University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Derivation, expansion and characterization of human
hippocampal primordial cells from normal and
diseased iPSCs
by
Simiao Wang
A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOCHEMISTRY AND MOLECULAR MEDICINE)
December 2018
Copyright 2018 Simiao Wang
1
Table of Contents
Abstract
.......................................................................................................................................................................................
3
Acknowledgments
..............................................................................................................................................................
4
Introduction
..............................................................................................................................................................................
5
1. CHARGE syndrome, neural crest cells and CHD7 mutations
.....................................................
5
2. CHD7 also regulates neurogenesis
.................................................................................................................
9
3. Neurogenesis and hippocampal neural stem cells
..............................................................................
10
4. Establishing WNT8B
en
-GFP reporter iPSCs to study human hippocampal
primordial cells
....................................................................................................................................................................
13
Materials and Methods
..................................................................................................................................................
26
Results
.......................................................................................................................................................................................
30
1. Defects in CHARGE WNT8B-GFP positive hippocampal neural stem cells
...................
30

2. CHIC-35 rescued the morphology of neural spheres, WNT8B enhancer activation
and development, migration and differentiation of neural crest cells
.........................................
34
2.1 CHIC-35, an inhibitor of SIRT1, rescued defects in CHARGE zebrafish model
...
34
2.2 CHIC-35 rescued the defects in CHARGE neural spheres and WNT8B-GFP
positive cells
..........................................................................................................................................................
34
2.3 CHIC-35 rescued the migration and differentiation of neural crest cells
...................
36
3. In vitro expansion of hippocampal primordial cells
............................................................................
42
3.1 The maintenance of WNT8B-GFP positive NSCs
....................................................................
42
3.2 In vitro expansion of WNT8B-GFP positive cells
.......................................................................
51
3.3 Expanded WNT8B-GFP positive cells expressed neural stem/progenitor
markers
........................................................................................................................................................................
53
2
3.4 WNT8B-GFP positive cells generated WNT8B-GFP negative neural
precursor cells
..............................................................................................................................................................
54
Discussion
...............................................................................................................................................................................
56
Citations
....................................................................................................................................................................................
58

3
Abstract
CHARGE syndrome is a rare genetic syndrome and the majority of CHARGE patients
have defective neural crest cells (NCCs) due to a mutation in the gene CHD7. However,
CHD7 is not only essential for the development of NCCs, it also regulates neurogenesis.
This led us to hypothesize that CHARGE patients may also have defective hippocampal
neural stem cells (hip-NSCs). To study hip-NSCs, our lab previously established a
WNT8B
en
-GFP reporter system in human stem cells. Using two WNT8B
en
-GFP CHARGE
patient iPSC lines, we showed the first evidence of defective human hip-NSCs in
CHARGE syndrome in vitro. A candidate drug, CHIC-35, rescued not only the migration
and differentiation of CHARGE NCCs, but also the defective CHARGE hip-NSCs
generated from the patient iPSCs. This study provides new insight into the mechanism of
CHARGE syndrome and a potential therapeutic for CHARGE patients that may improve
their brain functions. In addition, we also achieved the first in vitro expansion of human
WNT8B
en
-GFP positive hip-NSCs using normal WNT8B
en
-GFP iPSC lines. We found that
a potential Notch inhibitor, DAPT, promoted the self-renewal and maintenance of human
hip-NSCs in vitro, which indicates that the Notch ligand DLK1 may play an essential role
in self-renewal and maintenance of primary hip-NSCs. This study provides a promising
resource of human hippocampal primordial cells, not only for future stem cell
transplantation therapy, but also for further study of the earliest hip-NSCs. We implicate
the possible role of the DLK1 and Notch signaling pathway in self-renewal of the earliest
hip-NSCs. In conclusion, we utilized normal and CHARGE WNT8B
en
-GFP reporter lines
to study hip-NSCs in development and disease in vitro to establish potential therapeutics
and new insights into the mechanisms of human diseases.
4
Acknowledgements
I would like to express my sincere gratitude to my mentor Dr. Ruchi Bajpai for the
continuous support of my master study and research, and my thesis committee members
Dr. Pragna Patel and Dr. Michael Bonaguidi, and our collaborators Dr. Chetana
Sachidanadan and Zainab Asad for sharing unpublished data on compounds identified
from their screen to rescue CHARGE Syndrome. Also, I want to thank all lab members in
Dr. Bajpai’s lab, especially the WNT8B project team, Jennifer Oki and Krystal Mendez.
Finally, I want to thank my parents for everything they have done to help me to realize my
dream to become a scientist.

5
Introduction
1. CHARGE syndrome, neural crest cells and CHD7 mutations
1.1 CHARGE syndrome is a rare genetic syndrome with a constellation of
abnormalities
CHARGE syndrome (OMIM 214800) is a rare genetic syndrome diagnosed in
1/10,000 live births (Issekutz et al. 2005; Basson and van Ravenswaaij-Arts 2015).
CHARGE syndrome patients suffer from a combination of malformations involving
multiple organs. In 1981, Pagon et al. reported 21 cases of patients with choanal atresia,
ocular coloboma, or both, along with other abnormalities. They coined the acronym
CHARGE for this syndrome after the six major anomalies associated with it, namely
Coloboma, Heart defects, Choanal atresia, Retarded growth and development,
Genitourinary malformation, Ear anomalies, and deafness (Pagon et al. 1981). However,
other clinical features were also observed among CHARGE patients, including cleft lip,
cleft palate, esophageal atresia (EA), tracheoesophageal fistula (TEF), and facial nerve
paralysis. In most cases, CHARGE syndrome patients are diagnosed postnatally
according to criteria summarized by Hsu et al., as shown in Table 1 (Hsu et al. 2014),
which were first outlined by Blake et al. (Blake et al. 1998) and then modified by Verloes
et al. (Verloes 2005). Hsu et al. (2014) also summarized the multiple notable phenotypic
features of CHARGE syndrome (Table 2) and some typical facial features (Figure 1). So
far, the most common clinical features for diagnosis include coloboma and ophthalmic
features, cardiac malformations, choanal atresia and other upper airway abnormalities,
growth and developmental retardation, genital hypoplasia, ear malformation and hearing
6
defects, cranial nerve abnormalities, behavioral and sleep-related disabilities, other
potential endocrine issues, infections, and immune deficiency (Hsu et al. 2014).

Table 1. CHARGE syndrome diagnostic criteria, revised and updated. Table 2.
Phenotypic features of CHARGE syndrome.(Adapted from Hsu et al. 2014).

Figure 1. Clinical facial features of CHARGE syndrome. (a) Ten-year-old female patient
with cleft lip and palate (repaired), mild left facial palsy, and broad forehead (b) Six-year-
old male with cleft palate and square face with mild right facial palsy (c) 2.5-year-old male
with left-sided microphthalmia, square face with broad forehead, and broad nasal root
(Adapted from Hsu et al. 2014)
7
1.2 Defective neural crest cells due to CHD7 mutations cause CHARGE syndrome
In 1985, Siebert et al. first hypothesized that abnormalities in the development,
migration and cell-cell interaction of neural crest cells (NCCs) may result in the collection
of defects in multiple tissues involved in CHARGE syndrome (Siebert, Graham, and
MacDonald 1985). This was the first clue to the role of defective NCCs in CHARGE
syndrome. NCCs are multipotent cells originating from the embryonic ectoderm with
exceptional migratory potential that produce various tissues, including melanocytes,
peripheral nerves, ganglia, connective tissues, bones, and cartilage (Pauli, Bajpai, and
Borchers 2017; Dupin et al. 2018). The hypothesis that the constellation of abnormalities
of CHARGE syndrome is caused by defective NCCs was not supported until the CHD7
(chromodomain helicase DNA-binding proteins 7) gene was identified as being commonly
mutated in CHARGE patients. In 2004, Vissers et al. identified ten heterozygous
mutations in CHD7 in CHARGE patients, which led to the indication of haplosufficiency
of CHD7 as accounting for most cases of CHARGE (Vissers et al. 2004). Defective
tissues in the eye, heart, ear, and craniofacial region in Chd7 heterozygous mutant mice
were found to have similar phenotypes to those observed in CHARGE patients. In
zebrafish, the presence of chd7 has been shown to be essential for proper neural, retinal,
and vertebral development. Absence of chd7 in zebrafish causes abnormalities similar to
those observed in CHARGE patients. These in vivo studies using different animal models
further confirmed that mutations in CHD7 result in CHARGE syndrome (Randall et al.
2009; Patten et al. 2012; Bosman et al. 2005). Additionally, loss-of-function studies of
CHD7 firmly connected NCC defects to the abnormalities seen in CHARGE syndrome
both in vitro and in vivo, which at the same time demonstrated the essential role of CHD7
8
in the formation, development and migration of NCCs. Bajpai et al. (2010) first
demonstrated that CHD7 is crucial for the formation and migration of NCCs in both
humans and Xenopus. In conjunction with PBAF, CHD7 controls NCC formation by
activating the core components of neural crest transcriptional circuitry. They also showed
that morpholino-mediated knockdown of Chd7 in Xenopus results in defects similar to
major features of CHARGE syndrome (Bajpai et al. 2010). Conditional deletion of Chd7
in migrating NCCs resulted in severe craniofacial defects in mice (Sperry et al. 2014).
Recently, via a chd7-knockdown CHARGE zebrafish model, Asad et al. showed a wide
spectrum of defects in both NCCs and neural crest-derived cells and tissues, including
myelinating Schwann cells, pigment cells, neurons, and the craniofacial skeleton (Asad
et al. 2016). Therefore, CHD7 is the key to understanding the formation, development,
and migration of NCCs and how defects of NCCs due to CHD7 mutations lead to
CHARGE syndrome.

As an ATP-dependent chromatin remodeler, CHD7 utilizes ATP hydrolysis to regulate
transcription by chromatin remodeling (Li, Carey, and Workman 2007). The CHD7 gene,
located on human chromosome 8 (8q12), consists of 38 exons, has a genomic size of
188kb and its encoded protein contains 2997 amino acids (Schnetz et al. 2009). CHD7 is
highly conserved. The protein contains the following domains: a chromatin organization
modifier (chromodomain), helicase C, helicase N, DEAD-like helicase superfamily
(DEXDc), Switching-defective protein 3 (SANT domain) and Brahma and Kismet domain
(BRK domain). Several types of pathogenic mutations have been found in CHD7,
including nonsense mutations, frameshift deletions or insertions, splice site and missense
9
mutations, and small inframe deletions (Janssen et al. 2012). Previously, our lab
established two CHARGE patient iPSC lines: CHD7
1036X
and CHD7
1701X
for the study of
CHD7, NCCs, and CHARGE syndrome in vitro. The structure of CHD7 protein is shown
in Figure 2 and the mutations in the two CHARGE patient iPSC lines we use in our lab
are indicated with arrows.

Figure 2. The CHD7 protein and the mutations of CHARGE patient iPSC lines. The CHD7
protein contains 2997 amino acids and has a chromatin organization modifier
(chromodomain), helicase N-lobe (helicase N), DEAD-like helicase superfamily (DEXDc),
helicase C-lobe (helicase C), switching-defective protein 3 (SANT domain), Brahma and
Kismet domain (BRK domain), adaptor 2, nuclear receptor corepressor and transcription
factor IIIB domain. The mutations in the CHARGE patient iPSC lines used in our lab are
indicated with the arrows. (Adapted from Smart Protein:
http://smart.emblheidelberg.de/smart/set_mode.cgi?NORMAL=1)

2. CHD7 also regulates neurogenesis
The function of CHD7 is not limited to NCC formation, development, migration, and
differentiation. It is also involved in neurogenesis and the development of neural stem
cells (NSCs). CHD7 cooperates with SOX2, a neural stem cell regulator, as a
transcriptional cofactor to regulate genes that are mutated in human syndromes such as
CHARGE syndrome, indicating a role for CHD7 in neurogenesis (Engelen et al. 2011).
Micucci et al. (2014) demonstrated that Chd7 regulates the development of NSCs
localized in the subventricular zone using a mouse model of CHARGE syndrome. Chd7
deficiency resulted in self-renewal and proliferation defects in the subventricular zone
10
NSCs, indicating the essential role of Chd7 in the neurogenesis and NSC development
(Micucci et al. 2014). Feng et al. (2013) demonstrated the essential role of Chd7 in
neurogenesis in adult mice. They found that Chd7 is specifically enriched in the
neurogenic niches, namely the subgranular zone and subventricular zone. NSC-specific
Chd7 deficiency caused decreased neuronal differentiation in both the adult subgranular
zone and subventricular zone, providing new insight into CHARGE syndrome (Feng et al.
2013). Via an inducible knockout mouse model, Jones et al. (2015) showed the crucial
role of Chd7 in the maintenance of NSC quiescence. Deletion of Chd7 in adult NSCs in
mice resulted in the loss of hippocampal NSC quiescence, which led to the depletion of
the NSC pool. They demonstrated that the expression of Hes5, a downstream target of
the Notch signaling pathway, as well as some cell cycle genes are regulated by Chd7 as
an upstream effector, indicating that Chd7 maintains the quiescent NSCs by regulating
the cell cycle and Notch signaling pathway (Jones et al. 2015). All these previous studies
led us to hypothesize that CHARGE patients have defective hippocampal NSCs which
affect their neurogenesis and brain function due to the mutation of CHD7.

3. Neurogenesis and hippocampal neural stem cells
Neurogenesis in the hippocampal dentate gyrus region in the adult human brain
contributes to brain functions such as learning and memory throughout life (Goncalves,
Schafer, and Gage 2016; Eriksson et al. 1998). The hippocampus is roughly C-shaped
and is divided into four fields: the entorhinal cortex, dentate gyrus, hippocampus proper,
and subicular complex, as shown in Figure 3 (Amaral and Witter 1989).
11

Figure 3. The structure of the hippocampus and the dentate gyrus. The long axis of the
hippocampus bends into a C shape. DG: dentate gyrus region; EC: entorhinal cortex. The
hippocampus proper contains three regions: the cornu ammonis CA1, CA3 and CA2. The
subicular complex is divided into the subiculum (S), presubiculum (PrS), and
parasubiculum (PaS). (Adapted from Amaral and Witter 1989)

A previous study using a radioactive carbon dating method reported that
approximately 700 new neurons are added to the human dentate gyrus region every day.
These new neurons are generated throughout life from the NSCs localized in the stem
cell niche in the dentate gyrus, suggesting the potential contribution of adult hippocampal
neurogenesis to human brain functions (Spalding et al. 2013). The NSC population is
located near the baseline of the stem cell niche subgranular zone in the hippocampal
dentate gyrus in the adult human brain. During embryonic stages, NSCs produced by the
radial glial cells migrate to the subgranular zone in the dentate gyrus region and generate
granule neurons. During adult neurogenesis, quiescent NSCs in the dentate gyrus region
are activated to proliferate and generate prospero homeobox 1 (PROX1)
+
granule layer
12
neurons and glia via intermediate NSCs and transit progenitors (Figure 4). (Yu, Marchetto,
and Gage 2014; Altman and Bayer 1990)

Figure 4. Neurogenesis from the embryonic stages in the primordial hippocampus to the
postnatal period in the hippocampus. NSCs generate Prox1
+
granule neurons via
intermediate neural progenitors. AN, ammonic neuroepithelium; CH, cortical hem; CTX,
cortex; D, dorsal; DNe, dentate neuroepithelium; E, embryonic day; F, fimbria; INP,
intermediate neural progenitor; L, lateral; LGE, lateral ganglionic eminence; M, medial;
MGE, medial ganglionic eminence; P, postnatal day; V, ventral; SGZ, subgranular zone;
DG, dentate gyrus; CA cornu ammonis. (Adapted from Yu et al. 2014)

NSCs are heterogeneous and multiple distinct NSC populations have been identified
as coexisting during neurogenesis. It is extremely difficult to identify or isolate one
subpopulation of NSCs either in vitro or in vivo. Many efforts have been made to
characterize NSCs and progenitors; however, results from different studies sometimes
led to different conclusions about the character of NSCs (Lugert et al. 2010; Encinas et
al. 2011; Bonaguidi et al. 2011). Multiple different markers have been used for
identification of different types of cells involved in neurogenesis, as shown in Figure 5.
13

Figure 5. Markers useful for differentiating between the types of cells involved in
neurogenesis. During neurogenesis, SOX2
+
SOX1
-
quiescent stem cells are activated
and give rise to activated SOX2
+
SOX1
+
NSCs, which then proliferate and generate
granule layer neurons and glia via intermediate progenitors (Adapted from Venere et al.
2012).

4. Establishing WNT8B
en
-GFP reporter iPSCs to study human hippocampal
primordial cells
4.1 WNT8B is a putative hippocampal precursor marker in the developing human
fetal brain
To our knowledge, there is no genetic marker that can be utilized to uniquely identify
early hippocampal NSCs for further isolation or study. However, there are several
compelling reasons why having an in vitro model and genetic marker of hippocampal stem
cells would be beneficial. First, the characterization of neurogenic cells is required for the
study of adult neurogenesis. Without a reliable method for the identification of primary
NSCs, scientists will never be able to study this specific NSC population. What’s more,
patients suffering from neurodegenerative diseases such as Parkinson’s disease,
14
Alzheimer’s disease, and Huntington’s disease all have defective adult neurogenesis
(Winner and Winkler 2015). Other human conditions involving atrophy of the
hippocampus, including hypertension, epilepsy, schizophrenia, Cushing’s disease,
depression, and head trauma, are also associated with abnormal adult neurogenesis
(Anand and Dhikav 2012). Besides defective NSCs, CHARGE patients may also suffer
from defective neurogenesis due to abnormal NSCs during their adult life. Therefore, to
improve understanding of the mechanism underlying neurogenesis and explore potential
new medicines for human diseases associated with abnormal neurogenesis and defective
NSCs, an in vitro model of human hippocampal NSCs is a highly desirable resource.
Finally, generation of a pure population of NSCs or the ability to control their development
or differentiation process will provide an unprecedented opportunity for NSC
transplantation or cell replacement therapy. These in vitro expanded NSCs could be also
utilized for drug screening. A most recent study indicated that human hippocampal
neurogenesis is preserved throughout aging (Boldrini et al. 2018), which shows the
promise of studying adult neurogenesis and NSCs in human diseases as well as the
tremendous potential of the in vitro model of human hippocampal NSCs for translational
research.

Therefore, our lab established human ESC reporter lines to isolate hippocampal
precursors as an in vitro model of hippocampal (hip-)NSCs. We analyzed and compared
the microarray gene expression profiles in various parts of the human brain at embryonic
stage 8 (E8) to 13 (E13). Although we found that no one specific gene was expressed
only in the hippocampus, we did find 2000 genes co-expressed in the hippocampus that
15
may provide a good signature. Moreover, some of those 2000 genes were shown in the
microarray expression profile from the Allen Brain Atlas (Figure 6). Among these, WNT8B
(Figure 6, green arrow) was highly expressed in hippocampal tissue, but not in other
tissues, at E8-E13 which suggested that it may be a good candidate hip-NSC marker.

Figure 6. Microarray expression profiles of different human brain tissues. Different
colored boxes in second row represent different types of tissue in developing human brain.
Each column represents the gene expression level in that tissue. Here we only show a
small number of genes as examples. Blue indicates low expression and green indicates
high expression. Purple represents the hippocampus. The green arrow indicates WNT8B
gene expression in all these tissues. The black arrow indicates the MSTN gene, which is
also only expressed in hippocampal tissue. However, the expression level of this gene
was relatively constant across developmental stages. Zooming in on the WNT8B gene
shows that WNT8B is only highly expressed in the hippocampus, not in any other parts
of the brain, and the expression level decreased as the age of the embryo increased.
(Microarray gene expression profile from Allen Brain Atlas and analysis performed by R.
Bajpai)

Furthermore, we found that WNT8B was only expressed in the hippocampus and was
most highly expressed at E8. After E8 and until 4 months postnatally, WNT8B was always
expressed only in hippocampus and not in any other tissues; however, the expression
level of WNT8B was downregulated significantly after E12 (Figure 7). Since development
of the embryonic NSCs also decreases after E12, we hypothesized that downregulation
of WNT8B expression was associated with a decrease in hip-NSCs.
16

Figure 7. i. Temporal expression of human WNT8B in the postconceptional period from
day 8 to 4 months. ii. Different colored lines show different parts of the brain. Purple
regions represent the hippocampus. iii. WNT8B expression in all different tissues at
different ages. Purple arrows indicate hippocampal tissue. (Microarray gene expression
profile from Allen Brain Atlas and analysis performed by R. Bajpai)

To further confirm that WNT8B gene is specifically expressed in the hippocampus, we
checked stained sections of mouse embryonic specimens, neonatal specimens, and adult
brain specimens in the Allen Brain Atlas database (Figure 8). Wnt8b was highly
expressed at E11.5 and then became increasingly diffuse during the later development
of the embryo. Interestingly, all the Wnt8b expression was in the baseline of the dentate
gyrus (DG), which is supposed to be a site of hip-NSCs. In neonatal specimens at age
P4, we can see diffusion of Wnt8b
+
cells from the baseline of the DG. Moreover, in adult
brain tissue specimens, there were sparsely distributed Wnt8b
+
cells at the baseline of
the DG. As is known from previous research, hip-NSCs are only present in very small
numbers in the adult hippocampus, specifically located at the baseline of the DG. This
specific location, their diffusion from the DG, and their sparse distribution may suggest
that Wnt8b is expressed in hip-NSCs and it may be a good marker gene for this population.
17

Figure 8. In situ hybridization of Wnt8B on mouse embryo sections from Allen Brain Atlas
database. Embryonic specimens at age E11.5 (i, i-1), E13.5 (ii, ii-1) and E15.5 (iii, iii-1)
respectively. Arrows indicate Wnt8b positive cells located at the baseline of the DG. iv.
Neonatal specimen at age P4. Arrows indicate the distribution of the Wnt8b positive cells
in the DG. v. Adult specimen at age P28 showing sparse distribution of Wnt8b positive
cells all located at the baseline of DG. (Analysis by R. Bajpai)

18
Sox 2, DAPI, and Wnt8b immunofluorescence staining of brain section are shown in
Figure 9. The Wnt8b
+
Sox2
+
cells (yellow) were all in the baseline region of the DG and
they were surrounded by many other Wnt8b
-
Sox2
+
neural progenitors (red) and Wnt8b
-

Sox2
-
cells (blue). The staining results may indicate that Wnt8b
+
Sox2
+
cells were different
from Wnt8b
-
Sox2
+
cells and were located in the baseline of DG. Therefore, we
hypothesized that Wnt8b
+
Sox2
+
cells could be hip-NSCs and that Wnt8b could be a very
good candidate marker for these cells. Figure 10 is a schematic diagram showing the
position of two types of NSCs at different embryonic stages.

Figure 9. Immunofluorescence staining of
Sox2, Wnt8b, and DAPI. Sox2 appears
green. Wnt8b appears red and all the cells
are stained with DAPI blue. (Stained and
imaged by J. Oki)

Figure 10. Schematic of the positions of two types of NSCs at different stages. CTX:
cortex; CA1: cornu ammonis 1; CA3: cornu ammonis 3; DG: dentate gyrus; SGZ: sub-
granular zone; LGE: lateral ganglionic eminence; MGE: medial ganglionic eminence.

19
4.2 WNT8B
en
-GFP reporter ESC or iPSC lines
In the progression of embryonic stem cell (ESC) differentiation into NCCs, ESCs first
differentiate into neuroectodermal cells (NECs), and then NECs differentiate into a mixed
cell population including neural precursor cells (NPCs) and NCCs. We further confirmed
the transient expression of WNT8B in the neural precursor cell population, but not any of
the other three populations (Figure 11-A). Upon investigating the epigenomic profile, we
found an activated enhancer downstream which is only detected in the hippocampal
region (Figure 11-B). In order to confirm the interaction of the WNT8B promoter and
enhancer in the WNT8B
+
neural precursor cell population, we used a chromosome
conformation capture technique. We found that there was a ligation product using 3C
DNA as template and primers P and E as locus-specific primers. Therefore, we confirmed
the interaction between the WNT8B promoter and WNT8B enhancer, which is essential
for the expression of WNT8B (Figure 12). We constructed a lentiviral-enhancer reporter
to observe activation of the WNT8B enhancer in human tissue culture. Lentiviral-mediated
transduction of ESCs inserted the WNT8B enhancer tagged with a minimal promoter and
GFP into the ESC genome (Figure 13). Thus, we established the reporter cell line H9-
WNT8B
en
-GFP.

20
Figure 11. A. Transient expression of WNT8B during human stem cell differentiation from
day 0 to day 14. The peaks show the expression of WNT8B and the position of the gene
is shown at the bottom. B. Epigenomic profile identifying transiently activated enhancer.
H3K27me3: repressive histone mark; P300, H3K4me1 and H3K27Ac: enhancer marker;
H3K4me3: promotor and enhancer marker. Peaks show signal from the marker. TSS:
Transcription start site. (Data from R. Bajpai)

Figure 12. Chromosome conformation capture technique to detect cross-linking between
the WNT8B promoter and WNT8B enhancer. i. P and E were locus-specific primers we
designed and C1 and C2 were random primers upstream of the promoter and the
enhancer, respectively. ii. Schematic showing the principle of 3C tech. iii. Gel
electrophoresis analysis of the PCR products. This result shows the transient interaction
between the WNT8B promoter and its enhancer. (Data from K. Shevade)

Figure 13. Lentiviral construct for the ESC and iPSC WNT8B
en
-GFP reporter lines. This
construct functions as a reporter of the WNT8B enhancer tagged with GFP. (Data from J.
Oki)

As we expected, endogenous WNT8B was expressed in the hypothalamus and
hippocampus but WNT8B enhancer was only detected in the hippocampal region. Since
the WNT8B enhancer is not active in H9-WNT8B
en
-GFP ESCs, we did not detect GFP
21
expression in undifferentiated H9-WNT8B
en
-GFP ESCs, as expected (Figure 14- i). We
then cultured H9-WNT8B
en
-GFP ESCs in neural differentiation media, where they
differentiated to neuroectodermal spheres. At first, all neuroectodermal spheres were
GFP-negative, which means that the WNT8B enhancer was not activated in this cell
population (Figure 14- ii). However, 5 or 6 days after differentiation, the WNT8B enhancer
was activated in the spheres and they became a GFP-positive cell population (Figure 14-
iii). Around day 7 or 8, H9-WNT8B
en
-GFP spheres started attaching to the dish and NCCs
started to migrate from the spheres (Figure 14- iv). According to the IF staining (Figure
14- iv), the majority of cells in the spheres were GFP-negative and a small number were
GFP positive. Also, the NCCs that migrated out of the spheres were GFP-negative.
Therefore, the WNT8B enhancer was only activated in a specific population of NSCs, as
we expected.

Figure 14. Transient expression of WNT8B-GFP in the neurospheres generated from H9-
WNT8B
en
-GFP ESCs. i. H9-WNT8B
en
-GFP ESCs were all GFP negative, which means
the WNT8B enhancer was not activated. ii. All cells in the neuroectodermal spheres were
GFP negative at the beginning of differentiation. iii. Some cells in these mixed-cell-
population spheres were GFP positive, but not all of them. iv. GFP positive and negative
NPCs and the migrating P75
+
NCCs. (Data from J. Oki)

22
4.3 WNT8B
en
- GFP positive cells represent hippocampal primordial cells
We used RNA-sequencing to compare the gene expression of WNT8B-GFP positive
cells and NCCs (Figure 15-A). We annotated some representative upregulated genes in
both cell populations. The in situ hybridization of some of these genes on mouse embryo
sections were found in the Allen Brain Atlas. Some of them are shown in Figure 15-B.
Genes significantly upregulated in WNT8B-GFP positive cells were highly expressed in
the hippocampal region indicated by the red arrows, suggesting the WNT8B-GFP positive
cells localize in the hippocampus region and may correlate with hippocampal function.
23

Figure 15. A. RNA-seq of WNT8B
en
-GFP positive cells and NCCs derived from sorted
WNT8B
en
-GFP positive cells. X axis shows the expression level of a specific gene. Y axis
shows the fold change. Green puncta show WNT8B expression. Red puncta show genes
specifically expressed in WNT8B
en
-GFP positive cells. All known genes are labeled;
those not labeled are unknown genes or non-coding genes. (Data from R. Bajpai) B. In
situ hybridization of some different representative upregulated genes in two cell
populations. Wnt8b, Dlk1 and Lhx5, which were significantly upregulated in WNT8B-GFP
24
positive cells, were highly expressed in the hippocampal region. However, Pcdh12, Tgfb1
and Twist1, which were significantly upregulated in NCCs, were not expressed in the
hippocampal region at E11.5. (in situ hybridization data from Allen Brain Atlas and
analysis performed by R. Bajpai)

In a functional study of the genes specifically expressed in the WNT8B-GFP positive
cell population, gene ontology term analysis (Figure 16-A) showed that gene products of
WNT8B-GFP positive cells had functions in forebrain development, and some gene
products of WNT8B-GFP positive cells were closely related to the development of the
hippocampus. We also performed principal component analysis (Figure 16-B) comparing
400 upregulated genes in WNT8B-GFP positive neural precursor cells to different tissues
and repeated this analysis in 13 different embryos. For each one of them, the WNT8B-
GFP positive neural precursor cells we generated were most correlated to hippocampal
tissue than any other tissues.

Figure 16. A. Gene ontology of WNT8B-GFP positive cells. All genes specifically
expressed in WNT8B-GFP positive cells were associated with forebrain development.
Blue: Genes related to forebrain development but not hippocampus development. Green:
Genes related to hippocampus development. B. Principal component analysis comparing
WNT8B-GFP positive neural precursor cells to different embryonic tissues for 400
upregulated genes (n = 13 embryos). The distance between two tissues shows the
correlation between these two tissues. (Data from R. Bajpai)

25
Furthermore, we placed the WNT8B-GFP positive cells at the top of the hierarchy of
neurogenesis, as shown in Figure 17, according to the IF staining. We also found that
the WNT8B-GFP positive cells can generate a wide variety of neurons and astrocytes via
a series of intermediate progenitors (Figure 17), which provided further strong evidence
that the WNT8B-GFP positive cells were NSCs.

Figure 17. Top panel: IF staining of different cell markers on neural spheres, neurons,
and astrocytes generated from FACS-sorted WNT8B-GFP positive cells. Bottom panel:
Schematic showing the process of neurogenesis. (Data generated by R. Bajpai)

In this study, we wanted to determine the characteristics of hip-NSCs from CHARGE
patients and test a candidate drug to possibly correct their cellular defects. Further,
understanding the mechanism of self-renewal of hippocampal primordial cells is essential
for hip-NSC studies and future stem cell transplantation. However, this required a stable
population of WNT8B-GFP positive cells and the challenge was that they spontaneously
26
differentiated, like other kinds of stem cells. Therefore, we also focused on the self-
renewal and maintenance of WNT8B-GFP positive cell population and trying to
understand the mechanism of self-renewal of hippocampal primordial cells to finally
achieve in vitro expansion of human hippocampal primordial cells.

Materials and Methods
Induced pluripotent stem cell (iPSC) lines
Previously, our lab generated GFP reporter lines using lentiviral-mediated infection as
described in Bajpai 2011 with minor modifications. Our lab generated two CHARGE
patient iPSC WNT8B
en
-GFP reporter cell lines, namely CHARGE iPSC CHD7
1701X
and
CHARGE iPSC CHD7
1036X
. We also generated one control female iPSC WNT8B
en
-GFP
reporter cell line. All experiments described here used these three cell lines.

Induced pluripotent stem cell (iPSC) culture method
All tissue cultures were grown in a Sanyo CO
2
incubator at 37 ℃ with a [CO
2
] of 5.0%.
All iPSC lines were cultured in mTeSR
TM
(Stem Cell Technologies) media in a Falcon 6-
well flat bottom polystyrene plate (CORNING) coated with Cultrex® Basement Membrane
Matrix. According to the method described in (Bajpai et al. 2008), iPSC lines were
maintained by passaging using Accutase (Stem Cell Technologies).

27
Induction and differentiation of iPSCs into neural precursor cells
iPSCs were differentiated into neural precursor cells namely, neuroectodermal
spheres and neural crest cells in vitro using the method described by Bajpai et al. (2009).
This was done by first treating iPSCs with collagenase and then culturing in NCC media
composed of a 1:1 ratio of DMEM/F12 (CORNING cellgro): Neurobasal Medium (Life
technologies) with 0.5% N2 NeuroPlex (GEMINI), 0.5% Gem21 NeuroPlex (GEMINI), 0.5%
GlutaMAX
TM
(Life technologies), 5000 units/ml Penicillin (CELL CULTURE CORE
FACILITY, USC), 5ug/ml insulin, 20ng/ml bFGF, and 20ng/ml EGF. Medium was changed
every two days.

Fluorescence activated cell sorting of live cells (FACS sorting)
FACS sorting was done with a BD FACSAria
TM
cell sorter (BD Biosciences) at the
University of Southern California Flow Cytometry Core Facility. Excitation lines used were
488nm and 561nm. All processes were done at room temperature.

In vitro expansion of hippocampal primordial cells
FACS-sorted neuroectodermal single cells including WNT8B-GFP positive and
negative cells were cultured in different NSN2 media including different growth factors
(20ng/ml bFGF or 20ng/ml EGF or both), DMEM/F12 media with 0.5% N2 Neuroplex,
5000 units/ml Penicillin and 5ug/ml insulin. Ten thousand single cells per well were
seeded in 96 well V-bottom tissue culture dishes (Corning). Notch signaling pathway
inhibitor DAPT (N-[N-(3,5-Difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester,
28
Sigma) was used for the maintenance of the WNT8B-GFP positive cells. Dimethyl
sulfoxide-MAX (DMSO) was the solvent of the DAPT.

Treatment of hip-NSCs from CHARGE patients with the candidate drug CHIC-35
To rescue the defects in the hippocampal primordial cells and other differentiated cells in
CHARGE, we used candidate drug CHIC-35 ((6S)-2-Chloro-5,6,7,8,9,10-hexahydro-
cyclohept[b]indole-6-carboxamide, Sigma) to treat the neuroectodermal spheres at
differentiation day 7. At day 11, we conducted FACS sorting analysis of the
neuroectodermal spheres and cultured all WNT8B-GFP positive cells in complete NSN2
media composed of DMEM/F12 media with 0.5% N2 Neuroplex, 5000 units/ml Penicillin
and 5ug/ml insulin, 20ng/ml bFGF and 20ng/ml EGF.

Fixing cells with paraformaldehyde
We used 4% paraformaldehyde solution (PFA) to fix cells. For attached cells, we
incubated the cells in 4% PFA at room temperature for 20 minutes. For spheres, we
incubated the spheres in 4% PFA at room temperature for 1 hours. After fixation, we
washed cells with Dulbecco’s Phosphate-Buffered Saline (DPBS, CORNING) three times
and then kept the cells in DPBS at 4 ℃.

Differentiation of neural crest cells (NCCs)
NCCs were passaged with Trypsin-EDTA and cultured in NCC media for another three
days before adipogenic, osteogenic and chondrogenic differentiation. A high density of
NCCs was required for the adipogenic differentiation in media composed of DMEM with
29
10% Fetal Bovine Serum (FBS), 1mM dexamethasone, 5ug/ml insulin and 0.5mM IBMX
(3-Isobutyl-1-methylxanthine, Sigma). After 4 weeks, all cells were fixed in 4% PFA and
stained with Oil Red O for detection of lipid droplets within the cultured cells. Fixed cells
were gently washed with DPBS twice. All solution in the well was aspirated completely
before adding freshly prepared Oil Red O working solution. The cells were incubated in
Oil Red O working solution for 50-60 minutes at room temperature and then the staining
solution was removed and cells were washed with DPBS three times before being kept
in DPBS at 4 ℃. To make the Oil Red O working solution, first we made 0.5% Oil Red O
stock solution in isopropanol. The stock solution was diluted in DDH
2
O and filtered
through two layers of WHATMAN papers to make fresh working solution. A very low
density of NCCs was required for the osteogenic differentiation in media composed of
DMEM with 10% FBS, 10mM beta-glycerophosphate (BGP), 0.1μM Dex and 200μM
sodium L ascorbate. After 2 weeks, all cells were fixed in 4% PFA and stained with BM
purple. We used BM purple AP substrate precipitating (Roche Diagnostics) to stain the
alkaline phosphatase within cultured cells. We incubated the fixed cells in BM purple
solution for 15 minutes at room temperature in the dark, then washed the cells with DPBS
three times before keeping them in DPBS at 4 ℃.

Immunofluorescence staining
After fixing the cells or spheres, we washed them with DPBS three times. We then
incubated the cells or spheres in PBTX composed of DPBS with 0.1% Triton X-100
(Sigma) and 1% Bovine Serum Albumin (GEMINI). The cells or spheres were incubated
with primary antibody diluted in PBTX (following the instructions from the company)
30
overnight at 4 ℃. This was followed by washing the cells or spheres with PBTX three
times and incubating them with secondary antibody diluted in PBTX for 45 minutes at
room temperature. We washed the cells or spheres with PBTX three times before storing
them in DPBS at 4 ℃.

Imaging
Images of the tissue culture samples were taken with a fluorescence microscope
(Leica DMI3000 B) and a confocal microscope (Broadband Confocal Leica TCS SP5 II).
All photos were analyzed and assembled using Leica LAS AF Software.

Results
1. Defects in CHARGE WNT8B-GFP positive hippocampal neural stem cells
CHD7 has been shown to contribute to the control of neurogenesis and play a role in
maintenance of the hippocampal stem cell pool (Jones et al. 2015). Our lab hypothesized
that altering CHD7 expression would have an effect on WNT8B enhancer activation. To
test this hypothesis, a cell line with inducible CHD7 knockdown and the WNT8B
en
-GFP
reporter was created previously. In cells with high levels of CHD7 knockdown, there is
little or no WNT8B
en
-GFP reporter detected, indicating the silencing of WNT8B enhancer
(Figure 18).
31

Figure 18. CHD7 knockdown in the WNT8B-GFP reporter ESC line reduces WNT8B
enhancer activation. Neuroectodermal spheres with high knockdown levels of CHD7 do
not have WNT8B enhancer activation. In contrast, spheres with lower levels of
knockdown, which appear as less intense red spheres, can have a strongly activated
WNT8B enhancer. Green: GFP; Red: RFP; Blue: DAPI. (Data from J. Oki)

We therefore hypothesized that mutated CHD7 may lead to defective hippocampal
NSCs in CHARGE patients. Thus, we established two WNT8B
en
-GFP CHARGE patient
iPSC lines: CHD7
1701x
and CHD7
1036X
. After differentiating the CHARGE iPSCs into
neuroectodermal spheres, we found that the morphology of the spheres was significantly
different between the control and CHARGE groups. The percentage of WNT8B-GFP
positive cells in CHARGE neural spheres was significantly lower than in the controls
(Figure 19). Noticeably, both phenotypes were much more severe in CHARGE CHD7
1036X

neural spheres than in CHARGE CHD7
1701X
(Figure 20).
32

Figure 19. A. Neural spheres differentiated from control female WNT8B-GFP iPSCs at
day 7. The morphology of the spheres was spherical and the percentage of GFP positive
cells was as normal. i. Zoomed in view of one representative control neural sphere. B.
Neural spheres differentiated from CHARGE WNT8B-GFP iPSCs (CHD7
1701X
) at day 7.
The morphology of the spheres was irregular and the percentage of GFP positive cells
was much lower than in the control. ii. Zoomed in view of one representative CHARGE
neural sphere. Scale bar: 100um

Figure 20. A. Neural spheres differentiated from control female WNT8B-GFP iPSCs at
day 9. B. Neural sphere differentiated from CHARGE WNT8B-GFP iPSCs (CHD7
1036X
) at
33
day 9. Despite starting differentiation with many more iPSCs, the CHARGE neural
spheres were still much fewer in number and smaller than those of the control group. The
morphology of the CHARGE neural spheres was irregular and the spheres contained
many dead cells. What’s more, the WNT8B-GFP was not activated in CHARGE neural
spheres; however, the control neural spheres already turned on GFP expression. A’.
Representative neural sphere differentiated from control female WNT8B-GFP iPSCs at
day 9. The morphology of the sphere was spherical and the WNT8B-GFP was turned on.
B’. Representative neural sphere differentiated from CHARGE iPSCs (CHD7
1036X
). The
neural sphere was much smaller than the control sphere and the morphology of the
sphere was irregular. Most importantly, no GFP expression was turned on in the
CHARGE neural sphere. Scale bar: 100um

After 15 days of differentiation, FACS sorting was conducted to collect the WNT8B-
GFP positive cells from both control and CHARGE groups for further culturing. We
collected the same amount of WNT8B-GFP positive cells and cultured them in 96-well V-
bottom dishes at 10,000 cells per well in NSN2 complete media. However, the CHARGE
WNT8B-GFP positive cells did not survive after FACS sorting while the control WNT8B-
GFP positive cells did (Figure 21). This suggests that WNT8B-GFP hip-NSCs derived
from CHARGE patients may be defective.

Figure 21. (Left) Neural spheres formed from FACS-sorted control WNT8B-GFP positive
cells at day 7. (Right) Dead FACS-sorted CHARGE WNT8B-GFP positive cells at day 7.
No sphere has been formed from the FACS-sorted CHARGE WNT8B-GFP positive cells.
Scale bar: 100um

34
2. CHIC-35 rescued the morphology of neural spheres, WNT8B enhancer activation,
and development, migration and differentiation of NCCs.
2.1 CHIC-35, an inhibitor of SIRT1, rescued defects in a CHARGE zebrafish model
CHIC-35 is a cell-permeable, metabolically stable, and very potent inhibitor of Sirtuin
1 (SIRT1). Sirtuin1, also known as NAD-dependent deacetylase sirtuin-1, is a member of
the Sirtuin family of proteins. Sirtuins are protein deacetylases, which represent a new
class of histone deacetylases (HDAC) involved in gene silencing. SIRT modulators are
potential therapeutics for cancer, diabetes, muscle differentiation, heart failure,
neurodegeneration, and aging. Previously, our collaborator tested many drugs in a
CHARGE zebrafish model to identify drugs that could rescue their defects. The CHARGE
zebrafish model was established by injecting a morpholino into zebrafish embryos to
knockdown chd7. Knockdown embryos were treated with different candidate drugs. After
treatment, all chd7 knockdown zebrafish embryos were fixed for in situ hybridization. They
found that CHIC-35 significantly rescued the expression of crestin, a neural crest marker,
in CHARGE zebrafish embryos (Asad and Sachidanandan, unpublished results). Thus,
we wanted to test if CHIC-35 could rescue the defects in vitro using CHARGE patient
WNT8B-GFP iPSC lines.

2.2 CHIC-35 rescued the defects in CHARGE neural spheres and WNT8B-GFP
positive cells.
We tested the ability of CHIC-35 to rescue the defective CHARGE neural spheres and
CHARGE WNT8B-GFP positive cells in vitro. After treatment with 500nM CHIC-35, the
35
morphology of CHARGE patient (CHD7
1036X
) neural spheres was significantly rescued
(Figure 22).

Figure 22. CHIC-35 significantly rescued the morphology of CHARGE patient (CHD7
1036X
)
neural spheres. Without CHIC-35, the morphology of the CHARGE neural spheres was
irregular after 9 days’ differentiation and most of the spheres were much smaller than the
control neural spheres. Treated with CHIC-35, the morphology of the CHARGE neural
spheres was spherical, similar to the control neural spheres. What’s more, the size of the
CHARGE neural spheres treated with CHIC-35 was significantly increased, indicating that
CHIC-35 rescued the survival of CHARGE neural precursor cells.

CHIC-35 also rescued the CHARGE WNT8B-GFP positive cells. After 7 days of
differentiation, we treated CHARGE patient (CHD7
1071X
) neural spheres as well as control
female neural spheres with CHIC-35 and compared them to untreated spheres. To
36
estimate the percentage of WNT8B-GFP positive cells in the neural spheres, we
performed FACS sorting analysis. The percentage of WNT8B-GFP positive cells was
significantly increased by CHIC-35 treatment, indicating that CHIC-35 may either rescue
the survival of WNT8B-GFP positive cells or rescue the activation of the WNT8B
enhancer in the hippocampal primordial cells. (Figure 23)

Figure 23. Histograms depicting FACS sorting analysis of CHARGE patient (CHD7
1071X
)
neural spheres and control female neural spheres with or without CHIC-35 treatment. The
green rectangular area shows the WNT8B-GFP positive cells. Treatment with CHIC-35
significantly increased the percentage of WNT8B-GFP positive cells in CHARGE neural
spheres.

2.3 CHIC-35 rescued the migration and differentiation of NCCs.
NCCs have been shown to be defective in the majority of CHARGE patients. In 2017,
Okuno et al. generated two CHARGE patient iPSC lines and found that NCCs derived
37
from these lines showed defective delamination, migration, and motility in vitro (Okuno et
al. 2017). Therefore, we evaluated the migration of NCCs differentiated from neural
spheres treated with CHIC-35 to see whether the migration defect had been rescued. We
found that the migration of CHARGE NCCs was significantly rescued by CHIC-35, as
shown in Figure 24. Quantification of their migration rates is shown in Figure 25 and Table
3.

Control NCC
CHARGE NCC CHARGE NCC (CHIC-35)
Control NCC CHARGE NCC CHARGE NCC (CHIC-35)
38
Figure 24. CHIC-35 rescued the migration of CHARGE NCCs. i. Control female iPSC-
NCCs at differentiation day 12. ii. CHARGE patient iPSC (CHD7
1036X
)-NCCs at
differentiation day 12. CHARGE NCCs were not able to migrate out from the neural
spheres, unlike the control NCCs. iii. CHARGE patient iPSC (CHD7
1036X
)-NCCs at
differentiation day 12. Treatment with 500nM of CHIC-35 started on the first day of
differentiation. NCCs migrated out from the neural spheres, similar to the control NCCs,
indicating that CHIC-35 rescued the migration of CHARGE iPSC-NCCs. iv. Control
female iPSC-NCCs at differentiation day 12 showing the space between each NCC. v, vi.
CHARGE patient iPSC (CHD7
1036X
)-NCCs at differentiation day 12. NCCs were not able
to migrate and almost no space was present between cells. vii. CHARGE patient iPSC
(CHD7
1036X
)-NCCs treated with 500nM CHIC-35 at differentiation day 12. NCCs migrated
out from the spheres and the space between NCCs was similar to the control.

Figure 25. Quantification of NCC migration rate to compare the NCC migration distance.
Right panel shows the method used to calculate the migration rate (R/r). R = the distance
of the sphere center to edge of NCC migration circle; r = the radius of the sphere. Left
panel was drawn using Prism 7 software. Significance was calculated by Prism 7 via
Sidak’s multiple comparisons test. The migration rate of CHARGE patient iPSC
(CHD7
1036X
)-NCCs was significantly lower than that of the control NCCs. However, the
migration rate of CHARGE patient iPSC (CHD7
1036X
)-NCCs treated with 500nM CHIC-35
was significantly increased compared to the untreated CHARGE NCCs, and was not
significantly different from that of the control NCCs. Thus, CHIC-35 significantly rescued
the migration of CHARGE NCCs.

39

Table 3. Sidak’s multiple comparisons test result showing the significant differences
between the pairs of experimental groups.

Furthermore, adipogenic and osteogenic differentiation was conducted after
differentiation day 15. Both CHARGE patient iPSC (CHD7
1701X
)-NCCs and control iPSC-
NCCs were differentiated at the same time. For details regarding the differentiation, see
Materials and Methods section and Figure 26.
Figure 26. Schematic diagram of the
processes of NCC adipogenic and
osteogenic differentiation. After
differentiating iPSCs into neural spheres,
spheres were treated with 500nM or
1uM CHIC-35. Both CHARGE patient
iPSC-NCCs and control NCCs were
tested. Neural spheres gave rise to
NCCs and after day 11, NCCs were
passaged. Adipogenic and osteogenic
differentiation were performed on day 15.
After 4 weeks, adipogenic cells were
fixed for staining. Osteogenic cells were
fixed on osteogenic differentiation day
14.

40
After 4 weeks of adipogenic differentiation, control iPSC-NCCs generated adipocytes
with big oil drops (stained with Oil-Red). Compared to the control, no big oil drops were
formed from the CHARGE iPSC-NCCs. CHARGE iPSC-NCCs treated with CHIC-35
generated adipocytes with big oil drops similar to the control iPSC-NCCs. Thus, treatment
with CHIC-35 significantly rescued the adipogenic differentiation of CHARGE iPSC-NCCs
(Figure 27). Different concentrations of CHIC-35 did not produce significant a difference.
After 14 days of osteogenic differentiation, we found that CHIC-35 also significantly
changed the process of osteogenic differentiation of CHARGE iPSC-NCCs (Figure 28),
indicating that it may rescue the osteogenic differentiation of CHARGE iPSC-NCCs.
However, there are two possible explanations for this phenotype since we cannot
determine whether the osteoblasts have been already generated or not. BM purple
stained alkaline phosphatase (AP) with purple color. However, AP is not only present in
osteoblasts, but also in some stem-like cells. If the AP staining indicated the osteoblasts,
the CHARGE NCCs differentiated into osteoblasts much faster than the control NCCs,
and CHIC-35 rescued this phenotype by decelerating this process. On the other hand, if
the AP staining indicated stem-like cells, the CHARGE NCCs differentiated even slower
than the wild-type NCCs, and CHIC-35 accelerated this process. Therefore, we need
more evidence to show that the differentiated cells are osteoblasts. In conclusion, we
provide evidence to show that CHARGE hippocampal primordial cells derived from
patient iPSCs are defective, which indicates that CHARGE patients may have defects in
the hippocampal primordial cells and neurogenesis. CHIC-35 can rescue the morphology
of CHARGE neural spheres and WNT8B-GFP positive cells, as well as the migration and
differentiation of CHARGE iPSC-NCCs. Since adult neurogenesis persists throughout life,
41
rescuing hippocampal primordial cells may rescue the defective neurogenesis in
CHARGE patients. Thus, CHIC-35 may have clinical value in allowing CHARGE patients
to improve their health and quality of life.

Figure 27. Adipogenic differentiation of CHARGE iPSC-NCCs was rescued by CHIC-35.
Oil drops were stained with Oil-Red O. A. Adipogenic differentiation of CHARGE iPSC-
NCCs. B. Adipogenic differentiation of CHARGE iPSC-NCCs treated with 500nM CHIC-
35. C. Adipogenic differentiation of CHARGE iPSC-NCCs treated with 1 uM CHIC-35. D.
Adipogenic differentiation of control iPSC-NCCs. E. Adipogenic differentiation of control
iPSC-NCCs treated with 500nM CHIC-35. F. Adipogenic differentiation of control iPSC-
NCCs treated with 1uM CHIC-35. Big oil drops were formed in the adipocytes generated
by CHARGE iPSC-NCCs treated with CHIC-35, which was similar to the control iPSC-
NCCs. Almost no oil drops formed in CHARGE iPSC-NCC adipogenic differentiated cells.
The top right of each panel shows a zoomed-in view of the region circled in each panel.
Scale bars: 50um

CHARGE adipogenic
500nM CHIC-35 1uM CHIC-35
Control adipogenic
42

Figure 28. CHARGE iPSC-NCC osteogenic differentiation is rescued by CHIC-35.
Potential osteoblasts were stained with BM purple. A. Osteogenic differentiation of
CHARGE iPSC-NCCs. B. Osteogenic differentiation of CHARGE iPSC-NCCs treated with
500nM CHIC-35. C. Osteogenic differentiation of CHARGE iPSC-NCCs treated with 1uM
CHIC-35. D. Osteogenic differentiation of control iPSC-NCCs. E. Osteogenic
differentiation of control iPSC-NCCs treated with 500nM CHIC-35. F. Osteogenic
differentiation of control iPSC-NCCs treated with 1uM CHIC-35. The staining results
suggest that CHIC-35 improved the osteogenic differentiation of CHARGE NCCs to
become similar to that of the controls. Scale bar: 50um

3. In vitro expansion of hippocampal primordial cells
3.1 The maintenance of WNT8B-GFP positive NSCs.
As mentioned previously, the essential part of the hippocampal NSC in vitro model is
that we are able to maintain a pure population of NSCs. However, the FACS-sorted
WNT8B-GFP positive cells spontaneously lose GFP expression and differentiate into
progenitors, NCCs, and neurons. A schematic diagram of the standard method of
obtaining WNT8B-GFP positive cells is shown in Figure 29.
500nM CHIC-35 1uM CHIC-35
CHARGE osteogenic Control osteogenic
43

Figure 29. Schematic of the standard
method of obtaining WNT8B-GFP
positive cells. After differentiating the
WNT8B-GFP-iPSCs or ESCs into
neuroectodermal spheres, we
conducted FACS-sorting to obtain
WNT8B-GFP positive cells and
negative cells on day 14 or 15 and
culture them separately.

Previously, our lab tried many different methods to maintain WNT8B-GFP positive
cells and reduce their differentiation. However, all the three conditions we tried before led
to differentiation of the WNT8B-GFP positive cells (Figure 30-A). In the first 12 hours,
WNT8B-GFP positive cells could maintain WNT8B-GFP expression; however, after 12
hours they began to lose WNT8B-GFP expression. Theoretically, if all WNT8B-GFP
positive cells could self-renew and not differentiate, the hypothetical exponential growth
of a cell population that began with 10000 WNT8B-GFP positive cells would follow the
green dotted line in Figure 30-B. In reality, however, the amplification of WNT8B-GFP
positive cells had linear rather than exponential growth, which showed that the
proliferation of seeded GFP positive cells was much slower. In contrast, WNT8B-GFP
negative cells emerged from the differentiation of WNT8B-GFP positive cells and their
population grew exponentially.
44

Figure 30. A. Proportion of WNT8B-GFP positive cells that lose GFP expression in three
different media. Green: WNT8B-GFP positive; Red: WNT8B-GFP negative. NS+orn:
NSN2 media+ ornithine; NS + gtx: NSN2 media + gtx; NB + gtx: NCC media + gtx. B.
Amplification of WNT8B-GFP positive cells and WNT8B-GFP negative cells. Green
dotted line: hypothetical exponential growth of the seeded GFP positive cells. Green solid
line: actual linear amplification of GFP positive cells; Red solid line: rapid exponential
amplification of emerging GFP negative cells. (Data from J. Oki)

According to previous RNA-seq data, we noticed that DLK1 was highly expressed in
the WNT8B-GFP positive cells (Figure 15-A). Delta-like 1 homolog (DLK1) is an
noncanonical Notch ligand which can inhibit the Notch signaling pathway in vitro (Nueda
et al. 2007; Baladron et al. 2005). The Notch signaling pathway is involved in a wide range
of developmental process including stem cell self-renewal, proliferation, and
differentiation. Four Notch homologs (Notch receptors: Notch1,2,3,4) were identified in
vertebrates and five canonical Notch ligands have been described, namely JAGGED1,
JAGGED2, Delta-like 1 (DLL1), DLL3 and DLL4. In addition to the canonical ligands,
noncanonical ligands such as DLK1 can also bind to the Notch receptors. Three related
motifs have been identified in canonical Notch ligands, including the N-terminal Delta-
Serrate-LAG-2 (DSL) domain, specialized tandem EGF-repeats (DOS) domain, and a
variable number of EGF-like repeats. The structure of DLK1 is slightly different from the
canonical Notch ligands in that it is missing the DSL domain. The differences in the
structural motifs of different Notch ligands are shown in Figure 31 (Kopan and Ilagan
45
2009). In situ hybridization of Dlk1 on a mouse embryo section showed that the Dlk1 was
highly expressed in the primordial hippocampus region at E11.5. Interestingly, Dlk1 was
not expressed in the primordial hippocampus region at E13.5 or later (Figure 32). This
specific expression pattern might suggest a special function of Dlk1 during the early
development of NSCs.

Figure 31. The structural organization of
Notch ligands. Canonical Notch ligands
contain the DSL, DOS and EGF motifs. DLL3
and DLL4 are DSL-only ligands. DLK1 is a
DOS co-ligand (Adapted from Kopan and
Ilagan 2009).

Figure 32. In situ hybridization of Dlk1 on mouse embryo sections at different stages. At
E11.5, Dlk1 is highly expressed in the primordial hippocampus region. However, from
E13.5 or later E 15.5, Dlk1 is not expressed in the primordial hippocampus region. (Data
from Allen Brain Atlas)

Since WNT8B-GFP positive cells have been placed at the top of the hierarchy of
neurogenesis, as mentioned previously, we hypothesized that DLK1 promotes the
maintenance and self-renewal of WNT8B-GFP positive cells via inhibition of Notch
signaling pathway. To test this hypothesis, the chemical drug DAPT has been used as an
46
inhibitor of Notch signaling pathway in vitro. Previous studies showed that DAPT is a
potent inhibitor of Notch signaling pathway (Liu et al. 2014; Borghese et al. 2010) which
inhibits the cleavage of the intracellular domain of the Notch receptor that would ordinarily
activate Notch signaling pathway through γ-secretase (Figure 33).

Figure 33. Schematic illustrating the lateral inhibition of the Notch signaling pathway.
After DLK1 from a signal-sending cell binds to the Notch receptor on the signal-receiving
cell, γ-secretase cleaves that intracellular domain of the Notch receptor (NICD), which
then enters into the nucleus and binds to transcriptional factor RBPJ. However, γ-
secretase inhibitor DAPT inhibits the cleavage of the NICD, which leads to the inhibition
of Notch signaling pathway.

We tested three different media containing one or more growth factors, namely NSN2
Base + EGF + bFGF medium, NSN2 Base + bFGF medium, and NSN2 Base + EGF
medium for culturing FACS-sorted WNT8B-GFP positive cells. DAPT was used to treat
the WNT8B-GFP positive cells after FACS sorting and non-treated and DMSO-treated
47
groups were included as controls. After 35 days, all spheres were removed and imaged
using confocal microscopy. In culture with NSN2 media with EGF and bFGF growth
factors, neural spheres formed but the percentage of GFP positive cells in the spheres
was about 20% (Figure 34-A). When the cells were treated with 2uM DAPT, the
percentage of GFP positive cells in the spheres was significantly increased, but a high
percentage of GFP negative cells was still present (Figure 34-C). With only bFGF as the
growth factor, FACS-sorted WNT8B-GFP positive cells also formed big spheres and the
percentage of GFP positive cells was also relatively low (Figure 34-D). DAPT did not
significantly increase the percentage of GFP positive cells (Figure 34-F). With only EGF
as the growth factor, FACS-sorted WNT8B-GFP positive cells almost all dead or only one
or two very small spheres survived (Figure 34-J). However, when the cells were treated
with 2uM DAPT, many more, larger spheres survived and the percentage of the GFP
positive cells in each sphere reached 80% to 90% (Figure 34-I). DMSO did not
significantly affect the maintenance of the GFP-positive cells (Figure 34-B, E, H). A
schematic diagram illustrating the results is shown in Figure 35. The different percentages
of GFP-positive cells under different conditions was quantified and compared by
calculating the ratio of the number of WNT8B-GFP positive cells to the total number of
cells in neural spheres (Figure 36-A). The numbers of GFP-positive cells versus GFP-
negative cells were compared in Figure 36-B. Noticeably, however, DMSO had some
effect on the overall growth of the neural spheres. It seemed to diminish the size of the
spheres in the NSN2 Base + FGF media, which means it might have a negative effect on
the growth of the GFP negative cells (Figure 34-E). To demonstrate the importance of the
DAPT for WNT8B-GFP positive cell maintenance, we withdrew DAPT from the neural
48
spheres containing almost all GFP positive cells in NSN2 Base + EGF media. Significantly,
the WNT8B-GFP positive cells lost their GFP expression in only 12 hours; after 36 hours,
almost all cells became WNT8B-GFP negative (Figure 37). Therefore, DAPT seems to
be necessary for the maintenance of WNT8B-GFP positive cells. With only EGF as the
growth factor, DAPT significantly promoted the maintenance of this cell population and
reduced the differentiation of the WNT8B-GFP positive cells. Thus, we found the
conditions for culturing the WNT8B-GFP positive cells, successfully maintained this cell
population, and reduced their differentiation.

49

Figure 34. Neural spheres formed at 35 days after FACS sorting. A. Neural sphere in
NSN2 Base + EGF + bFGF media; B. Neural sphere in NSN2 Base + EGF +bFGF
+DMSO media; C. Neural spheres NSN2 Base + EGF + bFGF + 2μM DAPT media; D.
Neural sphere in NSN2 Base + bFGF media; E. Neural spheres in NSN2 Base + bFGF +
DMSO media; F. Neural spheres NSN2 Base + bFGF + 2μM DAPT media; G. Neural
spheres in NSN2 Base + EGF media; H. Neural sphere in NSN2 Base + EGF + DMSO
media; I. Neural sphere in NSN2 Base + EGF + 2μM DAPT media. Green: WNT8B-GFP
positive; Blue: Hoechst.
50

Figure 35. Schematic showing the growth of the
spheres in different conditions. DAPT
significantly increased the percentage of WNT8-
GFP positive cells with EGF or with both EGF
and bFGF as growth factors.

51
Figure 36. A. Percentage of GFP positive cells in neural spheres under different
conditions. Noticeably, in common NSN2 media, the percentage of WNT8B-GFP positive
cells was about 20% at 35 days after FACS sorting. However, after treatment with DAPT,
the percentage of GFP positive cells was significantly increased. With only EGF as the
growth factor, treatment of DAPT promoted the maintenance of WNT8B-GFP positive
cells and reduced their differentiation. The percentage of GFP positive cells was up to 80-
90%. B. Number of GFP negative cells and GFP positive cells in each condition.

Figure 37. DAPT withdrawal. A. Neural spheres in NSN2 Base + EGF + 2μM DAPT media
after 35 days; B.12 hours after removal of DAPT; C. 36 hours after removal of DAPT.
Green: WNT8B-GFP positive. Blue: Hoechst. Scale bar: 100um

3.2 In vitro expansion of WNT8B-GFP positive cells
Even though DAPT could support the self-renewal and maintenance of WNT8B-GFP
positive cells, it was difficult to obtain spheres and the size of the spheres was much
smaller than the neural spheres formed in NSN2 media with EGF and bFGF. We tried to
culture a large amount of WNT8B-GFP positive cells in flat-bottomed dishes with DAPT,
with very interesting results. WNT8B-GFP positive cells lost GFP expression in the flat-
bottomed dishes even with DAPT treatment and the size of the spheres was much bigger
than the spheres cultured in the 96-well plate (Figure 38).
52

Figure 38. WNT8B-GFP positive cells
lose GFP expression and generate GFP
negative cells in flat-bottomed dishes
even in the presence of DAPT.

This result led us to pay attention to the importance of cell-cell contact for the
development of NSCs. We hypothesized that the loss of WNT8B-GFP positive cells in
flat-bottomed dishes with DAPT was due to the lack of cell-cell contact in the flat-bottomed
dishes. Therefore, we tried a bead assay to culture the WNT8B-GFP positive cells. Before
12 hours of FACS sorting, we treated the beads with DAPT in EGF NSN2 media. After
collecting the WNT8B-GFP positive cells via FACS sorting, WNT8B-GFP positive cells
were cultured with beads with DAPT in EGF NSN2 media. We found that the bead assay
improved the growth and self-renewal of WNT8B-GFP positive cells, which suggested
that cell-cell contact was important for the self-renewal and development of the WNT8B-
GFP positive cells (Figure 39).
53

Figure 39. Bead assay improved the growth and self-renewal of WNT8B-GFP positive
cells. Scale bar (right): 50.5um

3.3 Expanded WNT8B-GFP positive cells expressed neural stem/progenitor
markers
Immunofluorescence staining was done on the expanded WNT8B-GFP positive cells
(Figure 40). The expanded WNT8B-GFP positive cells expressed neural stem/progenitor
markers such as Sox2 and Nestin, indicating they represent this population. They also
generated WNT8B-GFP
-
Sox2
+
neural progenitors, suggesting they maintained their
function as NSCs.

54

Figure 40. Expanded WNT8B-GFP positive cells expressed neural stem/progenitor
markers Sox2 and Nestin, indicating the expanded WNT8B-GFP positive cells are neural
precursors. They also generated WNT8B-GFP
-
Sox2
+
neural progenitors.

3.4 Expanded WNT8B-GFP positive cells generated WNT8B-GFP negative neural
precursor cells
The beads with WNT8B-GFP positive cells treated with DAPT in NSN2 with EGF
media were transferred into NSN2 media (with EGF and bFGF as growth factors). The
majority of the WNT8B-GFP positive cells lost the GFP expression and generated GFP-
negative cells. The morphology of the cells attached to the plate (Figure 41) resembled
neural precursors with extended processes forming lattices and networks. This suggested
that GFP positive cells were not arrested in the WNT8B-GFP expressing state and upon
DAPT withdrawal were able to initiate differentiation into progenitor states. Future
immunofluorescence staining of neural precursor markers will be needed for further
55
characterization. We will also perform neuronal differentiation to demonstrate that the
FACS-sorted WNT8B-GFP positive cells can generate all types of cells in the hierarchy
during neurogenesis in vitro, as we mentioned previously.

Figure 41. WNT8B-GFP positive cells lose GFP expression resulting in GFP negative
cells. The morphology of the attached cells resembles neural precursors with extended
processes forming lattices and networks. Scale bar: A. 10.1um; B. 50.1um
56
Discussion
By using an in vitro human hippocampal precursor reporter model, we found defects
in CHARGE hippocampal neural precursors. This is the first evidence implicating a defect
in hippocampal NSCs in CHARGE patients. The candidate drug CHIC-35 rescued the
defective CHARGE WNT8B-GFP positive cells and the morphology of CHARGE neural
spheres in vitro, which indicates that CHIC-35 may rescue the defective hippocampal
NSCs and neurogenesis in CHARGE patients. The migration and differentiation of
CHARGE NCCs generated from the CHIC-35 treated neural spheres were also rescued.
Immunostaining for NCC markers and Phalloidin staining could provide further evidence
for the rescue of NCCs. Since CHARGE WNT8B-GFP positive cells were too vulnerable
to survive the FACS sorting process, we attempted to culture the FACS-sorted CHARGE
WNT8B-GFP positive cells treated with CHIC-35 to see if they could survive the FACS
sorting process. During osteogenic differentiation, even though we found a significant
difference between the untreated and CHIC-35 treatment groups, we could not ensure
that the NCCs had already differentiated into osteoblasts. Therefore, immunostaining for
osteoblast markers could provide convincing evidence for a defect in osteogenic
differentiation in CHARGE. Just recently, Boldrini et al. have suggested that human
hippocampal neurogenesis persists throughout aging, which indicates that ongoing
hippocampal neurogenesis sustains human specific cognitive function throughout life.
Therefore, rescuing hippocampal NSCs may help promote brain function in CHARGE
patients during their adulthood.

57
We found that DAPT, a Notch signaling inhibitor, promoted the self-renewal and
maintenance of WNT8B-GFP positive cells generated from the WNT8B-GFP reporter
iPSC line, which indicates that DLK1 may play an essential role in self-renewal and
maintenance of primary human hippocampal NSCs. We have obtained evidence for the
importance of cell-cell contact in the self-renewal and maintenance of the WNT8B-GFP
positive cells, indicating that the cell-cell contact of the NSCs may also play an important
role in self-renewal of primary human hippocampal NSCs. However, since the chemical
inhibitor DAPT inhibits the Notch signaling pathway by inhibiting γ-secretase, which is
also involved in other signaling pathways, we could not conclude that the action of DLK1
on the Notch signaling pathway is essential for the self-renewal of human hippocampal
primordial cells. To test if DLK1 and Notch signaling pathway are indeed involved in the
self-renewal of WNT8B-GFP positive hippocampal primordial cells, we can knockdown
the transcriptional factor RBPJ-kappa in the Notch signaling pathway. We hypothesized
that RBPJ-kappa knockdown in WNT8B-GFP iPSCs may generate WNT8B-GFP positive
cells which are defective in differentiation and have a limited loss of GFP expression. The
immunofluorescence staining showed that the expanded WNT8B-GFP positive cells
expressed neural stem/progenitor markers such as Sox2 and Nestin, indicating the
expanded WNT8B-GFP positive cells are neural precursors. They also generated
WNT8B-GFP
-
Sox2
+
neural progenitors. To provide further evidence that the expanded
WNT8B-GFP positive cells are stem cells, we also need to do RNA sequencing of
WNT8B-GFP positive cells to see if the stem cell markers and hippocampal neural stem
cell-related genes are significantly unregulated. In addition, neuronal differentiation needs
to be carried out to show that expanded WNT8B-GFP positive cells have the function of
58
hippocampal NSCs. The Notch signaling pathway plays essential roles in the
development of different types of stem cells. This study suggests a potential role for the
Notch signaling pathway and the noncanonical Notch ligand DLK1 in the self-renewal of
primary human hippocampal primordial cells and in early neurogenesis. In vitro expansion
of human hippocampal primordial cells also provides a promising tool for future human
hippocampal NSC studies or stem cell replacement therapy.

Citations
Altman, J., and S. A. Bayer. 1990. 'Mosaic organization of the hippocampal
neuroepithelium and the multiple germinal sources of dentate granule cells', J
Comp Neurol, 301: 325-42.
Amaral, D. G., and M. P. Witter. 1989. 'The three-dimensional organization of the
hippocampal formation: a review of anatomical data', Neuroscience, 31: 571-91.
Anand, K. S., and V. Dhikav. 2012. 'Hippocampus in health and disease: An overview',
Ann Indian Acad Neurol, 15: 239-46.
Asad, Z., A. Pandey, A. Babu, Y. Sun, K. Shevade, S. Kapoor, I. Ullah, S. Ranjan, V.
Scaria, R. Bajpai, and C. Sachidanandan. 2016. 'Rescue of neural crest-derived
phenotypes in a zebrafish CHARGE model by Sox10 downregulation', Hum Mol
Genet, 25: 3539-54.
Bajpai, R., D. A. Chen, A. Rada-Iglesias, J. Zhang, Y. Xiong, J. Helms, C. P. Chang, Y.
Zhao, T. Swigut, and J. Wysocka. 2010. 'CHD7 cooperates with PBAF to control
multipotent neural crest formation', Nature, 463: 958-62.
59
Bajpai, R., G. Coppola, M. Kaul, M. Talantova, F. Cimadamore, M. Nilbratt, D. H.
Geschwind, S. A. Lipton, and A. V. Terskikh. 2009. 'Molecular stages of rapid and
uniform neuralization of human embryonic stem cells', Cell Death Differ, 16: 807-
25.
Bajpai, R., J. Lesperance, M. Kim, and A. V. Terskikh. 2008. 'Efficient propagation of
single cells Accutase-dissociated human embryonic stem cells', Mol Reprod Dev,
75: 818-27.
Baladron, V., M. J. Ruiz-Hidalgo, M. L. Nueda, M. J. Diaz-Guerra, J. J. Garcia-Ramirez,
E. Bonvini, E. Gubina, and J. Laborda. 2005. 'dlk acts as a negative regulator of
Notch1 activation through interactions with specific EGF-like repeats', Exp Cell
Res, 303: 343-59.
Basson, M. A., and C. van Ravenswaaij-Arts. 2015. 'Functional Insights into Chromatin
Remodelling from Studies on CHARGE Syndrome', Trends Genet, 31: 600-11.
Blake, K. D., S. L. Davenport, B. D. Hall, M. A. Hefner, R. A. Pagon, M. S. Williams, A. E.
Lin, and J. M. Graham, Jr. 1998. 'CHARGE association: an update and review for
the primary pediatrician', Clin Pediatr (Phila), 37: 159-73.
Boldrini, M., C. A. Fulmore, A. N. Tartt, L. R. Simeon, I. Pavlova, V. Poposka, G. B.
Rosoklija, A. Stankov, V. Arango, A. J. Dwork, R. Hen, and J. J. Mann. 2018.
'Human Hippocampal Neurogenesis Persists throughout Aging', Cell Stem Cell, 22:
589-99 e5.
Bonaguidi, M. A., M. A. Wheeler, J. S. Shapiro, R. P. Stadel, G. J. Sun, G. L. Ming, and
H. Song. 2011. 'In vivo clonal analysis reveals self-renewing and multipotent adult
neural stem cell characteristics', Cell, 145: 1142-55.
60
Borghese, L., D. Dolezalova, T. Opitz, S. Haupt, A. Leinhaas, B. Steinfarz, P. Koch, F.
Edenhofer, A. Hampl, and O. Brustle. 2010. 'Inhibition of notch signaling in human
embryonic stem cell-derived neural stem cells delays G1/S phase transition and
accelerates neuronal differentiation in vitro and in vivo', Stem Cells, 28: 955-64.
Bosman, E. A., A. C. Penn, J. C. Ambrose, R. Kettleborough, D. L. Stemple, and K. P.
Steel. 2005. 'Multiple mutations in mouse Chd7 provide models for CHARGE
syndrome', Hum Mol Genet, 14: 3463-76.
Dupin, E., G. W. Calloni, J. M. Coelho-Aguiar, and N. M. Le Douarin. 2018. 'The issue of
the multipotency of the neural crest cells', Dev Biol.
Encinas, J. M., T. V. Michurina, N. Peunova, J. H. Park, J. Tordo, D. A. Peterson, G.
Fishell, A. Koulakov, and G. Enikolopov. 2011. 'Division-coupled astrocytic
differentiation and age-related depletion of neural stem cells in the adult
hippocampus', Cell Stem Cell, 8: 566-79.
Engelen, E., U. Akinci, J. C. Bryne, J. Hou, C. Gontan, M. Moen, D. Szumska, C. Kockx,
W. van Ijcken, D. H. Dekkers, J. Demmers, E. J. Rijkers, S. Bhattacharya, S.
Philipsen, L. H. Pevny, F. G. Grosveld, R. J. Rottier, B. Lenhard, and R. A. Poot.
2011. 'Sox2 cooperates with Chd7 to regulate genes that are mutated in human
syndromes', Nat Genet, 43: 607-11.
Eriksson, P. S., E. Perfilieva, T. Bjork-Eriksson, A. M. Alborn, C. Nordborg, D. A. Peterson,
and F. H. Gage. 1998. 'Neurogenesis in the adult human hippocampus', Nat Med,
4: 1313-7.
61
Feng, W., M. A. Khan, P. Bellvis, Z. Zhu, O. Bernhardt, C. Herold-Mende, and H. K. Liu.
2013. 'The chromatin remodeler CHD7 regulates adult neurogenesis via activation
of SoxC transcription factors', Cell Stem Cell, 13: 62-72.
Goncalves, J. T., S. T. Schafer, and F. H. Gage. 2016. 'Adult Neurogenesis in the
Hippocampus: From Stem Cells to Behavior', Cell, 167: 897-914.
Hsu, P., A. Ma, M. Wilson, G. Williams, J. Curotta, C. F. Munns, and S. Mehr. 2014.
'CHARGE syndrome: a review', J Paediatr Child Health, 50: 504-11.
Issekutz, K. A., J. M. Graham, Jr., C. Prasad, I. M. Smith, and K. D. Blake. 2005. 'An
epidemiological analysis of CHARGE syndrome: preliminary results from a
Canadian study', Am J Med Genet A, 133A: 309-17.
Janssen, N., J. E. Bergman, M. A. Swertz, L. Tranebjaerg, M. Lodahl, J. Schoots, R. M.
Hofstra, C. M. van Ravenswaaij-Arts, and L. H. Hoefsloot. 2012. 'Mutation update
on the CHD7 gene involved in CHARGE syndrome', Hum Mutat, 33: 1149-60.
Jones, K. M., N. Saric, J. P. Russell, C. L. Andoniadou, P. J. Scambler, and M. A. Basson.
2015. 'CHD7 maintains neural stem cell quiescence and prevents premature stem
cell depletion in the adult hippocampus', Stem Cells, 33: 196-210.
Kopan, R., and M. X. Ilagan. 2009. 'The canonical Notch signaling pathway: unfolding the
activation mechanism', Cell, 137: 216-33.
Li, B., M. Carey, and J. L. Workman. 2007. 'The role of chromatin during transcription',
Cell, 128: 707-19.
Liu, Y., P. Li, K. Liu, Q. He, S. Han, X. Sun, T. Li, and L. Shen. 2014. 'Timely inhibition of
Notch signaling by DAPT promotes cardiac differentiation of murine pluripotent
stem cells', PLoS One, 9: e109588.
62
Lugert, S., O. Basak, P. Knuckles, U. Haussler, K. Fabel, M. Gotz, C. A. Haas, G.
Kempermann, V. Taylor, and C. Giachino. 2010. 'Quiescent and active
hippocampal neural stem cells with distinct morphologies respond selectively to
physiological and pathological stimuli and aging', Cell Stem Cell, 6: 445-56.
Micucci, J. A., W. S. Layman, E. A. Hurd, E. D. Sperry, S. F. Frank, M. A. Durham, D. L.
Swiderski, J. M. Skidmore, P. C. Scacheri, Y. Raphael, and D. M. Martin. 2014.
'CHD7 and retinoic acid signaling cooperate to regulate neural stem cell and inner
ear development in mouse models of CHARGE syndrome', Hum Mol Genet, 23:
434-48.
Nueda, M. L., V. Baladron, B. Sanchez-Solana, M. A. Ballesteros, and J. Laborda. 2007.
'The EGF-like protein dlk1 inhibits notch signaling and potentiates adipogenesis of
mesenchymal cells', J Mol Biol, 367: 1281-93.
Okuno, H., F. Renault Mihara, S. Ohta, K. Fukuda, K. Kurosawa, W. Akamatsu, T.
Sanosaka, J. Kohyama, K. Hayashi, K. Nakajima, T. Takahashi, J. Wysocka, K.
Kosaki, and H. Okano. 2017. 'CHARGE syndrome modeling using patient-iPSCs
reveals defective migration of neural crest cells harboring CHD7 mutations', Elife,
6.
Pagon, R. A., J. M. Graham, Jr., J. Zonana, and S. L. Yong. 1981. 'Coloboma, congenital
heart disease, and choanal atresia with multiple anomalies: CHARGE association',
J Pediatr, 99: 223-7.
Patten, S. A., N. L. Jacobs-McDaniels, C. Zaouter, P. Drapeau, R. C. Albertson, and F.
Moldovan. 2012. 'Role of Chd7 in zebrafish: a model for CHARGE syndrome',
PLoS One, 7: e31650.
63
Pauli, S., R. Bajpai, and A. Borchers. 2017. 'CHARGEd with neural crest defects', Am J
Med Genet C Semin Med Genet, 175: 478-86.
Randall, V., K. McCue, C. Roberts, V. Kyriakopoulou, S. Beddow, A. N. Barrett, F. Vitelli,
K. Prescott, C. Shaw-Smith, K. Devriendt, E. Bosman, G. Steffes, K. P. Steel, S.
Simrick, M. A. Basson, E. Illingworth, and P. J. Scambler. 2009. 'Great vessel
development requires biallelic expression of Chd7 and Tbx1 in pharyngeal
ectoderm in mice', J Clin Invest, 119: 3301-10.
Schnetz, M. P., C. F. Bartels, K. Shastri, D. Balasubramanian, G. E. Zentner, R. Balaji, X.
Zhang, L. Song, Z. Wang, T. Laframboise, G. E. Crawford, and P. C. Scacheri.
2009. 'Genomic distribution of CHD7 on chromatin tracks H3K4 methylation
patterns', Genome Res, 19: 590-601.
Siebert, J. R., J. M. Graham, Jr., and C. MacDonald. 1985. 'Pathologic features of the
CHARGE association: support for involvement of the neural crest', Teratology, 31:
331-6.
Spalding, K. L., O. Bergmann, K. Alkass, S. Bernard, M. Salehpour, H. B. Huttner, E.
Bostrom, I. Westerlund, C. Vial, B. A. Buchholz, G. Possnert, D. C. Mash, H. Druid,
and J. Frisen. 2013. 'Dynamics of hippocampal neurogenesis in adult humans',
Cell, 153: 1219-27.
Sperry, E. D., E. A. Hurd, M. A. Durham, E. N. Reamer, A. B. Stein, and D. M. Martin.
2014. 'The chromatin remodeling protein CHD7, mutated in CHARGE syndrome,
is necessary for proper craniofacial and tracheal development', Dev Dyn, 243:
1055-66.
64
Venere, M., Y. G. Han, R. Bell, J. S. Song, A. Alvarez-Buylla, and R. Blelloch. 2012. 'Sox1
marks an activated neural stem/progenitor cell in the hippocampus', Development,
139: 3938-49.
Verloes, A. 2005. 'Updated diagnostic criteria for CHARGE syndrome: a proposal', Am J
Med Genet A, 133A: 306-8.
Vissers, L. E., C. M. van Ravenswaaij, R. Admiraal, J. A. Hurst, B. B. de Vries, I. M.
Janssen, W. A. van der Vliet, E. H. Huys, P. J. de Jong, B. C. Hamel, E. F.
Schoenmakers, H. G. Brunner, J. A. Veltman, and A. G. van Kessel. 2004.
'Mutations in a new member of the chromodomain gene family cause CHARGE
syndrome', Nat Genet, 36: 955-7.
Winner, B., and J. Winkler. 2015. 'Adult neurogenesis in neurodegenerative diseases',
Cold Spring Harb Perspect Biol, 7: a021287.
Yu, D. X., M. C. Marchetto, and F. H. Gage. 2014. 'How to make a hippocampal dentate
gyrus granule neuron', Development, 141: 2366-75.

Abstract (if available)

Abstract CHARGE syndrome is a rare genetic syndrome and the majority of CHARGE patients have defective neural crest cells (NCCs) due to a mutation in the gene CHD7. However, CHD7 is not only essential for the development of NCCs, it also regulates neurogenesis. This led us to hypothesize that CHARGE patients may also have defective hippocampal neural stem cells (hip-NSCs). To study hip-NSCs, our lab previously established a WNT8Bᵉⁿ-GFP reporter system in human stem cells. Using two WNT8Bᵉⁿ-GFP CHARGE patient iPSC lines, we showed the first evidence of defective human hip-NSCs in CHARGE syndrome in vitro. A candidate drug, CHIC-35, rescued not only the migration and differentiation of CHARGE NCCs, but also the defective CHARGE hip-NSCs generated from the patient iPSCs. This study provides new insight into the mechanism of CHARGE syndrome and a potential therapeutic for CHARGE patients that may improve their brain functions. In addition, we also achieved the first in vitro expansion of human WNT8Bᵉⁿ-GFP positive hip-NSCs using normal WNT8Bᵉⁿ-GFP iPSC lines. We found that a potential Notch inhibitor, DAPT, promoted the self-renewal and maintenance of human hip-NSCs in vitro, which indicates that the Notch ligand DLK1 may play an essential role in self-renewal and maintenance of primary hip-NSCs. This study provides a promising resource of human hippocampal primordial cells, not only for future stem cell transplantation therapy, but also for further study of the earliest hip-NSCs. We implicate the possible role of the DLK1 and Notch signaling pathway in self-renewal of the earliest hip-NSCs. In conclusion, we utilized normal and CHARGE WNT8Bᵉⁿ-GFP reporter lines to study hip-NSCs in development and disease in vitro to establish potential therapeutics and new insights into the mechanisms of human diseases.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Chromatin remodeling factors Chd7 and Phf6 in craniofacial and heart development

PDF

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

PDF

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

PDF

Opposing function of chd7 and phf6 in zebrafish craniofacial development

PDF

An iPSC-based biomarker strategy to identify neuroregenerative responders to allopregnanolone

PDF

Understanding the role of APP and DYRK1A in human brain pericytes

PDF

An induced neuron cell model derived from human olfactory epithelial cells

PDF

Effects of a GSK-3 inhibitor on retroviral-mediated gene transfer to human CD34+ hematopoietic progenitor cells

PDF

IPS and CNS cell models of peroxisomal disorders

PDF

Characterization of human embryonic stem cell derived retinal pigment epithelial cells for age-related macular degeneration

PDF

Gene expression and genetic variation of ERG is associated with inflammation in endothelial cells and risk of coronary artery disease in humans

PDF

The role of LGR5 in the pathogenesis of Ewing sarcoma: a marker of aggressive disease and a contributor to the malignant phenotype

PDF

Molecular basis of mouse epiblast stem cell and human embryonic stem cell self‐renewal

PDF

Characterization of new stem/progenitor cells in skin appendages

PDF

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

PDF

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

PDF

Interaction of epigenetics and SMAD signaling in stem cells and diseases

PDF

The calcium-sensing receptor in the specification of normal and malignant hematopoietic cell localization in the bone marrow microenvironment

PDF

IFN-α treatment inhibits induction of type-2 innate lymphoid cell dependent airway hyperresponsiveness

PDF

Elucidating neurodevelopmental consequences of syngap1 mutations and inactivated functional regions in human iPSC-derived neurons

Asset Metadata

Creator Wang, Simiao (author)

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

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Medicine

Publication Date 10/17/2018

Defense Date 06/18/2018

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

Tag CHARGE syndrome,human hippocampal primordial cells,iPSCs,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bajpai, Ruchi (committee chair), Bonaguidi, Michael (committee member), Patel, Pragna (committee member)

Creator Email simiaowa@usc.edu,simiaowang0220@outlook.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-82994

Unique identifier UC11671403

Identifier etd-WangSimiao-6875.pdf (filename),usctheses-c89-82994 (legacy record id)

Legacy Identifier etd-WangSimiao-6875.pdf

Dmrecord 82994

Document Type Thesis

Format application/pdf (imt)

Rights Wang, Simiao

Type texts

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

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

Repository Name University of Southern California Digital Library

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