University of Southern California Dissertations and Theses

Chromatin remodeling factors Chd7 and Phf6 in craniofacial and heart development

(USC Thesis Other)

Chromatin remodeling factors Chd7 and Phf6 in craniofacial and heart development

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Chromatin Remodeling Factors Chd7 and Phf6
in Craniofacial and Heart Development

By

Yuhan Sun

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CRANIOFACIAL BIOLOGY)

August 2022

Copyright 2022 Yuhan Sun
ii

ACKNOWLEDGEMENTS
These six years of Ph.D. for me is not quite smooth but very precious.
In the first three years in Dr. Ruchi Bajpai’s lab, I learned a lot of technologies and mastered
the methods to study molecular biology in multiple models including human iPSC, Danio Rerio
and Xenopus. It is a good experience in understanding the experimental idea in different animals,
and therefore can help me to take advantage of different animal models. Dr. Ruchi Bajpai is a very
kind and passionate PI, I learned a lot from her. Lab members in the Bajpai lab also helped me a lot.
Susan Smith did all the western blot experiments and helped characterize the biomedical
mechanism of my project. Dr. Kaivalya Shevade generated the CHARGE corrected iPSC lines and
flag-tagged CHARGE iPSCs. Dr. Erin Moran collaborated with me in the Phf6 Xenopus
experiments, although which is mainly not included here. Simiao Wang and Krystal Mendez were
taking turns to feed the iPSCs with me during the weekend, so I can have breaks during the
weekend. Varsha Neelakantan and Dr. Casey Griffin also are good lab members and gave me
precious advice on the lab meeting.
After Dr. Ruchi Bajpai left USC, Dr. Ching-Ling Lien becomes my new PI. Dr. Ching-Ling
Lien is a person with a warm heart and is an expert in heart regeneration study. I am glad that I
become one of the family members in Lien’s lab. After learning more about the heart, I found that
heart is a very interesting organ. I still remember the excitement when I successfully induced the
iii
iPSCs into beating cardiomyocytes, I “showed off” my beating cells to all my friends, families,
and lab members. Xidi Feng being a good friend and perfect celebrator gave me a lot of wonderful
suggestions for my ideas as well as data analysis. She is also my teacher for AFOG staining and
cryoinjury the zebrafish’s heart. Dr. David Chee Ern Wong collaborated with me on many
experiments including some iPSC-cardiac lineage cell differentiation, immunostaining, and in situ
hybridization in zebrafish embryos. He helped to optimize many protocols and took perfect
staining images. Dr. Michael Harrison taught me to use an innovative device to take live images of
the fish heart in vivo. Zhiyu Tian collaborated with me in generating another phf6 zebrafish
mutants that completely delete the whole phf6 gene including the promoter region. She helped me
a lot in genotyping, RT-PCR, and collecting embryos. Haipeng Bai helped modify the 4 gRNA
co-injection CRISPER method; Subir Kapuria helped me looked at the pdgfrb:GFP phenotype in
fish; Stanislao Travisano asked me tons of questions and helped me think deeply about my project;
Raquel Rodriguez, Aaron Baugh and Siqi Tao also helped me a lot and gave me a lot of good
memories.
Besides these two labs, my thesis committee member Dr. Gage Crump as an expert in
craniofacial development in zebrafish, gave me plenty of help not only by giving me many reporter
fish lines but also gave me many invaluable suggestions. His lab members also helped me a lot. Dr.
Peter Fabian gave me two sox10:switch fish lines to study the sox10 lineage traced cells. Dr.
D’Juan Farmer, Dr. Pengfei Xu, and Olivia Hung-Jhen Chen helped me troubleshooting in alcian
blue staining. I would not get these beautiful staining results without their generous help. Megan
iv
Matsutani as an expert in handling and householding the zebrafish taught me to do the in vitro
fertilization and gave me several reporter fish lines. Without her help, I could not recover the
chd7(d6) mutations after the covid19 pandemic.
Dr. Megan Laura McCain collaborates with us in a cardiac study. Her lab member Dr. Patrick
Vigneault spent a lot of time teaching me how to differentiate the iPSCs into cardiomyocytes in
vitro. He is a good teacher and carefully explains all the tricks in cardiomyocytes’ culture. Natalie
Khalil and Mher Garibyan kindly helped us make the device for the zebrafish heart live images.
Dr. Jau-Nian Chen collaborates with us in the study of the cardiomyocytes in zebrafish heart.
She generously shared another sox10:switch fish line with us, and also gave me many good
suggestions. Dr. Adam Langenbacher in Chen’s lab also shared the rtf1 and p53 morpholino with
me.
G. Esteban Fernandez and the Cellular Imaging Core at the Saban Research Institute of
Children's Hospital Los Angeles gave me expert assistance with image acquisition. Narine
Harutyunyan and Andrew Salas in the Stem Cell Analytics Core at the Saban Research Institute of
Children's Hospital Los Angeles supported me a lot in the human cell culture and differentiation.
I would also like to thank my family. My parents have open minds and always stand by me
and support me no matter what I want. I would not be me without being unconsciously influenced
by my family. I will also show my appreciation to all my friends, especially Jiabo Xu and
Kaicheng Zhou, you are the best friends and neighbors. The covid19 pandemic changed our life
these years, but I never feel afraid of it since I know that you guys are together with me.
v
At last, I would like to thank my qualifying committing members Dr. Jian Xu, and my defense
committee member Dr. Jianfu Chen for their help in giving me suggestions to make my project
more intact. I would like also thanks to my previous director Dr. Michael Paine, and the program
members Janice Bea, Larissa Leach for always being there and giving me support in every aspect
during my Ph.D.
Thank you to everyone, without your kind help, I would not be able to make it.

Yuhan Sun
2022.5.31

vi
Table of Contents
ACKNOWLEDGEMENTS .................................................................................................... ii
List of Tables ........................................................................................................................... ix
List of Figures .......................................................................................................................... x
ABSTRACT ............................................................................................................................. xi
1. Chapter I: Rare Diseases CHARGE Syndrome and BFLS Are Caused by Mutations
in Chromatin Remodeling Factor Genes .............................................................................. 1
1.1. CHARGE Syndrome ............................................................................................. 1
1.1.1. CHARGE Syndrome Is the Leading Cause of Deaf-Blindness Disease ............. 1
1.1.2. CHARGE Patients Have Defects in Both Craniofacial and Cardiac Tissues ...... 1
1.1.3. CHARGE Patients Have Abnormalities in Neural Crest-Derived cells .............. 2
1.1.4. CHD7 Mutation Causes Two-Thirds of the CHARGE Syndrome Cases ........... 3
1.1.5. Previous CHARGE Mice Models ....................................................................... 6
1.2. Börjeson-Forssman-Lehmann Syndrome ........................................................... 7
1.2.1. BFLS Patients Have Opposite Symptoms to Patients with CHARGE
Syndrome ............................................................................................................. 7
1.2.2. BFLS Patients Have X Chromosome Linked PHF6 Mutations .......................... 8
1.2.3. PHF6 Directly Interacts with Modified Histone Tail and Double Strand
DNA ................................................................................................................... 10
1.2.4. Previous BFLS Mice Models ............................................................................ 11
1.3. CHD7 and PHF6 Might Have Antagonistic Functions .................................... 13
2. Chapter II: iPSCs to model human diseases, CHARGE and BFLS .......................... 14
2.1. Introduction ......................................................................................................... 14
2.1.1. Human iPSC – In Vitro Disease Model ............................................................. 14
2.1.2. Differentiation the iPSCs into NCCs In Vitro ................................................... 14
2.2. Results ................................................................................................................... 15
2.2.1. Characterization of CHARGE and BFLS Patient iPSC Lines .......................... 15
2.2.2. PHF6 Is a Negative Regulator of NCC Formation ............................................ 17
2.2.3. CHD7 Functions as a Positive Regulator in NCC Formation,
Migration, and Scattering ................................................................................. 22
2.2.4. CHARGE and BFLS iPSC Could Form Normal Cardiomyocytes ................... 24
2.3. Discussion ............................................................................................................. 26
2.3.1. CHD7 May Have Tissue Specific Functions ..................................................... 26
2.3.2. CHD7 and PHF6 Could Have Antagonistic Functions in NCC
Lineage Cells .................................................................................................... 27
2.4. Materials and Methods ....................................................................................... 28
CHARGE Patient iPSC Lines ......................................................................................... 28
Finger Printer Test .......................................................................................................... 29
vii
Mycoplasma Test ............................................................................................................ 29
iPSC Culture ................................................................................................................... 30
iPSC-NCC Differentiation .............................................................................................. 30
iPSC-Cardiomyocyte Differentiation ............................................................................. 31
Further Differentiation of NCCs into Adipocytes, Chondrogenic Cells,
Osteogenic Cells, and Mesenchymal Cells ..................................................................... 32
Immunostaining .............................................................................................................. 33
The List of Antibodies Used for Immunostaining: ......................................................... 33
Statistical Analysis ......................................................................................................... 35
3. Chapter III: Zebrafish In Vivo Disease Models of CHARGE Syndrome .................. 36
3.1. Introduction: ........................................................................................................ 36
3.1.1. Danio Rerio – In Vivo Animal Model ............................................................... 36
3.1.2. Neural Crest and Craniofacial Development in Zebrafish ................................ 37
3.1.3. Inner Ear Development in Zebrafish ................................................................. 39
3.1.4. Reverse Genetic Tools Used to Study Candidate Genes in Zebrafish .............. 40
3.1.5. Previous Chd7 Study in Zebrafish ..................................................................... 42
3.2. Results ................................................................................................................... 44
3.2.1. Co-Injecting 4 gRNAs With Cas 9 Protein Increased the Mutant
Rate in Generating chd7 Mutations .................................................................. 44
3.2.2. The chd7(d55) Mutant Fish Has Weaker CHARGE Like Phenotypes ............. 46
3.2.3. The chd7(d55) Allele Might Display Maternal Zygotic Phenotypes. ............... 48
3.2.4. The chd7(d55) -/- Mutants Show CHARGE-like Phenotypes in
Craniofacial Structure. ....................................................................................... 50
3.2.5. In Zebrafish, NCCs Contribute to the Ventral Aorta. ....................................... 54
3.2.6. The chd7(d55)-/- Fish Have Defects in Ventral Aorta. ..................................... 56
3.3. Discussion ............................................................................................................. 59
3.3.1. The Weaker Phenotype in chd7 Homozygous Fish .......................................... 59
3.3.2. Cardiac NCCs Lineage in Zebrafish Heart ........................................................ 60
Materials and Methods .................................................................................................... 61
Zebrafish Maintenance ................................................................................................... 61
List of Zebrafish Lines Used in This Study: ................................................................... 61
CRISPR/Cas9 Technology ............................................................................................. 62
Microinjection ................................................................................................................ 64
T7E1 Assay ..................................................................................................................... 64
Genotyping ..................................................................................................................... 65
RT-PCR .......................................................................................................................... 65
Whole-mount In Situ Hybridization ............................................................................... 66
Confocal Microscopy ..................................................................................................... 67
Chondrocytes’ Arrangement Analysis ............................................................................ 67
viii
Alcian blue and Alizarin Red Staining ........................................................................... 68
Kits Used in This Project ................................................................................................ 69
4. Chapter IV: In Vivo Disease Models of BFLS. ........................................................... 70
4.1. Introduction ......................................................................................................... 70
4.1.1. Xenopus as an In Vivo Animal Model ............................................................... 70
4.1.2. Zebrafish phf6 Mutants as an In Vivo BFLS Model. ......................................... 71
4.2. Results ................................................................................................................... 72
4.2.1. Phf6 Is Expressing in a Specific Subgroup of NCCs in Xenopus. .................... 72
4.2.2. BFLS Patient-Specific Mutant Fish Model Does Not Show
Obvious Phenotypes. ........................................................................................ 74
4.2.3. The phf6(C43Y)-/- Larvae Do Not Have Obvious Craniofacial Phenotype. .... 76
4.2.4. Phf6(C43Y)+/- and phf6(C43Y)-/- Can Partially Suppress the
chd7(d55)-/- Phenotype. ................................................................................... 78
4.3. Discussion ............................................................................................................. 80
4.3.1. The Weaker Phenotype in phf6 Homozygous Fish ........................................... 80
4.3.2. The Partially Suppressed chd7 Mutants’ Phenotypes in the chd7
and phf6 Double Mutants ................................................................................... 81
4.4. Materials and Methods ....................................................................................... 82
The gRNA Design for Generating phf6 CRISPR Mutants ............................................. 82
Xenopus Embryology ..................................................................................................... 82
Whole-mount In Situ Hybridization ............................................................................... 83
Probes for In Situ Hybridization ..................................................................................... 83
Generating chd7 and phf6 Double Mutants. ................................................................... 85
5. Chapter V: Future Directions ...................................................................................... 86
5.1. Discussion ............................................................................................................. 86
5.1.1. CHD7 and PHF6 May Have Antagonistic Functions ........................................ 86
5.1.2. The Possible Affected Cardiac Cell Lineages in CHARGE Mice Models ....... 87
5.1.3. Four Signal Pathways That are Associated With NCCs Formation
Are Interested to Be Studied in chd7 Mutants in the Future .............................. 89
5.1.4. Histone Citrullination Recognized by PHF6 Could Be a Novel
Aspect to Study .................................................................................................. 90
References .............................................................................................................................. 93

ix
List of Tables
Table 2-1: The results of the mycoplasma test ....................................................................... 30
Table 2-2: Antibodies used for immunostaining .................................................................... 33
Table 3-1: The forward primers used to generate the gRNAs targeting the chd7 gene. ........ 63
Table 3-2: Primers used to generating the RNA probes for in situ hybridizations in
zebrafish ................................................................................................................ 66
Table 4-1: The forward primers used to generate the gRNAs targeting the phf6 gene. ......... 82
Table 4-2: Phf6 probe sequences. ........................................................................................... 84

x
List of Figures
Figure 1-1: The overall structure of CHD7 protein .................................................................. 5
Figure 1-2: Symptoms of the patients with CHARGE syndrome and BFLS. .......................... 8
Figure 1-3: Mutations in BFLS patients and the interacting partner of PHF6. ........................ 9
Figure 2-1: The establishment of BFLS iPSC lines. .............................................................. 16
Figure 2-2: Accelerated differentiation in BFLS iPSC-NCCs. .............................................. 19
Figure 2-3: Accelerated differentiation in BFLS iPSC-NCCs’ further differentiation. ......... 21
Figure 2-4: iPSC-NCC defects in CHARGE patient cell lines. ............................................. 23
Figure 2-5: Immunostaining of iPSC-CMs. ........................................................................... 25
Figure 2-6: Sequencing of the wildtype control, CHARGE patients’, and BFLS
patients’ cell lines. ............................................................................................... 28
Figure 3-1: Generation of the chd7 zebrafish mutations. ....................................................... 45
Figure 3-2: T7E1 assay to test gRNAs targeting chd7. .......................................................... 46
Figure 3-3: Generation and preliminary characterizations of two chd7 mutant alleles. ......... 49
Figure 3-4: The craniofacial phenotypes characterization in chd7(d55) mutants. ................. 51
Figure 3-5: The chondrocytes’ and teeth phenotypes characterization in chd7(d55)
mutants. ............................................................................................................... 52
Figure 3-6: sox10:Cre lineage traced cells are observed in ventral aorta. ............................. 55
Figure 3-7: The phenotypes of chd7(d55) mutants in the cardiovascular .............................. 57
Figure 3-8: The high magnification images showed the pdgfrb positive cells in the
OFT did not have obvious defects in chd7(d55) mutants at 102 hpf ................... 58
Figure 4-1: The conservation of PHF6 across species and the expression pattern of
Phf6 in Xenopus embryos. ................................................................................... 73
Figure 4-2: Overall characterization of the phf6 zebrafish mutations. ................................... 75
Figure 4-3: The phf6(C43Y) mutants do not have obvious craniofacial phenotypes. ............ 77
Figure 4-4: The chd7 and phf6 double mutant can partially suppressed the chd7 fish
mutants’ phenotypes. ........................................................................................... 79
Figure 5-1: The schematic of our hypothesis that Chd7 and Phf6 work together as
antagonists specifically in the craniofacial NCCs but not in the cardiac NCCs. .. 86

xi
ABSTRACT
Background:
CHARGE syndrome is a rare genetic disease that has symptoms in many different organs,
including craniofacial and cardiac tissues. CHD7 mutations were found in two-thirds of CHARGE
patients and non-syndromic congenital heart disease patients but the molecular mechanisms
are yet to be fully understood. Börjeson-Forssman-Lehmann Syndrome (BFLS), caused by
mutations in PHF6, is another very rare genetic disease, which shows many opposite symptoms to
the CHARGE syndrome. Patients with either CHARGE syndrome or BFLS have abnormalities in
the neural crest cells (NCCs) derived tissues. Preliminary experiments showed that knockdown
CHD7 or PHF6 eliminated or accelerated the iPSC induced NCC (iPSC-NCC) differentiation,
respectively. Furthermore, inhibiting PHF6 expression in iPSC-NCCs could partially rescue the
CHD7 knockdown phenotypes in vitro.

Purpose:
Understand the functions of CHD7 and PHF6 in NCCs and NCCs derived tissues.

Methods:
Both human patients’ iPSC and zebrafish CRISPR Cas9 mutants were used to study the
function of CHD7 and PHF6 in vitro and in vivo.
xii

Results:
In patients’ iPSC lines, CHD7 works as a positive regulator in iPSC-NCC
differentiation, while PHF6 works as a negative regulator in vitro. In zebrafish model in vivo, chd7
homozygous mutations have weak CHARGE-like craniofacial defects and decreased the ventral
aorta. On the other hand, phf6 fish mutants do not show obvious craniofacial phenotypes.
Excitingly, the CHARGE-like craniofacial defects in chd7 homozygous mutants could be partially
suppressed in phf6, chd7 double mutants, suggesting these two chromatin remodeling factors
might have genetic interactions.

Conclusions:
Our results indicated that CHD7 and PHF6 have opposite functions in NCCs formation in
vitro. Moreover, in vivo zebrafish model demonstrated that Chd7 and Phf6 may have
as antagonistic functions in craniofacial development in vivo.

Key words ：
CHD7, PHF6, CRISPR/Cas9, neural crest, craniofacial, ventral aorta, iPSC, zebrafish

1
1. Chapter I: Rare Diseases CHARGE Syndrome and BFLS Are
Caused by Mutations in Chromatin Remodeling Factor Genes
1.1. CHARGE Syndrome
1.1.1. CHARGE Syndrome Is the Leading Cause of Deaf-Blindness Disease
CHARGE syndrome is a rare disease that occurs in approximately 1 in 8500 to 10,000
newborn babies, and it is the leading syndrome associated with deaf-blindness in school-aged
children (Flanagan et al., 2007). CHARGE syndrome is named by the abbreviation for six standard
features of this disorder: coloboma, heart disease, atresia choanae, retarded growth, genital
hypoplasia, and ear anomalies and/or deafness (Janssen et al., 2012). Other medical conditions
such as balance problems, swallowing/ breathing difficulties, kidney defects are also seen in
different patients (Pagon et al., 1981), which could be life-threatening and need to be treated as
soon as possible after birth. Thus, it is important to understand the underlying mechanism causing
the CHARGE syndrome to help us find out the potential therapeutic approach for the patients.
1.1.2. CHARGE Patients Have Defects in Both Craniofacial and Cardiac Tissues
CHARGE syndrome is extremely complex in terms of symptomatology that can vary in
severity from patient to patient. In addition to the symptoms described above, craniofacial
abnormalities including cleft palate/lip, broad forehead, reduced dysmorphic jaw, and asymmetric

2
face palsy are consistently found in patients with CHARGE syndrome. Nearly 25% of the
CHARGE patients have cleft palate and/or cleft lip (Isaac et al., 2018). The broad forehead,
reduced dysmorphic jaw, and asymmetric face palsy are even more common in CHARGE
syndrome, and these phenotypes together form a typical CHARGE face and can be used as a clue
in diagnosis of CHARGE syndrome (Isaac et al., 2018).
Another important feature is the cardiovascular malformations which have also been reported
in patients with CHARGE syndrome. Among those cardiovascular malformations, outflow tract
defects, atrioventricular septal defect (Corsten-Janssen et al., 2013), and aortic arch anomalies
were also frequently present (Corsten-Janssen and Scambler, 2017). In summary, CHARGE
patients have defects in both craniofacial and cardiac tissues, and in this project, I tried to mainly
focus on these two organs.
1.1.3. CHARGE Patients Have Abnormalities in Neural Crest-Derived cells
The concomitant phenotype of craniofacial and cardiac birth defects are common in patients
with CHARGE syndrome and some other syndromes (Nagelberg et al., 2015). Neural crest cells
(NCCs) may be one major cell type that contributes to this concomitant phenotype. Many tissues
affected in CHARGE syndrome, such as the deaf (Ritter and Martin, 2019), coloboma, cleft lip/
palate and hypopigmentation are also derived from NCC lineage.
NCCs are a group of transient cells in vertebrates, which migrate from the neural tube and
then give rise to diverse cell lineages that form many different tissues including craniofacial

3
skeleton, cartilage, cells in the great vessels, and septum of the hearts in humans (Achilleos and
Trainor, 2012; Huang and Saint-Jeannet, 2004). NCCs undergo an epithelial-to-mesenchymal
transition and migrate out as pharyngeal arches (PAs). In zebrafish, there are seven PAs. The first
PA contributes to the craniofacial cartilage including the Meckel’s cartilage and part of the
palatoquadrate; the second PA forms the dorsal hyosymplectics, the ventral ceratohyals, and the
interhyal cartilage; the 3-7 PAs produce pairs of ventrolateral ceratobranchial cartilages.
Specifically, the fifth ceratobranchials which form several ossified pharyngeal teeth is coming
from the seventh PA (Frisdal and Trainor, 2014; Mork and Crump, 2015). There are six pairs of
pharyngeal arch arteries (PAAs) that branch out from ventral aorta and are part of pharyngeal
arches. However, in humans and other mammals, there are only five pairs of PAAs since the fifth
pair is never fully formed (Bamforth et al., 2013). The posterior PAAs, PAA 3–6 contribute to the
formation of the great vessels – carotid arteries, the aortic arch, and the pulmonary arteries (Abrial
et al., 2017; Mao et al., 2019; Meng et al., 2017) while PAA 3-6 carry blood to gills in fish
(Anderson et al., 2008). The great vessels carry blood to and from the heart, and their abnormality
causes cardiac birth defects in humans.
1.1.4. CHD7 Mutation Causes Two-Thirds of the CHARGE Syndrome Cases
The first step to study a genetic disease is to identify the mutations that cause this disease.
Previously clinical research has established that haploinsufficiency in the chromodomain helicase
DNA binding protein 7 (CHD7) gene causes two-thirds of the CHARGE syndromes (Allen et al.,

4
2007). A study involving 299 patients with pathogenic CHD7 mutations shows that 74% of these
patients have heart defects and persistent ductus arteriosus are most prevalent (present 52% of the
heart defects in CHD7 mutant group (Corsten-Janssen et al., 2013). CHD7 is an ATP-dependent
chromatin remodeler and is a member of the SNF2-protein superfamily, which is highly conserved
from zebrafish to humans (Figure 1-1A). In Xenopus, Chd7 is expressed in the neural crest and
neural tissues (Bajpai et al., 2010). In mice, Chd7 has been shown to be required in the pharyngeal
ectoderm for normal PAA development (Randall et al., 2009). In addition to CHARGE syndrome
patients, CHD7 mutations have also been identified in non-syndromic patients with congenital
heart defects (Yan et al., 2020), which indicates that CHD7 plays an important role in human
heart development.
CHD7 contains double chromodomains, an ATPase domain [comprising DEXDc
(DEAD-like helicase superfamily) and HELICc (Helicase C) domains], one SLIDE (SANT-like
ISWI domain)/SANT (SWI3, ADA2, N-CoR, and TFIIIB) domain, and double BRK (Brahma and
Kismet) domains (Figure 1-1B). Previous research showed that the two BRK domains in both
CHD7 and CHD8 proteins can interact with the zinc-finger domains (Allen et al., 2007; Ishihara et
al., 2006). In humans, the chromodomains in CHD1 bind to methylated lysine 4 in the histone H3
tail (H3-K4me1) although there is no evidence of this kind of interaction in yeast and Drosophila
(Flanagan et al., 2007). In S. cerevisiae, the C-terminal region, which contains a SANT/SLIDE
domain contributes to DNA binding and is required for the function of Chd1 (Ryan et al., 2011).
The DEXDc domain and HELICc domain together form the ATPase domain, and the loss of this

5
ATPase domain may have a dominant-negative function in CHD7. One specific amino acid in the
ATPase domain is the 999th lysine of CHD7, which is in a highly conserved ATP-binding motif,
and ATPase and chromatin remodeling ability of several SNF2-superfamily proteins is abolished
when the equivalent lysine is mutated (Bouazoune and Kingston, 2012).

Figure 1-1: The overall structure of CHD7 protein. A, Phylogenetic analysis of CHD7 protein sequence between
Homo sapiens, Danio rerio, Xenopus tropicalis, Gallus gallu, Rattus norvegicus, Mus musculus, and Canis lupis
familiaris (Ishihara et al., 2006). The peaks indicate the similarity from 50% to 100%. CHD7 exons are highly
conserved from fish to human. Blue shows the peaks in exons, orange marks the peaks in introns, green are the not
consistent areas (less than 75%), yellow is the 3’UTR area. B, The protein domains of CHD7 and the predicting
binding partners of each domain.

6
1.1.5. Previous CHARGE Mice Models
To study the mechanisms underlying CHARGE syndrome in vivo, multiple Chd7-null mice
models have been generated. While no homozygous embryos were detected beyond E10.5 (Hurd
et al., 2007; Layman et al., 2010; Yan et al., 2020), the heterozygous Chd7 mice recapitulate
various symptoms of CHARGE syndrome such as coloboma abnormal, inner ear abnormalities,
cleft palate, choanal defects, cardiovascular defects, genital defects, and developmental growth
delays (Bosman et al., 2005; Gage et al., 2015; Layman et al., 2010; Layman et al., 2009; Ogier
et al., 2014; Schulz et al., 2014). Specific for the cardiovascular defects, Chd7 mice mutants
from an ENU mutagenesis project have been reported to have defects in the interventricular
septum (Bosman et al., 2005), hypocellular myocardial wall, and hypocellular atrioventricular
cushion (Liu et al., 2014a). To identify the temporal-spatial specific function of Chd7, many
conditional knockout mice also have been generated (Layman et al., 2010). Despite the
controversies of Chd7 functions in cardiac NCCs, the conotruncal defects are found not caused
by impaired migration of NCCs into PAAs but by the dramatically reduced migrated NCCs into
the proximal outflow tract (OFT) cushions at E11.5 in mutants (Yan et al., 2020) using an
improved Wnt1:Cre (Lewis et al., 2013). This finding supports the cell autonomous functions of
Chd7 in NCCs. However, conditional knockout Chd7 mice also died either before birth or just
after birth.

7
To understand the function of Chd7 in early development and overcome the lethality, in this
project, I generated two different zebrafish mutations as my disease animal model to study the
function of Chd7 in early development in both craniofacial and cardiac tissues. Interestingly,
cardiac NCCs have been shown to contribute to cardiomyocytes in zebrafish recently by multiple
labs (Cavanaugh et al., 2015; Tang et al., 2019). Therefore, I will also explore the potential
complex role of Chd7 in cardiac NCCs in fish.
1.2. Börjeson-Forssman-Lehmann Syndrome
1.2.1. BFLS Patients Have Opposite Symptoms to Patients with CHARGE Syndrome

8
Figure 1-2: Symptoms of the patients with CHARGE syndrome and BFLS. The opposite phenotypes in these
two diseases are shown in red and green, respectively (modified from Erin Moran’s figure). Blue arrows indicated the
symptoms that may relate to the NCCs derived tissues.
In addition to the CHARGE syndrome, another extremely rare disease, Börjeson-Forssman-
Lehmann Syndrome (BFLS) caught our eyes by its patients’ symptoms opposite to the CHARGE
syndrome (Figure 1-2). Clinically, BFLS is defined by a discrete set of varying penetrance features
such as intellectual disability, obesity with gynecomastia, microcephaly, characteristic
craniofacial features, hypogonadism, seizures, developmental delay, and occasionally digit
elongation (BORJESON et al., 1962; Turner et al., 2004; Zweier et al., 2013).
The patients with BFLS have phenotypes including long ear lobe, narrow forehead, broad jaw,
streaking pigmentation, and obesity are all opposite to the CHARGE patients’ abnormalities.
Among these opposing phenotypes, the craniofacial phenotype and streaking pigmentation in
BFLS patients are also related to the NCC-derived organs.
1.2.2. BFLS Patients Have X Chromosome Linked PHF6 Mutations
BFLS is caused by mutations in Plant Homeodomain (PHD) Figure Protein 6 (PHF6), which
is on the X chromosome in humans (Figure 1-3A) (Lower et al., 2002). Therefore, BFLS affects
males as hemizygous and rarely in females as heterozygous (Ardinger et al., 1984; BORJESON et
al., 1962; Gécz et al., 2006; Turner et al., 2004; Zweier et al., 2013). When compared the male and
female BFLS patients, mutations occurring in male patients are always be considered as the milder
ones such as missense mutations (8 in 12 mutations), small deletion (1 in 12 mutations), and

9
nonsense mutation (1 in 12 mutations) at the 3’ end, and nonsense mutation (1 in 12 mutations) at
the 5’ end of the gene follows by backup start codon (M46 could work as backup start codon).
However, female patients with these milder mutations are carriers without obvious BFLS
phenotypes.

Figure 1-3: Mutations in BFLS patients and the interacting partner of PHF6. A. Schematic of BFLS patient
mutations. White Circles: missense mutations, red circles: nonsense mutations, blue circles: frameshift mutations,
green circle: amino acid deletion, green bars: deletion over the area outlined by the bar, blue bar: duplication over the
area outlined by the bar. Male patient mutations are outlined above the protein schematic, with the precise residue

10
affected next to each circle, while female patient mutations are outlined below the protein schematic, with point
mutations having the precise residue affected next to each circle. (Berland et al., 2011; Chao et al., 2010; Crawford et
al., 2006; Di Donato et al., 2014; Ernst et al., 2015; Jahani-Asl et al., 2016; Lower et al., 2002; Turner et al., 2004;
Visootsak et al., 2004; Zweier et al., 2013; Zweier et al., 2014). B. Schematic for the binding affinity of PHF6 protein,
F1: ePHD1, F2: ePHD2 (Liu et al., 2014b). Schematics created using BioRender.
On the other hand, female BFLS patients always have more severe mutations such as large
deletions (4 in 9 mutations), large duplications (1 in 9 mutations), and frame-shift mutations (2 in 9
mutations) in PHF6. Furthermore, affected female patients also have extremely skewed
X-inactivation in blood samples compared to the carriers (Van Vlierberghe et al., 2010; Zweier et
al., 2013), which indicated that PHF6 may have an important function in blood cells. In addition to
the BFLS, somatic loss of PHF6 has been identified in several hematopoietic malignancies,
suggesting that PHF6 functions as a tumor suppressor gene.
1.2.3. PHF6 Directly Interacts with Modified Histone Tail and Double Strand DNA
PHF6 contains two extended PHD domains (ePHDs), each ePHD contains three zinc-fingers.
Previously published papers showed that the second ePHD (ePHD2) in PHF6 binds to the double
strand DNA randomly (Liu et al., 2014b). As a chromatin adapter protein (Todd et al., 2015), it is
important to understand the function of PHF6 in recognizing the epigenetic modified histone tails,
Susan Smith in Dr. Ruchi Bajpai’s lab did many immunoprecipitation experiments on PHF6 to
determine its functions, however, its role during early developmental is still unknown.
Previous results showed that endogenous PHF6 pulled down H3-K4me1 peptide by
immunoprecipitation, but not the triple-methylated 9th lysine residue of the histone H3 protein tail

11
(H3-K9me3). Interestingly, none of the Arginine methylated H3 histone peptides can be pulled
down by PHF6. To interpret this data, a modification of arginine residues by citrullination, which
is catalyzed by peptidylarginine deiminases (PADs) caught our eyes (Kan et al., 2012; Mastronardi
et al., 2006). Thrillingly, Citrullinated 8
th
Arginine residue of the histone H3 protein tail (H3-Cit8)
can be recovered by PHF6 efficiently. Moreover, it is the recombinant purified ePHD1 but not
ePHD2 that contributes to recognize the modified histone peptides and it has more efficiency in
binding to H3-K4me1 and H3-Cit8 dual modified peptide than the single modified peptides. Thus,
Susan Smith has characterized the function of the ePHD1 in PHF6 in direct association with
H3-K4me1 and H3-Cit8 dual modified peptide.
All together, these results made PHF6 the first citrulline binding protein to be identified. The
overall hypothesis for the function of PHF6 is that the ePHD1 binds to the H3-K4me1and H3-Cit8
dual modified histone and the ePHF2 binds to double strand DNA randomly, and therefore “locks”
the gene and prevents it from being transcribed (Figure 1-3B).
1.2.4. Previous BFLS Mice Models
To study the function of Phf6 in vivo, knocking down Phf6 using siRNA has been performed.
In developing murine central nervous system, knockdown experiments of Phf6 revealed its role in
neuronal cell migration that likely contributes to the microcephaly phenotype in some BFLS
patients (BORJESON et al., 1962; Franzoni et al., 2015; Kasper et al., 2017; Zhang et al., 2013).

12
Recently, three BFLS mice models were published. The first one is Phf6 C99F mice published
in 2018 (Cheng et al., 2018). The Phf6 C99F mice are viable with deficits in learning and memory,
social interactions, and emotionality. Furthermore, Phf6 has been shown to promote the expression
of neurogenesis genes and downregulate synaptic genes in these Phf6 C99F mice. The second
model is a Cre-mediated recombinant Phf6 allele with loxP sites flanking exon 4 and 5. After
recombination, the Phf6 mRNA is undetectable. These Phf6
-/Y
male mice have short stature which
may be caused by their reduced growth hormone. Nervous system-specific deletion of Phf6
indicates that Phf6 controls growth at least in part through neuroendocrine regulation.
Interestingly, Phf6
−/Y
males are viable in FVB/BALB/c background but perinatal lethal in
C57BL/6 mice background, which indicates that the genetic background is an important additional
mediator of the severity of phenotype presentation (McRae et al., 2020). PHF6 R342X is a
recurrent allele of BFLS in patients (Ahmed et al., 2021). The latest Phf6 R342X mice model is
consistently smaller in size with a significant decrease in body weight. The mutant animals have a
smaller pituitary gland, a hypoplastic anterior pituitary, and significantly reduced plasma levels of
prolactin which may be related to their small size. In contrast, Phf6 R342X mice do not have
changes in growth hormone levels. The Phf6 R342X mice show variations in brain volume and
altered hippocampus and hypothalamus relative brain volumes. Consistent with the C99F model,
behavior testing demonstrated Phf6 R342X mice have deficits in associative learning, spatial
memory, and an anxiolytic phenotype.

13
All these data suggested that Phf6 may have important functions in brain development and
body growth. To understand the function of Phf6 in early development in NCCs lineages, I
generated two phf6 mutant zebrafish as my animal model to compare to the CHARGE fish models.
1.3. CHD7 and PHF6 Might Have Antagonistic Functions
CHARGE syndrome and BFLS patients show the opposite symptoms (Figure 1-2). The
protein structures of CHD7 and PHF6 both contain domains that may interact with DNA (Liu et
al., 2014b; Ryan et al., 2011) and H3-K4me1 (Flanagan et al., 2007). Moreover, CHD7 has been
shown to bind to zinc-fingers, the main domains in the ePHDs of PHF6 protein. All these data
suggested that CHD7 and PHF6 may recognize similar epigenetic modified regions and may have
direct interactions. Dr. Ruchi Bajpai determined and identified a physical interaction between
CHD7 and PHF6 by pulling down the protein complex (personal communication and data are not
included here). Thus, we hypothesized that CHD7 and PHF6 may have opposite functions in
NCCs derived cells. To test this hypothesis, I utilized both human patients induced pluripotent
stem cells and zebrafish animal models to understand the function of CHD7 and PHF6 in NCCs
and NCC derived cells both in vitro and in vivo.

14
2. Chapter II: iPSCs to model human diseases, CHARGE and BFLS
2.1. Introduction
2.1.1. Human iPSC – In Vitro Disease Model
In 2006, Dr. Shinya Yamanaka pioneered the introduction of four genes (MYC, OCT3/4,
SOX2, and KLF4) encoding transcription factors and discovered that they could convert somatic
cells into induced pluripotent stem cells (iPSC) (Takahashi and Yamanaka, 2006). These induced
cells have ability to differentiate into cells of all three germ layers. The discovery of iPSC makes
it possible to study human patients’ cells directly. Human iPSCs derived from patients carry all
genetic information, thus allowing us to check whether patients’ symptoms are caused by the
mutations when recapitulating the developmental processes in vitro and by correcting the
mutations in patient iPSCs.
2.1.2. Differentiation the iPSCs into NCCs In Vitro
Dr. Ruchi Bajpai previously published the method to differentiate human embryonic stem
cells (ESC) into NCCs in vitro (Bajpai et al., 2010), and showed that using CHD7 shRNA
knockdown CHD7 disrupts the formation of the migratory NCCs. Consistently, in Xenopus,
injection of Chd7 MO, perturbed cell migration to NCC segments, which can be rescued by

15
co-injection of human CHD7 mRNA. Moreover, expression of multipotent, migratory NCC
specifiers Twist, Slug (Snail2), and Sox9 were severely diminished in Chd7 knockdown embryos.
Dr. Ruchi Bajpai also induced the fibroblast cells obtained from two patients with
CHARGE syndrome to generate stable iPSC lines carrying CHD7 p.Arg1036X and CHD7
p.Gln1701X mutations, respectively. Using these two cell lines, we could study the specific
defects in different CHARGE patients. While this work is in progress, another group published
their characterization of CHARGE patient-specific iPSC induced NCCs (iPSC-NCCs) (Okuno et
al., 2017) using two iPSC lines with CHD7 p.Gln 1391X and CHD7 p.Arg1493Ter mutations,
respectively. These two specific iPSC-NCCs showed abnormal migration, defective scattering,
and defective single-cell spontaneous motility.
2.2. Results
2.2.1. Characterization of CHARGE and BFLS Patient iPSC Lines
Dr. Ruchi Bajpai obtained the CHARGE and BFLS patients’ fibroblast cells from Stanford
University and induced them into iPSCs. Two CHARGE syndrome patients with de novo CHD7
p.Arg1036X and CHD7 p.Gln1701X mutations in ATPase domain are from different families; and
three BFLS patients (24B, 25B, and 26B) belong to a family carrying an inherited mutation in
PHF6 (p.C45Y) (Figure 2-1A). The mutated cysteine in BFLS patients is important for forming
the first zinc-finger in the first extended PHD domain in PHF6 protein (Figure 1-3Bi, Bii).

16

Figure 2-1: The establishment of BFLS iPSC lines. A. The family tree of the BFLS patients’ family. B. The C45Y
mutation in our BFLS patients is the Cysteine which contributes to the first zinc-finger of the extended PHD1
domain. C. sequencing data confirmed the C45Y mutation in 26B patient iPSCs. D. Finger printer sequencing shows
that 25B and 26B iPSCs are from different patients. E. MycoAlert Mycoplasma Detection Assays show that our
BFLS iPSCs do not have mycoplasma contamination. F-H. Immunostaining for the BFLS iPSCs and iPSC induced
cells. F . Differentiating the BFLS iPSC into NCCs, and staining the neuroectodermal markers in neuronal epithelial
spheres (first on the left), and NCCs (the rest three on the right). G. Differentiating the cells into mesenchymal cells,
and stain with mesendodermal markers. H. Pluripotency markers show that the BFLS iPSCs have the self renewal
ability.

17
The iPSCs from BFLS patients (iPSC-BFLS) had the specific PHF6 mutations and finger
printer pattern (Figure 2-1C and represent finger printer image in 2-1D), showing that they are
patient specific iPSCs. They are also free of mycoplasma contamination (compared to the positive
control, Figure 2-1E) and inducing vector (Done by Dr. Ruchi Bajpai, personal communication
and data not included), indicating that they are clean and can be used as BFLS disease models.
Importantly, iPSCs can both self-renew and differentiate into cells of different germ layers. To
test the self-renewal ability, immunostaining showed that pluripotency markers SSEA4, E
CADHERIN, OCT4, and SOX2 are expressed in iPSC-BFLS (Figure 2-1H). In parallel,
iPSC-BFLS was differentiated into different cell types to check the differentiating abilities.
Neuronal epithelial spheres and NCCs positive for neuroectodermal markers such as SOX1, βIII
TUBULIN, NESTIN, and TFAP2α (Figure 2-1F), and mesenchymal cells positive for
mesendodermal markers EOMES, NKX2.5, CARDIAC ACTIN, and SOX17 (Figure 2-1G), were
successfully induced from iPSC-BFLS. In summary, our iPSC-BFLS are pluripotent stem cells
with both self-renewal and differentiation abilities into three germ layers.
The two CHARGE patients’ iPSC lines were characterized by other lab members by similar
methods and data are not included here.
2.2.2. PHF6 Is a Negative Regulator of NCC Formation

18

19
Figure 2-2: Accelerated differentiation in BFLS iPSC-NCCs. A. Schematic of the accelerated differentiation in
NCCs formation. B. Attached neuronal epithelial spheres analysis at day4. C. Immunostaining shows that wildtype
and BFLS NCCs are similar. D. SOX2 and βIII TUBULIN staining in neural epithelial spheres in NEC self-renewal
media.
To understand PHF6’s function in NCC formation in BFLS patients, iPSC-BFLS was used to
do the differentiating experiments. The NCCs induction was performed following the protocol
established by Dr. Bajpai in 2010 (Bajpai et al., 2010). Using this method, I generated the BFLS
NCCs which co-express TFAP2α and P75 (Figure 2-2C). Interestingly, the iPSC-BFLS form
NCCs (~ 4 days in NCC inducing media) much faster than the healthy control iPSCs (9 days in
NCC inducing media) (Figure 2-2A). At day 4, less than 20% of the spheres are attached in the
control cells, however, more than 50% neuronal epithelial spheres are attached with NCCs
migrating out in the BFLS cell lines (Figure 2-2B).
To test whether these defects also happen in neuroepithelial cell (NEC) spheres, the
neuroepithelial spheres were collected and changed to the NEC self-renew media. BFLS patient
cells continually differentiate NCCs and even further to glia and neurons that express neuronal
marker βIII TUBULIN, while the healthy control stays at the neuroepithelial spheres stage
(Figure 2-2D). Moreover, the BFLS NEC lost the stem cell marker SOX2 in the NEC
self-renewal media, while the wildtype control cells mainly express the SOX2 (more than 80%
are still SOX2 positive). In conclusion, BFLS NEC spheres lost the stem cell marker and
expressed the differentiation marker, which indicated that the BFLS cells could differentiate to

20
NCCs even when there was a reduction of the growth factors compared to the NCC
differentiation media.
To check that whether the NCCs lineage cells also be affected in BFLS patients, NCCs were
further induced to form other cell types such as adipocytes, osteoblast, and chondrocytes. For
chondrogenic cell induction in pellet culture, after 22 days, the cells were fixed and stained with
prechondrogenic marker Aggrecan, and chondrogenic cell marker COL2. BFLS patient cells
form larger cell pellet with no statistically significant (data not shown). However, more BFLS
patient cells (~50%) expressed Aggrecan and COL2 than control (~10%) (Figure 2-3B). For
Adipocytes, oil-O-red staining showed that patient NCCs form adipocytes (start forming oil
drops at day6) earlier than the control (no oil drops at day 6 and starts forming oil drops at day
11) (Figure 2-3C). For osteogenic cells, Ki67, a proliferating marker; Periostin, a pre-osteoblast
marker; and COL1, an osteoblast marker, have been used in combination to distinguish the
osteogenesis (Figure 2-3D,E). Majority of the BFLS cells stopped proliferating at day 7 (>95%
in BFLS, ~70% in control). Additionally, BFLS cells expressed more Periostin than control at
day7 (~6 times Periostin intend density compared to the control). Moreover, at day 11, when
almost all cells are expressing COL1 in both control and BFLS osteogenic cells, much stronger
and more organized COL1 signal has been seen in BFLS cell lines. Consistently, the alkaline
phosphatase analysis also showed more expression in BFLS osteogenic cells than control. All
these data together showed that the BFLS patient specific NCCs have an accelerated
differentiation in NCCs differentiated to multiple cell lineages (Figure 2-3A).

21

Figure 2-3: Accelerated differentiation in BFLS iPSC-NCCs’ further differentiation. A. Schematic of
accelerated differentiation of BFLS NCCs into chondrocytes, osteogenic cells, and adipocytes. B. Immunostaining
of COL2 and Aggrecan in chondrogenic pellet cultured cells at day22. C. Oil-O-Red stained the oil drops in

22
adipocytes. D. Proliferation and differentiation markers staining in osteogenic cells. E. Quantification of the
expression of differentiation markers in osteogenic cells.
2.2.3. CHD7 Functions as a Positive Regulator in NCC Formation, Migration, and
Scattering
The CHARGE and control iPSC lines are stable and proliferating after more than thirty
passages and can form normal colonies (Figure 2-4B). Previous studies indicated that diminishing
CHD7 will affect the formation of NCCs (Bajpai et al., 2010). Another paper in 2017 reported
defects in two independent CHARGE patients’ iPSC lines having defects in iPSC induced NCCs
(iPSC-NCC), including abnormal scattering of the NCCs (Okuno et al., 2017). However,
different CHARGE patients might show different symptoms, and the severity of the
abnormalities are variable from patient to patient. So, it is important to determine patient specific
phenotypes using the iPSCs to facilitate phenotyping-genotyping correlation and the basic
mechanism behind the CHARGE syndrome. Thus, although there’s already one paper published
regarding the defects in CHARGE iPSC-NCCs, it is still important to check whether our
CHARGE patient cell lines have similar defects.
I performed the iPSC-NCCs differentiation from both control and CHARGE patient iPSCs.
CHARGE iPSCs form smaller and irregularly shaped neural epithelial spheres compared to the
control cell line (Figure 2-4A, quantification in Figure 2-4B,C). Defective migration has also
been observed in CHARGE iPSCs with fewer NCCs were migrating out with shorter migration
distance (data not shown here). Moreover, the scattering of the CHARGE NCCs are not as

23
dispersed as the control (Figure 2-4A). CHARGE NCCs tend to stay adjacent to each other and
migrate as a “sheet” which is consistent with the two previous published CHARGE iPSC-NCCs.

Figure 2-4: iPSC-NCC defects in CHARGE patient cell lines. A, Two different CHARGE patient cells used in
this experiment do not have defects in iPSCs but have smaller neural epithelial spheres and less NCCs formed. B,
Quantification of the circularity of the neuronal epithelial spheres. C, Quantification of the size of the neuronal
epithelial spheres.

24
Further differentiation of CHARGE NCCs into other cell types such as adipocytes and
osteogenic cells done by Simiao Wang in Bajpai lab demonstrated that CHD7 may work as a
positive regulator to adipocytes and a negative regulator for osteogenic cells (personal
communication and data are not included here). These results indicated that CHD7 may have
tissue specificity and have different functions in different cell types.
2.2.4. CHARGE and BFLS iPSC Could Form Normal Cardiomyocytes
One of the six most common pathologies in patients with CHARGE syndrome is heart
defects. However, the symptoms are related to the great vessels, atrioventricular septal,
conotruncal, and aortic arch defects (Corsten-Janssen et al., 2013; Corsten-Janssen and Scambler,
2017). There is no evidence that cardiomyocytes are affected in CHARGE syndrome. To
compare with the CHARGE iPSC-NCC phenotypes and determine the specificity of the NCC
phenotypes, I also differentiated CHARGE iPSCs into cardiomyocytes using a published method
to differentiate iPSCs into contractile cardiomyocytes (CMs) (iPSC-CMs) (Lian et al., 2013).
Utilizing this method, I successfully generated the beating CMs using both CHARGE patient
cells and control cell lines (videos not included here).

25

Figure 2-5: Immunostaining of iPSC-CMs. a, immunostaining on WTC11 control iPSC-CMs for cTnT and
GATA4 antibodies. b, immunostaining on CHARGE patient (CHD7 p.Arg1036X) cells with cTnT and NKX2.5
markers. c-e, the iPSC wildtype control from Dr. Ruchi Bajpai was differentiated into CMs. Cells were stained by
cTnT, NKX2.5, and α-SMA. f-g, another CHARGE patient cell line (CHD7 p.Gln1701X) was differentiated into
CMs, and stained with cTnT, NKX2.5, and α-SMA markers. All scale bars are indicated 50 µm. Nucleus was
stained by DAPI in all cells which are showing in blue color (In collaboration with Dr. David Chee Ern Wong who
helped in the immunostaining).
To check whether the beating cells are really CMs. I collaborated with Dr. David Chee Ern
Wong in Lien’s lab. We stained the control iPSC-CMs with antibodies targeting the earliest
markers of the cardiac lineage NKX2.5, early cardiac markers GATA4, myofibril marker
Cardiac Troponin T (cTnT), and myofibroblast marker α-Smooth muscle actin (α-SMA). The

26
result showed that the control iPSC-CMs express NKX2.5, GATA4, cTnT, and α-SMA which
indicated that they are cardiac myocytes (Figure 2-5 a,c,e). Similar to the control, CHARGE
patient iPSC-CMs also expressed NKX2.5, cTnT, and α-SMA (Figure 2-5 b,f,h), and showed
normal striations and sarcomeres under high magnification (Figure 2-5 d,g). In summary,
CHARGE patient iPSC-CMs could contract and express the same cardiac and myofibril markers
as the control cells.
2.3. Discussion
2.3.1. CHD7 May Have Tissue Specific Functions
Our two CHARGE patient cells have defects in NCCs’ formation and migration, which is
consistent with the two published CHARGE iPSCs (Okuno et al., 2017). Although CHARGE
patients carrying different mutations in CHD7 might show different disease symptoms, the
concomitant craniofacial and cardiac symptoms suggested that the NCC defects may play a key
role in the CHARGE syndrome because it is one of the major cell types that contribute to both
craniofacial and cardiovascular. Interestingly, in our project, CHD7 p.Gln1701X forms worse
spheres and fewer NCCs compared to the CHD7 p.Arg1036X cell line. This various severity of
the defects in neural epithelial spheres and NCCs in different cell lines may be related to the
various symptoms in CHARGE patients. Altogether, NCCs may be a very important cell type to
study to understand the basic mechanism of the CHARGE syndrome.

27
A previous paper showed that their CHARGE patient iPSC-NCCs could differentiate into
adipocytes, chondrocytes, and osteogenic cells (Okuno et al., 2017). Here, my data suggested
that although the CHARGE iPSC-NCCs have the ability to differentiate into different cell types,
the differentiation speed is variable. CHD7 may be a positive regulator of adipocytes but a
negative regulator for the osteogenic cells. These results indicated that CHD7 may have a
tissue-specific function, which is also consistent with the previous chromatin
immunoprecipitation sequencing (CHIP-seq) study in mouse embryonic stem cells (Schnetz et
al., 2010). Even though CHD7 has an important function in NCCs’ formation and migration,
CHD7 may have less function in iPSC-CMs or only affect a specific type of CMs.
2.3.2. CHD7 and PHF6 Could Have Antagonistic Functions in NCC Lineage Cells
Consistent with our patient iPSC data, previous in vitro study of human ESC showed that
strong PHF6 knockdown cells (~ 90% KD) are selected against, and weak knockdown cells (30%
KD) generated NCCs sooner than control cells (day 5 compared to day 7-9 in wildtype). In
contrast, partial CHD7 knockdown (~50%KD) in ESC recapitulates the haploinsufficiency in
CHARGE syndrome and generate fewer NCCs compared to the control group. Further, strong
knockdown of CHD7 (75%KD) results in complete absence of NCC formation. Interestingly,
these defects resulted from CHD7 knockdown can be rescued by a partial decrease in PHF6 levels
(Data generated by Dr. Ruchi Bajpai, personal communication and data not shown here).

28
These data suggested that, in vitro, the functional relationship between CHD7 and PHF6 is
antagonistic at least in NCCs. Reduction of one of these proteins disrupts the balance of positive
and negative regulation of NCC formation, which can be restored by downregulating the other
protein.
2.4. Materials and Methods
CHARGE Patient iPSC Lines

Figure 2-6: Sequencing of the wildtype control, CHARGE patients’, and BFLS patients’ cell lines. A. On the
left is the CHD7 DNA sequencing of the 1036 Arg position, and on the right is the 1701 Gln position. Same for the
Protein sequencing. B. The PHF6 sequencing at 45 cysteine position. On the right is the PHF6 protein sequence.
The CHARGE patient iPSC lines were induced by Dr. Ruchi Bajpai from two different
CHARGE patients’ fibroblast cells from Dr. Joanna Wysocka at Stanford University. The patients
A
B

29
have CHD7 p.Arg1026X and CHD7 p.Gln1701X mutations, respectively (Figure 2-6). Control
lines: wildtype (WT) control (induced by Dr. Ruchi Bajpai by using the same method as the
CHARGE patient cells) and WTC11 control cell line which was purchased from Coriell Institute.
Finger Printer Test
Short tandem repeat loci are the most informative DNA genetic markers for individualizing
biological material. To check the identity of the iPSCs from the two BFLS patients, I PCR
amplified the (AAAG)n repeat region in RUNX2 locus, (CA)n repeat region in RUNX2 locus,
plus the 13 genetic markers that form the core of the FBI Laboratory’s Combined DNA Index
System, which are CSF1PO, FGA, TH01, TPOX, VWA, D3S1358, D5S818, D7S820, D8S1179,
D13S317, D16S539, D18S51, and D21S11 (Foroughmand et al., 2014) in H9 control embryonic
cells, T54 iPSCs, and three different BFLS patient cell lines (26B from one patient, 25B and 25B-2
from the other patient). After the amplification, I send the samples to USC NGS Core to do the
high sensitivity DNA assay to confirm the finger printer of each cell line.
Mycoplasma Test
Spin at 200xg for 5 mins, remove 30 µl of the supernatant, add 30 µl MycoAlert Reagent,
and wait for 5 mins. Measure the luminescence (Read A). Add 30 µl MycoAlert Substrate to the
sample and wait for 10 mins. Measure the luminescence (Read B). Calculate the amount of B/A
and the result can be found in Table 2-1 below.

30
Table 2-1: The results of the mycoplasma test
B/A Result
<0.9 Negative for mycoplasma
0.9-1.2 Quarantine cells and retest in 24 hours
>1.2 Mycoplasma contamination
iPSC Culture
The iPSCs were cultured in mTeSR medium (Stemcell technologies). Medium was changed
every day and the cells were passaged when reaching 75-85% confluency by using Accutase
(Stemcell technologies) to detach the iPSC. Freezing media (40% FBS, 20% DMSO, 20%
mTeSR) was used to freeze down the iPSCs. Transfer the cells into cryovials (Thermofisher),
then keep them in an isopropanol chamber to decrease the temperature by approximately 1 ºC per
minute in -80 ºC freezer. Move the cells in liquid nitrogen for long-term storage.
iPSC-NCC Differentiation
iPSC-NCCs induction was performed following the method described by Dr. Ruchi Bajpai
in 2010 (Bajpai et al., 2010).
NCC inducing media (500 mL) incudes 250 mL DMEM F12 (Gibco), 250 mL Neuro basal
media (Gibco), 5mL GEM supplement, 2.5 mL N2 supplement (Gibcco), 2.5 mL glutamax, 5
mL pen/strep, 0.5 ml insulin stock, 0.5 mL EGF stock, 0.5 mL FGF stock.

31
On day 0, use collagenase IV to detach the cells for 30-45 minutes until the edge of the
colonies fold back. Wash with PBS once, and then add NCC inducing media to each well and
use 2 mL pipette gently scratch the cell colonies down. Passage cells to the 10 cm dish
containing NCC inducing 8-10 mL media. Feed with NCC media every 3 days. On day1, the cell
colonies will become floating spheres. For wildtype control, the spheres will attach on the plate
around day9. Since then, the NCCs will migrate out from the attached rosettes. To passage the
NCCs, aspirate the spheres out, wash with PBS once, and then use Trypsin to detach the NCCs.
Add 2-3 mL of NCC media to stop the reaction. Centrifuge 5 minutes to spin down the cells, and
then passage the cells into new plates with fresh NCC media.
To self-renew the neuroepithelial spheres, change the floating spheres into NEC
self-renewal media (500ml media: 500 mL DMEM F12 (Gibco), 5 mL N2 supplement (Gibcco),
5 mL pen/strep, 0.5 ml insulin stock, 0.5 mL EGF stock, 0.5 mL FGF stock).
iPSC-Cardiomyocyte Differentiation
Cardiomyocytes differentiation protocol was established by Dr. Xiaojun Lian (Lian et al.,
2013).
At day -3, passage and seed 80-100k iPSCs in each well in 12-well plate, feed with mTesR1
media. Change media every day. At day 0, the cells should reach 70-85% confluency. Feed the
cells with 1 ml warm 7.5 μM CHIR in RPMI/B27(-) (Gibco) for each well. Change the with the
same media on day1. On day 2 (exactly 48 hours after CHIR treatment), feed the cells with warm

32
7.5 μM IWP 2 in RPMI/B27(-), 1ml per well. Repeat once on day 3. Ensure 48 h have elapsed
since the first IWP2 feeding, gently add 2 mL 7.5 μM IWP 2 RPMI/B27(-) to each well for two
days. On day 6, media change to RPMI/B27(+) (Gibco), 2ml per well, and keep culturing the
cardiomyocytes in RPMI/B27(+) media. At day 7, start to check the beating cells. If we could
not find any beating cells till day 15, the differentiation will be considered as failed.
Video was taken by using Zeiss Axio Observer 7 fluorescence live cell imaging microscope.
Further Differentiation of NCCs into Adipocytes, Chondrogenic Cells, Osteogenic Cells,
and Mesenchymal Cells
The methods to further differentiate NCCs into adipocytes, chondrogenic cells, and
osteogenic cells were published by Dr. Gabsang Lee (Lee et al., 2010).
Adipogenic differentiation: Grow iPSC-NCCs to 75-85% confluence. Followed by exposure
to 1 mM dexamethasone, 10 µg/ml Insulin and 0.5 mM IBMX in DMEM medium containing
10% FBS for 2 weeks.
Chondrogenic differentiation: Need very high-density cells for differentiation (1 ×10
6
cells
per cm2), pellet iPSC-NCCs for 5 min at room temperature. Take out supernatant without
disruption of pellet structure. Then pellet culture can be performed with 10 ng/ml TGFβ-3 and
200 µM l-ascorbic acid in DMEM medium containing 10% FBS for 4 weeks.
Osteogenic differentiation: Plate iPSC-NCCs at low density (1 × 10
3
cells per cm2) on
tissue culture-treated Petri dishes in the presence of 10 mM β-glycerol phosphate, 0.1 µM

33
dexamethasone and 200 µM l-ascorbic acid in DMEM medium containing 10% FBS for 3–4
weeks.
Mesenchymal differentiation: Put the NCCs into mesenchymal stem cell Media (Gibco),
culture for 14 days.
Immunostaining
Wash the cells once with PBS, and then fix them with 4% PFA/FBS for 20 minutes on ice.
Wash the cells with PBS 3 times. Permeabilize the cell with 1% Triton X/PBS at room temperature
for 10 mins. Wash the cells once with PBS and block them with 1% BSA-PBS (Blocking buffer)
for 1 hour for one hour at room temperature. Stain with primary antibody overnight at 4 ºC in
blocking buffer. Wash the cells three times with PBS. Stain with secondary antibody diluted in
PBS at room temperature for 1 hour. Wash the cells with PBS three times. Stain with DAPI for 10
min at room temperature. Then wash in PBS and keep the cells in PBS at 4 ºC. The image was
taken later by using Zeiss LSM 710 Inverted Confocal Microscope.
The List of Antibodies Used for Immunostaining:
Table 2-2: Antibodies used for immunostaining
Target gene Full name of the antibody company
SOX1 Anti-SOX1 (C-20) Gt, sc-17318 Santa Cruz Biotechnology
βIII Tubulin Anti-βIII Tubulin Ms IgG1, sc-58888 Santa Cruz Biotechnology

34
NESTIN Anti-Nestin (10c2), sc-23927 Santa Cruz Biotechnology
TFAP2α Anti-AP2α Ms IgG2b, sc-12726 Santa Cruz Biotechnology
EOMES Anti-EOMES (1A8), sc-293481 Santa Cruz Biotechnology
SOX17 Anti-SOX17, sc-17355 Santa Cruz Biotechnology
SSEA4 Anti-SSAE-4 (813-70), sc-21704 Santa Cruz Biotechnology
E CADHERIN Anti-E-cadherin (67A4) ms IgG1, sc-21791 Santa Cruz Biotechnology
OCT4 Anti-OCT4(c-10) ms, sc-5279 Santa Cruz Biotechnology
SOX2 Anti-SOX2 Rb, sc-20098 Santa Cruz Biotechnology
P75 Anti-P75 Rb, G3231 Promega
COL2 Anti-COL2A1 Santa Cruz Biotechnology
Aggrecan Anti-Aggrecan Santa Cruz Biotechnology
Ki67 Anti-Ki67, sc-15402 Santa Cruz Biotechnology
Periostin Anti-Periostin1, sc-67233 Santa Cruz Biotechnology
COL1 Anti-COL1A1, sc-99357 Santa Cruz Biotechnology
c-TnT Anti-cardiac Troponin T monoclonal antibody
(13-11)
Invitrogen
α-SMA Anti-Monoclonal Ant-actin α Smooth Mucle-Cy3 Sigma
GATA4 Anti-GATA-4, sc-9053 Santa Cruz Biotechnology
NKX2.5 Anti-Nkx2.5, sc-14033 Santa Cruz Biotechnology

35
Statistical Analysis
Sample sizes, statistical tests, and P-values are specified in the figure legends. Data were
measured by using Image J (Schneider et al., 2012). Two-tailed statistical t-tests and One-way
ANOVA followed by Dunnett’s multiple comparisons tests were performed using GraphPad
Prism version 9.0.0 for Mac OS X, GraphPad Software, San Diego, California USA,
www.graphpad.comtest. Figures were drawn by using Prism Differences were considered
statistically significant at P < 0.05.

36
3. Chapter III: Zebrafish In Vivo Disease Models of CHARGE
Syndrome
3.1. Introduction:
3.1.1. Danio Rerio – In Vivo Animal Model
Zebrafish, Danio Rerio, is a member in Minnow family of the order Cypriniformes, which has
been suggested to use as a model organism by Dr. George Streisinger around 1981 (Li et al., 2013).
Due to the obvious advantages of zebrafish, such as rapid developmental period (72 hours to hatch
to free-swimming larvae), short generation time (around 3 months), a larger number of offspring,
external fertilization, transparent embryos (Teame et al., 2019), and strong regenerative capacity
(Goldshmit et al., 2012; Jopling et al., 2010), this tropical freshwater fish has become a unique
model especially for developmental biology and regeneration biology. The zebrafish whole
genome-sequencing project was initiated in 2001. The results show that 71.4% of human genes
have at least one zebrafish orthologue (Howe et al., 2013), not to mention that many important
genes are very conserved through fish to human. Furthermore, the genome of zebrafish can be
easily manipulated using reverse genetic technologies (Zu et al., 2013). Taking these advantages, I
chose to use zebrafish as my in vivo disease model in this project.
The external fertilization of zebrafish makes it easier to study the early developmental
process and all stages are well-documented (Kimmel et al., 1995). After fertilization, there is only

37
one cell adjacent to the yolk in the embryo (one-cell stage). 30 minutes after fertilization, the
single cell divides into two cells which continue to divide approximately every 15 minutes.
Around 3 hours post-fertilization (hpf), there are around 1,000 cells with 4 germ cells. Genetic
changes in these germ cells will be inherited by a large number of progenies. The cells then
migrate down along the yolk to form the epiboly. At 10 hours, the cells progress to the bud stage,
and somites are starting to emerge with an average speed of two somites per hour. Zebrafish larvae
have 30-somites at 24 hpf and go out of the oolemma at 72 hpf (Kimmel et al., 1995). The yolk
provides nutrition for the larvae and therefore shrinks in the meanwhile. After 5 days
post-fertilization (dpf), the mouth of the zebrafish is open to eat small food, such as paramecia.
Adult fish should reach reproductive maturity after three months.
3.1.2. Neural Crest and Craniofacial Development in Zebrafish
At the end of gastrulation, the ectoderm is subdivided into two parts: non-neural ectoderm and
neural ectoderm. Later, during the neurulation process, NCCs are formed within the dorsal-most
part of the neural ectoderm (also known as neuroepithelium) at the junction with the non-neural
ectoderm, a region called the “neural plate border” (Bhatt et al., 2013). In zebrafish, NCCs start to
migrate out of the neural tube as early as 3-somites and then begin to migrate in the cranial region
around 10-somites. Noticeably, the lateral positioned cells migrate out from the neural tube to
form the NCCs first, with only the later NCCs deriving from the dorsal-most aspect (Rocha et al.,
2020).

38
Internal gene expressions, such as transcriptional factors zic3, pax3a, dlx5, msx1/2, and tfap2a
are specific and broadly expressed at the neural plate border and have been shown to be needed for
the expression of bona fide neural crest markers (Rocha et al., 2020). Slug is one of the very first
steps in the delamination of NCCs. Following Slug, snail1 and snail2 can directly repress the cell
adhesion molecule E-cadherin by binding to its promoter, which is thought to facilitate NCC
migration. Other NCCs markers including foxd3, dlx2a, crestin, and sox10 then express in
pre-migrated and post-migrated NCCs, and Sox9 as an NCC-specific marker is the key regulator of
chondrogenic fate.
NCCs migrate as PAs and further differentiate to form other organs, including the craniofacial
cartilage and heart. After migration, a subset of NCCs from the midbrain migrate anteriorly to
contribute to the neurocranium, which provides support for the brain and sensory systems.
Whereas, posterior midbrain and hindbrain-derived NCCs migrate ventrally into the PAs and
produce the viscerocranium (Knight and Schilling, 2006), which serves as the feeding and
respiratory apparatus in zebrafish. The neurocranium and viscerocranium together form the
craniofacial cartilage at 5 dpf (Mork and Crump, 2015). On the other side, two distinct populations
of NCCs contribute to the zebrafish heart: one is the PA 1 and 2, which integrates into the heart
tube and adopts a myocardial fate; the other one is the PA 6 which contributes to the cells
surrounding the ventral aorta and invades the bulbus arteriosus (Cavanaugh et al., 2015).
Overall, the NCCs PAs were well studied in zebrafish, making zebrafish a good model to
understand the early development NCC derived craniofacial and cardiac tissues.

39
3.1.3. Inner Ear Development in Zebrafish
Patients with CHARGE syndrome also have defects in the inner ear. Zebrafish do not have
cochlea, that transduce sound in mammals. They depend on the otolith organs for balance and
hearing. These otolith organs are made of sensory hair cells, support cells, and some
biomineralized structures called otoliths (Fairall et al., 1993). Otoliths are required for the
sensation of gravity, linear acceleration, and sound in the zebrafish inner ear. Adult zebrafish have
three otolith organs in each ear: the sagitta of the saccule, the lapillus of the utricle and the
astericus of the lagena.
At 24 hpf, zebrafish have two hair-cell packed sensory maculae, each attached to an otolith
(Stooke-Vaughan et al., 2015). The anterior group develops into the utricle having roles in
equilibrium differentially (Inoue et al., 2013), whereas the posterior group becomes the saccule
contributing to hearing (Yao et al., 2016).
Instead of the NCCs, otoliths formation has been shown to be related to the placode in
zebrafish (Baxendale, 2014). Placodes arise from ectoderm at the anterior neural plate border, that
are discrete patches of thickened ectoderm (Whitlock and Westerfield, 2000). Like the neural
crest, cranial ectodermal placodes appear transiently in the head of all vertebrate embryos and give
rise to a very diverse array of cell types including sensory receptors, sensory neurons, supporting
cells, secretory cells, glia, neuroendocrine, and endocrine cells (Baker, 2005).

40
Placodes give rise to the bulk of the peripheral sensory nervous system in the head including
the inner ear. At the 18-somite stage, right after the otic placode cavitates to form the otic vesicle
(OV), dense, sticky precursor particles start to distribute throughout the OV lumen. These
precursor particles then aggregate at the anterior and posterior OV poles to form two otoliths. At
around the 26 somite stage, otoliths have been biomineralized by the crystallization of calcium
carbonate and stabilized by a protein and glycoprotein matrix (Baxendale, 2014).
3.1.4. Reverse Genetic Tools Used to Study Candidate Genes in Zebrafish
In addition to forward genetics that have been used as an unbiased screen to identify genes
involved in different biological processes, reverse genetic techniques allow us to knockdown or
mutate genes of interest. We can use these tools to investigate the functions of genes and establish
human genetic disease models.
In 1985, the site-specific DNA-binding properties of the zinc-finger proteins have been
discovered (Diakun et al., 1986). Each zinc-finger contains approximately 30 amino acids and
recognizes a specific three base pairs DNA sequence (Fairall et al., 1993). Scientists combined the
zinc-fingers and the FokI type II restriction endonuclease to form the zinc-finger nucleases (ZFNs)
to create double strand break (DSB) at the target site (Guo et al., 2010). Taking advantage of the
endogenous DNA repair machinery, small insertions or deletions (indels) will be generated at the
DSB. However, ZFN is difficult to synthesize, and the low affinity and high off-target effects
make it less popular than other new reverse genetic technologies.

41
Transcription activator-like (TAL) effectors were originally discovered in Xanthomonas
bacteria and are part of the mechanism used to infect plants (Boch and Bonas, 2010). TAL effector
unit is highly conserved. It has 33 to 35 amino acids with changes only in the 12
th
and 13
th
to
recognize different DNA nucleotides (Bogdanove and Voytas, 2011). TALEN (TAL Effector
Nucleases) is made by fusing a series of TAL effector DNA-binding units to Functional
endonuclease FokI (Li et al., 2011). Due to the one-to-one correspondence between TAL effector
units and contiguous nucleotides in the target site, TALEN is much easier to construct and has
better specificity and efficiency compared to ZFN.
Soon after TALEN, CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats),
one of the most important techniques so far in 21st-century biology has been discovered. Unlike
the previous protein-based DNA-binding methods, CRISPR recognizes and cuts a target gene in
an RNA-guided DNA cleavage in eukaryotic organisms (Marraffini and Sontheimer, 2010).
Among all CRISPR systems that have been reported, the type II CRISPR/Cas9 system depends on
a single Cas protein together with a referencing RNA and a structural RNA to target a particular
DNA sequence (Jinek et al., 2012). In 2013, the CRISPR/Cas 9 system in gene-editing of
mammalian cells has been published (Cong et al., 2013). Ever since then, CRISPR has become the
most popular reverse genetic technology.
Theoretically, CRISPR can target any sequence which fits in a 5’-GN-N(18-22)-NGG-3’
structure. The three bases of “NGG” are called the PAM sequence, and the N(18-22) is the
recognizing sequence. This type of sequence occurs randomly, approximately once every 32bp in

42
zebrafish (Blackburn et al., 2013). Since only one DNA strand is needed for recognition by
CRISPR/Cas9, the negative strand can also be targeted.
In many laboratories, researchers generate guide RNA (gRNA) that contains both the
referencing RNA and structural RNA to simplify the CRISPR/Cas9 system (Hwang et al., 2013).
To maximize the efficiency of generating insertions or deletions (Indels) to create biallelic
mutations, a previous paper described a method by co-injecting 4 gRNAs targeting the same gene
for a rapid-directed gene knockout in injected generation zero (G0) zebrafish (Wu et al., 2018). In
this project, I used the CRISPR/Cas9 system by co-injecting gRNAs together with Cas9 protein
into single-cell stage zebrafish embryos to generate chd7 and phf6 mutants.
3.1.5. Previous Chd7 Study in Zebrafish
The in situ hybridization experiments have shown that chd7 transcript is broadly expressed
in developing zebrafish, mainly in the central nervous system, gut, tail bud, and somite borders
(Jacobs-McDaniels and Albertson, 2011). At 24 hpf, chd7 is expressed in rhombomeres, where
some neural crest cells are derived; and at 48 to 96 hpf, chd7 was strongly expressed in
developing pharynx (Liu et al., 2018). In previous studies, morpholino (MO) was used to
diminish chd7 expression in zebrafish. High doses of chd7-MO (5–10 ng) induced lethality
within 24 hpf and low-doses chd7-MO (2.5 ng) injected fish showed phenotypes such as eye
abnormalities, otolith abnormalities, craniofacial defects (Asad et al., 2016; Patten et al., 2012),
heart defects, and pectoral fin hypoplasia (Balasubramanian et al., 2014). None of these

43
phenotypes can be rescued by co-injection of chd7 and p53 morpholinos (Balow et al., 2013;
Patten et al., 2012), indicating that these phenotypes are not caused by random cell apoptosis.
Due to the high off-target rate of the MO, several chd7 mutants were generated by different
groups to confirm. In human, loss of one allele of CHD7 can cause CHARGE syndrome.
However, in zebrafish, chd7 heterozygous fish do not have an obvious phenotype. By now, three
chd7 mutants have been generated and reported. The first is a 23bp deletion frameshift mutant.
The heterozygous mutants exhibit anxious-like and aggressive-like behavior (Liu and Liu, 2020).
The sox10‐positive cranial neural crest cells were scattered (Liu et al., 2019) and T-Cell and
thymus development are impaired in chd7 homozygous mutants (Liu et al., 2018). The second
mutant is a 2bp frameshift homozygous mutation with reported smaller eyes, enlarged heart with
edema, and failure to inflate the swim bladder phenotypes (Prykhozhij et al., 2017). The size of
the gastrointestinal tract also has been shown to decrease in this mutant (Cloney et al., 2018).
The last one has a 1bp insertion and therefore causes a frame-shift mutation. This chd7
homozygous mutant zebrafish larvae also displayed a CHARGE-like small head phenotype,
cranial cartilage malformations, and cranial nerve defects (Jamadagni et al., 2021). Moreover,
the chd7 homozygous mutant exhibited a low frequency of pericardial edema (20 %). More than
one group has shown that although the heterozygous fish are viable, the survival rate of the
homozygous larvae sharply dropped after 10 dpf.
I generated two new chd7 CRISPR mutants to determine whether potential CHARGE
syndrome phenotypes can be fully recapitulated in zebrafish since no cardiovascular phenotypes

44
have been reported for other chd7 mutants except for edema. chd7 mutant embryos show
decreased survival compared to sibling controls and display mild craniofacial phenotypes. Since
zebrafish have only one atrium, one ventricle, and no aortic-pulmonary septation, I do not expect
to observe phenotypes like atrioventricular septal defect in patients. I will focus my analyses on
ventral aorta for the cardiovascular phenotypes since cardiac NCCs have been shown to
contribute to the smooth muscle of the ventral aorta (Cavanaugh et al., 2015).
3.2. Results
3.2.1. Co-Injecting 4 gRNAs With Cas 9 Protein Increased the Mutant Rate in
Generating chd7 Mutations
CRISPR/Cas9 system was used to generate the chd7 mutations by co-injecting Cas9,
instead of Cas9 mRNA, with gRNAs to increase the mutant rate (Figure 3-1).
For targeting the 12
th
exon on chd7, I co-injected gRNA9 and gRNA10 with Cas 9 protein into
single-cell fish embryos and get around a 25% mutant rate (Figure 3-2B). Dr. Haipeng Bai in
Lien’s lab slightly modified the previous published rapid screening method (Wu et al., 2018); we
designed gRNAs targeting the 5
th
exon of chd7 to validate this method and as a proof of principle
to determine if this method can be used to generate CRISPR mutants to determine adult
phenotypes. All gRNAs were injected individually together with Cas9 protein, and T7
Endonuclease I (T7E1) assay then used to determine the efficiency in generating indels. Among

45
all the gRNAs targeting chd7 exon 5, gRNA2, gRNA4, gRNA5, and gRNA6 created mutations most
efficiently (Figure 3-2A).

Figure 3-1: Generation of the chd7 zebrafish mutations. 1. Using the specific gRNA primers, which contain T7
promoter-the target sequence-RNA scaffold forward primer, and a reverse RNA scaffold primer to do the PCR on
the RNA scaffold vector. 2. T7 in vitro transcription to generate the gRNAs. Purification steps need to be done for
the gRNAs. 3. Co-inject the gRNAs together with CRISPR Cas9 protein into one cell-stage zebrafish embryos. Cas9
protein and gRNA will bind together to create double strand DNAs and therefore generate mutations in the target
area. The gRNA targeting sequences are shown on the right (The PAM area is recognized by Cas9 protein but not
included in the targeting sequence). Schematics created using BioRender.
Therefore, I co-injected the gRNA2, gRNA4, gRNA5, and gRNA6 together with Cas9 protein
into zebrafish, and reached the highest mutant rate (Figure 3-2 A). In 24 4 gRNAs co-injected fish,
23 of them contained mutation bands. Quantification of the ratio of the intensity of the chd7 cut

46
mutant band to uncut wildtype band in the T7E1 assay shows the efficiency of the different gRNAs
and Cas9 injection combinations, respectively (Figure 3-2B). Overall, the four gRNAs
co-injection dramatically increased the mutant rate by nearly three folds.

Figure 3-2: T7E1 assay to test gRNAs targeting chd7. A, electrophoresis images show the wildtype uncut band
and mutant cut bands in each gRNA and Cas9 protein co-inject groups. B, Quantification of efficiency of different
chd7 gRNAs in creating Indels. The efficiency is estimated as the percentage of cut mutant bands/WT uncut bands.
The intensity of the bands was measured by imageJ. A standard curve was generated by measuring the overlap
intensity of two bands with the same grayscale. The total intensity of two mutant bands was calculated using the
mean intensity of the mutant bands and the standard curve. The efficiency of each injection group was calculated by
the percentage of mutant total intensity in the sum intensity of both mutant and wildtype bands.
3.2.2. The chd7(d55) Mutant Fish Has Weaker CHARGE Like Phenotypes
After outcrossing the injected G0 mosaic fish with wildtype fish and then followed by
incrossing the heterozygous offspring, I successfully generated two stable chd7 mutant zebrafish
lines (Figure 3-3A). The first mutant line is chd7(d55), which contains a 55-base pair (bp)

47
deletion in exon 5. After translating to amino acids, the mutant Chd7 protein has a frame-shift
mutation that starts at the 803
rd
Lysine in the first Chromo domain and creates a stop codon after
45 amino acids.
Survival rate analysis of the offspring of chd7(d55) heterozygous fish [chd7(d55)+/-] shows
that surviving chd7(d55) homozygous mutants [chd7(d55)-/-] are significantly less than 25%
compared to expected Mendelian ratio. The number of chd7(d55)-/- larvae dramatically
decreased after 10 dpf (~90% died); while the chd7(d55)+/- and wildtype offspring are
decreased gradually (~15% died till 30 dpf) (Figure 3-5B). The survival rate experiment has been
repeated once by separating the WT, chd7(d55)+/-, and chd7(d55)-/- into different tanks (from
different parent fish) to avoid the competition of the homozygous mutant fish with siblings. The
repeated results also showed a similar survival curve (data not shown here).
Gene expressions analysis of chd7 by RT-PCR at 4 dpf indicated that while the
chd7(d55)+/- larvae do not show a significant change in chd7 mRNA (figure 3-3C), the
chd7(d55)-/- larvae display dramatically decreased chd7 mRNA levels. This result suggests that
mutant chd7 mRNA is degraded in cells, likely due to the nonsense mediated decay and this
could trigger genetic compensation (El-Brolosy et al., 2019). Thus, chd7(d55)-/- line might
display weaker phenotypes and does not fully recapitulate the CHARGE disease phenotype. This
is consistent with the three published chd7 fish lines that only milder small head, small eye, and
cardiac edema have been reported (Jamadagni et al., 2021; Liu et al., 2018; Prykhozhij et al.,
2017) but with no significant cartilage defects shown in the morphants.

48
Mutations in the ATPase domain have a dominant-negative effect not by affecting the
activity of CHD7, but instead, by disrupting important protein-protein interactions in human
patients (Bouazoune and Kingston, 2012). To create mutant fish with stronger phenotypes, I
generated a chd7 mutant targeting the ATPase domain (Figure 3-3C). The chd7(d6) mutant has
an in-frame deletion that removed the 1082
nd
Tryptophan and 1083
rd
Threonine of the DEXDc
domain in the ATPase binding domain in Chd7 (Figure 3-3A). We will continue to determine if
the chd7(d6) allele shows stronger CHARGE-like phenotype.
3.2.3. The chd7(d55) Allele Might Display Maternal Zygotic Phenotypes.

49
Figure 3-3: Generation and preliminary characterizations of two chd7 mutant alleles. A. Schematics of Chd7
protein and chd7 gene structure. Chd7(d55) has 55 bo deletion in the fifth exon of chd7 which generated a truncation
after the first chromo domain. The deletion created a 44 amino acids frame-shift followed by a stop codon. Chd7(d6)
deletion has a mutation in the twelfth exon, which is in DEXDc domain, which causes a deletion in tryptophan and
threonine amino acids. B. Survival rates analyze the offspring of chd7(d55)+/- incross. The embryos were separated
into three groups randomly on day 0, each group has more than 50 embryos. Genotyping was done on day7, day14,
and day30. The survival rate has also been checked in the non-sibling WT, chd7(d55)+/-, chd7(d55)-/- fish and the
survival rate is similar. C. chd7 expression at 4 dpf from chd7(d55)+/- incross. The fold of chd7 expression compared
to the WT was shown (Collaborated with Zhiyu Tian who helped with the chd7(d6) RT-PCR). D. Phenotype of
chd7(d55)-/- fish. Chd7(d55)-/- (m-/-) fish have smaller swim bladder, pericardial edema, and small eyes. The
quantification showed that nearly 40% chd7(d55)-/- (m-/-) fish have heart edema, however, the chd7(d55)-/- (m+/-)
fish did not have heart edema. ns: p ≥ 0.05, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001
Interestingly, in addition to the possible genetic compensation, I also discovered that
maternal transcripts might help contribute to weak chd7 mutant phenotypes. When comparing
the pericardiac edema in both wildtype (n=879), chd7(d55)-/- fish from heterozygous mothers
(n=747) [chd7(d55)-/- (m+/-)], and chd7(d55)-/- fish from homozygous mother (n=751)
[chd7(d55)-/- (m-/-)], the chd7(d55)-/- (m-/-) larvae have significantly more fish with heart
edema (~30%) (Figure 3-3D). This data suggested that the maternal chd7 mRNA may help
chd7(d55)-/- mutant survive the early development and therefore the chd7(d55)-/- (m+/-) fish
have a mild phenotype. This might also account for the mild phenotypes for previously published
fish lines (Jamadagni et al., 2021; Liu and Liu, 2020; Prykhozhij et al., 2017).

50
3.2.4. The chd7(d55) -/- Mutants Show CHARGE-like Phenotypes in Craniofacial
Structure.
The structure of neurocranium cartilage in chd7(d55) fish do not have dramatic defects
compared to the wildtype (Figure 3-4A, A1). Patients with CHARGE syndrome have defects in
the inner ear. One of the diagnoses of CHARGE syndrome is to check the rudimentary
semicircular canals or an absence of the canals. The chd7(d55)-/- larvae from both homozygous
and heterozygous mothers (n=22 for each group) have significantly reduced size of otolith in the
utricles, although the chd7(d55)-/- (m-/-) have more severe phenotypes (Figure 3-4B). On the
other hand, the size of otolith in the saccule, which contributes to perceiving sound, is not
affected in the chd7(d55)-/- (m+/-) fish, but slightly reduced in the chd7(d55)-/- (m-/-) fish
(Figure 3-4C). These results indicated that chd7(d55) mutant fish might have CHARGE-like
inner ear abnormalities which have been reported in some Chd7 mouse models but not in
zebrafish models yet.
Many patients with CHARGE syndrome have a typical CHARGE face, which is
asymmetric with a broad prominent forehead, a prominent nasal bridge with square root,
prominent nasal columella, flat midface, small mouth, and occasional small chin. The
chd7(d55)-/- fish have mild defects in the viscerocranium cartilage structure (Figure 3-4D, D1).
The most obvious phenotype is the small eyes, which also have been reported in previously
reported chd7 mutant fish (Prykhozhij et al., 2017). My chd7(d55)-/- (m-/-) fish have smaller

51
eyes comparing to the wildtype (n=22 for each group), but the chd7(d55)-/- (m+/-) fish do not
show this phenotype (n=22) (Figure 3-4E). Using alcian blue staining to visualize the
viscerocranium cartilages, I found that the IV ceratobranchial cartilage is affected the most.
Quantification of the size of the IV ceratobranchial shows that the chd7(d55)-/- fish have shorter
IV ceratobranchial regardless of from homozygous (n=22) or heterozygous mothers (n=22)
(Figure 3-4F). These reduced chondrocytes in the IV ceratobranchial could be related to the
decreasing of the 6
th
NCC pharyngeal arches.

Figure 3-4: The craniofacial phenotypes characterization in chd7(d55) mutants. A. Neurocranium morphology
and the structure of the otoliths. Neurocraniums were dissected out from the alcian blue and alizarin red co-stained

52
WT, chd7(d55)-/- (m+/-), chd7(d55)-/- (m-/-) larvae in 7 dpf. No significant difference was found in the neurocranium
structure. A1. The schematic of neurocranium cartilage. ep, ethmoid plate; tr, trabecula; pch, parachordal; not,
notochord. (Kimmel et al., 2001)B, C. Data quantification of the size of otoliths in the utricle and saccule in WT,
chd7(d55)-/- (m+/-), chd7(d55)-/- (m-/-). Both chd7(d55)-/- (m+/-), chd7(d55)-/- (m-/-) have smaller utricle; however,
only chd7(d55)-/- (m-/-) have significant reduced saccule. D. Viscerocranium morphology. Viscerocraniums were
dissected out from the alcian blue and alizarin red co-stained WT, chd7(d55)-/- (m+/-), chd7(d55)-/- (m-/-) with or
without heart edema at 7 dpf. The structure of viscerocranium was mildly affected in chd7-/- without heart edema, and
severly affected in the mutants with heart edema. The IV ceratobranchial cartilage was reduced in the chd7-/-. D1. The
schematic of viscerocranium cartilage. bb, basibranchial; bh, basihyal; cb, ceratobranchial; ch, ceratohyal; hb,
hypobranchial; hs, hyosymplectic; ih, interhyal; m, Meckel’s; pq, palatoquadrate. (Kimmel et al., 2001) E. The
diameter of the eye. The longest diameter of the dissected eye was measured. The eye size of the chd7(d55)-/- (m-/-)
has significantly reduced. F. The size of the IV ceratobranchial. The outline of the IV ceratobranchial was drawn in
imageJ and the size was then measured. Chd7-/- mutants have reduced IV ceratobranchial regardless of the maternal
they have. ns: p ≥ 0.05, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001

Figure 3-5: The chondrocytes’ and teeth phenotypes characterization in chd7(d55) mutants. A. High
magnification image shows that the arrangement of chondrocytes was disrupted in chd7-/- mutants, especially in the
one with homozygous maternal. B. Schematic of chondrocytes’ angle measurement. C. Angles of three neighbor
chondrocytes in the I and II ceratobranchials. Each number indicated a different fish. The significance was calculated
between different groups. chd7(d55)-/- (m+/-) have milder disorganization in the chondrocytes arrangement; but

53
chd7(d55)-/- (m-/-) have more severe defects in the chondrocytes. D. Alizarin red stained teeth in 7 dpf fish and the
defects in the number of teeth in chd7-/- mutants. More than 22 fish in each group. Nearly 50% chd7-/- from
homozygous maternal fish have defects in teeth number. ns: p ≥ 0.05, *: p<0.05, **: p<0.01, ***: p<0.001, ****:
p<0.0001
Interestingly, when looking into the chondrocytes at higher magnification, the arrangements
of the chondrocytes are disorganized in chd7(d55) mutants (Figure 3-5A). I measured the angles
between the three neighbor chondrocytes (Figure 3-5B) in the I and II ceratobranchials (n=6 in
each group). The results indicated that chd7(d55) mutants have abnormal chondrocytes
arrangement (Figure 3-5C). Moreover, the chd7 maternal zygotic defects in early NCCs might
contribute to the later chondrocyte phenotypes, since the chd7(d55)-/- (m-/-) fish have
significantly severe defects in chondrocytes arrangement compared to the chd7(d55)-/- (m+/-)
fish (Figure 3-5C).
Like other vertebrates, the teeth of zebrafish arise by epithelial-mesenchymal interactions
(Verstraeten et al., 2010). chd7(d55)-/- (m+/-) fish do not have a significantly reduced number of
teeth, however, the chd7(d55)-/- (m-/-) fish have a large variation in the number of teeth (Figure
3-5D).
Overall, the chd7(d55) mutants have abnormalities in chondrocytes and osteocytes.
Moreover, some of these defects are more severe in the homozygous fish from homozygous
mothers. These results indicated that the early NCCs that are affected by the maternal chd7
mRNA may contribute to the later craniofacial cartilage, eye, and teeth formation.

54
3.2.5. In Zebrafish, NCCs Contribute to the Ventral Aorta.
Previous research has shown that in zebrafish, the PA 6 contributes to the ventral aorta
(Rocha et al., 2020). Some of the patients with CHD7 mutations also have defects in the OFT
and aortic arch (Corsten-Janssen et al., 2013). In zebrafish specifically, OFT developed into a
specific organ called bulbus arteriosus. As shown in the schematic of the vasculature and heart at 4
dpf (Figure 3-7C), OFT is the region connecting the heart and the ventral aorta. To examine the
NCCs derived cells via genetic lineage tracing, I first used the sox10:GALA-UAS-Cre, ubi:switch
(sox10:switch for short) double transgenic fish (Cavanaugh et al., 2015) from Dr. Jau-Nian Chen’s
lab (UCLA). Surprisingly, I did not observe NCCs in OFT in adult fish heart (Fig. 3-6). In contrast
to what is known for mammals and predicted in zebrafish (Rocha et al., 2020). To confirm this
result, I used two other sox10 lineage tracing lines from Dr. Gage Crump’s lab: Sox10:Cre/
bact:blue>Red [Tg(Mmu.Sox10-Mmu.Fos:Cre)
zf384
(Kague et al., 2012),
Tg(actab2:loxP-BFP-STOP-loxP-dsRed)
sd27
(Kobayashi et al., 2014) ] and sox10:Cre/
ubi:stop>Red [Tg(Mmu.Sox10-Mmu.Fos:Cre)
zf384
(Kague et al., 2012),
Tg(-3.5ubb:loxP-STOP-loxP-mCherry)
el818
(Fabian et al., 2020)]. Confocal images shown that
consistent sox10 lineage traced cells were found in the ventral aorta but not in bulbus arteriosus
(BA) using all these three sox10:switch lines (n>15) in adult hearts (Figure 3-6Ai, Aii, the
images for the sox10:Cre/ubi:stop>Red are not included here).

55
I went back to check the larvae using sox10 lineage tracing. In 4 dpf, sox10:Cre/
ubi:stop>Red larvae, specific fluorescence labeled cells were observed in the ventral aorta rather
than the OFT (Figure 3-6B). Thus, although no evidence showed that NCCs contribute to the
OFT in zebrafish, my results showed significant contribution of NCCs to the ventral aorta
consistent with the reported data (Cavanaugh et al., 2015).

Figure 3-6: sox10:Cre lineage traced cells are observed in ventral aorta. Ai. Wildtype sox10:GALA-UAS-Cre,
ubi:switch, cmlc2:nucGFP adult fish had expressions in ventral aorta as well as cardiomyocytes in ventricle and
atrium. Aii. Sox10:Cre/bact:blue>Red fish also have expressions in the ventral aorta, but not very consistent in the

56
ventricle and atrium in old fish. B. sox10:Cre/ubi:stop>Red fish had expressions in the cells in the ventral aorta at 4
dpf. OFT, outflow tract; V, ventricle; A, artium; *: ventral aorta.
3.2.6. The chd7(d55)-/- Fish Have Defects in Ventral Aorta.

57
Figure 3-7: The phenotypes of chd7(d55) mutants in the cardiovascular. A. Analysis of the width of the OFT at
day4 (102 hpf). B. Analysis of the length of the ventral aorta at day4 (102 hpf). C. Schematic of ventral view of the
zebrafish vasculature system at day4. AA, aortic arch; AA1, mandibular arch; AA2, hyoid arch; AA3, first branchial
arch; AA4, second branchial arch (Serrano et al., 2019); HA, hypobranchial artery; VA, ventral aorta, OFT, outflow
tract; V: ventricle; A: atrium. D. 20x ventral view of the kdrl:mTurquois fish at 102 hpf. E. 20x ventral view of the
kdrl:mTurquois fish at 78 hpf and 126 hpf. Fi. The elongation tendency of the ventral aorta from 78 hpf to 126 hpf.
Fii. The changing of the width of the OFT from 78 hpf to 126 hpf. White dash line, the ventral aorta; yellow dash
line, the OFT; white brace, the length of the ventral aorta; yellow brace, the width of the OFT.
To check whether there are defects in ventral aorta and OFT morphology, I used the
kdrl:Turquoise reporter line at 102 hpf in WT (n=34), chd7(d55)+/- (n=19), chd7(d55)-/- (m+/-)
(n=15), and chd7(d55)-/- (m-/-) (n=7) fish (Figure 3-7A). Kdrl is a vascular endothelial growth
factor receptor-like gene, which is expressed in the endothelial cells and therefore could be used
to observe the great vessel morphology (Harrison et al., 2019). I measured both the length of the
ventral aorta and the width of the OFT region by the thickest part (labeled in Figure 3-7E). The
results showed that only the chd7(d55)-/- (m-/-) significantly reduced the width of the endothelial
part in OFT (Figure 3-7A). We also examined OFT smooth muscle cells using the pdgfrb:GFP
reporter but did not find obvious differences between chd7 chd7(d55)-/- (m-/-) mutants and
control at 126 hpf (Figure 3-8). However, both chd7(d55)-/- (m+/-) and chd7(d55)-/- (m-/-)
mutants dramatically reduced the length of ventral aorta, even though the chd7(d55)-/- (m-/-) had
more severe phenotype (Figure 3-7B). Noticeably, the whole vascular vessels are thinner in the
chd7(d55)-/- mutants compared to the wildtype and heterozygous.
To further determine whether this reduced length of the ventral aorta is caused by
developmental delay, I also checked the wildtype (n=26 at 78 hpf, n=18 at 126 hpf),

58
chd7(d55)+/- (n=11 at 78 hpf, n=10 at 126 hpf), and chd7(d55)-/- (m+/-) (n=9 at 78 hpf, n=5 at
126 hpf) at 78 hpf and 126 hpf, respectively (Figure 3-7E). I determined the change of the length
of the ventral aorta (Figure 3-7Fi). Despite that both the wildtype and chd7(d55) larvae are
growing, the increasing of length of the chd7(d55) (m+/-)’s ventral aorta is slower than the
wildtype and heterozygous. Therefore, the reduction in the aorta is not simply caused by the
developmental delay. Moreover, a later time point at 10 dpf has also been checked in several fish
[1 WT fish, 2 chd7(d55)+/- fish, and 2 chd7(d55)-/- (m+/-) fish], and the vasculature is similar to
the length at 126 hpf and the chd7(d55)-/- (m+/-) did not catch up forming normal vasculature
structure (data not shown here). Interestingly, a narrowed kdrl positive endothelial cells in the
OFT has been obvious in chd7(d55)+/- mutants at 126 hpf (Figure 3-7 Fii), the implication and
consequence of this abnormality is still not known but might relate to the haploinsufficiency of
Chd7 in fish.

Figure 3-8: The high magnification images showed the
pdgfrb positive cells in the OFT did not have obvious
defects in chd7(d55) mutants at 102 hpf. The
pdgfrb:GFP line is used to track the pdgfrb positive cells.
Matenta cells are the pdgfrb positive smooth muscle cells
around the cyan kdrl positive endothelial cells in the OFT
region. No obvious differences had been found in these
pdgfrb positive smooth muscle cells.

59
3.3. Discussion
3.3.1. The Weaker Phenotype in chd7 Homozygous Fish
I compared my chd7(d55) mutants to the chd7 morphants published before. The chd7
morpholino I injected before in my master thesis project caused a severe defect in the sox10 (a
post migrated NCCs marker) positive craniofacial cartilage and caused pericardial edema (Asad
et al., 2016). However, in our chd7 mutants, except for the chd7(d55)-/- (m-/-) fish that
developed the heart edema slowly starting around 3 dpf and expanded till 6fpf with even larger
edema and destroyed cartilage structure, all other mutants, including the chd7(d55)+/-,
chd7(d55)-/- (m+/-), and the chd7(d55)-/- (m-/-) without edema, have much weaker cartilage
phenotype. In contrast, Dr. Zhi-Zhi Liu’s paper showed that chd7 MO and the offspring of
chd7+/- have a reduction in pharyngeal arches (Liu et al., 2018), which has not been found in our
sox10:NTR mcherry reporter line at 33-40 hpf in chd7(d55)+/- and chd7(d55)-/- (m+/-) mutants.
The first possible reason of this milder chd7 mutants phenotype compare to the morphants is
the off-target and toxicity-related effects of MO (Cunningham et al., 2020). The MO-induced
non-specific cell death could exaggerate the knockdown phenotype, and that is why we would
like to confirm the CHARGE-like fish phenotype by generating chd7 mutants. The second
possible reason is the genetic compensation induced by the nonsense mediated decay of chd7
RNA. To overcome this possibility, I also generated the second chd7(d6) mutant line that carries

60
an inframe 6 bp deletion without RNA degradation in the functional ATPase domain of Chd7
protein. The craniofacial cartilage phenotype will be checked soon in this fish line to see whether
without the genetic compensation we will get a stronger phenotype in chd7 mutant zebrafish.
The third possibility is the maternal contribution, which I have mentioned previously. My results
indicated that chd7 had a maternal zygotic phenotype in zebrafish, therefore in the future, to
study the function of chd7 in early development, we should use the chd7 homozygous mutants
from the homozygous mothers to fully eliminate the chd7 mRNA.
3.3.2. Cardiac NCCs Lineage in Zebrafish Heart
In mammals, NCCs have been shown to migrate into the great vessels and contribute to the
OFT septation (Bradshaw et al., 2009). However, in contrast to mammals, we did not see OFT
contribution in the sox10 lineage traced fish. One possible reason is fish hearts do not have septum
in OFT. Instead, NCCs are reported to contribute to ventricular cardiomyocytes and ventral aorta
in zebrafish (Cavanaugh et al., 2015). Whether Chd7 regulates the NCCs derived cardiomyocytes
remains unknown. The contribution of NCCs in ventral aorta is likely the smooth muscle cells. In
our chd7 fish mutants, thinner vasculature has also been observed, which may be related to the
shorter ventral aorta phenotype. We will continue to explore if this phenotype in ventral aorta is
related to NCCs derived smooth muscle cells or caused by chd7’s cell autonomous in endothelial
cells.

61
Materials and Methods
Zebrafish Maintenance
All procedures described have been approved by CHLA IACUC committee. To raise the
zebrafish, we keep the fish at 28.5 ºC with a 14 hours light/10 hours dark cycle. Brine shrimp and
commercial powder food smaller than 450µm have been used to feed the fish. In our system, we
feed the larvae with powder food smaller than 50µm at the beginning and switch to the powder
food smaller than 100µm together with brine shrimp for the juvenile fish. The fish raising and
maintenance are under standard conditions of care and with CHLA IACUC oversight. After
mating the fish, fertilized embryos were collected and changed into the E3 media (5 mM NaCl,
0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4, 10
-5
% Methylene Blue). An incubator at 28.5
ºC is used to raise the embryos, plus and minus 5 ºC may be used to adjust the speed of the
development.
Tricaine-s / syncaine (ms 222) fish anesthetic (Syndel) was used to anesthesia (0.168mg/ml in
fish water) and euthanize (0.4 mg/ml in fish water for more than 15 minutes) the fish.
List of Zebrafish Lines Used in This Study:
AB line (ZIRC and UCLA) is used as wildtype control in this experiment.
Tg(kdrl:mTurquoise) (Harrison et al., 2019), Tg(sox10:GAL4-UAS-Cre,ubi:switch),

62
Tg(sox10:GAL4-UAS-Cre,UAS:Nfsb-mCherry) (Cavanaugh et al., 2015),
Tg(pdgfrb:GFP,fli:dsRed), transgenic lines have been described before.
Another two sox:switch fish lines are obtained from Dr. Gage Crump’s lab, which are
Tg(Sox10:Cre/bact:Blue>Red) and Tg(Sox10:Cre/ubi:stop>Red, cdh1:mlanYFP). The allele
references are : Tg(Mmu.Sox10-Mmu.Fos:Cre)zf384 (Kague et al., 2012),
Tg(actab2:loxP-BFP-STOP-loxP-dsRed)sd27 (Kobayashi et al., 2014),
Tg(-3.5ubb:loxP-STOP-loxP-mCherry)el818 (Fabian et al., 2020), Tg(cdh1:mlanYFP)xt17Tg
(Cronan and Tobin, 2019).
CRISPR/Cas9 Technology
1. Order gRNA scaffold oligo (PAGE purified):
5’-gatccgcaccgactcggtgccactttttcaagttgataacggactagccttattttaacttgctatttctagctctaaaac-3’
2. Go to Syntheno website to help design the gRNAs (https://design.synthego.com/#/). It will
recommend 8 or more gRNA sequence options. Choose the top 8 gRNA target sequence. Order
gene-specific oligos: [T7 promoter]-[Target Sequence]-[start of gRNA sequence].
5’-aattaatacgactcactata-[18-22bp Target Sequence]-gttttagagctagaaatagc-3’
3. Perform PCR. The two oligos serve both as primers and templates. Phusion polymerase is
recommended but other polymerases should work as well. Recommend 100 µl system to make
sure that you have enough template to do the in vitro transcription

63
4. Clean the PCR reaction with QIAuick® PCR Purification Kit (Qiagen). At this point, the
concentration can be checked by using NanoDrop machine and the product can be analyzed on a
gel. It should be a single 125 bp band.
5. Use as a template for in vitro transcription with either, T7 RiboMAX
TM
express large scale
RNA production system (Promega) or MEGAscript® T7 Kit (Invitrogen). Perform the optional
DNAse step.
6. Recover the gRNA product with mirVana
TM
miRNA Isolation Kit (Invitrogen). Aliquot and
store the gRNAs in -80ºC.
Forward primers used in this project to generate the gRNAs are shown in Table 3-1
Table 3-1: The forward primers used to generate the gRNAs targeting the chd7 gene.
NO. T7 promoter Targeting sequence RNA scaffold forward primer
gRNA1 AATTAATACGACTCACTATA GACTACAGAAGAGACGATCT
AGT
GTTTTAGAGCTAGAAATAGC
gRNA2 AATTAATACGACTCACTATA GAAGAGACGATCTAGTCGGC GTTTTAGAGCTAGAAATAGC
gRNA3 AATTAATACGACTCACTATA GATCTAGTCGGCAGGTGAAG GTTTTAGAGCTAGAAATAGC
gRNA4 AATTAATACGACTCACTATA GAATTTCAGATGACGACGA GTTTTAGAGCTAGAAATAGC
gRNA5 AATTAATACGACTCACTATA GACGATGGAGATGATTCTGC GTTTTAGAGCTAGAAATAGC
gRNA6 AATTAATACGACTCACTATA GCTCAGATGTAGTTGGAGAC
TT
GTTTTAGAGCTAGAAATAGC
gRNA7 AATTAATACGACTCACTATA GACCTGCTGCTCAGATGTAGT GTTTTAGAGCTAGAAATAGC
gRNA8 AATTAATACGACTCACTATA GTCTCCAACTACATCTGAGCA
GC
GTTTTAGAGCTAGAAATAGC

64
gRNA9 AATTAATACGACTCACTATA GGGAGAGGGAGTTCAGGACA GTTTTAGAGCTAGAAATAGC
gRNA10 AATTAATACGACTCACTATA GGCCAGCCGAAAGACCATTC GTTTTAGAGCTAGAAATAGC
Microinjection
Mate the fish one day before injection. On the injecting day, thaw the gRNA, make the mix
with 1/10 EnGen Spy Cas9 Nucleases (Cas9 protein, BioLabs), 1/10 10X Buffer 3.1(NEB),
200-300 ng/µl for single gRNA injection or 125 ng/µl each for multiple gRNAs injection, and
RNase free ddH2O. Place the mix at 37ºC for 10 minutes. Co-injected the gRNA together with
Cas9 protein into single-cell stage zebrafish embryos, 2 nl of injection mixture per embryo.
T7E1 Assay
To check the efficiency of creating mutations, collect the genomic DNA of the injected fish at
2 dpf. PCR was done expanding the targeting sequences. Purified the PCR product and then melt
and anneal the PCR product by slowly reducing the temperature (95 ºC : 5min, cool to 85 ºC : -2 ºC
/min, cool to 25 ºC : -1 ºC /10 seconds, hold on 4 ºC ). T7 Endonuclease I (BioLabs) then was used
to distinguish and cut at the mismatched DNA loops (Vouillot et al., 2015). WT band cannot be cut
by T7 Endonuclease I, while the mutant mismatched bands can be cut. Thus, the mutant bands are
smaller than the WT band.

65
gRNAs were injected together with Cas9 protein individually, then choose the most efficient
four gRNAs and do the co-injection to gain the highest mutant efficiency. Raised the fish, and
genotyping was done and sequenced by Genewiz from Azenta.
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.
RT-PCR
Real-time reverse transcription PCR was done in 24 hpf fish and 4 dpf fish. At 4 dpf, after
euthanizing the fish, fish were cut at the middle of the yolk. The head part was used for extracting
the total mRNA and the tail was used to do the genotyping. Total mRNA was extracted by using
Trizol RNA extraction process (Invitrogen). RNeasy kit (Qiagen) was used to purify the total
mRNA. cDNA was generated by using the SuperScriptTM III Reverse Transcriptase kit
(Invitrogen), and then real-time PCR (RT-PCR) was done by using LightCycler 480 SYBR green I
master mix (Roche). RT-PCR data were adjusted by householding gene EF1α. The fold of chd7
expression compared to the WT was shown in results.

66
Whole-mount In Situ Hybridization
Whole-mount in situ hybridization was done following the published Nature protocol (Thisse
and Thisse, 2008). Primers used to generate the probes are in table 3-2:
Table 3-2: Primers used to generating the RNA probes for in situ hybridizations in zebrafish
Gene Forward primer Reverse primer bp
chd7 TGTCTACCGAGAAAGCCTCAG GTTCAGTCAGCGTCTCCTCA 616
phf6 GGAATCTATCTTGTTTACTGTCGC AGCTGGCCCCTTTCTTGTTT 823
Hand2 CCTCATTGATTCCACAACGTGCTC TAAGCGATATAACTGGTAGCTAAACGA 809
Crestin CCAACCTGCTGACACAGACCCACT TGTTGACCACTGCAATATGCCGAT 991
Foxd3 CAAAACAAGCCCAAGAGC CTGATGCCTCTTGAAGCG 330
Sox9a TGAACGAGGTGGAAAAGCGT GTGGCTGTCGGAATAGTCGT 743
Sox2 TGGGCACAACAGGACCTAAG TTCGTCGATGAATGGTCGCT 421

Phusion High-Fidelity DNA Polymerase (Thermoscientific) was used to PCR the target
sequence. Zero Blunt™ TOPO™ PCR Cloning Kit (Incitrogen), One Shot TOP 10 chemically
competent cells E.coli (Invitrogen), and QIAprep Spin Miniprep Kit (Qaigen) were used to
generate and purify the vectors. EcoRV or NotI (New England Biolabs) were used to cut different
vectors. T7 or SP6 MEGAscript Kit (ambion by life technologies) and DIG RNS Labeling Mix
(Roche) were used to synthesize the probes.

67
Confocal Microscopy
Embryos from 18 hpf to 5 dpf were anesthetized in 0.1% Tricaine S (Syndel) and then
mounted in 1% agarose with tricaine in fish water. Adult fish were euthanized by using ice for
more than 15 minutes. Fish hearts were dissected by using forceps and spring scissors (Fine
science tools). Dissected hearts were then washed in PBS and fixed in 4% PFA/PBS for less than 5
minutes. Hearts were mounted in 1% agarose in PBS and then imaged by confocal. Confocal
microscopes (Leica STELLARIS 5 inverted confocal microscope and Zeiss LSM 710 Inverted
Confocal Microscope) were used to image the embryos, fish hearts, and immunostaining sections.
Chondrocytes’ Arrangement Analysis
To analyze the chondrocytes’ arrangement, the angle between three neighbor chondrocytes
was measured. 1. Since the cells connecting to the basibranchial are round and not as organized as
the chondrocytes in the other chondrocytes in the ceratobranchials, skip the 1/4 to 1/3
chondrocytes in the basibranchial side. Start measuring the angles from the first more organized
chondrocytes if possible. If could not distinguish the organized chondrocytes, just skip 1/3 of the
chondrocytes on the basibranchial side. Treat the basibranchial side as the base and the other end
as the tip. 1. Draw the line vertical to the cell membrane to connect the cell to the adjacent cell
(choose the one near the base side if there are more than one cell). 2. Repeat step 1 and measure the

68
angle between these two lines. 3. Choose the unmeasured cell, which is most close to the base end,
and repeat steps 1 and 2. Until measuring all the cells or there are less than three cells in the tip end.
Alcian blue and Alizarin Red Staining
For embryos and larvae, anesthetize larvae with tricaine and then fix in 2% PFA at room
temperature for 1 hour. Then washed with 100 mM Tris (pH 7.5), 10 mM MgCl2 for 10 minutes
(min). Incubate overnight in Alcian stain (Stock: 0.1g Alcian blue 8GX, 2.6ml H2O, bring the
volume to 50ml by using 95% EtOH; working solution: 10ml Alcian blue stock, 32.6ml 95%
EtOH, 5ml Tris-HCl, pH7.5, 0.5ml 1M MgCl2, 0.9ml H2O) at room temperature. Destain the fish
by 80% EtOH/100 mM Tris (pH 7.5)/10 mM MgCl2, 50% EtOH/100 mM Tris (pH 7.5)/10 mM
MgCl2, 25% EtOH/100 mM Tris (pH 7.5)/10 mM MgCl2 for 5 min each. Move the fish in a well
of 24-well plates, bleach the fish by 3% H2O2/0.5% KOH under a light source until the eyes turn
to light brown. Wash twice with 1 ml 25% glycerol/0.1% KOH for 10 min each. Nutate in Alizarin
stain (Stock: 0.25g Alizarin Red S to 50ml H2O, working solution: 1ml Alizarin Red stock, 12.5ml
Glycerol, 5ml Tris-Hcl, pH 7.5, 31.5ml H2O) at room temperature for 1 hour. Destain by washing
with 1 ml 50% glycerol/0.1% KOH for 10 min twice. Gradient moves into 100 glycerol and
images the fish in glycerol by using Zeiss Axioplan Upright Microscope in a bright field.

69
Kits Used in This Project
1. QIAquick® PCR purification Kit, MinElute® Gel Extraction Kit (by QIAGEN) was used
for gel purification.
2. QIAprep® Miniprep Kit (by QIAGEN) was used for plasmid extracting.
3. MAXIscript® T7/T3/SP6 Kit, MEGAscript® T7/T3/SP6 Kit, and mMESSAGE
mMACHINE T7/T3/SP6 (by Invitrogen) were used for in vitro transcription.
4. Zero blunt TOPO® PCR cloning kit and Zero TA TOPO® PCR cloning kit
5. TRIzol (Invitrogen), chloroform, RNeasy column (Qiagen kit), mirVanaTM miRNA
Isolation Kit
6. LightCycler® 480 SYBR Green I Master (Roche)
7. T7 RiboMAXTM Express Large Scale RNA Production System (Technical Bulletin)
8. MicroPoly(A) Purist
TM
Kit (by Thermo Fisher Scientific) was used for purifying mRNA.
9. SuperScript® III First-Strand Synthesis System for RT-PCR (by Invitrogen) was used to
synthesize cDNA from purified total mRNA.
10. GoTaq® Flexi DNA Polymerase (Promega)
11. CloneJET PCR Cloning Kit (thermo scientific)
12. Mut Express MultiS Fast Mutagenesis Kit V2 (vazyme)

70
4. Chapter IV: In Vivo Disease Models of BFLS.
4.1. Introduction
4.1.1. Xenopus as an In Vivo Animal Model
In Xenopus, the NCC segregates early during neurulation and remains adjacent to the neural
plate for many hours before initiating migration. At Nieuwkoop and Faber stage (NF) 16, the NCC
masses can be distinguished as a deep layer of the ectoderm for the first time. The premigratory
cephalic NCCs segmented into three groups of cells that are destined to be the mandibular (MCS),
hyoid (HCS) and branchial (BCS). By NF 21-22, the first MCS, moves ventrally to the optic
vesicle; the second HCS and the third BCS originate from the anterior and posterior
rhombencephalon. By NF 23, BCS splits into two portions: an anterior one and a posterior one. By
NF 24, two portions of the MCS have already rejoined (Sadaghiani and Thiébaud, 1987).
Tracking the destination of these segments showed that MCS takes part in the formation
including Meckel’s cartilage; HCS contributes to the formation of the interhyoideus and two
muscles connected to the ceratohyal cartilage; and BCS has been observed in gills (in
undifferentiated cartilages and in the mesenchyme of the gills). Moreover, some crest cells from
the BCS penetrate the wall of the truncus arteriosus, indicating that they may have functions in the
arteriosus (Sadaghiani and Thiébaud, 1987).

71
4.1.2. Zebrafish phf6 Mutants as an In Vivo BFLS Model.
Xenopus has some advantages as an animal model including that their embryos are big and
easy to inject. Injection can be done into one cell at the two-cells stage using the other side as
control, and the large embryos can be used to do ex vivo transplantation experiments. However,
there are also many disadvantages for Xenopus. First, unlike zebrafish, usually Xenopus would not
mate naturally. Male Xenopus need to be sacrificed to get the testes, and two previous injections
need to be done on the female Xenopus to induce egg production. Second, according to IACUC
protocol, Bajpai’s lab was not allowed to raise the Xenopus tadpoles to adults. Third, in this project,
I would like to generate the phf6 mutants to compare to our chd7 mutants, especially on the
cartilage phenotypes. Zebrafish, due to its transparent body in larvae, the structure of craniofacial
cartilage is also well studied compared to the Xenopus. Thus, I transitioned to zebrafish and tried to
generate the phf6 mutants in zebrafish in Bajpai’s lab and continued to characterize the mutants in
the Lien’s lab. Furthermore, I can even generate double mutants to see whether one mutant can
suppress the other’s phenotypes.
Only one phf6 heterozygous zebrafish mutant has been generated so far with a 10 base pair
deletion in the second exon of the phf6 gene (Loontiens et al., 2020). The thymus size was
increased in the phf6 heterozygous fish, and notch1 expression during thymopoiesis was reduced.
No other phenotypes of phf6 mutants have been reported. To understand the function of Phf6 in
early development in vivo and to compare its role to the Chd7 in parallel, I generated two phf6

72
mutant fish lines to study the molecular mechanisms behind Phf6. Our BFLS iPSCs have a PHF6
C45Y mutant allele, which is a recurrent mutation that has been found in multiple BFLS families.
Here, in zebrafish, one of the phf6 mutants I generated has a mutation in the counterpart cystine,
which is the 43rd amino acid of Phf6 in zebrafish.
4.2. Results
4.2.1. Phf6 Is Expressing in a Specific Subgroup of NCCs in Xenopus.
PHF6 is highly conserved from zebrafish to humans (Figure 4-1A). To characterize PHF6 in
more details at the molecular levels and identify an in vivo model to study its functions, I
performed in situ hybridization to determine its expression patterns in Xenopus embryos.
I utilized two Phf6 probes which target the 3’UTR and the coding region of Phf6 in Xenopus,
respectively. The whole-mount in situ hybridization was done on the late neurula stage and late
tailbud stage embryos (Figure 4-1Bi, Ci). Compared to Twist1 which is expressed in migrated
NCC segments and Sox2 which is expressed in neural tubes and not in the NCCs streams our Phf6
probes stained both the neural tube and migrated NCCs. However interestingly, it is clearly shown
in the cross section that Phf6 is only expressed in the shallow layer of the NCCs’ segments
compared to the Twist1 stained embryos (Figure 1-4Bii, Cii). Therefore, I hypothesized that Phf6
maybe only expressed in a specific subtype of the NCCs in Xenopus.

73

Figure 4-1: The conservation of PHF6 across species and the expression pattern of Phf6 in Xenopus embryos.
A, Phylogenetic analysis of PHF6 protein sequence between Homo sapiens, Danio rerio, Xenopus tropicalis, Gallus
gallu, Rattus norvegicus, Mus musculus, and Canis lupis familiaris (Ovcharenko et al., 2004). The peaks indicate the
similarity from 50% to 100%. CHD7 exons are highly conserved from fish to human. Blue shows the peaks in exons,
orange marks the peaks in introns, green is the not consistent area (less than 75%), yellow is the 3’UTR area. B,

74
C.Whole mount in situ hybridization in Xenopus embryos at Bi, NF19 late neurula stage, and Bii, NF29 late tailbud
stage. Bii, Cii. Zoomed in image and cross section through the trunk region shown in Bi, Ci. The filled arrows point
out present neural crest segments’ staining, red: PAA2, yellow: PAA3, green: PAA4. Model of embryos, tissue types
are shown. Orange: Neural Tube. Green: Neural Crest. Purple: Placode. Dark Red: Somites.
4.2.2. BFLS Patient-Specific Mutant Fish Model Does Not Show Obvious Phenotypes.
I then tried to generate the phf6 mutants in zebrafish due to the ease of generating CRISPR
mutants. Our BFLS iPSCs have a PHF6 C45Y mutated allele, which is a recurrent mutation that
has been found in multiple families with inherited BFLS. Here, in zebrafish, I tried to generate the
mutation in a similar Cystine. In zebrafish, the counterpart of that cysteine is the 43
rd
amino acid of
Phf6. I co-injected gRNA 1 and 2 together with Cas 9 protein into zebrafish embryos to get the
patient counterpart fish mutations (Figure 4-2A). After outcross and incross, a stable phf6(C43Y)
mutant fish line, which has a three base pair deletion having exchanged the 43
rd
cystine and 44
th

lysine into serine (Figure 4-2B,C), was generated.
The second phf6 fish line has 8 bp deletions right before the ePHD2 [phf6(d8)] to cause
frame-shift mutation and stop after 15 amino acids (Figure 4-2C) by co-injected the gRNA 3 and 4
with Cas 9 protein (Figure 4-2A). It has been reported that deletion of the ePHD2 may have a
dominant-negative function in PHF6 female patients, therefore this second line could have a
severer phenotype compared to the phf6(C43Y) and may mimic the mechanism in BFLS female
patients.

75

Figure 4-2: Overall characterization of the phf6 zebrafish mutations. A. Co-inject the gRNAs together with
CRISPR Cas9 protein into one cell-stage zebrafish embryos that target phf6 (PAM area is recognized by Cas9
protein but not included in the targeting sequence). B. Structure of the mutated ePHD1 in phf6(C43Y) mutants. C.
Genotyping and protein sequences in WT control and phf6 mutations. D. Percentage of the genotypes of the phf6
incrossed offspring at 7 dpf and 90 dpf, respectively. (Collaborated with Zhiyu Tian who helped with the genotyping
for part of the 7 dpf fish.) Chi-squared test was used to check the statistical significance between different groups in
Prism 9. E. RT-PCR experiments to analyze the gene expression of phf6 at 4 dpf embryos. The fold of phf6
expression compared to the WT was shown (Collaborated with Zhiyu Tian who helped with collecting and generating
the RT-PCR data).

76
In both our phf6(C43Y) and phf6(d8) homozygous fish, no obvious phenotypes have been
found including the fish size, eye size, survival rate, and heart edema. Percentage of the genotypes
in phf6(C43Y)+/- and phf6(d8)+/- incrossing offspring indicated that the phf6(C43Y)-/- (n>100)
and phf6(d8)-/- (n>100) are not lethal in zebrafish (Figure 4-2D). Further, gene expression analysis
showed that both phf6(C43Y)-/- and phf6(d8)-/-have a significant reduction in the phf6 expression
(Figure 4-2D). In summary, we do not see any obvious gross phenotype in our phf6 larvae.
4.2.3. The phf6(C43Y)-/- Larvae Do Not Have Obvious Craniofacial Phenotype.
To check whether our phf6(C43Y)-/- has opposite phenotype with our CHARGE fish model,
alcian blue and alizarin red double staining had been done in wildtype control (n=16) and
phf6(C43Y)-/- (n=11) fish to check the craniofacial in detail. Neurocranium structure (Figure
4-3A) and the size of otoliths (Figure 4-3Bi,Bii,Biii) did not show any significant defects.
Similarly, the structure of the viscerocranium (Figure 4-3C) and the chondrocytes arrangement in
the ceratobranchials (n=6 for each group) (Figure 4-3D), which are affected in the chd7(d55)-/-
fish, did not have any obvious abnormalities (Figure 4-3E). Together, our data suggested that
phf6(C43Y)-/- does not have any craniofacial cartilage defects.

77

Figure 4-3: The phf6(C43Y) mutants do not have obvious craniofacial phenotypes. A. Neurocranium
morphology of WT and phf6(C43Y)-/- mutants. Neurocraniums were dissected out from the alcian blue and alizarin
red co-stained larvae in 7 dpf. No significant difference was found in the neurocranium structure. Bi. The structure of
the otoliths. Bii. Data quantification of the size of utricle and saccule in WT and phf6(C43Y)-/-. No significant defects
have been found. C. Viscerocranium morphology in WT and phf6(C43Y)-/-. Viscerocraniums were dissected out from
the alcian blue and alizarin red co-stained larvae at 7 dpf. No obvious abnormalities have been found in the structure of
viscerocranium in phf6(C43Y)-/-. D. High magnification image shows the arrangement of chondrocytes. cb,
ceratobranchial. E. Chondrocytes’ arrangement analysis in ceratobranchial I and II. Angles of three neighbor

78
chondrocytes was measured. Each number indicated a different fish. The significance was calculated between
different groups. No significant differences have been found in phf6(C43Y)-/- chondrocytes. ns: p ≥ 0.05, *: p<0.05,
**: p<0.01, ***: p<0.001, ****: p<0.0001
4.2.4. Phf6(C43Y)+/- and phf6(C43Y)-/- Can Partially Suppress the chd7(d55)-/-
Phenotype.
In vitro, Dr. Ruchi Bajpai’s previous data suggested that knocking down PHF6 could partially
rescue the defects in the CHARGE patients’ iPSC-NCCs. To further determine the opposite
function of Phf6 and Chd7 in vivo, I generated the phf6(C43Y), chd7(d55) double mutants. The
phf6(C43Y)-/-, chd7(d55)-/- (n=4) mutants are viable till adult stage. I incrossed the
phf6(C43Y)+/-, chd7(d55)+/- double heterozygous fish to get the double mutants and stained the
cartilage and bone with alcian blue and alizarin red. Consistent with the phf6(C43Y)-/- and
chd7(d55)-/- (m+/-) data, no defects had been found in the neurocranium (Figure 4-4A) and the
saccule (Figure 4-4Bii). In contrast, the decreased size of the utricle in chd7(d55)-/- (m+/-) (n=13)
is rescued in both phf6(C43Y)+/-, chd7(d55)-/- (n=27) and phf6(C43Y)-/-, chd7(d55)-/- (n=14)
fish (Figure 4-4Bi).
However, on the other hand, both chd7(d55)-/- (m+/-); phf6(C43Y)+/-, chd7(d55)-/-; and
phf6(C43Y)-/-, chd7(d55)-/- larvae have mild abnormalities in the viscerocranium (Figure 4-4C),
and the arrangement of the chondrocytes in the ceratobranchials I and II (n=6 for each group)
(Figure 4-4Di, Dii). Quantification of the three neighbor chondrocytes also indicated no rescue of
phf6(C43Y) mutation in chd7(d55)-/- fish. Therefore, my data suggested a partial rescue of

79
CHARGE-like chd7 mutants’ phenotype by reducing the phf6 expression in the double mutant
fish.

Figure 4-4: The chd7 and phf6 double mutant can partially suppressed the chd7 fish mutants’ phenotypes. A.
Neurocranium morphology of WT; chd7(d55)-/-; phf6(C43Y)+/-, chd7(d55)-/-; and phf6(C43Y)-/-, chd7(d55)-/-
mutants. All these four types of fish are siblings from the incross of double heterozygous fish. Neurocraniums were
dissected out from the alcian blue and alizarin red co-stained larvae in 7 dpf. No significant difference was found in the

80
neurocranium structure. Bi, Bii. The structure of the otoliths. No significant defects have been found in the size of
saccule, and the defects of chd7(d55)-/- in uricle can be rescued by phf6(C43Y) mutations. C. Viscerocranium
morphology in WT; chd7(d55)-/-; phf6(C43Y)+/-, chd7(d55)-/-; and phf6(C43Y)-/-, chd7(d55)-/- mutants.
Viscerocraniums were dissected out from the alcian blue and alizarin red co-stained larvae at 7 dpf. All mutants have
mild abnormalities in the structure of viscerocranium. D. High magnification image shows the arrangement of
chondrocytes. cb, ceratobranchial. E. Chondrocytes’ arrangement analysis in ceratobranchial I and II. Angles of three
neighbor chondrocytes were measured. Each number indicated a different fish. The significance was calculated
between different groups. All mutants have mild defects in the arrangement of the chondrocytes. ns: p ≥ 0.05, *:
p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001
4.3. Discussion
4.3.1. The Weaker Phenotype in phf6 Homozygous Fish
Although no previous phf6 mutant fish data has been published, our phf6 mutants still seem
to have a weaker phenotype compared to the human patients (no homozygous female patients
have been found), and some lethal mice models.
The first reason for this weaker phenotype is caused by the genetic background, which is
suggested by the previous research that the same phf6 knockout males are viable in FVB/BALB/c
background but not C57BL/6 mice background (McRae et al., 2020). Similar to chd7 mutants, the
mRNA degradation had been found both in the frame-shift phf6(d8) mutants and the in-frame
phf6(C43Y) mutants, therefore the genetic compensation could also exist in phf6 mutants. Zhiyu
Tian in our lab is working on generating two new phf6 mutations which delete the whole phf6 gene
as well as the 100bp or 1000bp upstream of the phf6 gene to also delete the promoter region. In this
way, we could eliminate phf6 completely and get rid of the genetic compensation. Lastly, both the
Xenopus phf6 gene expression data (from Xenbase) and the zebrafish phf6 gene expression data

81
(RT-PCR done by Zhiyu Tian in AB wildtype) indicated that there is a high phf6 expression at the
gastrulation stage which dropped later. These data suggested that maternal contribution may also
occur in phf6 which needs to be checked in the future.
4.3.2. The Partially Suppressed chd7 Mutants’ Phenotypes in the chd7 and phf6 Double
Mutants
So far mutating chd7 caused defects in ceratobranchials, teeth, inner ear, and the ventral aorta
in zebrafish, indicating that Chd7 have functions in different PAs. I found that in our chd7(d55),
phf6(C43Y) double mutants, the inner ear phenotype has been suppressed but not the
ceratobranchial phenotype. The otoliths are derived from otic placode, and the ceratobranchials are
from NCC PA3-7. Therefore, one possible reason is that unlike chd7 which may have functions in
the majority of the NCCs, phf6 may only have functions in a subset of the NCCs and placodes,
which is also consistent with our Xenopus Phf6 in situ results. However, the otoliths defects in
chd7 mutant fish suggested that Chd7 may also have functions in the placodes, which is also
interesting to study in the future.

82
4.4. Materials and Methods
The gRNA Design for Generating phf6 CRISPR Mutants
The gRNAs used to generate the phf6 fish mutants is shown in Table 4-1. The gRNA1 and 2
are co-injected with CRISPR Cas9 protein to generate the phf6(C43Y) mutants and the gRNA3
and 4 are co-injected to generate the phf6(d8) mutants.

Table 4-1: The forward primers used to generate the gRNAs targeting the phf6 gene.
NO. T7 promoter Targeting sequence RNA scaffold forward primer
gRNA1 AATTAATACGACTCACTATA GCTAACATACCATGCACTTG GTTTTAGAGCTAGAAATAGC
gRNA2 AATTAATACGACTCACTATA CACTTGTGGTGGGCTGCC GTTTTAGAGCTAGAAATAGC
gRNA3 AATTAATACGACTCACTATA GACATTAAAACAGTCATCC GTTTTAGAGCTAGAAATAGC
gRNA4 AATTAATACGACTCACTATA GTCATCCAGGAAATCAAGAG GTTTTAGAGCTAGAAATAGC
Xenopus Embryology
Animals were acquired and kept according to IACUC protocols. Females were pre-primed
using 50 units pregnant mare serum gonadotropin 7-10days prior to priming with 500 units human
chorionic gonadotropin. Gently squeezed the female Xenopus to collect the eggs. Fertilized the
eggs with the testis’s slurry extracted from the male frog which can be kept at 4 degrees for

83
maximum 14 days. Embryos were treated with a 2% cysteine solution to remove the jelly coat and
develop in 0.3x modified Barth solution.
Whole-mount In Situ Hybridization
Whole-mount in situ hybridization was performed using the standard protocol presented in
Sive et al. (Sive et al., 2010) with some modifications. Briefly, embryos were rehydrated, washed
using PBST, treated with proteinase K for 12 minutes, treated with acetic anhydride twice, re-fixed
using a 4% PFA solution and chased using 0.1M glycine, washed using PBST, pre-hybridized for
6 hours, and the probe was hybridized using 1.5 μg/ml probe in hybridization buffer overnight at
65
o
C. Embryos were then washed using 2x SSC, treated with RNase for 30 minutes, washed using
2x SSC, 0.2x SSC, MAB, and a 2% Boeringer-Manheim Block (BMB) for 1 hour, followed by
overnight antibody incubation (anti-dig AP, Roche). Embryos were then washed twice in 2%
BMB, washed extensively in MAB, transferred to alkaline pH using alkaline phosphatase buffer,
and BM Purple (Roche) was used as a substrate for the alkaline phosphatase color reaction for
approximately 6 hours. Stain was fixed using Bouin’s Fixative for 2 hours, washed using 70%
ethanol, rehydrated, and bleached using a 30% Hydrogen Peroxide solution under light.
Probes for In Situ Hybridization
The probes used for Phf6 in situ hybridization are sown below in Table 4-2.

84
Table 4-2: Phf6 probe sequences. Both probes are anti-sense to the respective targets, cloned into pCS2+,
linearized using BamHI and transcribed using T7 megascript polymerase.
Probe Sequence
Phf6
3’UTR
5’-ATTTCTAGACTTTGGAATTTATTTTCCCTAAAACTAGAACATTAAAGTTTGCAATCA
GCCAAACGCTTTGCAATGATACACCATGTATTCATATTGTGTAATAAAGGTAACTGTA
CATTACCGCAGAACAAAAGACACATATTGACGTGTATGAAGGTCATTAGTAACAGCCC
ACATTGTATACATTTACAAACTGAGCTGCGTGGGGGCCAAACAGACTTTTTACGCACC
AAAGTCTTTCCTCGCAATATTTTGCAGCATGTTGCTGGGCAGTTTTTAAATAAAGAAA
GTAAATAGCTATCTTGTGCTATGTCAACAGTAAGCAACTGGCAGCACAGCCATAGTTT
AAACAGCAGTGTGACAGCCTGTCTAACCCTTTCCATTGGCTTTAAACTGCTTAAAACG
TGTGAAATGCATAAAGCATAAAAAGTAATAACCAATGGGTACAAACCAAGACATATGT
AGGATAACAACCGCTGCCCTTTTTGTGATTATCACATTTTCACTAACTGGTGAAAAAA
ATACATATTTCCCATTTCCCCGCAGTCCTTTTCACAATCCTAGATAAATATATGACTTT
TATCTAGGAAATTGAAACTGTGTCTGCATTCCCACAACTGGCAGACTGGGCACTGGCT
TATATGACAGCTTTGATCCTTAGTTCCCATTGAGCTGCTGCACAGTGCTCTGATTGTG
CCCTACTGTTGCATTGCTCTTGCTTCTGATTTCACGTTCCTCGTCCTCTTCATCTCTTT
CATCGTTTCCACTGTGATTTTTACAATAAAGT-3’
Phf6
Coding
5’-TCACAAAATACTACACATCAATTCAAAAAACAAAACCTGGAAAAATACCATTCTCTT
CCCACGTTTCATCTCTTGGATCACAGTTTTGATATCAAAGTCTCCAAATTCTGCTCTTG
ATGTTGTGGTGAGCTGTACTGTGCCAGAGGAGAAAAGCATGCATTTGTAATGAGCCG
CGGCCTTTTTTGCATTAAATACATGGAGCTTGCCTCGTGTCTCATTTTCTTCATCTCCT
GCGTGACAGAAACCACATTTAGGTCTAGTGTCACTGGGACTGCTTCTGTGTGGAGAA
CGATCTCTACTAGAGCTGCTTTCGGTTTCTTCTGTACAATGTGAGGATGAAAGTCTGC
TCTCCTCAGACTGCCCTCTGGAATTTGATGTCCGTGATTTTCCTCTAGACCCTCTTCCT
CGGCCACGTTGAGGAGATGGAGTCAAACCTCTACGTGTAAAACTACCTTCTAATTCAT
CGTCTGAATTCCGCACCTGTTCCTTATGCTTTCTGCAAAAAATCGTATACATTCCCTGT
GATGGATTTTCCTGAATATGAGCTTTATCACGCAGGGCACAGTGGTAATGGTAGCTTC
TATGGCAGGATTTAACATCACACCCAATTGTGGCTCCTGGGCAGTGACAAAGTGAGC
ACATCAGCTTTTTTCCTCTCTTAAGCTCCTTTTGGATATCTTCTATTGAAAAGCCCCCG
AGATTCTCACTGTCAGACTGTGATGATACCAGTGCTGATGAAAACAGCATACATCTGT
GATGTGCTGCCACCTTTTGGTTGCTTGATATTAGTAGATGTCCACATTCCTTTTCTCG
GTTAGATCTGCAGAATCCACATTTGTGTTGCTGATGTGAAGAGCCCTTTCTCTGTCCA
GTGGAGCTAGACAT-3’

85
Generating chd7 and phf6 Double Mutants.
The chd7(d55) mutants was used to cross to the phf6(C43Y) and phf6(d8) mutant fish.
Genotyping was done to screen the chd7(d55)+/-, phf6(C43Y)+/- and chd7(d55)+/-,
phf6(d8)+/- double heterozygous fish. These double heterozygous fish then incossed to generate
offspring for the experiments.

86
5. Chapter V: Future Directions
5.1. Discussion
5.1.1. CHD7 and PHF6 May Have Antagonistic Functions

Figure 5-1: The schematic of our hypothesis that Chd7 and Phf6 work together as antagonists specifically in
the craniofacial NCCs but not in the cardiac NCCs. Gray, PA1 and tissues derived from PA1; orange, PA2 and
tissues derived from PA2; yellow, PA3 and tissues derived from PA3; pink, PA4 and tissues derived from PA4; cyan,
PA5 and tissues derived from PA5; blue, PA6 and tissues derived from PA6; purple, PA7 and tissues derived from
PA7.

87
In this project, I mainly focused on the characterization of CHD7 and PHF6’s function both
in iPSCs in vitro and in zebrafish models in vivo. Our data suggested both in human iPSC-NCCs
(done by Dr. Ruchi Bajpai) and in zebrafish craniofacial cartilage, diminishing phf6 expression
could partially rescue the defects in CHD7 knockdown cells and chd7 homozygous mutants. Plus,
the CHD7 and PHF6 are both chromatin remodelers and have the possibility to recognize the
similar epigenetically modified histone tails, not to mention that CHD7 and PHF6 could directly
bind to each other (done by Dr. Ruchi Bajpai). It is highly possible that CHD7 and PHF6 may
work as antagonists.
Due to the fact that BFLS patients do not have the congenital heart diseases, and the tissue
specificity of both CHD7 and PHF6, we hypothesized that CHD7 and PHF6 work together as
antagonist specifically in the craniofacial NCCs and cranial ectodermal placodes but not in the
cardiac NCCs (the schematic of our hypothesis in zebrafish is shown in Figure 5-1).
5.1.2. The Possible Affected Cardiac Cell Lineages in CHARGE Mice Models
The heart is a very complex organ that contains many cell types such as cardiomyocytes,
endothelial cells, fibroblast cells, and mural cells. Cells from all three germ layers: the endoderm,
mesoderm, and ectoderm (cardiac NCCs migrated out from the ectoderm) and contribute to heart
development (Moorman et al., 2003).
Endocardial-specific Chd7 ablation mice line was generated by utilizing the Tie2-Cre
marker. Great vessels, septation defects, and myocardial non-compaction defects are present.

88
However, in the Tie2-Cre, Chd7 knockout mice, fewer showed edema and arterial pole septation
are normally formed. Therefore, Chd7 plays an important role in the endocardium for the AV
canal septation (Payne et al., 2015).
Ablation of Chd7 in the anterior mesoderm, by using Mesp1-Cre marker, is lethal with
severe edema and/or hemorrhaging. Double outlet left ventricle, absent or poorly formed venous
valves, interrupted aortic arch type B, absent or reduced vestibular spine, and myocardial
non-compaction defect, but not the hypocellular atrioventricular cushion defects are seen in the
Mesp1 conditional knockout mice. PAAs, which are related to the great vessel defects, are also
formed normally in this mice model. Indicated that the ventricle and valve defects may be related
to the defects in the mesodermal cell lineage (Payne et al., 2015).
NCCs specific Chd7 ablation mice were generated by using Wnt1-Cre or Foxg1-Cre driven
knockouts. Results showed that the conditional knockout mice are perinatally lethal and have
defects in the upper and lower airway, palatal, bone, and conotruncal but not in heart and PAAs
(Sperry et al., 2014). However, when an improved NCC Cre line, Wnt1-Cre2 was used, severe
heart phenotypes in the conotruncal region were observed (Yan et al., 2020). Moreover, using
Chd7 gene-trap mice to restore Chd7 expression in NCCs could not rescue the abnormal PAA
phenotype in Chd7 heterogenic embryos (Randall et al., 2009). Together, these data indicated
that the chd7 mutant NCCs have more function in craniofacial rather than in the ventricle and
great vessels in mice. However, in zebrafish, there is evidence showing that NCCs contribute to
the ventral aorta and cardiomyocytes (George et al., 2020). According to the differences between

89
mice and zebrafish, it is still interesting to test the phenotypes of the ventral aorta in the chd7
mutants.
5.1.3. Four Signal Pathways That are Associated With NCCs Formation Are Interested
to Be Studied in chd7 Mutants in the Future
During NCCs’ fate choice, to become NCCs rather than neurons, the cells need to get some
extracellular signaling molecules. These kinds of inductive signals are contact-mediated and
secreted from the adjacent non-neural ectoderm and underlying mesoderm (Bhatt et al., 2013).
After getting the induced signals, neural crest cells undergo an epithelial-to-mesenchymal
transition and then delaminate from the neuroepithelium. Previous studies have shown four major
inducers: bone morphogenetic protein (BMP), fibroblast growth factor (FGF), Wnt signaling
families, and Notch signal pathway.
First, for BMP, a current NCC induction model shows that the BMP antagonists diffusing
from the ectoderm generate a gradient of BMP activity. Along with this BMP gradient, low BMP
signaling is required for the development of the neural plate; intermediate levels of BMP signaling
is required for NCC formation, and high BMP signaling is required for the neural ectoderm
(Huang and Saint-Jeannet, 2004). The second extracellular signaling molecule is FGF, which is
secreted by the paraxial mesoderm. Mayor in 1997 showed that expressing the dominate-negative
Fgf receptor in ectoderm explants blocks neural crest induction, and the Fgf signaling pathway is
required in inducing Slug expression (Mayor et al., 1997) in Xenopus. Wnt signaling is the third

90
inductive signal derived from both the surrounding epidermis as well as the underlying paraxial
mesoderm. As mentioned previously, one of the first steps of NCC determination is the expression
of the transcription factor Slug. Wnt has been shown directly involved in the regulation of Slug
during the NCC formation in many species (Vallin et al., 2001). Last, membrane-bound protein
Delta and its receptor Notch also implicate the neural crest formation. In zebrafish, Delta is
required for trunk NCC formation but not the cranial NCC (Cornell and Eisen, 2002). Other
research have shown that reducing Notch signaling (by expressing a Notch construct lacking the
ligand-binding domain) induced the cells to become neurons rather than NCCs (Coffman et al.,
1993), whereas activating Notch during gastrulation dramatically expands the NCC population
without causing defects on neural plate or mesoderm (Glavic et al., 2004). All in all, all these four
signal pathways are associated in NCCs formation. Therefore, it is interesting to study the changes
in these pathways in our chd7 CHARGE fish models.
5.1.4. Histone Citrullination Recognized by PHF6 Could Be a Novel Aspect to Study
Histone citrullination is an epigenetic post-translational modification that converts histone
arginine to citrulline, an amino acid found in proteins, thereby impacting chromatin structure.
Once the arginin becomes citrulline, it cannot be methylated anymore. The enzymes in charge of
citrullinating the proteins are the PADI family, which includes PADI1, PAID2, PADI3, PADI4,
and PADI6. Among these families, PADI4 and PADI2 are the only two has been shown that have
the ability to reside in the nucleus. PADI4 has the ability to citrullinate the histone at H1-R54,

91
H2A-R3, H3-R2, H3-R8, H3-R17, H3-R26, and H4-R3; while PADI2 has only been shown to
citrullinate at the H3-R2, H3-R8, H3-R12, and H3-R26 position (Muth et al., 2017).
Nowadays, the most popular study on histone citrullination is its function in Neutrophil
extracellular traps (NET). As a novel effector mechanism of the host’s innate immune response
against microbial infections, NET structures studded with antimicrobial peptides, histones, and
proteases are released to entrap, kill or immobilize the different parasites. Global histone
citrullination is an important step in NET to decondense the chromatin (Muñoz Caro et al., 2014).
Similar to the global histone citrullination, specific histone citrullination also is able to
increase the accessibility of chromatin at a specific position (Fuhrmann and Thompson, 2016;
Zhang et al., 2012). Notability, histone citrullination may have important functions during early
development. Using Cl-amidine to block the PADI can cause the mice embryos arresting at 4 or
8-cell stage (Kan et al., 2012). Citrullination of histone 3 (H3-Cit) is detectable in ES and iPSCs as
well as an early embryo. In pluripotent stem cells, H1 Cit54 leads to extensive chromatin
decondensation. In ESCs, over-expressing PADI4 up-regulated many pluripotency genes (such as
Klf2, Tcl1, Tcfap2c, Kit, and Nanog). In contrast, using DAPI inhibitor Cl-admin treated the cells
up-regulate the differentiation makers (such as Prickle 1, EphA1, and Wnt8a) markers
(Christophorou et al., 2014). Thus, H3-Cit has functions in regulating gene expression, specifically,
H3-Cit26 plays this role may through a negative cross-talk with the nearby lysine methylation
(Clancy et al., 2017; Fuhrmann and Thompson, 2016).

92
PHF6, as the first recognized protein that recognizes H3-Cit8, could be useful to study the
epigenetic function of H3-Cit.

93
References
Abrial, M., Paffett-Lugassy, N., Jeffrey, S., Jordan, D., O'Loughlin, E., Frederick, C.J., Burns,
C.G., and Burns, C.E. (2017). TGF-β Signaling Is Necessary and Sufficient for Pharyngeal Arch
Artery Angioblast Formation. Cell Rep 20, 973-983. 10.1016/j.celrep.2017.07.002.
Achilleos, A., and Trainor, P.A. (2012). Neural crest stem cells: discovery, properties and potential
for therapy. Cell Res 22, 288-304. 10.1038/cr.2012.11.
Ahmed, R., Sarwar, S., Hu, J., Cardin, V ., Qiu, L.R., Zapata, G., Vandeleur, L., Yan, K., Lerch, J.P.,
Corbett, M.A., et al. (2021). Transgenic mice with an R342X mutation in Phf6 display clinical
features of Börjeson-Forssman-Lehmann Syndrome. Hum Mol Genet 30, 575-594.
10.1093/hmg/ddab081.
Allen, M.D., Religa, T.L., Freund, S.M., and Bycroft, M. (2007). Solution structure of the BRK
domains from CHD7. J Mol Biol 371, 1135-1140. 10.1016/j.jmb.2007.06.007.
Anderson, M.J., Pham, V .N., V ogel, A.M., Weinstein, B.M., and Roman, B.L. (2008). Loss of
unc45a precipitates arteriovenous shunting in the aortic arches. Dev Biol 318, 258-267.
10.1016/j.ydbio.2008.03.022.
Ardinger, H.H., Hanson, J.W., and Zellweger, H.U. (1984). Börjeson-Forssman-Lehmann
syndrome: further delineation in five cases. Am J Med Genet 19, 653-664.
10.1002/ajmg.1320190405.
Asad, Z., Pandey, A., Babu, A., Sun, Y ., Shevade, K., Kapoor, S., Ullah, I., Ranjan, S., Scaria, V .,
Bajpai, R., and Sachidanandan, C. (2016). Rescue of neural crest-derived phenotypes in a
zebrafish CHARGE model by Sox10 downregulation. Hum Mol Genet 25, 3539-3554.
10.1093/hmg/ddw198.
Bajpai, R., Chen, D.A., Rada-Iglesias, A., Zhang, J., Xiong, Y ., Helms, J., Chang, C.P., Zhao, Y .,
Swigut, T., and Wysocka, J. (2010). CHD7 cooperates with PBAF to control multipotent neural
crest formation. Nature 463, 958-962. 10.1038/nature08733.
Baker, C. (2005). Neural crest and cranial ectodermal placodes. .
Balasubramanian, R., Choi, J.H., Francescatto, L., Willer, J., Horton, E.R., Asimacopoulos, E.P.,
Stankovic, K.M., Plummer, L., Buck, C.L., Quinton, R., et al. (2014). Functionally compromised
CHD7 alleles in patients with isolated GnRH deficiency. Proc Natl Acad Sci U S A 111,
17953-17958. 10.1073/pnas.1417438111.

94
Balow, S.A., Pierce, L.X., Zentner, G.E., Conrad, P.A., Davis, S., Sabaawy, H.E., McDermott,
B.M., and Scacheri, P.C. (2013). Knockdown of fbxl10/kdm2bb rescues chd7 morphant phenotype
in a zebrafish model of CHARGE syndrome. Dev Biol 382, 57-69. 10.1016/j.ydbio.2013.07.026.
Bamforth, S.D., Chaudhry, B., Bennett, M., Wilson, R., Mohun, T.J., Van Mierop, L.H.,
Henderson, D.J., and Anderson, R.H. (2013). Clarification of the identity of the mammalian fifth
pharyngeal arch artery. Clin Anat 26, 173-182. 10.1002/ca.22101.
Baxendale, S.a.W., T.T. (2014). Zebrafish inner ear development and function (Academic Press.).
Berland, S., Alme, K., Brendehaug, A., Houge, G., and Hovland, R. (2011). PHF6 Deletions May
Cause Borjeson-Forssman-Lehmann Syndrome in Females. Mol Syndromol 1, 294-300.
10.1159/000330111.
Bhatt, S., Diaz, R., and Trainor, P.A. (2013). Signals and switches in Mammalian neural crest cell
differentiation. Cold Spring Harb Perspect Biol 5. 10.1101/cshperspect.a008326.
Blackburn, P.R., Campbell, J.M., Clark, K.J., and Ekker, S.C. (2013). The CRISPR
system--keeping zebrafish gene targeting fresh. Zebrafish 10, 116-118. 10.1089/zeb.2013.9999.
Boch, J., and Bonas, U. (2010). Xanthomonas AvrBs3 family-type III effectors: discovery and
function. Annu Rev Phytopathol 48, 419-436. 10.1146/annurev-phyto-080508-081936.
Bogdanove, A.J., and V oytas, D.F. (2011). TAL effectors: customizable proteins for DNA targeting.
Science 333, 1843-1846. 10.1126/science.1204094.
BORJESON, M., FORSSMAN, H., and LEHMANN, O. (1962). An X-linked, recessively
inherited syndrome characterized by grave mental deficiency, epilepsy, and endocrine disorder.
Acta Med Scand 171, 13-21. 10.1111/j.0954-6820.1962.tb04162.x.
Bosman, E.A., Penn, A.C., Ambrose, J.C., Kettleborough, R., Stemple, D.L., and Steel, K.P.
(2005). Multiple mutations in mouse Chd7 provide models for CHARGE syndrome. Hum Mol
Genet 14, 3463-3476. 10.1093/hmg/ddi375.
Bouazoune, K., and Kingston, R.E. (2012). Chromatin remodeling by the CHD7 protein is
impaired by mutations that cause human developmental disorders. Proc Natl Acad Sci U S A 109,
19238-19243. 10.1073/pnas.1213825109.
Bradshaw, L., Chaudhry, B., Hildreth, V ., Webb, S., and Henderson, D.J. (2009). Dual role for
neural crest cells during outflow tract septation in the neural crest-deficient mutant Splotch(2H). J
Anat 214, 245-257. 10.1111/j.1469-7580.2008.01028.x.

95
Cavanaugh, A.M., Huang, J., and Chen, J.N. (2015). Two developmentally distinct populations of
neural crest cells contribute to the zebrafish heart. Developmental biology 404, 103-112.
10.1016/j.ydbio.2015.06.002.
Chao, M.M., Todd, M.A., Kontny, U., Neas, K., Sullivan, M.J., Hunter, A.G., Picketts, D.J., and
Kratz, C.P. (2010). T-cell acute lymphoblastic leukemia in association with
Börjeson-Forssman-Lehmann syndrome due to a mutation in PHF6. Pediatr Blood Cancer 55,
722-724. 10.1002/pbc.22574.
Cheng, C., Deng, P.Y ., Ikeuchi, Y ., Yuede, C., Li, D., Rensing, N., Huang, J., Baldridge, D.,
Maloney, S.E., Dougherty, J.D., et al. (2018). Characterization of a Mouse Model of
Börjeson-Forssman-Lehmann Syndrome. Cell Rep 25, 1404-1414.e1406.
10.1016/j.celrep.2018.10.043.
Christophorou, M.A., Castelo-Branco, G., Halley-Stott, R.P., Oliveira, C.S., Loos, R.,
Radzisheuskaya, A., Mowen, K.A., Bertone, P., Silva, J.C., Zernicka-Goetz, M., et al. (2014).
Citrullination regulates pluripotency and histone H1 binding to chromatin. Nature 507, 104-108.
10.1038/nature12942.
Clancy, K.W., Russell, A.M., Subramanian, V ., Nguyen, H., Qian, Y ., Campbell, R.M., and
Thompson, P.R. (2017). Citrullination/Methylation Crosstalk on Histone H3 Regulates ER-Target
Gene Transcription. ACS Chem Biol 12, 1691-1702. 10.1021/acschembio.7b00241.
Cloney, K., Steele, S.L., Stoyek, M.R., Croll, R.P., Smith, F.M., Prykhozhij, S.V ., Brown, M.M.,
Midgen, C., Blake, K., and Berman, J.N. (2018). Etiology and functional validation of
gastrointestinal motility dysfunction in a zebrafish model of CHARGE syndrome. FEBS J 285,
2125-2140. 10.1111/febs.14473.
Coffman, C.R., Skoglund, P., Harris, W.A., and Kintner, C.R. (1993). Expression of an
extracellular deletion of Xotch diverts cell fate in Xenopus embryos. Cell 73, 659-671.
10.1016/0092-8674(93)90247-n.
Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X., Jiang, W.,
Marraffini, L.A., and Zhang, F. (2013). Multiplex genome engineering using CRISPR/Cas systems.
Science 339, 819-823. 10.1126/science.1231143.
Cornell, R.A., and Eisen, J.S. (2002). Delta/Notch signaling promotes formation of zebrafish
neural crest by repressing Neurogenin 1 function. Development 129, 2639-2648.
10.1242/dev.129.11.2639.

96
Corsten-Janssen, N., Kerstjens-Frederikse, W.S., du Marchie Sarvaas, G.J., Baardman, M.E.,
Bakker, M.K., Bergman, J.E., Hove, H.D., Heimdal, K.R., Rustad, C.F., Hennekam, R.C., et al.
(2013). The cardiac phenotype in patients with a CHD7 mutation. Circ Cardiovasc Genet 6,
248-254. 10.1161/CIRCGENETICS.113.000054.
Corsten-Janssen, N., and Scambler, P.J. (2017). Clinical and molecular effects of CHD7 in the
heart. Am J Med Genet C Semin Med Genet 175, 487-495. 10.1002/ajmg.c.31590.
Crawford, J., Lower, K.M., Hennekam, R.C., Van Esch, H., Mégarbané, A., Lynch, S.A., Turner,
G., and Gécz, J. (2006). Mutation screening in Borjeson-Forssman-Lehmann syndrome:
identification of a novel de novo PHF6 mutation in a female patient. J Med Genet 43, 238-243.
10.1136/jmg.2005.033084.
Cronan, M.R., and Tobin, D.M. (2019). Endogenous Tagging at the. Zebrafish 16, 324-325.
10.1089/zeb.2019.1746.
Cunningham, C.M., Bellipanni, G., Habas, R., and Balciunas, D. (2020). Deletion of morpholino
binding sites (DeMOBS) to assess specificity of morphant phenotypes. Sci Rep 10, 15366.
10.1038/s41598-020-71708-1.
Di Donato, N., Isidor, B., Lopez Cazaux, S., Le Caignec, C., Klink, B., Kraus, C., Schrock, E., and
Hackmann, K. (2014). Distinct phenotype of PHF6 deletions in females. Eur J Med Genet 57,
85-89. 10.1016/j.ejmg.2013.12.003.
Diakun, G.P., Fairall, L., and Klug, A. (1986). EXAFS study of the zinc-binding sites in the protein
transcription factor IIIA. Nature 324, 698-699. 10.1038/324698a0.
El-Brolosy, M.A., Kontarakis, Z., Rossi, A., Kuenne, C., Günther, S., Fukuda, N., Kikhi, K.,
Boezio, G.L.M., Takacs, C.M., Lai, S.L., et al. (2019). Genetic compensation triggered by mutant
mRNA degradation. Nature 568, 193-197. 10.1038/s41586-019-1064-z.
Ernst, A., Le, V .Q., Højland, A.T., Pedersen, I.S., Sørensen, T.H., Bjerregaard, L.L., Lyngbye, T.J.,
Gammelager, N.M., Krarup, H., and Petersen, M.B. (2015). The PHF6 Mutation c.1A>G; pM1V
Causes Börjeson-Forsman-Lehmann Syndrome in a Family with Four Affected Young Boys. Mol
Syndromol 6, 181-186. 10.1159/000441047.
Fabian, P., Tseng, K.C., Smeeton, J., Lancman, J.J., Dong, P.D.S., Cerny, R., and Crump, J.G.
(2020). Lineage analysis reveals an endodermal contribution to the vertebrate pituitary. Science
370, 463-467. 10.1126/science.aba4767.

97
Fairall, L., Schwabe, J.W., Chapman, L., Finch, J.T., and Rhodes, D. (1993). The crystal structure
of a two zinc-finger peptide reveals an extension to the rules for zinc-finger/DNA recognition.
Nature 366, 483-487. 10.1038/366483a0.
Flanagan, J.F., Blus, B.J., Kim, D., Clines, K.L., Rastinejad, F., and Khorasanizadeh, S. (2007).
Molecular implications of evolutionary differences in CHD double chromodomains. J Mol Biol
369, 334-342. 10.1016/j.jmb.2007.03.024.
Foroughmand, A.M., Jari, M., Kazeminezhad, S.R., Abdollahi, A., Ahmadi, L., and Heidari, M.
(2014). Genetic analysis of two STR loci (VWA and TPOX) in the Iranian province of Khuzestan.
Iran J Basic Med Sci 17, 583-587.
Franzoni, E., Booker, S.A., Parthasarathy, S., Rehfeld, F., Grosser, S., Srivatsa, S., Fuchs, H.R.,
Tarabykin, V ., Vida, I., and Wulczyn, F.G. (2015). miR-128 regulates neuronal migration,
outgrowth and intrinsic excitability via the intellectual disability gene Phf6. Elife 4.
10.7554/eLife.04263.
Frisdal, A., and Trainor, P.A. (2014). Development and evolution of the pharyngeal apparatus.
Wiley Interdiscip Rev Dev Biol 3, 403-418. 10.1002/wdev.147.
Fuhrmann, J., and Thompson, P.R. (2016). Protein Arginine Methylation and Citrullination in
Epigenetic Regulation. ACS Chem Biol 11, 654-668. 10.1021/acschembio.5b00942.
Gage, P.J., Hurd, E.A., and Martin, D.M. (2015). Mouse Models for the Dissection of CHD7
Functions in Eye Development and the Molecular Basis for Ocular Defects in CHARGE
Syndrome. Invest Ophthalmol Vis Sci 56, 7923-7930. 10.1167/iovs.15-18069.
George, R.M., Maldonado-Velez, G., and Firulli, A.B. (2020). The heart of the neural crest:
cardiac neural crest cells in development and regeneration. Development 147.
10.1242/dev.188706.
Glavic, A., Silva, F., Aybar, M.J., Bastidas, F., and Mayor, R. (2004). Interplay between Notch
signaling and the homeoprotein Xiro1 is required for neural crest induction in Xenopus embryos.
Development 131, 347-359. 10.1242/dev.00945.
Goldshmit, Y ., Sztal, T.E., Jusuf, P.R., Hall, T.E., Nguyen-Chi, M., and Currie, P.D. (2012).
Fgf-dependent glial cell bridges facilitate spinal cord regeneration in zebrafish. J Neurosci 32,
7477-7492. 10.1523/JNEUROSCI.0758-12.2012.
Guo, J., Gaj, T., and Barbas, C.F. (2010). Directed evolution of an enhanced and highly efficient
FokI cleavage domain for zinc finger nucleases. J Mol Biol 400, 96-107.
10.1016/j.jmb.2010.04.060.

98
Gécz, J., Turner, G., Nelson, J., and Partington, M. (2006). The Börjeson-Forssman-Lehman
syndrome (BFLS, MIM #301900). Eur J Hum Genet 14, 1233-1237. 10.1038/sj.ejhg.5201639.
Harrison, M.R., Feng, X., Mo, G., Aguayo, A., Villafuerte, J., Yoshida, T., Pearson, C.A.,
Schulte-Merker, S., and Lien, C.L. (2019). Late developing cardiac lymphatic vasculature
supports adult zebrafish heart function and regeneration. Elife 8. 10.7554/eLife.42762.
Howe, K., Clark, M.D., Torroja, C.F., Torrance, J., Berthelot, C., Muffato, M., Collins, J.E.,
Humphray, S., McLaren, K., Matthews, L., et al. (2013). The zebrafish reference genome sequence
and its relationship to the human genome. Nature 496, 498-503. 10.1038/nature12111.
Huang, X., and Saint-Jeannet, J.P. (2004). Induction of the neural crest and the opportunities of life
on the edge. Dev Biol 275, 1-11. 10.1016/j.ydbio.2004.07.033.
Hurd, E.A., Capers, P.L., Blauwkamp, M.N., Adams, M.E., Raphael, Y ., Poucher, H.K., and
Martin, D.M. (2007). Loss of Chd7 function in gene-trapped reporter mice is embryonic lethal and
associated with severe defects in multiple developing tissues. Mamm Genome 18, 94-104.
10.1007/s00335-006-0107-6.
Hwang, W.Y ., Fu, Y ., Reyon, D., Maeder, M.L., Tsai, S.Q., Sander, J.D., Peterson, R.T., Yeh, J.R.,
and Joung, J.K. (2013). Efficient genome editing in zebrafish using a CRISPR-Cas system. Nat
Biotechnol 31, 227-229. 10.1038/nbt.2501.
Inoue, M., Tanimoto, M., and Oda, Y . (2013). The role of ear stone size in hair cell acoustic
sensory transduction. Sci Rep 3, 2114. 10.1038/srep02114.
Isaac, K.V ., Ganske, I.M., Rottgers, S.A., Lim, S.Y ., and Mulliken, J.B. (2018). Cleft Lip and
Palate in CHARGE Syndrome: Phenotypic Features That Influence Management. Cleft Palate
Craniofac J 55, 342-347. 10.1177/1055665617738994.
Ishihara, K., Oshimura, M., and Nakao, M. (2006). CTCF-dependent chromatin insulator is linked
to epigenetic remodeling. Mol Cell 23, 733-742. 10.1016/j.molcel.2006.08.008.
Jacobs-McDaniels, N.L., and Albertson, R.C. (2011). Chd7 plays a critical role in controlling
left-right symmetry during zebrafish somitogenesis. Dev Dyn 240, 2272-2280.
10.1002/dvdy.22722.
Jahani-Asl, A., Cheng, C., Zhang, C., and Bonni, A. (2016). Pathogenesis of
Börjeson-Forssman-Lehmann syndrome: Insights from PHF6 function. Neurobiol Dis 96,
227-235. 10.1016/j.nbd.2016.09.011.

99
Jamadagni, P., Breuer, M., Schmeisser, K., Cardinal, T., Kassa, B., Parker, J.A., Pilon, N., Samarut,
E., and Patten, S.A. (2021). Chromatin remodeller CHD7 is required for GABAergic neuron
development by promoting PAQR3 expression. EMBO Rep, e50958. 10.15252/embr.202050958.
Janssen, N., Bergman, J.E., Swertz, M.A., Tranebjaerg, L., Lodahl, M., Schoots, J., Hofstra, R.M.,
van Ravenswaaij-Arts, C.M., and Hoefsloot, L.H. (2012). Mutation update on the CHD7 gene
involved in CHARGE syndrome. Hum Mutat 33, 1149-1160. 10.1002/humu.22086.
Jinek, M., Chylinski, K., Fonfara, I., Hauer, M., Doudna, J.A., and Charpentier, E. (2012). A
programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337,
816-821. 10.1126/science.1225829.
Jopling, C., Sleep, E., Raya, M., Martí, M., Raya, A., and Izpisúa Belmonte, J.C. (2010). Zebrafish
heart regeneration occurs by cardiomyocyte dedifferentiation and proliferation. Nature 464,
606-609. 10.1038/nature08899.
Kague, E., Gallagher, M., Burke, S., Parsons, M., Franz-Odendaal, T., and Fisher, S. (2012).
Skeletogenic fate of zebrafish cranial and trunk neural crest. PLoS One 7, e47394.
10.1371/journal.pone.0047394.
Kan, R., Jin, M., Subramanian, V ., Causey, C.P., Thompson, P.R., and Coonrod, S.A. (2012).
Potential role for PADI-mediated histone citrullination in preimplantation development. BMC Dev
Biol 12, 19. 10.1186/1471-213X-12-19.
Kasper, B.S., Dörfler, A., Di Donato, N., Kasper, E.M., Wieczorek, D., Hoyer, J., and Zweier, C.
(2017). Central nervous system anomalies in two females with Borjeson-Forssman-Lehmann
syndrome. Epilepsy Behav 69, 104-109. 10.1016/j.yebeh.2017.01.022.
Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., and Schilling, T.F. (1995). Stages of
embryonic development of the zebrafish. Dev Dyn 203, 253-310. 10.1002/aja.1002030302.
Kimmel, C.B., Miller, C.T., and Moens, C.B. (2001). Specification and morphogenesis of the
zebrafish larval head skeleton. Dev Biol 233, 239-257. 10.1006/dbio.2001.0201.
Knight, R.D., and Schilling, T.F. (2006). Cranial neural crest and development of the head skeleton.
Adv Exp Med Biol 589, 120-133. 10.1007/978-0-387-46954-6_7.
Kobayashi, I., Kobayashi-Sun, J., Kim, A.D., Pouget, C., Fujita, N., Suda, T., and Traver, D.
(2014). Jam1a-Jam2a interactions regulate haematopoietic stem cell fate through Notch signalling.
Nature 512, 319-323. 10.1038/nature13623.

100
Layman, W.S., Hurd, E.A., and Martin, D.M. (2010). Chromodomain proteins in development:
lessons from CHARGE syndrome. Clin Genet 78, 11-20. 10.1111/j.1399-0004.2010.01446.x.
Layman, W.S., McEwen, D.P., Beyer, L.A., Lalani, S.R., Fernbach, S.D., Oh, E., Swaroop, A.,
Hegg, C.C., Raphael, Y ., Martens, J.R., and Martin, D.M. (2009). Defects in neural stem cell
proliferation and olfaction in Chd7 deficient mice indicate a mechanism for hyposmia in human
CHARGE syndrome. Hum Mol Genet 18, 1909-1923. 10.1093/hmg/ddp112.
Lee, G., Chambers, S.M., Tomishima, M.J., and Studer, L. (2010). Derivation of neural crest cells
from human pluripotent stem cells. Nat Protoc 5, 688-701. 10.1038/nprot.2010.35.
Lewis, A.E., Vasudevan, H.N., O'Neill, A.K., Soriano, P., and Bush, J.O. (2013). The widely used
Wnt1-Cre transgene causes developmental phenotypes by ectopic activation of Wnt signaling.
Dev Biol 379, 229-234. 10.1016/j.ydbio.2013.04.026.
Li, H.H., Huang, P., Dong, W., Zhu, Z.Y ., and Liu, D. (2013). [A brief history of zebrafish
research--toward biomedicine]. Yi Chuan 35, 410-420. 10.3724/sp.j.1005.2013.00410.
Li, T., Huang, S., Jiang, W.Z., Wright, D., Spalding, M.H., Weeks, D.P., and Yang, B. (2011). TAL
nucleases (TALNs): hybrid proteins composed of TAL effectors and FokI DNA-cleavage domain.
Nucleic Acids Res 39, 359-372. 10.1093/nar/gkq704.
Lian, X., Zhang, J., Azarin, S.M., Zhu, K., Hazeltine, L.B., Bao, X., Hsiao, C., Kamp, T.J., and
Palecek, S.P. (2013). Directed cardiomyocyte differentiation from human pluripotent stem cells by
modulating Wnt/β-catenin signaling under fully defined conditions. Nat Protoc 8, 162-175.
10.1038/nprot.2012.150.
Liu, H., and Liu, Z.Z. (2020). Aggressive-like behavior and increased glycine transporters in a
zebrafish model of CHARGE syndrome. Behav Brain Res 378, 112293.
10.1016/j.bbr.2019.112293.
Liu, Y ., Harmelink, C., Peng, Y ., Chen, Y ., Wang, Q., and Jiao, K. (2014a). CHD7 interacts with
BMP R-SMADs to epigenetically regulate cardiogenesis in mice. Hum Mol Genet 23, 2145-2156.
10.1093/hmg/ddt610.
Liu, Z., Li, F., Ruan, K., Zhang, J., Mei, Y ., Wu, J., and Shi, Y . (2014b). Structural and functional
insights into the human Börjeson-Forssman-Lehmann syndrome-associated protein PHF6. J Biol
Chem 289, 10069-10083. 10.1074/jbc.M113.535351.
Liu, Z.Z., Guo, J., Lu, Y ., Liu, W., Fu, X., Yao, T., Zhou, Y ., and Xu, H.A. (2019). Sema3E is
required for migration of cranial neural crest cells in zebrafish: Implications for the pathogenesis
of CHARGE syndrome. Int J Exp Pathol 100, 234-243. 10.1111/iep.12331.

101
Liu, Z.Z., Wang, Z.L., Choi, T.I., Huang, W.T., Wang, H.T., Han, Y .Y ., Zhu, L.Y ., Kim, H.T., Choi,
J.H., Lee, J.S., et al. (2018). Chd7 Is Critical for Early T-Cell Development and Thymus
Organogenesis in Zebrafish. Am J Pathol 188, 1043-1058. 10.1016/j.ajpath.2017.12.005.
Loontiens, S., Dolens, A.C., Strubbe, S., Van de Walle, I., Moore, F.E., Depestel, L., Vanhauwaert,
S., Matthijssens, F., Langenau, D.M., Speleman, F., et al. (2020). PHF6 Expression Levels Impact
Human Hematopoietic Stem Cell Differentiation. Front Cell Dev Biol 8, 599472.
10.3389/fcell.2020.599472.
Lower, K.M., Turner, G., Kerr, B.A., Mathews, K.D., Shaw, M.A., Gedeon, A.K., Schelley, S.,
Hoyme, H.E., White, S.M., Delatycki, M.B., et al. (2002). Mutations in PHF6 are associated with
Börjeson-Forssman-Lehmann syndrome. Nat Genet 32, 661-665. 10.1038/ng1040.
Mao, A., Zhang, M., Liu, J., Cao, Y ., and Wang, Q. (2019). PDGF signaling from pharyngeal
pouches promotes arch artery morphogenesis. J Genet Genomics 46, 551-559.
10.1016/j.jgg.2019.11.004.
Marraffini, L.A., and Sontheimer, E.J. (2010). CRISPR interference: RNA-directed adaptive
immunity in bacteria and archaea. Nat Rev Genet 11, 181-190. 10.1038/nrg2749.
Mastronardi, F.G., Wood, D.D., Mei, J., Raijmakers, R., Tseveleki, V ., Dosch, H.M., Probert, L.,
Casaccia-Bonnefil, P., and Moscarello, M.A. (2006). Increased citrullination of histone H3 in
multiple sclerosis brain and animal models of demyelination: a role for tumor necrosis
factor-induced peptidylarginine deiminase 4 translocation. J Neurosci 26, 11387-11396.
10.1523/JNEUROSCI.3349-06.2006.
Mayor, R., Guerrero, N., and Martínez, C. (1997). Role of FGF and noggin in neural crest
induction. Dev Biol 189, 1-12. 10.1006/dbio.1997.8634.
McRae, H.M., Eccles, S., Whitehead, L., Alexander, W.S., Gécz, J., Thomas, T., and V oss, A.K.
(2020). Downregulation of the GHRH/GH/IGF1 axis in a mouse model of
Börjeson-Forssman-Lehman syndrome. Development 147. 10.1242/dev.187021.
Meng, Z.Z., Liu, W., Xia, Y ., Yin, H.M., Zhang, C.Y ., Su, D., Yan, L.F., Gu, A.H., and Zhou, Y .
(2017). The pro-inflammatory signalling regulator Stat4 promotes vasculogenesis of great vessels
derived from endothelial precursors. Nat Commun 8, 14640. 10.1038/ncomms14640.
Moorman, A., Webb, S., Brown, N.A., Lamers, W., and Anderson, R.H. (2003). Development of
the heart: (1) formation of the cardiac chambers and arterial trunks. Heart 89, 806-814.
10.1136/heart.89.7.806.

102
Mork, L., and Crump, G. (2015). Zebrafish Craniofacial Development: A Window into Early
Patterning. Curr Top Dev Biol 115, 235-269. 10.1016/bs.ctdb.2015.07.001.
Muth, A., Subramanian, V ., Beaumont, E., Nagar, M., Kerry, P., McEwan, P., Srinath, H., Clancy,
K., Parelkar, S., and Thompson, P.R. (2017). Development of a Selective Inhibitor of Protein
Arginine Deiminase 2. J Med Chem 60, 3198-3211. 10.1021/acs.jmedchem.7b00274.
Muñoz Caro, T., Hermosilla, C., Silva, L.M., Cortes, H., and Taubert, A. (2014). Neutrophil
extracellular traps as innate immune reaction against the emerging apicomplexan parasite
Besnoitia besnoiti. PLoS One 9, e91415. 10.1371/journal.pone.0091415.
Nagelberg, D., Wang, J., Su, R., Torres-Vázquez, J., Targoff, K.L., Poss, K.D., and Knaut, H.
(2015). Origin, Specification, and Plasticity of the Great Vessels of the Heart. Curr Biol 25,
2099-2110. 10.1016/j.cub.2015.06.076.
Ogier, J.M., Carpinelli, M.R., Arhatari, B.D., Symons, R.C., Kile, B.T., and Burt, R.A. (2014).
CHD7 deficiency in "Looper", a new mouse model of CHARGE syndrome, results in ossicle
malformation, otosclerosis and hearing impairment. PLoS One 9, e97559.
10.1371/journal.pone.0097559.
Okuno, H., Renault Mihara, F., Ohta, S., Fukuda, K., Kurosawa, K., Akamatsu, W., Sanosaka, T.,
Kohyama, J., Hayashi, K., Nakajima, K., et al. (2017). CHARGE syndrome modeling using
patient-iPSCs reveals defective migration of neural crest cells harboring CHD7 mutations. Elife 6.
10.7554/eLife.21114.
Ovcharenko, I., Nobrega, M.A., Loots, G.G., and Stubbs, L. (2004). ECR Browser: a tool for
visualizing and accessing data from comparisons of multiple vertebrate genomes. Nucleic Acids
Res 32, W280-286. 10.1093/nar/gkh355.
Pagon, R.A., Graham, J.M., Zonana, J., and Yong, S.L. (1981). Coloboma, congenital heart disease,
and choanal atresia with multiple anomalies: CHARGE association. J Pediatr 99, 223-227.
10.1016/s0022-3476(81)80454-4.
Patten, S.A., Jacobs-McDaniels, N.L., Zaouter, C., Drapeau, P., Albertson, R.C., and Moldovan, F.
(2012). Role of Chd7 in zebrafish: a model for CHARGE syndrome. PLoS One 7, e31650.
10.1371/journal.pone.0031650.
Payne, S., Burney, M.J., McCue, K., Popal, N., Davidson, S.M., Anderson, R.H., and Scambler, P.J.
(2015). A critical role for the chromatin remodeller CHD7 in anterior mesoderm during
cardiovascular development. Developmental biology 405, 82-95. 10.1016/j.ydbio.2015.06.017.

103
Prykhozhij, S.V ., Steele, S.L., Razaghi, B., and Berman, J.N. (2017). A rapid and effective method
for screening, sequencing and reporter verification of engineered frameshift mutations in zebrafish.
Dis Model Mech 10, 811-822. 10.1242/dmm.026765.
Randall, V ., McCue, K., Roberts, C., Kyriakopoulou, V ., Beddow, S., Barrett, A.N., Vitelli, F.,
Prescott, K., Shaw-Smith, C., Devriendt, K., et al. (2009). Great vessel development requires
biallelic expression of Chd7 and Tbx1 in pharyngeal ectoderm in mice. J Clin Invest 119,
3301-3310. 10.1172/JCI37561.
Ritter, K.E., and Martin, D.M. (2019). Neural crest contributions to the ear: Implications for
congenital hearing disorders. Hear Res 376, 22-32. 10.1016/j.heares.2018.11.005.
Rocha, M., Singh, N., Ahsan, K., Beiriger, A., and Prince, V .E. (2020). Neural crest development:
insights from the zebrafish. Dev Dyn 249, 88-111. 10.1002/dvdy.122.
Ryan, D.P., Sundaramoorthy, R., Martin, D., Singh, V ., and Owen-Hughes, T. (2011). The
DNA-binding domain of the Chd1 chromatin-remodelling enzyme contains SANT and SLIDE
domains. EMBO J 30, 2596-2609. 10.1038/emboj.2011.166.
Sadaghiani, B., and Thiébaud, C.H. (1987). Neural crest development in the Xenopus laevis
embryo, studied by interspecific transplantation and scanning electron microscopy. Dev Biol 124,
91-110. 10.1016/0012-1606(87)90463-5.
Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. (2012). NIH Image to ImageJ: 25 years of
image analysis. Nat Methods 9, 671-675. 10.1038/nmeth.2089.
Schnetz, M.P., Handoko, L., Akhtar-Zaidi, B., Bartels, C.F., Pereira, C.F., Fisher, A.G., Adams,
D.J., Flicek, P., Crawford, G.E., Laframboise, T., et al. (2010). CHD7 targets active gene enhancer
elements to modulate ES cell-specific gene expression. PLoS Genet 6, e1001023.
10.1371/journal.pgen.1001023.
Schulz, Y ., Wehner, P., Opitz, L., Salinas-Riester, G., Bongers, E.M., van Ravenswaaij-Arts, C.M.,
Wincent, J., Schoumans, J., Kohlhase, J., Borchers, A., and Pauli, S. (2014). CHD7, the gene
mutated in CHARGE syndrome, regulates genes involved in neural crest cell guidance. Hum
Genet 133, 997-1009. 10.1007/s00439-014-1444-2.
Serrano, M.L.A., Demarest, B.L., Tone-Pah-Hote, T., Tristani-Firouzi, M., and Yost, H.J. (2019).
Inhibition of Notch signaling rescues cardiovascular development in Kabuki Syndrome. PLoS
Biol 17, e3000087. 10.1371/journal.pbio.3000087.
Sive, H.L., Grainger, R.M., and Harland, R.M. (2010). Microinjection of Xenopus embryos. Cold
Spring Harb Protoc 2010, pdb.ip81. 10.1101/pdb.ip81.

104
Sperry, E.D., Hurd, E.A., Durham, M.A., Reamer, E.N., Stein, A.B., and Martin, D.M. (2014). The
chromatin remodeling protein CHD7, mutated in CHARGE syndrome, is necessary for proper
craniofacial and tracheal development. Dev Dyn 243, 1055-1066. 10.1002/dvdy.24156.
Stooke-Vaughan, G.A., Obholzer, N.D., Baxendale, S., Megason, S.G., and Whitfield, T.T. (2015).
Otolith tethering in the zebrafish otic vesicle requires Otogelin and α-Tectorin. Development 142,
1137-1145. 10.1242/dev.116632.
Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse
embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676.
10.1016/j.cell.2006.07.024.
Tang, W., Martik, M.L., Li, Y ., and Bronner, M.E. (2019). Cardiac neural crest contributes to
cardiomyocytes in amniotes and heart regeneration in zebrafish. Elife 8. 10.7554/eLife.47929.
Teame, T., Zhang, Z., Ran, C., Zhang, H., Yang, Y ., Ding, Q., Xie, M., Gao, C., Ye, Y ., Duan, M.,
and Zhou, Z. (2019). The use of zebrafish (Anim Front 9, 68-77. 10.1093/af/vfz020.
Thisse, C., and Thisse, B. (2008). High-resolution in situ hybridization to whole-mount zebrafish
embryos. Nat Protoc 3, 59-69. 10.1038/nprot.2007.514.
Todd, M.A., Ivanochko, D., and Picketts, D.J. (2015). PHF6 Degrees of Separation: The
Multifaceted Roles of a Chromatin Adaptor Protein. Genes (Basel) 6, 325-352.
10.3390/genes6020325.
Turner, G., Lower, K.M., White, S.M., Delatycki, M., Lampe, A.K., Wright, M., Smith, J.C., Kerr,
B., Schelley, S., Hoyme, H.E., et al. (2004). The clinical picture of the
Börjeson-Forssman-Lehmann syndrome in males and heterozygous females with PHF6 mutations.
Clin Genet 65, 226-232. 10.1111/j.0009-9163.2004.00215.x.
Vallin, J., Thuret, R., Giacomello, E., Faraldo, M.M., Thiery, J.P., and Broders, F. (2001). Cloning
and characterization of three Xenopus slug promoters reveal direct regulation by Lef/beta-catenin
signaling. J Biol Chem 276, 30350-30358. 10.1074/jbc.M103167200.
Van Vlierberghe, P., Palomero, T., Khiabanian, H., Van der Meulen, J., Castillo, M., Van Roy, N.,
De Moerloose, B., Philippé, J., González-García, S., Toribio, M.L., et al. (2010). PHF6 mutations
in T-cell acute lymphoblastic leukemia. Nat Genet 42, 338-342. 10.1038/ng.542.
Verstraeten, B., Sanders, E., van Hengel, J., and Huysseune, A. (2010). Zebrafish teeth as a model
for repetitive epithelial morphogenesis: dynamics of E-cadherin expression. BMC Dev Biol 10, 58.
10.1186/1471-213X-10-58.

105
Visootsak, J., Rosner, B., Dykens, E., Schwartz, C., Hahn, K., White, S.M., Szeftel, R., and
Graham, J.M. (2004). Clinical and behavioral features of patients with
Borjeson-Forssman-Lehmann syndrome with mutations in PHF6. J Pediatr 145, 819-825.
10.1016/j.jpeds.2004.07.041.
V ouillot, L., Thélie, A., and Pollet, N. (2015). Comparison of T7E1 and surveyor mismatch
cleavage assays to detect mutations triggered by engineered nucleases. G3 (Bethesda) 5, 407-415.
10.1534/g3.114.015834.
Whitlock, K.E., and Westerfield, M. (2000). The olfactory placodes of the zebrafish form by
convergence of cellular fields at the edge of the neural plate. Development 127, 3645-3653.
10.1242/dev.127.17.3645.
Wu, R.S., Lam, I.I., Clay, H., Duong, D.N., Deo, R.C., and Coughlin, S.R. (2018). A Rapid
Method for Directed Gene Knockout for Screening in G0 Zebrafish. Dev Cell 46, 112-125.e114.
10.1016/j.devcel.2018.06.003.
Yan, S., Thienthanasit, R., Chen, D., Engelen, E., Brühl, J., Crossman, D.K., Kesterson, R., Wang,
Q., Bouazoune, K., and Jiao, K. (2020). CHD7 regulates cardiovascular development through
ATP-dependent and -independent activities. Proc Natl Acad Sci U S A 117, 28847-28858.
10.1073/pnas.2005222117.
Yao, Q., DeSmidt, A.A., Tekin, M., Liu, X., and Lu, Z. (2016). Hearing Assessment in Zebrafish
During the First Week Postfertilization. Zebrafish 13, 79-86. 10.1089/zeb.2015.1166.
Zhang, C., Mejia, L.A., Huang, J., Valnegri, P., Bennett, E.J., Anckar, J., Jahani-Asl, A., Gallardo,
G., Ikeuchi, Y ., Yamada, T., et al. (2013). The X-linked intellectual disability protein PHF6
associates with the PAF1 complex and regulates neuronal migration in the mammalian brain.
Neuron 78, 986-993. 10.1016/j.neuron.2013.04.021.
Zhang, X., Bolt, M., Guertin, M.J., Chen, W., Zhang, S., Cherrington, B.D., Slade, D.J., Dreyton,
C.J., Subramanian, V ., Bicker, K.L., et al. (2012). Peptidylarginine deiminase 2-catalyzed histone
H3 arginine 26 citrullination facilitates estrogen receptor α target gene activation. Proc Natl Acad
Sci U S A 109, 13331-13336. 10.1073/pnas.1203280109.
Zu, Y ., Tong, X., Wang, Z., Liu, D., Pan, R., Li, Z., Hu, Y ., Luo, Z., Huang, P., Wu, Q., et al. (2013).
TALEN-mediated precise genome modification by homologous recombination in zebrafish. Nat
Methods 10, 329-331. 10.1038/nmeth.2374.

106
Zweier, C., Kraus, C., Brueton, L., Cole, T., Degenhardt, F., Engels, H., Gillessen-Kaesbach, G.,
Graul-Neumann, L., Horn, D., Hoyer, J., et al. (2013). A new face of Borjeson-Forssman-Lehmann
syndrome? De novo mutations in PHF6 in seven females with a distinct phenotype. J Med Genet
50, 838-847. 10.1136/jmedgenet-2013-101918.
Zweier, C., Rittinger, O., Bader, I., Berland, S., Cole, T., Degenhardt, F., Di Donato, N.,
Graul-Neumann, L., Hoyer, J., Lynch, S.A., et al. (2014). Females with de novo aberrations in
PHF6: clinical overlap of Borjeson-Forssman-Lehmann with Coffin-Siris syndrome. Am J Med
Genet C Semin Med Genet 166C, 290-301. 10.1002/ajmg.c.31408.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Opposing function of chd7 and phf6 in zebrafish craniofacial development

PDF

Characterization of zebrafish phf6 CRISPR mutant phenotypes

PDF

Epigenetic checks and balances: PHF6 activity restricts neural crest migration

PDF

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

PDF

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

PDF

Zebrafish as a blueprint for cardiac regeneration

PDF

Craniofacial skull joint and temporomandibular joint (TMJ) in homeostasis and disease

PDF

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

PDF

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

PDF

Characterization of Cxcr7 in zebrafish cardiac lymphatic vessel development

PDF

Functionalization of scaffolds with anti-BMP-2 antibody: role in antibody mediated osseous regeneration (AMOR)

PDF

Amelogenin-ameloblastin protein interaction and function in dental enamel formation

PDF

Adaptive patterning during skin formation: assembly of vasculature and adipose tissue

PDF

Interaction of epigenetics and SMAD signaling in stem cells and diseases

PDF

Proliferation and maturation events in second heart field cells during cardiovascular development activated by the Delta like ligand-4 mediated notch signaling

PDF

Role of MMP-20 in preventing protein occlusion in enamel apatite crystals: relevance in enamel biomineralization and biomimetics

Asset Metadata

Creator Sun, Yuhan (author)

Core Title Chromatin remodeling factors Chd7 and Phf6 in craniofacial and heart development

School School of Dentistry

Degree Doctor of Philosophy

Degree Program Craniofacial Biology

Degree Conferral Date 2022-08

Publication Date 07/25/2022

Defense Date 06/09/2022

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

Tag CHD7,craniofacial,CRISPR/Cas9,iPSC,neural crest,OAI-PMH Harvest,phf6,ventral aorta,zebrafish

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Lien, Ching-Ling (committee chair), Chen, Jianfu (committee member), Crump, Gage (committee member)

Creator Email yuhansun@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC111375415

Unique identifier UC111375415

Legacy Identifier etd-SunYuhan-10973

Document Type Dissertation

Format application/pdf (imt)

Rights Sun, Yuhan

Type texts

Source 20220728-usctheses-batch-962 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu