (Table 2). For this reason, I attempted to
make chd7 and phf6 mutant zebrafish.

Figure 17: Comparison of the neural crest migration in phf6 and chd7 mutant fish at
24hpf. Foxd3 is labeled by GFP fluorescence. A. Phf6 gRNA was injected together with Cas9
mRNA into the embryos at the one-cell stage. The arrow indicates the migration of neural
crest cells, and phf6 injected fish shows no difference with the control one. It can be seen that
many neural crest cells are migrating. B. Chd7 gRNA was injected together with Cas9 mRNA
into the embryos at the one-cell stage. The arrow indicates the migration of neural crest cells.
In chd7 mutant fish, the neural crest cells stop migrating.
University of Southern California Master ’s Thesis Results
32

Serial
number
Parental
fish
Injected
gRNA
Embryos with
predicted
CHARGE-like
defects
Embryos with
predicted
BFLS-like
defects
Comments
1 WT phf6
gRNA1+3
– ++

2 WT phf6 gRNA
1+4
– +

3 WT phf6 gRNA
2+3
– –
gRNA2 may
not work
4 WT phf6 gRNA
2+4
– –
gRNA2 may
not work
5 WT phf6 gRNA
1,2,3,4
– ++

6 WT chd7 gRNA
1,2
+ –

Table 2: Phenotypes of the chd7- and phf6-injected fish are different. The presence of
CHARGE-like defects means that the cessation of NCC migration was observed under
fluorescence microscopy. The presence of a BFLS like-defect was indicated by a small or
absent head. “–” means the defect was absent, “+” means it was present but not severe, “++”
means present and severe.

Comparing the F2 phenotypes of chd7 and phf6 fish (Figure 18), chd7 fish have reduced
eye and jaw structures as well as a larger amount of opaque brain tissue. On the other hand,
the eyes and jaws of the phf6 mutants appear normal, but they present with a transparent head,
which indicates they may have less brain tissue.
University of Southern California Master ’s Thesis Results
33

Figure 18: The contrasting phenotypes of chd7 and phf6 fish.

phf6 fish have a small brain (microcephaly)
I injected phf6 gRNA 1, 2, 3, 4 together to generate fish with a deletion of phf6. In the
founder generation (F
0
), I observed some fish with a small head (Figure 19), though this
phenotype never occurred in other chd7, foxd5, chd6 or chd8 gRNA-injected fish. This
suggests that phf6 deletion is the cause of this phenotype. Affected fish can live for up to six
University of Southern California Master ’s Thesis Results
34

days. The question we want to discuss is this microcephaly is caused by loss of brain or loss of
facial tissue, which are formed by NCCs.

Figure 19: The phf6 affected fish 2d. phf6 gRNA was injected into sox10-dsRed transgenic
fish. The red arrows indicate that in phf6 fish, the head is much smaller than in wild type fish.

Figure 20: Comparison of the migrated neural crest cells in phf6 mutant fish vs. wild
type fish at 36hpf. Sox10 is labeled with dsRed. A. After injection of phf6 gRNA together
with Cas9 protein, some phenotypes were observed in the injected fish. In comparison with the
University of Southern California Master ’s Thesis Results
35

wild type fish (B), we saw that, in spite of the overall smaller size of phf6 embryos, the
embryos were not significantly developmentally delayed. Sox10-dsred expression, which
begins at about the 8-somite stage, was robustly detected in mutant embryos, indicating an
abundant number of migrating neural crest cells that go on to form facial tissue. However, the
phf6 fish have smaller heads than the wild types, and the brain tissue is reduced or even absent
in these mutant fish.

In order to investigate this further, I injected phf6 gRNA with Cas9 mRNA into
sox10-dsRed fish. Sox10 is expressed in post-migratory NCCs
[35]
. Figure 20 demonstrates that
microcephalic phf6 fish also express sox10, which indicates that the microcephaly is not due
to developmental delay, and that the face of the fish is similar to that of the wild type fish. In
summary, the microcephalic phf6 fish have a much smaller or even absent brain.
I also injected phf6 gRNA into wnt1-GFP fish, in which both the neural tube and NCCs
are fluorescently labeled
[36]
. The brain structure of zebrafish is shown in Figure 21 (courtesy
of the Hazel Sive lab
[37]
). In the wnt1-GFP fish, I found that the brain is defective in
phf6-injected zebrafish (Figure 22). This fish have a normal appearance; however, both the
forebrain and midbrain are affected. The neural tube is not open like in a wild type fish. Thus,
I concluded that phf6 inhibits NCC formation. In phf6 mutant fish, the neural tube cells
differentiate to form NCCs and migrate to form other types of cells, such as facial tissue.

University of Southern California Master ’s Thesis Results
36

a

b

Figure 21: The brain structure of wild type zebrafish at 24 hpf. a. Wild type zebrafish
embryo stained with laminin antibody, which outlines the neural tube in green, and
counterstained with propidium iodide to label the nuclei in red (figure courtesy of the Hazel
Sive Lab). b. The expression of hWnt1::GFP in the early stages of neural crest migration
(18hpf). Note that both the dorsal neural tube (NT) and migrating neural crest in pharyngeal
arches I-III and the supraorbital streams are detected by GFP. E-developing eye.
University of Southern California Master ’s Thesis Results
37

Figure 22: The neural tube structure in phf6-injected fish and wild type fish at 24 hpf.
Dorsal neural tube cells and NCC cells are labeled by wnt1-GFP. The red dotted line shows the
lumen of the neural tube. The neural tubes of the wild type and chd7 fish are opened and form
folds. However, the neural tube of the phf6 mutant fish is closed and straight. More strikingly,
the migration of neural crest cells is not adversely affected.

chd7 fish have an asymmetric face
When chd7 gRNA1 and 2 were injected together with Cas9 mRNA, many fish exhibited
phenotypes such as heart edema, ear defects, microcephaly, kinked tail and asymmetric face
University of Southern California Master ’s Thesis Results
38

(Figure 23). The forehead and midbrain thicknesses were measured in wild type and
chd7-injected fish. The results show that chd7 mutant fish have a significantly smaller
forehead, while the thicker midbrain seen in these fish did not reach statistical significance.

Figure 23: Affected chd7 fish. The yellow star indicates the ear defect; the blue arrow shows
the heart edema; the red dotted line shows the NCCs derived craniofacial tissue. The forehead
length and the midbrain thickness of chd7 mutant fish and wild type fish are compared in the
graph on right. The error bar indicates the SEM. *: p value less than 0.05.

The F
0
fish were raised and incrossed to obtain F
1
fish. In the F
1
fish, some embryos had a
developmental delay. I also injected chd7 gRNA together with Cas9 protein into the
sox10-daRed, RARG-GFP fish, attempting to make the phenotype apparent at an earlier
developmental stage. A severe developmental delay was indeed observed in the injected fish
(Figure 24), which appeared similar to some of the F
1
fish. Developmental delay is also
characteristic of CHARGE syndrome patients; therefore, we believe that this is one of the
dominant phenotypes of chd7 mutants. I separated the F
1
fish into two groups according to
University of Southern California Master ’s Thesis Results
39

whether they had a development delay, and then raised the groups separately.

Figure 24: The developmental delay of chd7-injected fish. At 30 hpf, the wild type fish is
much bigger than the chd7 delayed fish. The developmentally delayed fish appears similar to a
wild type embryo at 19 to 22 hpf.

After the Cas9 mRNA F
1
fish matured, I incrossed them to obtain F
2
fish. In this
generation, I found some fish had a severe phenotype – some had only one eye or extremely
small eyes (Figure 25). Identically, in Cas9 Protein F
1
fish, there were some fish shared a
similar eyeless phenotype. In these F
1
fish, some NCC-derived cells are labeled by
RARG-GFP
[38]
, besides the cartilage structure which is tagged by sox10-dsRed. The results of
this experiment showed that the cartilage structure was defective in these fish, and the chd7
mutant fish only had fewer eyes. I arranged the fish in phenotype spectrum, from mild to
severe (Figure 26). The smaller eyes the fish has, the worse cartilage structure it has, and the
less NCC-derived cells it has. Therefore, chd7 mutant fish have fewer cells that come from
University of Southern California Master ’s Thesis Results
40

NCCs, which means they have less NCCs. In contrast, phf6 mutation increases the number of
NCCs. We also found that the chd7 fish shared some similar phenotype criteria; however, no
one was same with others. This is also uniform with human CHARGE patients.

Figure 25: One-eye fish compared with wild type fish. RARG-GFP marks some cells that
derive from NCCs. Sox10-dsRed labels cartilage structure, which also derives from NCCs. The
chd7 F
2
fish only had one eye. The wild type control is shown on the right.

University of Southern California Master ’s Thesis Results
41

Figure 26: chd7 F
2
phenotypic spectrum of defects in zebrafish models. RARG-GFP marks
some cells that derive from NCCs. Sox10-dsRed labels cartilage structure, which also derives
from NCCs. As the phenotype become severe, the fish has less RARG-GFP labeled cells.

University of Southern California Master ’s Thesis Discussion
42

Discussion

Figure 27: The zebrafish models for CHARGE syndrome and BFLS syndrome. The red
and green texts indicate the phenotypes of patients with CHARGE syndrome and BFLS,
respectively. Arrows indicate the corresponding phenotypes in the zebrafish models.

Taken together, my results suggest that chd7 and phf6 mutant fish can be used as animal
models for CHARGE syndrome and BFLS, respectively (Figure 27). The two mutants both
have some similar phenotypes as the human patients. Therefore, we can use these two fish to
University of Southern California Master ’s Thesis Discussion
43

study these diseases in more detail, or do some drug screening experiments. Moreover, we can
also try to use them to control each other since chd7 and phf6 have opposing functions.
The chd7 fish family tree is shown below (Figure 28). As shown below, injected Cas9
protein can generate fish with the desired phenotype much faster than injected Cas9 mRNA.
We also found that, incrossed chd7+Cas9 protein fish have severe phenotypes such as the
development of only one eye and these fish died around 6d. However, when we outcross them
with wild type fish, no progeny have severe phenotypes. The electrophoresis results show that
the fish which have severe phenotypes are also heterozygous. Therefore, we hypothesize that
there may be some maternal or paternal influences on the chd7 protein.

Figure 28: chd7 fish family tree. Blue indicates WT cells, and red indicates chd7 mutant cells.
Mild phenotypes include: NCCs stopping migration or migrating slower than the NCCs in
normal fish; mild dysmorphic face; heart edema; kinked tail. Severe phenotypes primarily
include severe face dysmorphias such as loss of one eye or extremely small eyes. If the fish
had fewer than 18 somites at 24hpf, it was considered to have severe developmental delay. In
the F
1
generation of chd7 and Cas9 protein coinjected fish, 15% of fish from one tank had a
severe phenotype, and 2% of fish in the other tank had a severe phenotype. In the F
2

University of Southern California Master ’s Thesis Discussion
44

generation of chd7 and Cas9 mRNA coinjected fish, 4% of fish had a severe phenotype.

The phf6 mutant family tree is shown in Figure 29. In the F
0
generation, some fish had
severe microcephaly or acephaly, and they did not survive beyond 6dpf. However, in the
phf6-injected fish that did survive, the phf6 deletion band was hardly detectable. One possible
reason is that phf6 deletion has a negative dominant function, and the cells which have phf6
deletion are selective-against. Because some F
1
of phf6 and Cas9 protein coinjected fish do not
survive, it is possible that the germline cells of some F
0
fish contain the phf6 deletion gene.
Since the germ line cells stayed silence for a long time and phf6 might not necessary for them.
So, these cells did not selectively kill and passed on to the next generation. However, for F
1

fish, the phf6 is essential to them. Therefore, the F
1
fish that contains phf6 mutated gene
cannot live through the gastrulation stage.
Because phf6 deletion cells seems like cannot survive to adulthood, it may be beneficial
to generate a phf6 deletion in chd7 mutant fish. Such cell line was successfully generated in
human IPS cells.
University of Southern California Master ’s Thesis Discussion
45

Figure 29: phf6 fish family tree. Blue indicates WT cells, and yellow indicates phf6 mutant
cells. Mild phenotypes include: heart edema and smaller head. Severe phenotypes primarily
included severely microcephalic or acephalic fish. In the F
0
generation, some fish had a very
severe phenotype. In the F
1
generation of phf6 and Cas9 protein coinjected fish, 15% of fish in
one tank had a mild phenotype, and all fish died in the other one. In the F
2
generation of phf6
and Cas9 mRNA coinjected fish, almost all fish died.
University of Southern California Master ’s Thesis Further Study
46

Further Study
In the future, there are some additional experiments that need to be done.
First, I will genotype the chd7 and phf6 mutant fish with abnormal phenotypes, to make
sure these phenotypes are caused by these gene mutations. Our preliminary analysis suggests
that chd7 mutants have a big brain with a small face, and phf6 mutants have a small brain with
expanded facial tissue. To confirm these findings, in situ hybridization will be performed using
laminin and other genes that are expressed in developing neural crest or brain as probes.
A rescue will also be attempted by injecting human mRNAs to make sure the observed
phenotypes are caused by the mutations under investigation rather than other off-target effects.
I would also like to inject phf6 gRNA plus Cas9 in fish with a chd7 mutant background, to see
that if the phf6 deletion fish are able to survive.
Finally, I would like to cross confirmed chd7 and phf6 mutants to test whether phf6
mutation can truly rescue CHARGE-like defects in developing embryos.
University of Southern California Master ’s Thesis Works Cited
47

Works Cited
[1] Froese R and Pauly D eds. (2007). "Danio rerio" in FishBase. March 2007 version.
[2] In Memory of George Streisinger, "Founding Father" of Zebrafish Developmental and
Genetic Research. University of Oregon. Retrieved September 23, 2015.
[3] Spence R, Gerlach G, Lawrence C and Smith C. (2007). The behaviour and ecology of
the zebrafish, Danio rerio. Biological Reviews. 83 (1): 13–34.
[4] Goldshmit Y , Sztal TE, Currie PD, et al. (2012). Fgf-Dependent Glial Cell Bridges
Facilitate Spinal Cord Regeneration in Zebrafish. The Journal of Neuroscience. 32 (22):
7477–92.
[5] Jopling C, Sleep E, Izpisú a Belmonte JC, et al. Zebrafish heart regeneration occurs by
cardiomyocyte dedifferentiation and proliferation. Nature. 464, 606–609 (25 March
2010).
[6] Kimmel et al. 1995. Developmental Dynamics. 203:253-310.
[7] Kimmel CB, Ballard WW, Kimmel SR, et al. Stages of embryonic development of the
zebrafish. Developmental Dynamics. 1995, 203:253-310
[8] Dahm R, Geisler R, Nü sslein-V olhard C, et al. Zebrafish (Danio rerio) genome and
genetics. Encyclopedia of molecular cell biology and molecular medicine. 2005, 15:
593-626.
[9] Nusslein-V olhard C and Dahm R edit, zebrafish [M], page 226.
[10] Meng X, Noyes MB, Wolfe SA, el al. Targeted gene inactivation in zebrafish using
engineered zinc-finger nucleases. Nat Biotechnol. 2008 Jun; 26(6):695-701.
[11] Cathomen T and Young JK. Zinc-finger nucleases: the next generation emerges. Mol
Therapy. 2008, 16:1200–1207.
University of Southern California Master ’s Thesis Works Cited
48

[12] Porteus MH and Carroll D. Gene targeting using zinc finger nucleases. Nat Biotechnol.
2005, 23: 967–73.
[13] Huang P, Xiao A, Zhou M, et al, Heritable gene targeting in zebrafish using customized
TALENs. Nat Biotechnol. 2011, 29(8): 699-700.
[14] Sigma-Aldrich website :
http://www.sigmaaldrich.com/life-science/zinc-finger-nuclease-technology/learning-cen
ter/what-is-zfn.html
[15] Boch J and Bonas U. (September 2010). XanthomonasAvrBs3 Family-Type III Effectors:
Discovery and Function. Annual Review of Phytopathology. 48: 419–36.
[16] Zu Y , Tong X, Wang Z, et al. TALEN-mediated precise genome modification by
homologous recombination in zebrafish. Nat Methods. 2013, 10: 329-331.
[17] Marraffini LA and Sontheimer EJ. (March 2010). "CRISPR interference: RNA-directed
adaptive immunity in bacteria and archaea". Nature Reviews Genetics. 11 (3): 181–190.
[18] GeneCopoeia website: http://www.genecopoeia.com/product/talen-tal-effector/
[19] Barrangou R, Fremaux C, Horvath P, et al (March 2007). "CRISPR provides acquired
resistance against viruses in prokaryotes". Science. 315 (5819): 1709–1712.
[20] Martin J, Krzysztof C, Ines F, et al. Programmable Dual-RNA–Guided DNA
Endonuclease in Adaptive Bacterial Immunity. Science. 2012, 337:816-821.
[21] Woong YH, Yanfang F, Deepak R, et al. Efficient genome editing in zebrafish using a
CRISPR-Cas system. Nature Biotechnology. 2013, 31:227-229.
[22] Nannan C, Changhong S, Lu G, et al. Genome editing with RNA-guided Cas9 nuclease in
Zebrafish embryos. Cell Research. 2013, 23:465-472.
[23] Heasman J. (2002). Morpholino oligos: making sense of antisense. Dev. Biol. 243 (2):
209–14.
University of Southern California Master ’s Thesis Works Cited
49

[24] Summerton J. (1999). Morpholino Antisense Oligomers: The Case for an RNase-H
Independent Structural Type. Biochimica et Biophysica Acta. 1489 (1): 141–58.
[25] Stancheva I, Collins AL, Meehan RR, et al (2003). A mutant form of MeCP2 protein
associated with human Rett syndrome cannot be displaced from methylated DNA by
notch in Xenopus embryos. Mol Cell. 12 (2): 425–35.
[26] Vetrini F, Tammaro R, Marigo V , et al(2006). Aberrant splicing in the ocular albinism
type 1 gene (OA1/GPR143) is corrected in vitro by morpholino antisense
oligonucleotides. Hum Mutat. 27 (5): 420–6.
[27] Huang X and Saint-Jeannet JP. (2004). Induction of the neural crest and the opportunities
of life on the edge. Dev. Biol. 275, 1–11.
[28] Le Douarin NM. (2004). The avian embryo as a model to study the development of the
neural crest: a long and still ongoing story. Mech Dev. 121, 1089–102.
[29] Kopinke et al. Retinoic acid is required for endodermal pouch morphogenesis and not
for pharyngeal endoderm specification. Developmental Dynamics. 235(10):2695-2709.
[30] Ferreira T and Yelick PC. Alk8 Is Required for Neural Crest Cell Formation and
Development of Pharyngeal Arch Cartilages. Developmental Dynamics. 12/2003;
228(4):683-96.
[31] Daniel M and Marianne BF. Gene-Regulatory Interactions in Neural Crest Evolution and
Development. Developmental cell. V olume 7, Issue 3, p291–299, September 2004.
[32] Sauka-Spengler T, Meulemans, Jones M, and Bronner-Fraser M. (2007). Ancient
evolutionary origin of the neural crest gene regulatory network. Dev Cell. 13, 405–20.
[33] Cox, et al, 2012 - An essential role of variant histone h3.3 for ectomesenchyme
potential of the cranial neural crest. PLoS Genetics. 8(9):e1002938
[34] Stewart RA, Arduini BL, Look AT, et al. (2006) Zebrafish foxd3 is selectively required
University of Southern California Master ’s Thesis Works Cited
50

for neural crest specification, migration and survival. Dev Biol. 292, 147-188.
[35] Drerup CM, Wiora HM, Topczewski J and Morris JA. (2009) Disc1 regulates foxd3 and
sox10 expression, affecting neural crest migration and differentiation. Development. 136,
2623-2632.
[36] Jessica LL, Jennifer B, David WR. Reiterated Wnt signaling during zebrafish neural
crest development. Development. 2004 131: 1299-1308.
[37] Hazel Sive lab, http://sivelab.wi.mit.edu/Ongoing%20Projects2.htm
[38] Brenda LB and Alon K. Thyroid Hormone and Retinoic Acid Interact to Regulate
Zebrafish Craniofacial Neural Crest Development. Dev Biol. 2013 Jan 15; 373(2):
300–309.

Abstract (if available)

Abstract Background: In our studies of neural crest cells (NCCs) and craniofacial development, our lab has identified a physical interaction between two human proteins: CHD7, a chromatin remodeling protein, and PHF6, a dual PHD finger protein of unknown function. Mutations in these genes result in distinct syndromes in humans with contrasting craniofacial defects. Patients with CHARGE syndrome, with CHD7 mutations, have dysmorphic faces with reduced neural crest derived structural and sensory tissues. On the other hand, patients with Borjeson Forssman Lehmann syndrome (BFLS), caused by plant homeodomain Finger Protein 6 (PHF6) mutations, have a thick calvarium and broad jaw. ❧ Purpose: Using zebrafish as a model system, in this project we generated novel tools for understanding BFLS and CHARGE syndromes and defined the functions of chd7 and phf6 in craniofacial development. ❧ Methods: A clustered regularly interspaced short palindromic repeats (CRISPR) /CRISPR-associated (Cas) system was used to knock out these two genes in zebrafish embryos. PCR was used to identify the mutants. We also observed effects of mutation or knockdown on activation of known (sox10) and novel (RARG, CCND1) neural crest enhancers using confocal microscopy. In-situ hybridization will be performed in future to check the expression of some markers of NCCs and craniofacial tissue. ❧ Results: Chd7 mutant fish have phenotypes including developmental delay, reduced amounts of facial tissue, small heads, small eyes and dysmorphic craniofacial cartilage, similar to defects seen in CHARGE patients. In contrast, phf6 mutants have a significantly smaller or absent brain with a normal face. ❧ Conclusions: Chd7 and phf6 have opposing functions in craniofacial development in zebrafish. Mutations in these genes cause phenotypes similar but more severe to CHARGE and BFLS patients, respectively.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Chromatin remodeling factors Chd7 and Phf6 in craniofacial and heart development

PDF

Epigenetic checks and balances: PHF6 activity restricts neural crest migration

PDF

Characterization of zebrafish phf6 CRISPR mutant phenotypes

PDF

Elucidating the functional role of CHD7 associated nuclear PDH complex and other associated proteins on neural crest development

PDF

Characterization of Cxcr7 in zebrafish cardiac lymphatic vessel development

PDF

Derivation, expansion and characterization of human hippocampal primordial cells from normal and diseased iPSCs

PDF

Characterization of transgenic zebrafish lines for studying role of platelet derived growth factor (PDGF) signaling on coronary vessel formation in regenerating zebrafish heart after cryoinjury

PDF

Retinoic acid and TGFβ signaling regulate cardiovascular development

PDF

Modeling SynGAP1 truncating mutations in neurodevelopmental disease using iPSC-derived neurons

PDF

Understanding the role of APP and DYRK1A in human brain pericytes

PDF

Generation of human ESC reporter-lines to isolate hippocampal precursors and study hippocampal development

PDF

Functional analysis of a prostate cancer risk enhancer at 7p15.2

PDF

A novel role for post-migratory neural crest cells in cardiac outflow tract alignment: insights into the molecular basis of concomitant congenital cardiac and cleft malformations

PDF

TWIST1 functions in both mesoderm and neural crest derived cranial tissues to establish and maintain coronal suture patency

PDF

Utilizing zebrafish and mouse models to uncover the underlying genetics of human craniofacial anomalies

PDF

Development of ECM for the preservation of adult mouse pancreatic islet function in vitro

PDF

Characterization of cytochrome c peroxidase of Marinobacter aquaeolei

PDF

Development and evolution of skeletal joints: lessons learned from studying zebrafish

PDF

Neuroendocrine regulation of the transcription factor SKN-1/Nrf2 in oxidative stress response

PDF

Mixed lineage leukemia proteins (MLLs), their effect as coregulators on target gene expression and global histone methylation

Asset Metadata

Creator Sun, Yuhan (author)

Core Title Opposing function of chd7 and phf6 in zebrafish craniofacial development

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Biology

Publication Date 07/17/2018

Defense Date 05/10/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag CRISPR/Cas9,Morpholino,neural crest,OAI-PMH Harvest,zebrafish

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bajpai, Ruchi (committee chair), Lien, Ching Ling (committee member), Tokes, Zoltan (committee member)

Creator Email sunyh@cau.edu.cn,yuhansun@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-270749

Unique identifier UC11281265

Identifier etd-SunYuhan-4554.pdf (filename),usctheses-c40-270749 (legacy record id)

Legacy Identifier etd-SunYuhan-4554.pdf

Dmrecord 270749

Document Type Thesis

Format application/pdf (imt)

Rights Sun, Yuhan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA