Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
IPS and CNS cell models of peroxisomal disorders
(USC Thesis Other)
IPS and CNS cell models of peroxisomal disorders
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
IPS AND CNS CELL MODELS OF PEROXISOMAL DISORDERS
by
Xiaoming Wang
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
(MOLECULAR BIOLOGY)
August 2012
Copyright 2012 Xiaoming Wang
ii
ACKNOWLEDGEMENTS
This dissertation would not have been possible without the support of
many important people in my life. I would like to take this opportunity to
acknowledge their contributions to my work and my life.
I would like to thank my mentor, Dr. Joseph Hacia, for his guidance,
encouragement, trust, and support throughout my graduate studies, which will
also benefit my future. He guided me to be open-minded in scientific view and
vigorous in research work. He encouraged me to be an independent thinker. He
trusted me and gave me great freedom to pursue my interests in science. Most
important, I respect him as a great person, who is always a good listener and
always trying his best to help people in pain with his knowledge and kind heart.
I would also like to thank my committee members Dr. Michael Stallcup, Dr.
Andrew Smith, and Dr. Wange Lu for their suggestions and help.
My dissertation work has been greatly strengthened through
collaborations. I would like to thank Dr. Steven Steinberg and Ann Moser, in the
Kennedy Krieger Institute for providing the patient fibroblasts and performing the
lipid analysis. I would like to thank Dr. Wange Lu, Dr. Qilong Ying, Dr. Peilin
Zhang and Cunye Qu for helping me initiate the reprogramming system. In
particular, I would like to greatly thank Dr. Peilin Zhang for helping me conquer
many critical problems in my research. I would also like to thank Dr. Darryl
Shibata, Dr. Robert Hsiu-Ping Chow, and Madison Zitting for their time and
expertise.
iii
I would also like to take this opportunity to express my appreciation to the
former and current lab members in Dr. Hacia’s lab. Dr. Patricia Dranchak
generated the global gene expression data for CCALD and healthy control
fibroblasts. I would like to deliver my special thank to Winnie Win-Yan Yik, who
has been my companion throughout my graduate school life. Winnie and I
together generated and maintained all the iPS cell lines. We shared all our
frustrations and happiness together in the past years. She also brought great joy
and comfort to my life. Our friendship is one of the most precious things I have in
my life. Again, I must give my special gratitude to Dr. Joe Haica, Winnie Win-Yan
Yik, Dr. Richard Pelikan, and Ying Huang for walking me through great difficulties.
I also owe my deepest gratitude to my parents, Yuhua Liu and Han Wang,
my parents-in-law, Guanghui Zhou and Zhiqiang Li, and all the other family
members and friends for their love and support.
Finally, I would like to thank my beloved husband, Dr. Peng Li, for his love
and support, for seeing and helping me grow up, for making me strong, and for
everything you did for me in the past thirteen years. I wouldn’t be here without
you.
iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES vi
LIST OF FIGURES vii
ABSTRACT x
CHAPTER 1: GENERAL INTRODUCTION 1
1.1 Peroxisome overview 1
1.2 Peroxisome biogenesis and assembly 7
1.3 Peroxisomal disorders 10
1.4 Mouse models of peroxisomal disorders 17
1.5 Disease model induced pluripotent stem cells 20
CHAPTER 2: EXPERIMENTAL DESIGN AND PREPARATION 25
2.1 Cell culture and cell lines 25
2.2 Retrovirus preparation 27
2.3 Cellular reprogramming 28
2.4 Immunocytochemistry and enzyme activity assays 29
2.5 Gene expression profiling 30
2.6 DNA methylation profiling and bisulfite sequencing
confirmation
31
2.7 DNA sequencing of mutations in PBD and CCALD 32
2.8 Cytogenetic analysis 32
2.9 In vitro differentiation and teratoma assays 33
2.10 Lipid analysis 33
2.11 CNS lineage differentiation 34
CHAPTER 3: GENERATION AND CONFIRMATION OF HEALTHY
CONTROL, PBD AND CCALD PATIENT IPS CELLS
37
3.1 Abstract 37
3.2 Introduction 37
3.3 Results 39
3.3.1 Derivation of candidate iPS cells 39
3.3.2 Genetic characteristics of candidate iPS cells 40
3.3.3 Candidate iPS cells express hES markers 47
3.3.4 Candidate iPS cells have unique DNA methylation
profiles
51
3.3.5 Candidate iPS cells are pluripotent 55
3.4 Discussion 58
v
CHAPTER 4: FUNCTIONAL CHARACTERIZATION OF PBD PATIENT,
CCALD PATIENT AND HEALTHY CONTROL IPS CELLS
61
4.1 Abstract 61
4.2 Introduction 62
4.3 Results 66
4.3.1 Lipid profiles of patient and healthy control cells 66
4.3.2 Differential gene expression among PBD-ZSD
patient and healthy control cells
71
4.3.3 Differential gene expression among CCALD patient
and healthy control cells
76
4.3.4 Differential DNA methylation among patient and
healthy control cells
90
4.4 Discussion 90
CHAPTER 5: DIFFERENTIATION AND CHARACTERIZATION OF PBD
PATIENT CNS CELLS
102
5.1 Abstract 102
5.2 Introduction 102
5.3 Results 109
5.3.1 Variable neural differentiation potency of iPS cells 109
5.3.2 Differentiation and characterization of motor
neurons
116
5.3.3 Differentiation and characterization of
oligodendrocytes
121
5.4 Discussion 127
CHAPTER 6: CONCLUSIONS AND PERSPECTIVES 132
6.1 Reprogramming, pluripotency and peroxisomal disorders 133
6.2 Differential gene expression in PBD and CCALD iPS cells 134
6.3 Influence of copy number variance and epigenetic
variance in iPS cells
137
6.4 Defects in PBD iPS cell derived CNS cells 140
6.5 Future directions 142
BIBLIOGRAPHY 145
vi
LIST OF TABLES
Table 2.1: Skin fibroblast and iPS cell information 26
Table 3.1: Summary of copy number variations in iPSCs 43
vii
LIST OF FIGURES
Figure 1.1: Comparative schematic of mitochondrial and peroxisomal β-
oxidation pathways in human cells
3
Figure 1.2: Phytanic acid catabolism in mammals 4
Figure 1.3: Plasmalogen structure 6
Figure 1.4: Model for peroxisome biogenesis and proliferation 8
Figure 1.5: Model for peroxisomal membrane protein and matrix protein
import
9
Figure 1.6: Schematic representation of the relationships among
different classes of peroxisomal disorders
11
Figure 3.1: Human ES marker expression of representative PBD,
CCALD and healthy control iPSCs
48
Figure 3.2: The gene expression profile of fibroblasts and iPSCs show
dramatic differences that are not strongly influenced by
mutation status
49
Figure 3.3: The DNA methylation profiles and bisulfite sequencing
confirmation of fibroblasts and iPSCs
53
Figure 3.4: Embryoid body and teratoma differentiation of representative
PBD, CCALD and healthy control iPSCs
56
Figure 4.1: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control
fibroblasts and iPSCs
69
Figure 4.2: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in CCALD patient and healthy control
fibroblasts and iPSCs
70
Figure 4.3: DEGs with higher expression in all PBD-ZSD patient iPSCs
relative to healthy control iPSCs enriched for mitochondria
genes
81
viii
Figure 4.4: DEGs with higher expression in all PBD-ZSD patient iPSCs
relative to healthy control iPSCs enriched for ketone body
metabolism
82
Figure 4.5: DEGs with less expression in all PBD-ZSD patient iPSCs
relative to healthy control iPSCs enriched for Wnt signaling
pathway
83
Figure 4.6: DEGs in all PBD-ZSD patient iPSCs relative to healthy
control iPSCs enriched for Wnt signaling pathway
84
Figure 4.7: DEGs between all PBD-ZSD patient iPSCs and healthy
control iPSCs enriched for inflammation
85
Figure 4.8: Less expression of PEX1 gene in PBD-ZSD patient iPSCs
with homozygous PEX1 frameshift mutation relative to
healthy control iPSCs
86
Figure 4.9: qRT-PCR confirmation of differentially expressed genes in
CCALD patient and healthy control iPSCs
87
Figure 4.10: DEGs between CCALD patient iPSCs and healthy control
iPSCs involved in peroxisome turnover
88
Figure 4.11: DEGs between CCALD patient iPSCs and healthy control
iPSCs involved in inflammation
89
Figure 5.1: Scheme of the neural progenitor (NP), motor neuron (MN)
and oligodendrocyte (OD) in vitro differentiation from human
induced pluripotent stem cells (iPSCs)
108
Figure 5.2: Induction of neural progenitors from representative PBD-
ZSD iPSCs
112
Figure 5.3: Variable neural differentiation potency of PBD-ZSD patient
and healthy control iPSCs
113
Figure 5.4: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control
iPSC derived neural progenitors (NP)
115
Figure 5.5: Differentiation and maturation of representative PBD-ZSD
patient iPSC-derived motor neuron progenitors (MNP) and
motor neurons (MN)
118
ix
Figure 5.6: Neurons derived from representative PBD-ZSD patient
iPSCs display typical neuron electrophysiological
characteristics
119
Figure 5.7: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control
iPSC derived motor neuron progenitors (MNP)
120
Figure 5.8: Oligodendrocyte precursors (OP) derived from
representative PBD-ZSD patient and healthy control iPSCs
125
Figure 5.9: Oligodendrocytes (OD) derived from representative PBD-
ZSD patient and healthy control iPSCs
126
x
ABSTRACT
Peroxisomal disorders are a group of genetically heterogeneous metabolic
diseases caused by defects in peroxins, proteins encoded by PEX genes that
function in peroxisome biogenesis, or in a single peroxisomal protein that has a
more targeted effect on specific peroxisome functions. In general, peroxisome
disorders can affect almost every organ system, with especially devastating
effects on the nervous, hepatic, and adrenocortical systems.
Currently, there is no curative therapy or long-term effective treatment
available for peroxisomal disorders. Ongoing pathomechanism studies,
diagnostics, and drug testing are mainly established on patient-derived primary
fibroblasts and Pex gene knockout mouse models, which do not represent the
exact human mutations and most clinical aspects of the human disease.
In this thesis, I describe a new model system which we established for
studying the pathology of peroxisomal disorders and testing new therapeutic
agents. We generated induced pluripotent stem cells (iPSCs) from primary skin
fibroblasts of multiple healthy controls and patients with peroxisomal biogenesis
disorders (PBD), caused by genetic defects in PEX genes, or the childhood
cerebral form of X-linked adrenoleukodystrophy (CCALD), caused by genetic
defects in the ABCD1 gene that encodes a peroxisome membrane protein
involved in very long chain fatty acid (VLCFA) metabolism. Candidate iPSCs
were subject to global expression, DNA methylation, and genotyping analysis
and tested for pluripotency through in vitro embryoid body differentiation and in
xi
vivo teratoma formation. We characterized the gene expression and biochemical
profiles of these patient-specific iPSCs and further differentiated these iPSCs into
pathologically related central nervous system cell (CNSC) lineages, including
neural progenitors, motor neurons, and oligodendrocytes.
Our molecular characterization of iPSCs and CNSCs provided a novel
perspective into disease mechanisms that supports leading hypotheses
regarding disease pathogenesis including the pivotal roles of neuroinflammation,
lipid metabolism, and aberrant mitochondrial function. Our novel resources also
provide a first step required for the development and interpretation of patient-
specific model systems that investigate non-cell autonomous processes relevant
to the etiology of peroxisomal disorders. These iPSC and CNS cell resources
could also have applications for high content screening (HCS) of chemical
libraries for candidate drugs that directly address the cell type specificity of
disease and the nature of the mutations found in the patient population.
1
CHAPTER 1
GENERAL INTRODUCTION
1.1 Peroxisome overview
Peroxisomes are ubiquitous organelles that present in almost all human
cells, with the exception of the mature erythrocyte (Hruban, Gotoh et al. 1974),
The number of peroxisomes per cell can range from a few hundred to a few
thousand (Hruban, Gotoh et al. 1974). Based on electron microscopy analysis,
they appear as single, membrane-bound, spherical organelles ranging in size
from 0.1μm to 1μm in diameter (Moser 1991).
Peroxisomes were first described as ‘microbodies’ by a Swedish PhD
student and later electron microscopy scientist and pioneer, Johannes Rhodin, in
his thesis on the ultrastructural studies of proximal tubule cells from mouse
kidney (Rhodin 1954). However, these microbodies were not characterized as a
cellular organelle until the end of the 1960s when the Nobel Prize laureate (1974),
Christian de Duve, and his team succeeded in isolation of subcellular organelles,
such as microbodies from rat liver and studied their biochemical functions (De
Duve and Baudhuin 1966). De Duve’s group discovered the co-localization of
H
2
O
2
-producing oxidases as well as H
2
O
2
-degrading catalase in the organelle
matrix and established the name ‘peroxisome’.
Since then, an increasing number of catabolic and anabolic biochemical
functions have been associated with peroxisomes, which demonstrates the
2
indispensable role of peroxisomes in lipid, amino acid, polyamine, and bile acid
metabolism.
Fatty acid (FA) β-oxidation is a general feature of virtually all types of
peroxisomes among eukaryotic species, from yeasts to plants to mammals.
Peroxisomes are the only site of FA β-oxidation in yeast and plants, whereas in
human and other mammals β-oxidation occurs in both mitochondria and
peroxisomes (Figure 1.1). Mitochondrial and peroxisomal β-oxidation are
mechanistically similar; nevertheless, defects in these processes result in
different clinical phenotypes (Rinaldo, Matern et al. 2002; Wanders 2004). The
peroxisomal and mitochondrial β-oxidations differ in their substrate specificities
and means of fatty acid transportation (Wanders 2004). The very long-chain FA
(VLCFAs), especially the saturated hexacosanoic acid (C26:0 FA), can only be β-
oxidized in peroxisomes. In contrast, the short- and medium-chain FAs are
exclusively subject to mitochondrial -oxidation. Likewise, long-chain FAs are
predominantly subject to mitochondrial -oxidation.
As mentioned above, the first step of β-oxidation in mammalian
peroxisomes is catalyzed by acyl-CoA oxidases. The next step is the hydration of
the bond between C-2 and C-3 followed by the oxidation of the hydroxyl group
into a keto group by bifunctional proteins with both enoyl-CoA hydratase and 3-
hydroxy-acyl-CoA dehydrogenase activities. The final step is a thiolysis process
catalyzed by thiolases (Wanders and Waterham 2006).
3
Figure 1.1: Comparative schematic of mitochondrial and peroxisomal β-
oxidation pathways in human cells.
(A) Mitochondria: the FADH2 and NADH, generated in the first and third steps of
fatty acid -oxidation, are directly reoxidized by the respiratory chain. (B)
Peroxisomes: O
2
is the electron acceptor in the first step of beta-oxidation,
resulting in the formation of hydrogen peroxisde, which is reconverted into O
2
by
catalase. Figure is adapted from reference (Wanders and Waterham 2006).
Prior to undergoing β-oxidation, FAs with a methyl-group at the C3
position must go through α-oxidative decarboxylation to produce (n-1) FAs with
the methyl-group at C2 position. This α-oxidation process is strictly confined to
peroxisomes (Wanders and Waterham 2006). Phytanic acid (3,7,11,15-
tetramethyl hexadecanoic acid) is a branched chain fatty acid subject to alpha-
oxidation in the peroxisome (Figure 1.2). The accumulation of large stores of
4
phytanic acid in various tissues can lead to a peripheral neuropathy and loss of
sensory neural functions that affect vision, hearing, taste, and smell.
Figure 1.2: Phytanic acid catabolism in mammals.
Phytanic acid in ruminant fats is derived from phytol produced during the
bacterial degradation of chlorophyll in their rumen (first stomach). After
conversion to its CoA thioester, phytanic acid undergoes -oxidation, yielding
pristanic acid. This fatty acid then undergoes three subsequent rounds of -
oxidation in the peroxisome. The resulting medium chain fatty acid exits the
peroxisome and translocates to the mitochondrion where the remaining carbon
chain is degraded by -oxidation. Abbreviations: HACL1 = 2-hydroxyphytanoyl-
CoA lyase; PHYH = phytanoyl-CoA -hydroxylase; PDH = pristanal
dehydrogenase, whose gene is not yet known. Phytanic acid can also be
degraded by -oxidation; however, this activity of this pathway is relatively minor.
Adapted from reference (Watkins, Moser et al. 2010)
2-hydroxy-
phytanoyl-CoA
phytanic acid
pristanal
pristanic acid
phytanoyl-CoA
PHYH
HACL1
PDH
medium chain fatty acids
mitochondrial -oxidation to carbon dioxide and water
multiple rounds
of -oxidation
O
C S CoA
OH
O
C S CoA
O
C OH
C H
O
C OH
O
2-hydroxy-
phytanoyl-CoA
PHYH
HACL1
PDH
multiple rounds
of -oxidation
O
C S CoA
O
C OH
5
Another important peroxisomal function is the disposal of cellular
metabolism by-products, including oxygen, reactive oxygen species (ROS), and
reactive nitrogen species. A number of oxidases in peroxisomes reduce
molecular oxygen (O
2
) to hydrogen peroxisome (H
2
O
2
), which can be disposed
via the activity of catalase, glutathione peroxidase and peroxiredoxin v (also
known as PMP20). The resulting superoxide anions are inactivated by
superoxide dismutases (Singh 1996; Wanders 2004; Wanders and Waterham
2006).
In addition to the catabolism of very long chain and branched chain fatty
acid, peroxisomes also play a major role in the biosynthesis of plasmalogens
(Figure 1.3), ether phospholipids that compose up to 80% of the membrane
phospholipids in the white matter of the brain. In fact, the first two steps of ether-
phospholipid biosynthesis happen exclusively in the peroxisome. In human, 18%
of the total phospholipid mass is made up of plasmalogens. The sn-1 position of
plasmalogens is predominantly occupied by C16:0, C18:0 and C18:1 fatty
alcohols and the sn-2 position usually contains polyunsaturated FAs.
Plasmalogens are the major component of myelin in the central nervous system
(CNS) and are also enriched in liver, kidney, lung, heart, skeletal muscle and
testis. Plasmalogens are thought to play essential roles in oxidative damage
protection, membrane fluidity, intracellular signaling, cholesterol transport and
metabolism. (Nagan and Zoeller 2001; Maeba and Ueta 2003; Maeba and Ueta
2003; Steinberg, Dodt et al. 2006; Wanders and Waterham 2006)
6
Figure 1.3: Plasmalogen structure.
(A) Plasmalogens are glycerophospholipids characterized by the presence of a
vinyl-ether linkage at the sn-1 position and an ester-linkage at the sn-2 position.
R1 and R2 represent straight-chain carbon groups. At the sn-1 position, the
chemical moiety highlighted in red is an alkenyl group, which we use to measure
plasmalogen abundance and molecular composition. These alkenyl groups are
most commonly derived from C16:0, C18:0, or C18:1 fatty alcohols. The sn-2
position of plasmalogens is occupied typically by polyunsaturated fatty acids. X
represents the head group, typically ethanolamine or choline for plasmalogens.
In contrast, (B) diacylglycerophospholipids have ester-linkages at their sn-1 and
sn-2 positions. As above, R1 and R2 represent straight-chain carbon groups and
X represents the head group. Adapted from reference (Moser, Steinberg et al.
2011)
Besides the critical catabolic and biosynthetic functions mentioned above,
peroxisomes are also involved in degradation of pipecolic acid, synthesis of bile
acid, cholesterol, dolichol, and other isoprenoids. Peroxisomes also contain
enzymes involved in glycoxylate, amino acid, and purine and pyrimidine
metabolism. (Moser 1991; Wanders and Waterham 2006)
C
C
C C C
C
C
C
C
C
O
O
O
O
O
O
C
O
O
O
O
O
O
O
O
O
R
R R
R
X X H
H
H H
H
H
H H
P P
2 2
2 2
sn-1
position
sn-2
position
A B
1 1
2 2
head
group
7
1.2 Peroxisome biogenesis and assembly
There are three key stages of the development of mature peroxisomes:
peroxisomal membrane formation and membrane protein insertion, peroxisome
proliferation and the import of proteins into the peroxisomal matrix, the internal
compartment of this organelle. The peroxisomal proteins involved in the
peroxisome assembly are called peroxins, which are encoded by a group of PEX
genes. (Platta and Erdmann 2007)
Two pathways co-exist for peroxisome biogenesis, the fusion pathway and
the fission pathway (Figure 1.4) (Fagarasanu, Fagarasanu et al. 2007). In the
fusion pathway (also called de novo biogenesis), the peroxisomal membrane
originates from specialized compartments of the endoplasmic reticulum (ER)
(Novikoff and Shin 1964) which fuses to form immature peroxisomes, which lack
most of their internal matrix proteins. As discussed below, these matrix proteins
are imported into the peroxisome through the activity of a specific group of PEX
genes. The PEX3, PEX16 and PEX19 proteins serve to facilitate the insertion of
peroxisomal membrane proteins. However, it is believed that new peroxisomes
arise primarily from the division of pre-existing mature peroxisomes, which is
fulfilled by the fission pathway. In this pathway, PEX11 assists in the elongation
and constriction of mature peroxisome, followed by a fission event facilitated by
dynamin-related proteins (DRPs) (Figure 1.4). (Steinberg, Dodt et al. 2006;
Fagarasanu, Fagarasanu et al. 2007; Platta and Erdmann 2007)
8
Figure 1.4: Model for peroxisome biogenesis and proliferation.
The processes of de novo peroxisome biogenesis from the ER and the
peroxisome fission are depicted. This figure is adapted from reference
(Fagarasanu, Fagarasanu et al. 2007).
Peroxisomal matrix proteins are encoded in the nucleus, synthesized on
the free ribosomes in the cytosol and imported posttranslationally into the pre-
existing peroxisomes (Figure 1.5) (Lazarow, Robbi et al. 1982; Lazarow and
Fujiki 1985). Free matrix proteins in the cytosol are recognized by their
peroxisomal targeting signals (PTSs), PTS1 (a C-terminal tripeptide:
(S/A/C)(K/R/H)(L/M) sequence) via PEX5 and PTS2 (a N-terminal nonapeptide:
(R/K)(L/V/I)X(5)(H/Q)(L/A) sequence) via PEX7 attached on PEX5. In both
pathways, the cargo proteins are transported to the surface of the peroxisome
which houses the docking complex of PEX13 and PEX14. Then the cargo
binding PEX5 are integrated to the peroxisomal membrane and the cargo is
imported with the assistance of the really interesting new gene (RING)-finger
9
complex, composed of the PEX2, PEX10 and PEX12 proteins. After releasing the
cargo, PEX5 is monoubiquitinated by the RING-finger complex, recognized and
recycled to the cytosol by the recycling complex of the cytosolic AAA proteins
PEX1 and PEX6 and the membrane protein PEX26 for the next round of import.
(Steinberg, Dodt et al. 2006; Platta, El Magraoui et al. 2007; Platta, El Magraoui
et al. 2009; Prestele, Hierl et al. 2010)
Figure 1.5: Model for peroxisomal membrane protein and matrix protein
import.
De novo peroxisome biogenesis starts with PEX3, PEX16 and PEX19. The
import of membrane proteins into existing peroxisomes requires PEX19 for
recognition, targeting and insertion via docking at PEX3. Nascent peroxisome
matrix proteins are recognized by their amino acid targeting signals (PTS1 via
PEX5 and PTS2 via PEX7) and transported to the PEX13/PEX14 docking
complex. PEX5 integrates into the peroxisome membrane and releases its cargo
for matrix import, mediated by a PEX2/PEX10/PEX12 complex. Later, PEX5 exits
the membrane with the help of a PEX1/PEX6/PEX26 complex, and then cycles
back to the cytosol, ready for another round of import. Adapted from reference
(Steinberg, Dodt et al. 2006).
10
The metabolic functions of peroxisomes require the active transportation
of metabolites across the peroxisomal membrane. A variety of peroxisomal
transporters serve as the metabolite carriers to import and export the metabolites,
like different kinds of FAs and plasmalogens, ADP/ATP and NAD
+
/NADH
(Wanders and Waterham 2006). The peroxisomal ABC transporters are a major
group of peroxisomal metabolite carriers and are involved in transportation of
VLCFAs. A functional ABC transporter is composed of a homo- or heterodimer of
half ABC transporters, each having six transmembrane alpha helices and one
conserved hydrophilic nucleotide-binding domain. The mammalian peroxisomes
contain four types of half ABC transporters, the adrenoleukodystrophy protein
(ALDP) (also called ABCD1), the ALDRP (also called ABCD2), PMP70 (also
called ABCD3), and the PMP70R (also called ABCD4). (Dean and Annilo 2005)
1.3 Peroxisomal disorders
The significance of peroxisomes for normal development and growth has
been revealed by the existence of a group of inherited human diseases, the
peroxisomal disorders. In general, they arise as a result of two types of defects,
those involving peroxisome biogenesis (PBDs) and those involving single
peroxisomal protein deficiencies that do not affect peroxisomal assembly (Figure
1.6).
PBDs are autosomal recessive diseases with an incidence of about 1 in
50,000 at births. PBDs include the Zellweger spectrum disorders (ZSD) and the
rhizomelic chondrodysplasia punctata type 1 (RCDP1). PBD-ZSD represents a
11
continuum of disorders including Zellweger syndrome (ZS), neonatal
adrenoleukodystrophy (NALD), and infantile Refsum disease (IRD), with ZS
being the most and IRD the least severe disorder in the spectrum. Mutations in
12 PEX genes are known to be associated with PBD-ZSD, including PEX1, 2, 3,
5, 6, 10, 13, 14, 16, 19 and 26. These genes play critical roles in the import of
approximately 50 peroxisome matrix protein through the PTS1 pathway.
Figure 1.6: Schematic representation of the relationships among different
classes of peroxisomal disorders.
It has been estimated that over 90% of PBD-ZSD patients carry mutations
in the PEX1, PEX6, PEX10, PEX12 or PEX26 genes (Steinberg, Chen et al.
2004). PEX1 and PEX6 mutations account for about 70% and 10% of PBD-ZSD
patients, respectively (Steinberg, Chen et al. 2004; Yik, Steinberg et al. 2009).
Most recently, a novel defect of peroxisome division caused by a homozygous
12
nonsense mutation in PEX11B gene was identified in a 26-year-old man with
symptoms atypical for PBD, which extended the spectrum of PBD genotypes and
phenotypes (Ebberink, Koster et al. 2012; Thoms and Gartner 2012). PBD-ZSD
patients are often characterized by the elevated levels of VLCFAs and decreased
levels of plasmalogens in blood, urine and cultured primary skin fibroblasts. Due
to the ubiquitous nature of peroxisomes in human cells, almost every organ
system is affected in PBD-ZSD patients, especially the liver and the central
nervous system (CNS). Affected children typically manifest skeletal and
craniofacial abnormalities at birth. In the newborn stage, these children are also
hypotonic, show poor appetites and absorption of fat soluble vitamins, and also
have hepatomegaly, renal cysts and seizures (Steinberg, Dodt et al. 2006;
Wanders and Waterham 2006). Patients typically show variable neuro-
developmental delay, neural migration defects, delayed myelination,
demyelination, retinopathy and perceptive deafness (Barth, Majoie et al. 2004;
Faust, Banka et al. 2005). PBD-ZSD patient life spans are severely shortened,
with ZS patients usually surviving less than one year of age. In contrast, NALD
patients usually live until late infancy and the IRD patients may survive beyond
infancy and even reach young adulthood. The genotype-phenotype correlations
are imperfect and individuals with the same PEX gene mutations can show any
of these three different forms of disease. This indicates possible environmental
contributions to disease as well as the existence of modifier genes.
RCDP1 patients carry mutations in their PEX7 gene, the most common of
which is the L292X nonsense mutation. Their major clinical features include
13
disproportionally short proximal long bones (rhizomelia), punctate calcification of
the cartilage of the rounded ends of long bones (chondrodysplasia punctata), and
similar symptoms found in other PBDs (Steinberg, Dodt et al. 2006; Wanders and
Waterham 2006). RCDP1 differs from PBD-ZSD in that the import of only three
peroxisome matrix proteins are compromised (ACAA1, PHYH, and AGPS).
ACAA1 is a thiolase involved in peroxisomal -oxidation, but is functionally
redundant and thus not directly involved in disease. PHYH is a critical enzyme in
peroxisomal -oxidation (Figure 1.2) and thus RCDP1 patients cannot
metabolize branched chain fatty acids such as phytanic acid. As discussed above,
this can impair functions of the peripheral and sensory nervous systems. AGPS
plays a pivotal role in plasmalogen biosynthesis. Defects in AGPS import into the
peroxisome matrix results in plasmalogen deficiencies, which are primarily
responsible for the clinical manifestations of RCDP1, as discussed above.
The second type of peroxisome disorder is the single peroxisomal enzyme
deficiencies (Figure 1.6). There are more than 10 types of single enzyme
peroxisomal disorders. X-linked adrenoleukodystrophy (X-ALD) is the most
common type, with an estimated incidence of 1 in 17,000 newborn males. X-ALD
is an X-linked disorder caused by mutations in the ABCD1 gene that encodes an
integral peroxisome membrane protein belonging to the ATP-binding cassette
transporter superfamily involved in peroxisome β-oxidation (Berger and Gartner
2006; Moser, Mahmood et al. 2007; Moser, Moser et al. 2007; Berger, Pujol et al.
2010; Kemp and Wanders 2010). The patients are characterized by the elevated
VLCFAs. X-ALD primarily affects the nervous system, adrenal cortex, and testes
14
with highly variable clinical presentations that are influenced by modifier genes
and the environment. Males with ABCD1 mutations develop the childhood
cerebral ALD (CCALD) form of disease approximately 33% of the time and adult
onset adrenomyeloneuropathy (AMN) approximately 45% of the time. (Berger
and Gartner 2006; Berger, Pujol et al. 2010)
CCALD patients typically show symptoms between 5 and 9 years of age
with rapid cerebral demyelination and adrenocortical atrophy. Within a few years
of onset, they suffer dementia and progressive neurological deficits that
eventually lead to death. In contrast, AMN patients show a later onset of disease
(20 - 40 years of age) and present with adrenal insufficiency, a distal axonopathy
in the spinal cord, and peripheral neuropathy (Berger and Gartner 2006; Berger,
Pujol et al. 2010) that results in progressive paralysis of the lower limbs with
debilitating end stage disease. Furthermore, an estimated 10% of male mutation
carriers develop primary adrenocortical insufficiency (Addison’s disease) with no
evidence of nervous system dysfunction. Approximately 50% of female ABCD1
mutation carriers develop AMN-like symptoms later in life (Berger and Gartner
2006; Berger, Pujol et al. 2010).
The molecular mechanism underlying the inflammatory brain
demyelination found in PBD and CCALD patients are not fully known. It has been
hypothesized to be related to the accumulation of saturated very long chain fatty
acids (VLCFAs) in specific CNS cell types (e.g. oligodendrocytes and microglial
cells) and/or lipid classes (e.g. ganglioside, phosphatidylcholine, and cholesterol
ester fractions) (Garashi, Belchis et al. 1976; Theda, Moser et al. 1992; Asheuer,
15
Bieche et al. 2005). Other hypotheses have focused on the roles of oxidative
stress (Gilg, Singh et al. 2000; Vargas, Wajner et al. 2004; Powers, Pei et al.
2005; Fourcade, Lopez-Erauskin et al. 2008; Singh and Pujol 2010), myelin
sheath integrity (Ho, Moser et al. 1995), oligodendrocyte apoptosis and microglial
cell activation (Berger, Pujol et al. 2010; Kemp and Wanders 2010), and CNS cell
membrane receptors (Singh, Brogan et al. 1992).
The first step in peroxisome disorder diagnosis is the biochemical testing
of blood and urine lipid levels, usually including VLCFAs (saturated C26:0 and
monounsaturated C26:1), branched chain fatty acids (phytanic acid and pristanic
acid), plasmalogens, pipecolic acid and bile acids, and confirmation in cultured
patient fibroblasts. Usually these biochemical assays are sufficient for PBD
diagnosis (Steinberg, Raymond et al. 2003-2012). Additional complementation
analysis and genetic screening strategies based on sequencing analysis can be
applied to infer which specific gene is defective and which nucleotides are
mutated (Steinberg, Chen et al. 2004; Steinberg, Dodt et al. 2006).
Currently, there is no curative therapy or long-term effective treatment
available for PBD. Treatments are focusing on symptomatic therapy and
palliative measures. Vitamin K and other fat-soluble vitamin supplements are
recommended, as well as a diet low in phytanic acid and VLCFA. Bile acid may
improve liver function by reducing cholestanoic acid accumulation (Setchell,
Bragetti et al. 1992). Other peroxisomal metabolites, such as the omega-3 fatty
acid docosahexaenoic acid (DHA) and plasmalogen precursors, are
administrated to patients with variable clinical benefits. Seizures may be
16
controlled with the use of appropriate standard antiepileptic drugs (AEDs).
Cataract removal surgery is performed in early infancy to preserve vision and
hearing aids are provided to children having hearing impairment (Steinberg,
Raymond et al. 2003-2012).
There is also no general curative therapy for X-ALD patients. The most
efficient treatment for CCALD is the hematopoietic stem cell transplantation and
gene therapy of autologous hematopoietic stem cells. However, these treatments
are not accessible to all X-ALD patients due to the lack of markers for predicting
cerebral demyelination (Cartier and Aubourg 2010). In a subset of asymptomatic
patients with normal brain myelination, dietary therapy with “Lorenzo’s Oil”, a
mixture of glycerol trioleate and glyceryl trierucate that inhibits endogenous
VLCFA synthesis, helps prevent the onset of cerebral disease (Moser, Raymond
et al. 2005; Moser, Moser et al. 2007; Semmler, Kohler et al. 2008). However, it is
ineffective for symptomatic patients. Likewise, the significance of decreased
plasmalogen levels in the RBCs and white matter of patients is debatable (Wilson
and Sargent 1993; Moser 1999; Khan, Singh et al. 2008). Other ongoing studies
and trials include gene therapies targeting upregulation of ABCD1 homolog,
ABCD2, antioxidative drug treatment, histone deacetylase inhibitors like valproic
acid, butyrates and suberoylanilide hydroxamic acid (SAHA), and other
neuroprotective agents (Moser, Moser et al. 2007; Berger, Pujol et al. 2010;
Fourcade, Ruiz et al. 2010; Singh, Khan et al. 2011).
Ongoing biochemical studies and drug testing (Dranchak, Di Pietro et al.
2010) on the peroxisomal disorders primarily utilize patient primary cell cultures,
17
particularly primary skin fibroblast cultures due to their easy accessibility.
Although informative, skin fibroblast has limited peroxisome activity and does not
represent the defects in other affected tissues or organs, like hepatocytes or
CNS cell types. Also the studies based on in vitro skin fibroblast culture cannot
explain the association between a certain peroxisomal dysfunction and a specific
pathophysiology.
1.4 Mouse Models of Peroxisome Disorders
A number of knockout mouse models targeting Pex genes and
peroxisomal enzymes have been generated to obtain more insights into the
peroxisomal disorders. Pex2 (Faust and Hatten 1997; Faust, Su et al. 2001),
Pex5 (Baes, Gressens et al. 1997; Gressens, Baes et al. 2000) and Pex13
(Maxwell, Bjorkman et al. 2003) knockout mouse models have the PTS1-
dependent protein import completely blocked. These mice show biochemical
defects similar to the PBD patients with increased levels of VLCFA, phytanic acid
and decreased plasmalogen levels in liver and brain, as well as most but not all
phenotypical defects like severe hypotonia, growth retardation and neuronal
development defects. The utility of these models is lessened due to the fact that
they are either embryonic lethal or die shortly after birth.
Pex7 knockout mice (Brites, Motley et al. 2003) have increased phytanic
acid and decreased plasmalogen and similar phenotypic characterizations as the
RCDP1 patients. The presence of two Pex7 null alleles results in mortality shortly
after birth. However, a new mouse model carries hypomorphic Pex7 alleles such
18
that Pex7 transcript levels are reduced to less than 5% of wild type
(http://www.ncbi.nlm.nih.gov/pubmed/20060764). Intriguingly, these mice are
fertile and have a normal life span; however, they show growth retardation and
develop early cataracts. Furthermore, these Pex7 hypomorphic mice showed
delayed endochondral ossification and abnormalities in their lens fibers.
Inactivation of the Pex11α gene in mice does not result in noticeable
phenotypic abnormalities (Li, Baumgart et al. 2002). However, complete
inactivation of Pex11β in mice is neonatal lethal, decreases the number of
peroxisomes in brain, impairs neuronal migration, and enhances neuronal
apoptosis, developmental retardation and hypotonia without abrogating
peroxisome function (Li, Baumgart et al. 2002). Furthermore, deletion of a single
allele of Pex11β gene is sufficient to cause oxidative stress and neuronal death,
but slightly increases the number of peroxisomes in brain (Ahlemeyer, Gottwald
et al. 2012).
Selective rescue of peroxisomes in liver or brain by tissue-selective
overexpression of Pex5p demonstrate the significance of peroxisomal
metabolism in brain and extraneuronal tissues to the normal development of
mouse neocortex (Janssen, Gressens et al. 2003). The generation of Pex5
conditional knockout mice enables selective inactivation of peroxisomes in
specific neuronal lineages and established the significance of peroxisomal
functions in oligodendrocytes to the axonal integrity and neurological functioning
(Kassmann, Lappe-Siefke et al. 2007; Bottelbergs, Verheijden et al. 2010;
Bottelbergs, Verheijden et al. 2012). Peroxisome-deficiency in oligodendrocytes
19
(Kassmann, Lappe-Siefke et al. 2007), but not the defect in ether lipid synthesis
(Bottelbergs, Verheijden et al. 2012), is causally involved in axonal loss and
neuroinflammation (Baes and Van Veldhoven 2006; Wanders and Waterham
2006).
Abcd1 null mice accumulate saturated VLCFAs in their brains and adrenal
glands, but do not show demyelination or CNS inflammatory responses (Forss-
Petter, Werner et al. 1997; Kobayashi, Shinnoh et al. 1997; Lu, Lawler et al.
1997). Myelin and axonal abnormalities appear in their spinal cords and sciatic
nerves with age, which mimics the neuropathy that occurs in AMN patients (Pujol,
Hindelang et al. 2002). Mice carrying null Abcd2 alleles alone or in concert with
Abcd1 null alleles primarily show spinal axonal degeneration; however, limited
cerebellar involvement (Pujol, Ferrer et al. 2004; Ferrer, Kapfhammer et al. 2005)
and adrenal lesions (Lu, Barron-Casella et al. 2007) have been observed.
Several mouse knockouts targeting peroxisomal genes have been
generated, although corresponding to no identified human peroxisomal disorders,
including L-bifunctional enzyme (Qi, Zhu et al. 1999), thiolase B (Chevillard,
Clemencet et al. 2004) and catalase (Ho, Xiong et al. 2004). Their biochemical
and phenotypical characterizations will help understand the functions of their
human counterparts.
Although genetically engineered mouse models have provided insights
into peroxisome disorder pathophysiology, they do not recapitulate all aspects of
disease. Furthermore, there is no transgenic mouse carrying specific point
mutations of their human counterparts in peroxisomal disorders. Therefore, these
20
established knockout mouse models are not very useful for drug screening or
preclinical drug testing. There is still a need for a patient-specific multi-organ
model system in the disease investigation, drug screening and drug testing.
1.5 Disease model induced pluripotent stem cells
Induced pluripotent stem cells (iPS cells or iPSCs) are pluripotent stem
cells artificially generated by reprogramming somatic cells into an embryonic-like
state. IPS cells were first generated in 2006 from mouse embryonic and adult
fibroblast cultures by introducing four factors, Oct4, Sox2, c-Myc, and Klf4, using
retrovirus transduction and reprogramming under embryonic cell (ES) culture
conditions (Takahashi and Yamanaka 2006). In the following two years, iPS cells
were produced from human fibroblast cells using the same approach and similar
factors (Takahashi, Tanabe et al. 2007; Yu, Vodyanik et al. 2007; Lowry, Richter
et al. 2008; Park, Lerou et al. 2008). Since then, iPS cells have been generated
using a variety of somatic cell sources with improved efficiency and safety and
applied in studies of different human diseases.
iPS cells have been derived from different somatic cells other than
embryonic and adult skin fibroblasts, such as keratinocytes (Grinnell, Yang et al.
2007; Aasen, Raya et al. 2008; Dimos, Rodolfa et al. 2008), hepatocytes
(Stadtfeld, Nagaya et al. 2008), mature B lymphocytes (Stadtfeld, Nagaya et al.
2008), adipose stem cells (Sun, Panetta et al. 2009), and neural stem cells
(Duinsbergen, Eriksson et al. 2008; Eminli, Utikal et al. 2008; Kim, Zaehres et al.
2008; Kim, Sebastiano et al. 2009) from both mouse and human. iPS cells have
21
also been successfully derived from adult cells obtained from rat and non-human
primates (Liu, Zhu et al. 2008; Li, Wei et al. 2009; Liao, Cui et al. 2009).
The first reported and still the widely used method for deriving induced
pluripotent stem cells is conducted by retrovirus mediated ectopic expression of a
set of four transcription factors, OCT4, SOX2, KLF4 and C-MYC (Takahashi and
Yamanaka 2006), or a different set of four transcription factors, OCT4, SOX2,
NANOG and LIN28 (Yu, Vodyanik et al. 2007).
Although efficient and reliable, this method poses several safety concerns.
The addition of C-MYC significantly improves cell proliferation and
reprogramming efficiency. However, C-MYC is involved in cancer formation. A
translocation event t(8;14) involving MYC is commonly associated with Burkitt’s
lymphoma. Temporary inhibition of Myc selectively kills mouse lung cancer cells,
making it a proto-oncogene (Ruf, Rhyne et al. 2001). Furthermore, using
retroviruses to induce transduction results in random insertions of genes into a
genome, causing uncontrolled gene activation, gene silencing and chromosome
instability. The concerns of tumorigenesis hinder the clinical application of human
iPS cells. As a result, many techniques have been tested to generate iPS cells
more safely and efficiently.
Different combinations of these transcription factors have been tested,
demonstrating that KLF4 and C-MYC are dispensable for somatic
reprogramming to the pluripotent state depending on the types of somatic cells,
and their genetic and epigenetic status (Huangfu, Osafune et al. 2008; Wernig,
Meissner et al. 2008; Kim, Sebastiano et al. 2009).
22
A drug inducible transgenic system has also been applied in somatic
reprogramming to control the ectopic expression of transcription factors (Wernig,
Lengner et al. 2008). Non-integration strategies have been successfully applied
in somatic reprogramming, including adenovirus (Stadtfeld, Nagaya et al. 2008),
repetitive transfection of plasmids (Okita, Nakagawa et al. 2008; Yu, Hu et al.
2009), single polycistronic vectors (Carey, Markoulaki et al. 2009), piggyback
transposon system (Woltjen, Michael et al. 2009), synthetic modified mRNA
(Warren, Manos et al. 2010), and synthetic protein transduction (Gump and
Dowdy 2007; Bosnali and Edenhofer 2008).
Shortly after the first publication of iPS cells, this technique started to be
applied in generating disease-specific iPSC models. The number of applications
is growing exponentially. In general, iPS cells can be used to model human
diseases, especially genetic diseases, and provide new angles to study the
disease mechanism and pathogenesis in specific cell types at the cellular level.
Combined with in vitro differentiation and transplantation, iPSC cells can help
understand the progress of complex genetic disorders and the impact during
early development, such as PBD and X-ALD. Furthermore, iPS cells can be used
in high through-put drug screening and drug testing in specific genotypes.
Eventually, iPS cells may be applied in personalized medication by monitoring
the drug safety and efficacy for individuals. With the advance of the gene therapy
and the improvement of safety, human iPS cells could be a reliable source for cell
therapy. With iPS cells derived from the patient to be treated, genetically
23
corrected, and transplanted back to the same patient, it would avoid the risk of
immune rejection.
The first disease iPS model was derived from a patient with amyotrophic
lateral sclerosis (ALS) and further differentiated into functional motor neurons
(Dimos, Rodolfa et al. 2008). This provided a proof-of-principle that it is feasible
to generate disease-specific iPS cells and to differentiate them into cell lineages
with pathological phenotypes. Since then, iPS cells have been generated to
model a variety of diseases, including Parkinson’s disease (Park, Arora et al.
2008; Wernig, Zhao et al. 2008; Soldner, Hockemeyer et al. 2009), spinal
muscular atrophy (Ebert, Yu et al. 2009), adenosine deaminase deficiency-
related severe combined immunodeficiency (ADA-SCID), Shwachman-Bodian-
Diamond syndrome (SBDS), Gaucher disease (GD) type III, Duchenne (DMD),
Becher muscular dystrophy (BMD), Huntington disease (HD), juvenile-onset, type
1 diabetes mellitus (JDM), Down syndrome (DS)/trisomy 21 (Park, Arora et al.
2008), Long-QT syndrome (Moretti, Bellin et al. 2010), myeloprolierative
disorders (MPDs) (Ye, Zhan et al. 2009), and Hutchinson-Gilford progeria
syndrome (HGPS) (Liu, Barkho et al. 2011). These model iPS cells feature
defects in multiple tissues and organs, like CNS cells, eye, heart, liver, blood
system and smooth muscle cells. Furthermore, in combination with gene-editing
technologies, the mutations can be efficiently and safely corrected in iPS cells
(Zou, Maeder et al. 2009; Khan, Hirata et al. 2010; Liu, Suzuki et al. 2011; Yusa,
Rashid et al. 2011; Zou, Mali et al. 2011). Recently, iPS cells were applied to
model complex neurodevelopmental diseases, like Rett syndrome (Marchetto,
24
Carromeu et al. 2010) and Timothy syndrome (Pasca, Portmann et al. 2011;
Yazawa, Hsueh et al. 2011). The pathological phenotypes of affected cell
lineages, like mature neurons and cardiac cells, were elaborately characterized
and successfully treated in culture dish.
iPS cells can serve as a powerful model for disease investigation and drug
development for peroxisomal disorders with the complex genetic variations and
multi-organ defects. With technological improvements, it will become more
common to generate patient-specific iPS cells, differentiate these personalized
iPS cells into the affected cell types, and use them to test drug efficacy in vitro
and in mice transplanted with affected patient cells.
In this dissertation, I established iPS cell models for a broad spectrum of
peroxisomal disorders, including PBD-ZSD representing 5 different PEX gene
mutations, RCDP1 and CCALD, as well as several healthy controls. I also began
to characterize the pathological phenotypes of the iPS cells and the differentiated
CNS lineages, including neural progenitors, motor neuron progenitors,
oligodendrocyte progenitors, mature motor neurons and mature oligodendrocytes.
In long term, we will utilize these iPSC resources for robust high through-put drug
screening and drug testing in cell culture and in mouse models transplanted with
these cells. In the near future, we hope that this line of research will contribute to
community efforts to improve the quality of life of children with peroxisomal
disorders.
25
CHAPTER 2
EXPERIMENTAL DESIGN AND PREPARATION
2.1 Cell culture and cell lines
Primary dermal fibroblast cultures of healthy controls were obtained from
Coriell Institute Cell Repositories. PBD and CCALD patient dermal fibroblast
cultures were obtained from the Peroxisomal Disease Laboratory at the Kennedy
Krieger Institute. The mutations and patient information were detailed in Table 2.1.
Primary mouse embryonic fibroblasts (MEFs) were collected from CF-1 strain
mouse embryos 12.5 dpc (days post-coitum). The pregnant female mice were
obtained from the Jackson laboratory. Primary MEFs were expanded according
to the following split ratio, 1:5 for P1, 1:4 for P2 and 1:3 for P3. At 100%
confluence, passage 3 MEFs were inactivated by incubating with 10ug/ml
mitomycin C (Sigma) for 3 hours and the inactivated MEFs (iMEFs) were
harvested. Alternatively, MEFs were inactivated using a cell irradiator to deliver
5,000 rads. For co-culture, iMEFs were plated at a density of 50,000-60,000
cells/cm
2
in 0.1% gelatin coated plates and use within 2 days. For storage, iMEFs
were frozen at a density of 2×10
6
/ml in -80°C for short term or liquid N2 for long
term (Klimanskaya, Chung et al. 2007; Lerou, Yabuuchi et al. 2008). Each batch
of iMEFs was tested by growing hES or iPS cells for at least 4 passages. GP2-
293 cell was a kindly gift from Dr. Qilong Ying’s lab.
26
Table 2.1: Skin fibroblast and iPS cell information.
M=Male, F=Female, U=unknown gender.
Patient ID Cell line ID Status
Number
of iPS
cells
Genotype
PBD721 PBD_PEX1fs1 ZS patient, M 2 Homozygous PEX1 n.2097_2098insT p.I700fs
PBD702 PBD_PEX1fs2 ZS patient, F 3
Heterozygote PEX1 n.2097_2098insT p.I700fs; n.2916delA
p.G973fs
PBD615 PBD_PEX1ms1 ZS patient, M 5 Homozygous PEX1 n. 2527G>A p.843G>D
PBD643 PBD_PEX1ms2 ZS patient, F 5 Homozygous PEX1 n. 2527G>A p.843G>D
PBD673 PBD_PEX12 ZS patient, M 3 Homozygous PEX12 n.959C>T p.320S>F
PBD604 PBD_PEX26 ZS patient, M 2
Homozygous PEX26 n.292C>T p.98R>W; heterozygous PEX1
n.2843G>A p.948R>Q
PBD687 PBD_PEX10 ZS patient, F 5
Heterozygote PEX10-n.337delC(fs), n.880A>G p.294T>A;
n.890T>C p.297L>Q
XALD216212 CCALD1 CCALD patient, M 4 Homozygous ABCD1 n. 252delC p. P84X
XALD306463 CCALD2 CCALD patient, M 1 Homozygous ABCD1 n. 1847C>T p. 616A>V
PBD080 RCDP-1 RCDP patient, U 4 Homozygous PEX7 n. 875T>A p.L292fs
PBD127 RCDP-2 RCDP patient, U 4 Homozygous PEX7 n. 875T>A p.L292fs
AG05838 Control1 Healthy donor, F 6 Presumed wild type
AG09599 Control2 Healthy donor, F 4 Presumed wild type
AG13153 Control3 Healthy donor, M 1 Presumed wild type
27
All cells described herein were cultured at 37 C with 5% CO
2
. Human
primary dermal fibroblasts and MEFs and iMEFs were cultured in fibroblast
media (DMEM supplemented with 10% FBS, L-glutamine, penicillin/streptomycin,
vitamin solution, essential and nonessential amino acids (Invitrogen)). IPS cells
were co-cultured with MEF feeder cells in iPSC medium (DMEM/F12 medium
supplemented with 20% KSR, L-glutamine, penicillin/streptomycin, nonessential
amino acids, -mecaptoethanol and bFGF (Invitrogen)). GP2-293 cells were
maintained in 293 medium (DMEM supplemented with 10% heat inactivated FBS,
L-glutamine, penicillin/streptomycin, vitamin solution, essential and nonessential
amino acids (Invitrogen)).
2.2 Retrovirus preparation
Five pMX retroviral vectors designed to deliver the jellyfish GFP and
human OCT4, SOX2, KLF4, and C-MYC cDNA sequences were obtained from
Addgene (http://www.addgene.org/) (Takahashi, Okita et al. 2007; Park, Lerou et
al. 2008). For retrovirus packaging, 5ug VSVG envelope protein vector and 5ug
retroviral vector were co-delivered into GP2-293 cells using the lipofectamine®
LTX kit (Invitrogen). The retrovirus-containing medium were collected 36 hours
and 48 hours after the transfection, pooled and filtered through 0.45µm filter. The
GFP retrovirus was used for calculating virus titer and determining the multiplicity
of infection (Tiscornia, Singer et al. 2006).
28
2.3 Cellular reprogramming
On day 0, primary human skin fibroblasts at 80% confluence were seeded
in 6-well plates (1x10
5
cells/well). Next day, cells were transduced with a mixture
of all five retroviruses with MOI of 5 along with 5ug/ml protamine sulfate in the
medium, centrifuged at 3,000 rpm for 45 minutes and incubate at 37°C, 5% CO2
overnight. Day 2, aspirate medium, wash cells with 1xDPBS for 3 times and add
2ml fresh fibroblast medium. Day 3, perform second transduction, repeat the
procedure on day 1. Day 4, repeat the procedure on day 2. Day 5, prepare
gelatin-coated tissue culture plates and plate iMEFs (2x10
6
cells/6-well plate).
Incubate at 37°C, 5% CO2 for 24 hours. Day 6, 1:3 split the infected cells into
tissue culture plates with iMEFs in fibroblast medium. Day 7, change to iPS
medium supplemented with 10ng/ml bFGF and change the medium every day.
Medium is supplemented with 1mM VPA for two weeks to increase the
reprogramming efficiency. Switch to one-day iMEF conditioned medium starting
from day 21. By day 10 after infection, human fibroblasts start to shorten and
become round. By day 14-28 after infection, hES cell-like colonies should start to
appear. However, certain fibroblast cell lines form colonies after 30 days. By 4
weeks, hES cell-like colonies were confirmed by Tra-1-60 or Tra-1-81 live-
staining before manually picked and expanded. (Takahashi, Okita et al. 2007;
Park, Lerou et al. 2008)
29
2.4 Immunocytochemistry and enzyme activity assays
For fluorescent immunostaining, cells were fixed in 4% paraformaldehyde
for 20 minutes, permeabilized with 0.1% Triton X-100 for 5 minutes except for
surface marker staining, and blocked in 1% BSA in 1×PBS for 1 hour at room
temperature. Primary antibody staining was performed at 4 C overnight with
antibodies against OCT4 and NANOG (goat polyclonal IgG against human,
RandD Systems), SOX2 (goat polyclonal IgG against mouse, rat and human)
and SSEA4 (mouse monoclonal IgG
3
against human, Santa Cruz Biotechnology),
TRA-1-60 (mouse monoclonal IgM against human, Millipore), TuJ1 (rabbit
monoclonal IgG
1
against mammalian, Covance Research Products), -SMA
(mouse monoclonal IgG
2a
against human, mouse, et al, Sigma), AFP (mouse
monoclonal IgG
2a
against human, pig and canine, Invitrogen), PAX6 (mouse
monoclonal IgG1 against human), NKX2.2 (mouse monoclonal IgG2b against
human), HB9 (mouse monoclonal IgG1 against human, Developmental Studies
Hybridoma Bank), OLIG2 (RandD systems), SOX10 (rabbit polyclonal IgG
against human), PDGFRα (rabbit polyclonal IgG against human, Chemicon), O4
(mouse monoclonal IgM against human, RandD systems) and MBP (rat
monoclonal IgG against human, Abcam). Secondary antibody staining was
performed at room temperature for 1 hour with appropriate fluorescence
conjugated secondary antibodies from Invitrogen and Jackson ImmunoResearch.
Nuclei were visualized by staining with 100ng/ml DAPI (Invitrogen). Alkaline
phosphatase (AP) staining was performed with the leukocyte AP kit (Sigma-
Aldrich).
30
2.5 Gene expression profiling
Total RNA samples (100ng per sample) were converted into biotin-labeled
cRNA targets (Affymetrix GeneChip® IVT Labeling Kit), processed, and analyzed
on Affymetrix Human Genome 133A 2.0 or 133 Plus 2.0 GeneChips, as
previously described (Karaman, Houck et al. 2003). Using WebArray software,
we applied the RMA algorithm to generate log
2
-transformed gene expression
values and linear model statistical analysis (limma) to identify differentially
expressed genes (DEGSs) with false discovery rates (FDRs) calculated using the
spacings LOESS histogram (SPLOSH) method (Xia, McClelland et al. 2005;
Wang, McClelland et al. 2009). We performed hierarchical clustering analysis
and analyzed gene relationships using Partek Genomics Suite and Ingenuity
Pathway Analysis (IPA) software (Ingenuity Systems), respectively. We
conducted GeneOntology (GO) and Kyoto Encyclopedia of Genes and Genomes
(KEGG) pathway analyses using WebGestalt software
(http://bioinfo.vanderbilt.edu/webgestalt/) (Zhang, Kirov et al. 2005; Kirov, Zhang
et al. 2007; Tazi, Quioc et al. 2009). We used the DAVID (Database for
Annotation, Visualization and Integrated Discovery) v6.7 Bioinformatics resource
(http://david.abcc.ncifcrf.gov/) for additional gene function annotation (Huang da,
Sherman et al. 2009). Scaled gene expression scores and .cel files are available
at the National Center for Biotechnology Information (NCBI) Gene Expression
Omnibus (GEO) repository (http://www.ncbi.nih.gov/geo/). Quantitative RT-PCR
was conducted for confirmation of differentially expressed genes, PEX11B and
IMPA1, in CCALD and control iPSCs. The following primers were used:
31
PEX11B-Forward: 5’-CCCAGCGTTCATTCAGGTAC-3’,
PEX11B-Reverse: 5’-CCTCCAGAACCTTTCAGTCG-3’,
IMPA1-Forward: 5’-CTGGAGAGGTAGTTTGTGAAGC-3’,
and IMPA1-Revers: 5’-TTTTTCAACTTTTTGGTCCGTAG-3’.
2.6 DNA methylation profiling and bisulfite sequencing confirmation
Genomic DNA was extracted from cultured cells as described (Pike,
Greiner et al. 2008; Wang, Greiner et al. 2010) and analyzed on 450K Infinium
Methylation BeadChips (Illumina), which interrogate the methylation status of
over 485,000 cytosine guanine dinucleotides (CpG) sites distributed across the
human genome. The resulting data was analyzed using GenomeStudio software
(Illumina) with DNA methylation levels summarized as β-values ranging from 0 to
1 (i.e. from no to complete DNA methylation) for each locus. Bisulfite DNA
sequencing was conducted as previously described (Pike, Greiner et al. 2008;
Wang, Greiner et al. 2010).
The differentially methylated loci (DML) proximal to pluripotency genes,
OCT4 and NANOG, were confirmed by bisulfite sequencing. First, bisulfite
conversion was performed on 500ng genomic DNA samples using EZ DNA
Methylation kit (Zymo Research) according to manufacturer’s protocol. The
converted DNA samples served as templates for PCR reactions using the
following primers:
OCT4-Forward: 5’-TAGTTGGGATGTGTAGAGTTTGAGA-3’,
OCT4-Reverse: 5’-TAAACCAAAACAATCCTTCTACTCC-3’,
32
NANOG-Forward: 5’-GAGTTAAAGAGTTTTGTTTTTAAAAATTAT-3’,
and NANOG-Reverse: 5’-TCCCAAATCTAATAATTTATCATATCTTTC-3’ (Freberg,
Dahl et al. 2007). The following parameter was used for the amplification reaction:
(Taq Platinum Polymerase, invitrogen): 94ºC for 2 minutes; 38 cycles of 94ºC for
20 seconds, 54ºC for 20 seconds, and 68ºC for 1 minute, and followed by 68ºC
for 10 minutes. The PCR products were confirmed with electrophoresis, sub-
cloned into pCR-TOPO vector (TOPO TA cloning kit, Invitrogen) and at least 12
colonies were sequenced for each selected CpG site with T3 and T7 primers.
2.7 DNA sequencing of mutations in PBD and CCALD
Genomic DNA was extracted from cultured iPS cells and amplified using
the Repli-G Mini Kit (Qiagen) following the manufacturer’s recommended
protocols. Primer design, exon amplification and sequencing was performed as
described (Yik, Steinberg et al. 2009). PEX1 exons 13 and 15, PEX10 exon 5,
PEX12 exon 3 and ABCD1 exons 1, 8, and 9 were sequenced in the patient-
derived iPS cells carrying the corresponding mutations for confirmation.
2.8 Cytogenetic analysis
Genomic DNA from cultured cells was also analyzed using Human
CytoSNP-12 Infinium HD BeadChips (Illumina) that interrogate the genotypes of
299,140 human single nucleotide polymorphisms (SNPs). Data cleaning,
processing and exporting were performed in GenomeStudio (Illumina). Copy
number analysis was performed using CNVPartition version2.4.4 with a
33
confidence threshold set at 50 and a minimum SNP probe count per CNV region
at 10. (Laurent, Ulitsky et al. 2011)
2.9 In vitro differentiation and teratoma assays
For in vitro differentiation, iPS cells derived from patients and healthy
controls were detached from their culture dishes by treating with CTK (1mg/ml
collagenase IV, 0.25% trypsin, 20% KSR, 1mM CaCl2 in 1×DPBS) for 5 minutes
and maintained in iPSC medium without FGF2 in low adherent plates for 7 days
to induce embryoid body (EB) formation. The embryoid bodies were transferred
to cell culture plates coated with Matrigel® in DMEM/F12 medium supplemented
with 15% FBS for endoderm / mesoderm differentiation or in DMEM/F12 medium
supplemented with 10% N2 and 10% B27 for neural differentiation. For teratoma
analysis, iPS cells from a confluent 10cm
2
plates were harvested, resuspended
in Matrigel® and subcutaneously injected to the dorsal flanks of immunodeficient
(SCID) mice (Jackson Laboratory). Nine weeks after injection, teratomas were
excised, fixed in 10% formalin, sectioned, and stained with hematoxylin and
eosin. (Itskovitz-Eldor, Schuldiner et al. 2000; Takahashi, Tanabe et al. 2007)
2.10 Lipid analysis
Lipid analysis was performed by Dr. Steven Steinberg’s lab at the
Peroxisomal Diseases Labroatory located at the Kennedy Krieger Institute. Cell
extracts were processed and total fatty acids analyzed using capillary GC with
flame ionization detection, as described (Moser and Moser 1991). C16:0 DMA
34
(dimethyl acetyl) and C18:0 DMA levels, representing the substituent of the sn-1
position of plasmalogens, also were measured by capillary GC with flame
ionization detection, as described. (Bjorkhem, Sisfontes et al. 1986; Steinberg,
Jones et al. 2008)
2.11 CNS lineage differentiation
EBs were formed and maintained as described above for 4 days, then
switched to neural induction medium (NIM) containing DMEM/F12 supplemented
with 1% N2, L-glutamine, penicillin/streptomycin, nonessential amino acids and
2µg/ml heparin (Invitrogen) for 3 days. On day 7, EBs were attached to Matrigel®
coated cell culture plates and maintained in the same medium for an extra 7 to
10 days for neural epithelia (NE) induction. The small columnar-like neural
rosette structure of NE will start to appear around day 10.
For motor neuron differentiation, 1µM retinoic acid (RA) will be added to
NIM for caudalization starting from day 10. After 5 more days, the NE rosettes
were gently blown off by a 1-ml pipette and gently triturated to form motor neuron
progenitors (MNP) / neural spheres (NS) in neural differentiation medium (NDM)
containing DMEM/F12 supplemented with 1% N2, 2% B27, L-glutamine,
penicillin/streptomycin, nonessential amino acids, 2µg/ml heparin, 1µM RA and
100ng/ml sonic hedgehog (SHH) until day 28 (Invitrogen). For terminal motor
neuron differentiation, NSs were triturated and attached to laminin coated cell
culture plates and maintained in NDM supplemented with 1µM cAMP, 200µg/ml
35
ascorbic acid and neurotrophic factors, BDNF, GDNF and IGF1, at 10ng/ml each
for up to 7 weeks. (Hu and Zhang 2009; Xia and Zhang 2009)
For oligodendrocyte differentiation, iPS cells derived from patients and
healthy controls were detached from their culture dishes by treating with CTK for
5 minutes and resuspended in transition medium (TM) containing 50% iPSC
medium and 50% glial restrictive medium (GRM) with 5ng/ml FGF2 and 20ng/ml
EGF in low adherent plates for 2 days. GRM contains DMEM/F12, 2% B27
(Invitrogen), 25µg/ml insulin, 6.3ng/ml progesterone, 10µg/ml putrescin, 50ng/ml
sodium selenite, 50µg/ml holotransferin and 40ng/ml tri-iodo-thyroidin (Sigma).
On day 3, unattached EBs were switched to GRM supplemented with 20ng/ml
EGF and 5µM RA for 8 days with medium changed every day. Yellow spheres will
start to appear during the RA treatment. On day 11, the yellow spheres are
manually selected, cut to small pieces and maintained in GRM supplemented
with 20ng/ml EGF. On day 28, the yellow spheres were cut to small pieces and
attached to 1:30 diluted growth-factor-reduced Matrigel® (BD Biosciences) in the
same medium. After one week, the attached clusters of cells were dissociated by
incubating in 1xHBSS for 10-15 minutes and attached to poly-L-
ornithine/fibronectin double coated plates. For oligodendrocyte progenitor
expansion, cells were maintained in GRM supplemented with 1% N2, 10ng/ml
FGF2 and 20ng/ml EGF (Invitrogen). For terminal oligodendrocyte differentiation,
cells were maintained in GRM supplemented with 1% N2, 50ng/ml noggin
(RandD systems), 5ng/ml FGF2 and 10ng/ml EGF for 2-3 days. Then FGF2 and
EGF were removed from the medium along with the addition of 1mM cAMP,
36
200nM ascorbic acid (Sigma), 20ng/ml IGF, GDNF and CNTF (Peprotech).
(Nistor, Totoiu et al. 2005; Zhang, Izrael et al. 2006; Izrael, Zhang et al. 2007;
Hatch, Nistor et al. 2009; Sharp, Frame et al. 2010)
37
CHAPTER 3
GENERATION AND CONFIRMATION OF HEALTHY CONTROL, PBD AND
CCALD PATIENT IPS CELLS
3.1 Abstract
We generated induced pluripotent stem (iPS) cells from primary skin
fibroblasts obtained from PBD-ZSD, RCDP1, and CCALD patients and multiple
healthy controls by ectopic expression of OCT4, SOX2, KLF4, and C-MYC
mediated by retroviruses. Both patient and control iPS cells are similar to human
embryonic stem (hES) cells in their morphology, surface and nuclear marker
presentation, gene expression, and global epigenetic status, as well as the
capability to differentiate into all three germ layers. Furthermore, we detected
copy number changes (CNCs) and incomplete demethylation in some patient
and control iPS cells, which did not result in differential gene expression in iPS
cells.
3.2 Introduction
iPSCs have been generated to model many human diseases with defects
in multiple tissues and organs using a variety of reprogramming techniques.
iPSC models have been used to investigate disease pathogenesis, test
candidate drugs, and explore the application of targeted gene correction for gene
therapy applications (Nishikawa, Goldstein et al. 2008; Lengerke and Daley 2009;
Patel and Yang 2010).
38
As discussed in the introduction, peroxisomal disorders are a group of
genetically heterogeneous metabolic diseases resulting from dysfunctional
peroxisome activity. These diseases affect multiple organ sytems, including the
central nervous system and liver. It has been estimated that over 90% of PBD-
ZSD patients carry mutations in the PEX1, PEX6, PEX10, PEX12 or PEX26
genes (Steinberg, Chen et al. 2004). PEX1 mutations account for about 70% of
PBD-ZSD patients (Steinberg, Chen et al. 2004; Yik, Steinberg et al. 2009).
PEX1 and PEX6 belong to the AAA (ATPase Associated with diverse cellular
Activities) ATPase family of proteins and form heterodimers that are involved in
the recycling of PEX5, the receptor for proteins with peroxisomal targeting
signals (PTS) (Tamura, Yasutake et al. 2006; Thoms and Erdmann 2006).
Ongoing investigations on disease mechanisms and drug tests for peroxisome
disorders (Dranchak, Di Pietro et al. 2010) are established mainly on patient
primary cell cultures, particularly primary skin fibroblast cultures due to its easy
accessibility and a number of knockout mouse models targeting Pex genes and
peroxisomal enzymes.
Although informative for diease diagnostics, cultured skin fibroblasts have
limited peroxisome metabolic activity and do not represent the defects in other
affected tissues or organs, like hepatocytes or CNS cell types. As discussed
earlier, genetically engineered mouse models have provided insights into
peroxisome disorder pathophysiology, but they do not recapitulate all aspects of
disease and there is no transgenic mouse carrying specific point mutations of
their human counterparts in peroxisomal disorders.
39
Given the drawbacks of skin fibroblast cell culture models as well as the
limitations of PBD and X-ALD mouse models that do not develop all aspects of
disease found in humans, iPS cells may serve as a highly valuable alternative
model for disease investigation and drug development. Here, we derived multiple
iPS cell lines from primary fibroblasts obtained from seven PBD-ZSD, two
RCDP1 and two CCALD patients with a variety of mutations, and three healthy
control donors. Given that approximately 70% of PBD-ZSD cases involve PEX1
mutations, so we chose to reprogram fibroblasts from four individuals with PEX1
mutations. Two of these patients had two severe null PEX1 mutations and two
had two mild PEX1 missense mutations. We also reprogrammed skin fibroblasts
from PBD-ZSD patients with PEX10, PEX12 or PEX26 mutations.
Then, we evaluated the qualities of the iPSCs by their genomic integrity,
stem cell signature gene expression, global epigenetic status, and pluripotency,
as well as the confirmation of the expected mutations.
3.3 Results
3.3.1 Derivation of candidate iPS cells
Primary skin fibroblast cultures from seven PBD-ZSD patients, two
RCDP1 patients, two CCALD patients and three healthy donors (Table 2.1) were
infected twice in three days with retroviruses expressing human OCT4, SOX2,
KLF4 and c-MYC genes. Seven to ten days after infection, human fibroblasts
start to shorten and become round. By two weeks, we started to observe
colonies with hES cell-like morphological properties. However, some fibroblast
40
cell lines form colonies after 30 days. TRA-1-60 positive candidate iPSC colonies
were clonally expanded by four weeks. We didn’t observe any significant
difference in reprogramming efficiency, morphological features, and proliferation
rate between genders or phenotypes.
3.3.2 Genetic characteristics of candidate iPS cells
By dideoxysequencing, we confirmed that all the patient iPSCs have the
expected mutant PEX gene alleles and that control iPSCs lacked these specific
mutations in PEX1, PEX10, PEX12, PEX26, PEX7 or ABCD1 genes.
As determined by CytoSNP analysis, the genotypes of over 290,000 SNPs
in iPSCs and original fibroblasts were >99.9% concordant. Based on data from
same CytoSNP analysis, we did not detect de novo copy number variations
(CNVs) in patient PBD_PEX1fs1-iPS2, PBD_PEX1ms1-iPS5, PBD_PEX1ms2-
iPS1, PBD_PEX10-iPS1, PBD_PEX26-iPS1, CCALD1-iPS1, CCALD1-iPS2, and
CCALD2-iPS1 or Control1-iPS3, Control1-iPS4, Control1-iPS5 and Control2-
iPS1. However, 21 out of 33 (63.6%) iPSCs analyzed showed CNVs (deletions
and/or amplifications) based on molecular Karyotype analysis (Table 3.1). In
patient PBD_PEX1fs2-iPS1, there is an amplification of a large region of
chromosome 1 and a deletion of a large region of chromosome 10. Therefore,
this sample is excluded from further functional analysis to avoid bias caused by
these large deletion and amplification.
Molecular karyotype analysis also revealed the recurrence of the
chr12:p11.21 and chr20:q11.21 amplification. The chromosome 12 region was
41
amplified in 4 out of 33 iPSCs, including PBD_PEX1ms1-iPS1/2/3/4. This region
encompasses gene DDX11, a DEAD box protein involved in embryogenesis,
spermatogenesis, and cellular growth and division (Inoue, Li et al. 2007; Feeney,
Saade et al. 2011). DDX11 is expressed in all the iPS cell lines at similar level.
The chromosome 20 region was amplified in 6 out of 33 iPSCs, including
PBD_PEX1ms1-iPS1, PBD_PEX1fs2-iPS3, PBD_PEX10-iPS2, PBD_PEX26-
iPS2, CCALD1-iPS3 and Control3-iPS1. Within the chromosome 20 region,
HM13, ID1 and BCL2L1, are known to be expressed in human embryonic stem
cells (Amps, Andrews et al. 2011) and PDRG1 is reported as a novel tumor
marker (Jiang, Luo et al. 2011). This result is consistent with prior reports of
reprogrammed human cells and human embryonic stem cells (Amps, Andrews et
al. 2011; Laurent, Ulitsky et al. 2011).
To address the influence that CNVs present in iPSC have on their
transcriptome, we focused on the expression profiles of genes residing in the
affected genomic regions in a subset of iPSCs, 5 CCALD and 9 healthy control
iPSCs. For all these samples, we obtained CytoSNP data in duplicates and have
gene expression profiles and DNA methylation profiles generated from RNA and
DNA prepared simultaneously from one sample. A total of 11 amplified segments
containing 22 unique genes were found in eight iPSCs. Only 6 of these unique
genes showed elevated expression in the amplified relative to the diploid
samples (i.e. >1.2-fold increase relative to the mean expression scores of diploid
samples and falling outside their range). This included the ID1 gene in CCALD1-
iPS3, WWC1 gene in CCALD1-iPS4, and IQCA1, CXCR7, SQLE and KIAA0196
42
genes in Control1-iPS1. Three iPSCs (Control2-iPS2, Control2-iPS4, and
Control3-iPS1) showed evidence of having at least one genomic deletion, with
evidence in each case that one allele was retained. Collectively, 5 unique genes
were present in the 4 deleted genomic regions in these iPSCs. There was no
evidence of reduced expression in the samples with reduced copy number
(i.e. >1.2-fold decrease relative to the mean expression scores of diploid samples
and falling outside their range). A total of 745 DNA methylation assays
interrogated loci located within amplified regions present in control or patient
iPSCs. In all cases, the DNA methylation status of such genomic regions in all
samples was similar regardless of whether it was in the diploid or amplified state.
In fact, we observed no evidence of a block of DNA methylation change
associated with a CNC (i.e. 3 or more contiguous assays wherein the β-value of
the amplified segment was greater than 0.2 units outside the range of the diploid
samples). Next, we assessed the methylation status of genomic regions subject
to a loss of copy number in iPSCs. A total of 79 DNA methylation assays
interrogate loci with the genomic regions of heterozygous deletion. The affected
samples included Control2-iPS2 (chr13:q14.2), Control2-iPS4 (chr2:q33.3), and
Control3-iPS1 (chr3:p14.2 and chr5:p15.2). Again, we observed no evidence of a
block of DNA methylation change associated with a CNC (i.e. 3 or more
contiguous assays wherein the β-value of the deleted segment fell more than 0.2
units outside the range of the diploid samples).
43
Table 3.1: Summary of copy number variations in iPSCs.
* iPSCs with CytoSNP data obtained only one time
# Cytogenetic changes appeared in multiple iPSCs
$ The pericentromeric region is defined as the 2-5 megabases in the center of the human DNA (She, Horvath et al. 2004);
The subtelomeric region is defined as the most distal 0.5 megabases in human DNA (Ambrosini, Paul et al. 2007).
Cell Lines
Clone
No.
Cytogenetic Change
Cytogenetic Band Start End Length Type No. of Markers Location
$
PBD_PEX1fs1 2* No changes detected
PBD_PEX1ms1 5 No changes detected
PBD_PEX1ms2 1* No changes detected
PBD_PEX10 1* No changes detected
PBD_PEX26 1* No changes detected
CCALD1 1 No changes detected
CCALD1 2 No changes detected
CCALD2 1 No changes detected
Control1 3 No changes detected
Control1 4 No changes detected
Control1 5 No changes detected
Control2 1 No changes detected
CCALD1 3 chr20:q11.21 29306843 30300592 993749 duplication 191 Pericentromeric
CCALD1 4 chr5:q35.1 167667709 168012853 345144 duplication 78 internal
Control1 2*
chr1:q44 244541684 244936916 395232 duplication 52 internal
chr2:q13
#
108779391 108922797 143406 duplication 28 internal
44
Table 3.1, Continued
Cell Lines
Clone
No.
Cytogenetic Change
Cytogenetic Band Start End Length Type
No. of
Markers
Location
$
Control1 1
chr1:q44 244592242 244929211 336969 duplication 49 internal
chr2:p24.3 14797141 15290013 492872 duplication 81 internal
chr2:q13
#
108779391 108922797 143406 duplication 28 internal
chr2:q37.2-q37.3 236359405 237730720 1371315 duplication 203 internal
chr8:q24.13-q24.21 125835452 129778467 3943015 duplication 625 internal
Control2 3 chr10:q21.1 53451316 53904193 452877 duplication 40 internal
Control2 2
chr13:q14.2 47286549 47936678 650129 hemizygous deletion 78 internal
chr20:p12.1 14956470 15825027 868557 duplication 126 internal
Control2 4*
chr2:q33.3 205665802 205848049 182247 hemizygous deletion 18 internal
chr6:q26 162672040 162834976 162936 duplication 11 internal
chr14:q11.2 19302319 19456237 153918 duplication 10 Pericentromeric
Control3 1*
chr3:p14.2 60447249 60742492 295243 hemizygous deletion 45 internal
chr5:p15.2 14238484 14478062 239578 hemizygous deletion 36 internal
chr20:q11.21 29306843 30419769 1112926 duplication 217 Pericentromeric
chrY:q11.23 26057950 26521000 463050 homozygous deletion 36 internal
PBD_PEX1fs1 1*
chr7:q11.21 61631605 62065560 433955 hemizygous deletion 48 Centromeric
chrX:p11.4 41308093 41383746 75653 homozygous deletion 10 internal
45
Table 3.1, Continued
Cell Lines
Clone
No.
Cytogenetic Change
Cytogenetic Band Start End Length Type
No. of
Markers
Location
$
PBD_PEX1fs2 1*
chr1:q25.3-q44 182843940 247169378 64325438 duplication 6313 internal
chr8:p23.2 4044803 4775868 731065 hemizygous deletion 190 Subtelomeric
chr8:q24.23 137739710 137919193 179483 hemizygous deletion 22 internal
chr10:p15.3-
p12.33
125708 19128078 19002370 hemizygous deletion 2532 internal
PBD_PEX1fs2 2*
chr22:q12.3 31797057 32033156 236099 duplication 26 internal
chr8:q24.23 137799462 137916243 116781 hemizygous deletion 18 internal
chr3:p14.2 59933428 60163645 230217
homozygous
deletion
53 internal
PBD_PEX1fs2 3*
chr10:q11.22 46504244 47230072 725828 duplication 32 internal
chr20:q11.21 29306843 30443249 1136406 duplication 221 Pericentromeric
chr1:q44 243548969 243630065 81096 hemizygous deletion 12 internal
chr8:q24.23 137807857 137919193 111336 hemizygous deletion 19 internal
PBD_PEX1ms1 1*
chr12:p11.21 31160567 31292645 132078 duplication 20 Pericentromeric
chr20:q11.21 29306843 30234744 927901 duplication 179 Pericentromeric
PBD_PEX1ms1 2*
chr12:p11.21 31160567 31292645 132078 duplication 20 Pericentromeric
chr14:q21.1 38386980 38984948 597968 duplication 45 internal
PBD_PEX1ms1 3* chr12:p11.21 31116977 31292645 175668 duplication 22 Pericentromeric
PBD_PEX1ms1 4*
chr1:q32.1 201505191 204180471 2675280 duplication 226 internal
chr12:p11.21 31044304 31292645 248341 duplication 29 Pericentromeric
PBD_PEX1ms2 2* chr11:p11.12 50231935 51398585 1166650
homozygous
deletion
122 Centromeric
46
Table 3.1, Continued
Cell Lines
Clone
No.
Cytogenetic Change
Cytogenetic Band Start End Length Type No. of Markers Location
$
PBD_PEX10 2* chr20:q11.21 29306843 30109620 802777 duplication 154 Pericentromeric
PBD_PEX12 1* chr7:p21.1 16294156 16404707 110551 hemizygous deletion 22 internal
PBD_PEX12 2*
chr1:p31.1 79831917 82717159 2885242 duplication 231 internal
chr7:p15.3 22121901 22987197 865296 duplication 94 internal
chr8:q21.3-q22.1 92502885 93941173 1438288 duplication 202 internal
chr7:p21.1 16294156 16404707 110551 hemizygous deletion 22 internal
PBD_PEX26 2*
chr7:q31.1 109009400 110109932 1100532 duplication 69 internal
chr20:q11.21 29306843 30209909 903066 duplication 174 Pericentromeric
47
3.3.3 Candidate iPS cells express hES markers
All candidate iPSC colonies maintained the expected morphological
features similar to that of hES cells and express hESC-specific surface markers,
SSEA3, SSEA4, tumor-related antigen (TRA)-1-60, TRA-1-81 and TRA-2-49/6E,
alkaline phosphatase, as well as the nuclear proteins OCT4, SOX2, and NANOG.
(Figure 3.1)
We also validated the robust expression of previously reported iPS cell
signature genes in healthy control and peroxisomal disorder patient-derived
iPSCs and skin fibroblasts (Lowry, Richter et al. 2008) based on a subset of the
data generated from global expression profiling of over 22,000 transcripts.
Unsupervised hierarchical clustering analysis based on the expression of
preselected pluripotency biomarkers (Fig 3.2A) or the most variable transcripts
(Fig 3.2B) (i.e. coefficient of variation (CV)>0.25 across all samples) produced
two distinct clusters consisting of skin fibroblasts and iPSCs. However, neither
the patient-derived iPSCs nor the patient fibroblasts are distinguishable from the
corresponding cells derived from the healthy individuals according to their gene
expression profiles.
48
Figure 3.1: Human ES marker expression of representative PBD-ZSD,
CCALD and healthy control iPSCs.
Red and green colors represent ES marker staining, blue color represents DAPI
nuclear counterstaining.
PBD_PEX1fs1-iPS1
CCALD1-iPS2 Control2-iPS2
49
Figure 3.2: The gene expression profile of fibroblasts and iPSCs show
dramatic differences that are not strongly influenced by mutation status.
We provide dendrograms generated based on the unsupervised hierarchical
clustering analysis of gene expression data from PBD patient, CCALD patient
and healthy control fibroblasts and iPSCs using Euclidean distances as the
distance measure and average-linkage as clustering method. (A) based on the
gene expression of 36 pluripotency genes, conducted. (B) based on the gene
expression of 569 probe sets with coefficient of variation great than 0.25. The
heat maps represent log2 gene expression scores colored coded according to
the color bar provided at the bottom of each panel.
A.
50
Figure 3.2, Continued
B.
51
3.3.4 Candidate iPS cells have unique DNA methylation profiles
We performed global DNA methylation analysis interrogating over 485,000
CpG sites of all starting fibroblasts and reprogrammed iPSCs (Table 2.1).
Unsupervised hierarchical clustering analysis demonstrated that the iPSCs and
fibroblasts have distinct DNA methylation profiles that were all independent of the
disease gene mutation status (Figure 3.3A). The differences in fibroblast and
iPSC DNA methylation profiles are expected given prior reports (Nishino, Toyoda
et al. 2011).
The clustering analysis also revealed incomplete demethylation in some
PBD and healthy control iPSCs, including PBD_PEX1fs1-iPS1, PBD_PEX1fs1-
iPS2, PBD_PEX1fs2-iPS1, PBD_PEX1fs2-iPS1, PBD_PEX1ms2 iPSCs,
PBD_PEX0-iPS1, PBD_PEX26 iPS1, PBD_PEX26-iPS2, Control1-iPS4,
Control2-iPS1 and Control2-iPS3. Gene Ontology (GO) analysis shows that the
genes proximal to these incomplete demethylation regions are involved in
transcription regulations, central nervous system (CNS) development, organ
morphogenesis, and anterior/posterior pattern formation. Genes involved in CNS
development include PRKG1, BCL2, NOTCH3, CA10, MKKS, IKZF1, CTNNA2,
JARID2 and several homeobox genes, NKX6-1, LHX6, LMX1B, and EN2. The
gene expression profiles of 167 gene transcripts proximal to these incomplete
demethylation regions were compared between iPSCs with or without incomplete
demethylation. We did not observe any gene with significant differential
expression in the iPS cells. Bisulfite sequencing analyses confirmed the
significant demethylation of the CpG sites in the promoter regions of the
52
pluripotent genes, OCT4 and NANOG, in the PBD_PEX1ms1 iPSCs as
compared to the parental PBD_PEX1ms1 fibroblast culture.
53
Figure 3.3: The DNA methylation profiles and bisulfite sequencing
confirmation of fibroblasts and iPSCs.
We provide a dendrogram generated based on the unsupervised hierarchical
clustering analysis of DNA methylation data from PBD patient, CCALD patient
and healthy control fibroblasts and iPSCs using Pearson dissimilarity as the
distance measure and average-linkage as clustering method. Data from 5840
DNA methylation assays interrogating autosomal CpG loci was used in this
analysis. The heat map depicts b-value obtained from each assay, with red
depicting hypermethylated and green depicting hypomethylated alleles,
respectively. B. Bisulfite sequencing of the promoter regions of OCT4 and
NANOG. Open and closed circles indicate unmethylated and methylated CpG
sites in the promoters of these genes.
A.
54
Figure 3.3, Continued
B.
55
3.3.5 Candidate iPS cells are pluripotent
To evaluate the differentiation ability of the patient and healthy control iPS
cells, we first performed embryoid body (EB) differentiation. Immunostaining
detected germ layer-exclusive markers, TuJ1 for ectoderm, α-SMA for mesoderm
and AFP for mesoderm, in differentiated cells. (Figure 3.4A) Using this approach,
we confirmed that all the PBD, CCALD and control iPSCs differentiate into all
three germ layers in vitro.
To further test pluripotency in vivo, we performed teratoma formation
assays by subcutaneously injecting one PBD-donor derived iPSC
(PBD_PEX1fs1-iPS1), two CCALD-donor derived iPSCs (CCALD1-iPS4 and
CCALD2-iPS1) and one healthy control iPSC (Control2-iPS3) into dorsal flanks
of immune-deficient (SCID) mice. Nine weeks after injection, we observed tumor
formation from all four iPSCs above. Histological analysis showed the presence
of various tissues derived from all three germ layers in all teratomas, including
neural rosette structures (ectoderm), retinal pigment epithelium (ectoderm),
cartilage and bone (mesoderm), non-keratinizing squamouse epithelium
(endoderm) and glandular tissue (endoderm) (Figure 3.4B).
56
Figure 3.4: Embryoid body and teratoma differentiation of representative
PBD, CCALD and healthy control iPSCs.
(A) Fluorescent immunostaining for three germ layers from embryoid body-
mediated differentiation of representative PBD, CCALD and healthy control
iPSCs. Red and green colors represent ES marker staining, blue color
represents DAPI nuclear counterstaining. (B) Teratomas derived from
representative PBD, CCALD and healthy control iPSCs. Teratomas developed
three germ layers. N, Neural rosettes; P, pigmented neuroepithelium; C, Cartilage
tissue; G, Glandular tissue. Scale bar= 50µm.
A.
CCALD1-iPS2 Control2-iPS2 PBD_PEX1fs1-iPS1
57
Figure 3.4, Continued
B.
CCALD1-iPS4 Control2-iPS3 PBD_PEX1fs1-iPS2
58
3.4 Discussion
Peroxisomal disorders are a group of complex metabolic disorders with
variable expressivity and different genetic mutations. Although the primary
genetic basis for peroxisomal diseases has been known for years, the exact
nature of their pathogenesis and genetic and environmental modifiers are not
fully understood. Here, we generated iPSC resources for the long-term purpose
of developing novel tissue culture models for elucidating the pathogenesis of
PBD-ZSD, RCDP1, and CCALD and screening for more effective drug therapies.
In keeping with prior reports, skin fibroblasts from PBD and CCALD patients can
be successfully reprogrammed to form iPSCs with the hallmark molecular
properties of pluripotency including the expression of hES-specific gene and
protein biomarkers and changes in DNA methylation levels, as shown in the
results. Patient iPSCs, like the control iPSCs, can be stably maintained for more
than one hundred passages. Patient and control iPSCs could form embryoid
bodies and spontaneously differentiate in vitro into representative cell types of all
three germ layers. Most importantly, patient iPSCs, as well as the control iPSCs,
formed teratomas with evidence of cell types from all three germ layers.
Consistent with prior reports, we identified de novo CNCs in some of our
iPSCs. The observed genomic deletions always affect only one allele and
genomic amplifications always result in duplication in copy number. We ran
replicate CytoSNP array assays for a subset of the patient and control iPSCs
using DNA samples from either the same or different preparation. We observed
that although some CNVs exist in both samples or only in later passage, some
59
CNVs only appear in one sample but not the other. This may due to experimental
error. Therefore, we strongly suggest that all CNVs should be confirmed by
duplicate CytoSNP array assays. Furthermore, these analyses should be
performed respectively to monitor CNVs occurred along extensive passaging.
We detected the reoccurrence of regional duplications of chromosome 12
and chromosome 20, which encompass genes involved in embryogenesis,
cellular growth and division. We speculate that these two reproducible gains of
chromosome regions may confer benefitial growth properties for these iPSCs.
However, our gene expression microarray data did not show increased
expression levels of genes within the duplicated regions in samples with the
indicated CNVs.
Due to the fact that we conducted global expression and DNA methylation
analyses on all the fibroblasts and most iPSCs, we could investigate the effects
that these CNVs have on the expression of genes located within affected
genomic segments. In almost all circumstances, their expression levels were
within range of diploid samples. Although there are several possible explanations
for these observations, this could reflect the effects of selection whereby CNVs
are only tolerated or benefit in iPSCs if they involve genomic regions that do not
influence the initiation of reprogramming or maintenance of pluripotency. It is also
possible that epistatic interactions reduce the effects of CNVs on the gene
expression profiles of iPSCs.
The global DNA methylation analyses detected iPSC-specific methylation
and demethylation as compared to their parental fibroblasts, which are features
60
of pluripotency cells, including iPSCs and hESCs. We also detected incomplete
demethylation in some iPSCs, including those from patients and control donors.
It has been reported that low-passage (p<20) iPS cells retain an epigenetic
memory of the paternal cells (Ohi, Qin et al. 2011). But it does explain our
observations since the iPSCs with incomplete demethylation range from passage
30 to passage 78. Some genes proximal to these incomplete demethylation
regions are involved in transcription regulation and early development. Although
their expression levels are comparable among all the iPSCs, extra caution needs
to be taken when we characterize disease relevant functions of differentiated
cells, since the incomplete demethylation associated genes may be differentially
expressed in certain cell lineages.
Here, we successfully generated iPS cells from PBD, CCALD patient and
healthy control iPS cells and performed vigorous characterization on the genomic,
epigenomic, protein, and functional levels. The patient and control iPS cells
provide a novel platform for studying the disease phenotype and pathology in
different type of cells, especially the CNS cells. However, some iPS cells bear
copy number changes or incomplete demethylation. Even though these
variances did not result in detectable difference at gene expression level in iPS
cells, it could result in differences in their ability to produce a variety of
differentiated cell types or even the properties of these differentiated cell types
themselves. Therefore, further confirmation steps are required when evaluating
disease related phenotypes in differentiated cells obtained from iPSCs that
acquired CNCs or experienced incomplete epigenetic reprogramming.
61
CHAPTER 4
FUNCTIONAL CHARACTERIZATION OF PBD PATIENT, CCALD PATIENT
AND HEALTHY CONTROL IPS CELLS
4.1 Abstract
PBD-ZSS and CCALD are diagnosed by the change of lipid levels,
especially very long chain fatty acids (VLCFA) and plasmalogen (PL), in patients’
blood, urine and fibroblast culture. Here, we observed elevated VLCFA levels in
both PBD-ZSS and CCALD patient fibroblasts, as in previous reports. In contrast,
patient iPSCs have similar VLCFA levels and variable plasmalogen levels as
compared to control iPSCs.
Global gene expression analysis identified higher expression of genes in
mitochondrial β-oxidation and ketone body metabolism and lower expression of
genes in the Wnt signaling pathway in all PBD-ZSS patient iPSCs relative to
healthy control iPSCs; lower expression of PEX1 in PBD-ZSS patient iPSCs
carrying PEX1 frameshift mutation; as well as differentially expressed genes
(DEGs) enriched for lipid metabolism oxidative stress and inflammation,
especially neural inflammatory response, in both PBD-ZSS and CCALD patient
iPSCs relative to healthy control iPSCs.
However, global DNA methylation analyses did not reveal any robust
differential methylated loci (DMLs) correlated to DEGs between patient and
healthy control cells and thus no evidence supportive of epigenetic modifiers of
peroxisomal disorders on the level of DNA methylation.
62
4.2 Introduction
Peroxisomal biogenesis disorders (PBD) and X-linked
adrenoleukodystrophy (X-ALD) are two major groups of peroxisomal disorders
(Figure 1.6). These disorders have variable severity and expressivity.
Patients with the most severe form of PBDs, Zellweger syndrome (ZS),
most often do not survive beyond their first year of life, whereas less severe
neonatal adrenoleukodystrophy (NALD) patients usually live until late infancy and
the infantile Refsum disease (IRD) patients may survive beyond infancy and
even reach young adulthood. Most patients in the Zellweger syndrome spectrum
(ZSS) have liver disease, variable neuro-developmental delay, retinopathy and
perceptive deafness since birth.
X-ALD primarily affects the nervous system, adrenal cortex, and testis with
highly variable clinical presentations that are influenced by modifier genes and
the environment (Berger and Gartner 2006; Berger, Pujol et al. 2010). Males with
ABCD1 mutations develop childhood cerebral ALD (CCALD) approximately 33%
of the time and adult onset adrenomyeloneuropathy (AMN) approximately 45% of
the time (Berger and Gartner 2006; Berger, Pujol et al. 2010). CCALD patients
typically show symptoms between 5 and 9 years of age with rapid cerebral
demyelination and adrenocortical atrophy. In contrast, AMN patients show a later
onset of disease (20 - 40 years of age) and present with adrenal insufficiency, a
distal axonopathy in the spinal cord, and peripheral neuropathy (Berger and
Gartner 2006; Berger, Pujol et al. 2010) that results in progressive spastic
paraparesis with debilitating end stage disease. In contrast, approximately 10%
63
of male mutation carriers develop primary adrenocortical insufficiency (Addison’s
disease) with no evidence of nervous system dysfunction. Although male carriers
typically show the most severe clinical manifestations of disease, about half of
female ABCD1 mutation carriers develop AMN-like symptoms later in life (Berger
and Gartner 2006; Berger, Pujol et al. 2010).
Individuals with peroxisomal disorders are frequently diagnosed based on
the lipid levels of various body fluids and cultured fibroblasts. These lipid
biomarkers include measuring the levels of very long chain fatty acids (VLCFA),
branched chain fatty acids (pristanic acid and phytanic acid), di- and
trihydroxycholestanoic acid (D/THCA), and plasmalogens in patients’ red blood
cells, plasma, urine and cultured fibroblasts. PBD-ZSD patient cells tend to have
elevated VLCFA, pristanic acid, phytanic acid, and D/THCA levels, but decreased
plasmalogen levels relative to those of controls. RCDP1 patients have normal
VLCFA, D/THCA and pristanic levels, but increased phytanic acid and decreased
plasmalogen levels. As discussed in the Introduction, this is due to defects in the
import of the PHYH and AGPS proteins. Male ABCD1 mutation carriers who are
predisposed to the various forms of X-ALD have elevated VLCFA, but normal
levels of plasmalogens and branched chain fatty acids. Nevertheless, we note
that the biochemical analyses of lipid components have been primarily limited to
patient blood, urine and skin biopsies. Their levels in other cell types and tissues
have not been explored as fully.
Microarray-based transcriptome profiling has been used to identify genes
involved in peroxisome assembly and function in a number of different model
64
systems. This includes studies performed in yeast grown on oleate (peroxisome
induction) or glucose (peroxisome repression) containing media (Smith, Marelli et
al. 2002). Gene expression profiling of a Drosophila melanogaster Pex1-mutant
model for PBD-ZSD uncovered gene clusters differentially expressed between
wild-type and mutant larvae, revealing peroxisomal functions in neuronal
development, innate immunity, lipid and protein metabolism, gamete formation,
and meiosis (Mast, Li et al. 2011). A comprehensive comparative study of human
and great ape peroxisomal functions also employed gene expression microarray
assays and reported cross-species differentially expressed genes (DEGs)
involved in plasmalogen biosynthesis in multiple tissues. For example, DEGs in
brain (AGPS, FAR1, and FAR2) and heart (AGPS and PEX7) are more abundant
in humans than in chimpanzees. In contracts, DEGs in kidney (AGPS, GNPAT,
and PEX7) and liver (AGPS and PEX7) are more abundant in chimpanzees as
compared to humans. (Moser, Steinberg et al. 2011)
To date, the direct application of microarray gene expression profiling in
diagnosis and gene discovery of peroxisomal disorders has not been as
informative with respect to providing useful leads for deciphering disease
pathogenesis. One of our previous studies performed comprehensive
comparisons of the gene expression profiling of cultured fibroblasts of multiple
PBD-ZSD, RCDP1, CCALD and AMN patients with a variety of mutations. The
Hacia laboratory has also compared skin fibroblasts grown under various culture
conditions, including different culture temperatures and nutrition compositions
(unpublished data). None of these comparisons revealed any differentially
65
expressed genes whose identity correlated with hypotheses relevant to
peroxisome function and disease pathogenesis. A similar study using microarray-
based gene expression profiling of cultured fibroblasts of a broad group of
inherited metabolic diseases, including PBD-ZSD, detected no characteristic
gene signatures, except decreased mRNA expression of the defective genes with
disease-associated nonsense and frameshift mutations (Hernandez, Schulz et al.
2010). Overall, we hypothesize that these negative results obtained from different
research groups are primarily influenced due to the fact that skin fibroblasts have
very limited peroxisomal activity and thus are relatively weak models for
uncovering the mechanistic basis for the pathological phenotypes in the CNS and
other organ systems.
It has been reported that drugs that effect the epigenome, such as valproic
acid, butyrates and suberoxylanilide hydroxamic acid (SAHA), effectively
normalized the VLCFA levels in XALD patient fibroblasts and reduced expression
of proinflammatory cytokines in mouse astrocytes that were silenced for Abcd1
and Abcd2 gene activity. However, the mechanistic basis for the rescue of
peroxisomal lipid metabolic function in response to epigenetic regulators is
unclear. Our DNA methylation BeadArray data from the patient and healthy
control fibroblasts and iPSCs provides an opportunity to further investigate the
epigenetic contribution to the pathophysiology of peroxisomal disorders.
66
4.3 Results
4.3.1 Lipid profiles of patient and healthy control cells
First, we compared pertinent lipid profiles of cultured skin fibroblasts from
PBD-ZSD patient and healthy control donors. Consistent with prior reports
(Steinberg, Dodt et al. 2006), cultured PBD-ZSD patient skin fibroblasts grown in
fibroblast growth media showed significantly elevated VLCFA levels (two-tailed
Student’s t-test p-value<0.05) (Figure 4.1A). The PBD-ZSD patient skin
fibroblasts carrying PEX1 frameshift mutations have abolished peroxisomal
activities and showed >10-fold elevated VLCFA levels; the patient fibroblasts
carrying PEX1 or PEX26 missense mutations have reduced peroxisomal
activities and showed >7-fold elevated VLCFA levels. The patient fibroblasts with
PEX10 mutations have similar VLCFA levels as the healthy control fibroblasts,
consistent with previous report of normal VLCFA and plasmalogen levels
(Steinberg, Snowden et al. 2009). The patient fibroblasts with PEX12 mutations
also have comparable VLCFA levels as the healthy control ones, reflecting the
temperature sensitive feature of this cell line with peroxisomal assembly
improved at 37 and defective at 40°C (Steinberg, Dodt et al. 2006). Similarly, we
found elevated VLCFA levels in patient fibroblasts compared to control fibroblast
grown under the same conditions in iPS medium.
We observed more variations in the plasmalogen levels among PBD-ZSD
patient and healthy control fibroblasts grown in fibroblast medium or iPSC
medium. (Figure 4.1B) However, the plasmalogen levels of patient fibroblasts
carrying mutations in PEX10, PEX12 or PEX26 are consistently lower than
67
healthy control fibroblasts grown under the same conditions in fibroblast medium
or iPSC medium.
Then, we compared the lipid profiles of iPSCs derived from PBD-ZSD
patients and healthy control donors. In contrast to patient fibroblasts, which have
higher VLCFA levels than healthy control fibroblasts, all patient iPSCs have
slightly lower VLCFA levels as compared to healthy control iPSCs. In addition,
the plasmalogen levels are highly variable among patient iPSCs with different or
same genotypes as well as the healthy control iPSCs (Figure 4.1C, D). This is
not due to the different components in fibroblast medium and iPSC medium,
since the PBD-ZSD patient fibroblasts show consistent higher VLCFA levels than
healthy control fibroblasts in both medium conditions. However, we cannot
exclude the supportive effects of MEFs in the iPSC-feeder co-culture system.
Nevertheless, all healthy control iPSCs have higher VLCFA levels than all the
PBD-ZSD patients.
We also compared the lipid profiles of cultured skin fibroblasts from
CCALD patients and healthy controls. Consistent with prior reports (Corzo,
Gibson et al. 2002), cultured CCALD patient skin fibroblasts grown in fibroblast
growth media showed 4-fold elevated VLCFA levels, but similar plasmalogen
levels relative to fibroblasts from healthy donors (Figure 4.2A). Similarly, we
found about 4-fold elevated VLCFA levels, but comparable plasmalogen levels, in
patient and control fibroblasts grown under the same conditions in iPSC media
(Figure 4.2B). In contrast, no significant differences were found for either VLCFA
68
or plasmalogen levels in patient and control iPSCs cultured under the same
conditions in iPSC media (Figure 4.2C, D).
Therefore, we observed consistent elevation of VLCFA levels in all PBD-
ZSD and CCALD patient fibroblasts grown in fibroblast medium, as well as in
iPSC medium. We also observed decreased plasmalogen levels in PBD-ZSD
patient fibroblasts carrying PEX10, PEX12, or PEX26 mutations. In contrast, no
significant differences were found for either VLCFA or plasmalogen levels in
either PBD-ZSD patient iPSCs or CCALD patient iPSCs as compared to healthy
control iPSCs.
69
Figure 4.1: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control fibroblasts and
iPSCs. (A) Relative VLCFA levels in PBD-ZSD patient and healthy control
fibroblasts grown in fibroblast medium or iPSC medium, as represented by the
C26:0/C22:0 ratio; (B) Relative plasmalogen levels in patient and healthy control
fibroblasts grown in fibroblast medium or iPSC medium, as represented by the
total phosphatidyl ethanolamine (PE) plasmalogen/C20:0 ratio; (C) Relative
VLCFA levels in patient and healthy control iPSCs grown in iPSC medium, again
represented by the C26:0/C22:0 ratio; (D) Relative plasmalogen levels in patient
and healthy control iPSCs grown in iPSC medium again represented by the total
PE plasmalogen/C20:0 ratio. In (C) and (D), the VLCFA and plasmalogen levels
were measured in multiple iPSCs corresponding to each parental fibroblast.
A B
C D
70
Figure 4.2: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in CCALD patient and healthy control fibroblasts and
iPSCs. (A) Relative VLCFA levels in CCALD patient and healthy control
fibroblasts grown in fibroblast medium or iPSC medium, as represented by the
C26:0/C22:0 ratio; (B) Relative plasmalogen levels in patient and healthy control
fibroblasts grown in fibroblast medium or iPSC medium, as represented by the
total phosphatidyl ethanolamine (PE) plasmalogen/C20:0 ratio; (C) Relative
VLCFA levels in patient and healthy control iPSCs grown in iPSC medium, again
represented by the C26:0/C22:0 ratio; (D) Relative plasmalogen levels in patient
and healthy control iPSCs grown in iPSC medium again represented by the total
PE plasmalogen/C20:0 ratio.
A B
C D
71
4.3.2 Differential gene expression among PBD-ZSD patient and healthy
control cells
We began the analyses by comparing the gene expression profiles of
PBD-ZSD patient and healthy control fibroblast. Among the six PBD-ZSD patient
and three control fibroblasts used for reprogramming, there were no differentially
expressed gene (DEGs) found (>1.2-fold change, FDR<0.1), likely due to the
limited number of healthy control samples analyzed.
Then, we studied the gene expression profiles of 12 control iPSCs derived
from 3 healthy donors and 12 PBD-ZSD iPSCs derived from 6 PBD-ZSD donors
(two with PEX1 frameshift mutant, one with PEX1 missense mutant, one with
PEX10 mutant, one with PEX12 mutant, and one with PEX26 mutant) using the
same criteria. From this study, we discovered 151 DEGs, with 93 genes more
and 58 genes less expressed in PBD-ZSD patient relative to control iPSCs.
Based on GeneOntology (GO) analysis, we found a total of 21 functional
categories enriched (≥4 genes, B-H corrected P<0.05) for DEGs that were more
highly expressed in PBD-ZSD patient iPSCs relative to the control iPSCs. These
enriched functional categories included MHC class I protein complex involved in
antigen processing and presentation of peptide antigen (HLA-B, HLA-E, HLA-F,
and HLA-G), mitochondria components especially mitochondria envelop proteins
(ACAT1, AGK, BCL2L2, COX5B, CPT1A, MCL1, SLC25A36, BSG, CDK7, COQ3,
COQ7, ETHE1, MRPL48, and PTRH2) (Figure 4.3), and ketone body metabolic
process (ACLY, AKR1A1, ATF4, BSG, CACNA1A, COQ3, COQ7, CPT1A,
72
ELOVL5, FOLR1, GRHPR, and LIPA) (Figure 4.4). Ketone bodies are mainly
generated in liver mitochondria as by-products of β-oxidation, during which fatty
acids are broken down to acetyl-CoA and enter the citric acid cycle for energy.
Ketone bodies are synthesized with excess acetyl-CoA that were unable to enter
the citric acid cycle and transported to brain and heart as energy source (Laffe
1999). KEGG analysis showed enriched pathways involved in endocytosis, cell
adhesion molecules, antigen processing and presentation, autoimmune thyroid
disease, viral myocarditis, allograft rejection, Type I diabetes mellitus, and graft-
versus-host disease, all related to the major histocompatibility complex, HLA-B,
HLA-E, HLA-F, and HLA-G (≥4 genes, B-H corrected P<0.05).
For the DEGs that are less expressed in PBD-ZSD patient relative to
control iPSCs, GO analysis identified no enriched functional categories (≥4
genes, B-H corrected P<0.05). With the same criteria, KEGG analysis showed
enriched Wnt signaling pathway (DKK1, FZD4, PPP2R1B, PPP2R5E, SOX17,
and WNT3) (Figure 4.5) and oocyte meiosis (CDC27, PPP2R1B, PPP2R5E, and
YWHAQ). WNT3 is a member of the WNT gene family which encodes secreted
signaling proteins implicated in oncogenesis and in developmental processes,
including regulation of cell fate in embryogenesis. Overexpression of WNT3
promotes cell proliferation and increased neuronal differentiation of neural stem
cells and neurite outgrowth. (Yin, Zhang et al. 2007; David, Canti et al. 2010;
Shruster, Eldar-Finkelman et al. 2011) FZD4 is a member of the frizzled gene
family encoding seven-transmembrane domain receptor proteins that are
coupled to the β-catenin canonical signaling pathway as a positive regulator.
73
FZD4 has been shown to be up-regulated during mouse embryonic stem cell
neural differentiation in response to retinoic acid (RA) (Verani, Cappuccio et al.
2007). SOX17 encodes a member of the SOX (SRY-related HMG-box) family of
transcription factors involved in the regulation of embryonic development and cell
fate determination. In rodents, Sox17 antagonizes the Wnt/β-catenin signaling
pathway, guides endoderm differentiation (Hudson, Clements et al. 1997; Sinner,
Rankin et al. 2004) and regulates oligodendrocyte maturation by promoting the
expression of myelin basic protein (MBP) (Chew, Shen et al. 2011). DKK1 is a
member of the dickkopf family which encodes secreted proteins with two cysteine
rich regions. DKK1 is involved in embryonic development through its inhibition of
WNT signaling pathways. Expression of Dkk1 is required for neural induction in
mouse embryonic stem cells (Verani, Cappuccio et al. 2007) and differentiation
into astrocytes, oligodendrocytes and neurons (Ahn, Byun et al. 2008). However,
down-regulation of Wnt/β-catenin signaling with the addition of Dkk1 during mid
and late stages of neurogenesis inhibited neuronal production (Munji, Choe et al.
2011). PPP2R1B and PPP2R5E encode the regulatory subunits A and B of
protein phosphatase 2, one of the four major serine-threonine phosphatases.
They are regulators in Wnt signaling, involved in the negative control of cell
growth and division, and required for dorsal development, midbrain-hindbrain
boundary formation and for closure of the neural tube (Yang, Wu et al. 2003).
The Ser/Thr phosphatase PP2A binds to Axin/APC/GSK-3 complex to antagonize
the effects of the kinase GSK-3 on β-catenin (Hsu, Zeng et al. 1999).
74
Given the possible gaps in public databases of gene functions relevant to
peroxisome biology and PBD pathogenesis, we used the DAVID Bioinformatics
resource to annotate the function of DEGs and manually searched for genes
relevant to peroxisome biology, lipid metabolism, oxidative stress,
neuroinflammation, and embryonic development. Six DEGs are involved in lipid
metabolism, AGK, AKR1A1, ACAT1, CPT1A, and ELOVL5 are more highly
expressed and ALDH3A2 is less highly expressed in PBD-ZSD patient iPSCs
relative to control iPSCs. Acetyl-Coenzyme A acetyltransferase 1(ACAT1) and
carnitine palmitoyltransferase 1A (CPT1A) are involved in mitochondria β-
oxidation. ELOVL5 plays a role in elongation of polyunsaturated long chain fatty
acids on the endoplasmic reticulum (ER) (Leonard, Bobik et al. 2000). COX5B
encodes a subunit of the cytochrome C oxidase, terminal enzyme of the
mitochondrial respiratory chain. Seven genes up-regulated, CRYGD, EGR1, FOS,
GLRX, MCL1, NEF2L1, and STK24, and five genes down-regulated, ALDH3A2,
CLN8, DKK1, HAS2, and KPNA1, in PBD-ZSD patient iPSCs relative to control
iPSCs are involved in oxidative stress (Figure 4.6). In addition to the four major
histocompatibility complex genes involved in inflammatory, we also identified up-
regulation of BSG, LIPA, M6PR, and MCAM expression, and down-regulation of
CTSC involved in the inflammatory process (Figure 4.7). Finally, we identified
genes involved in formation of embryonic tissue including COQ7, DKK1, FKBP1A,
FLI1, FOLR1, FOS, HAS2, SOX17 and WNT3.
Next, we focused on the gene expression profiles of PBD-ZSD patient
iPSCs with PEX1 mutations to identify disease-genotype specific DEGs. We
75
identified 34 DEGs, with 25 genes more and 9 genes less expressed in four
PBD-ZSD patient iPSCs with PEX1 frameshift (fs) mutations relative to nine
healthy control iPSCs. Among the 34 DEGs, more expression of 6 genes,
AKAP17A, ATG12, FOS, MCL1, MARK3, and ITM2C, and less expression of 4
genes, ALDH3A2,CDC27, GAPVD1, and TMEM149, are common in all the PBD-
ZSD patient iPSCs regardless of the mutation genotypes. ALDH3A3 is involved
in lipid metabolism. FOS and MCL1 respond to oxidative stress, and ATG12 is
involved in the regulation of autophagy through RIG-I-like receptor signaling
pathway. Among the homozygous PEX1 fs mutant iPSC unique-DEGs, PEX1
gene shows significantly less expression (>2.4-fold) in patient relative to control
iPSCs (Figure 4.8), correlated with the absence of peroxisomal activities in the
patient fibroblasts (Yik, Steinberg et al. 2009).
For genes with higher expression in the PBD-ZSD iPSCs with
homozygous PEX1 fs mutations, GO (GeneOntology) and KEGG analysis
identified no enriched functional categories or pathways (≥4 genes, B-H
corrected P<0.05). For genes with lower expression in the patient iPSCs with
homozygous PEX1 frameshift mutaitons, GO analysis identified two broad
biological process, organelle organization and cellular component organization,
and KEGG analysis showed no enriched pathways under the same criteria.
Therefore, we used the DAVID Bioinformatics resource to annotate the function
of DEGs and manually searched for genes relevant to peroxisome biology and
peroxisomal disorder pathology. Besides the DEGs shared by all the PBD-ZSD
patient iPSCs mentioned above, the increased expression of SPTLC1 is shown
76
to be associated with elevated sphingolipid metabolism, which has been
demonstrated to be a diagnostic criterion for PBD (Pettus, Baes et al. 2004);
PGK1 expression in the glycolysis pathway is elevated in response to oxidative
stress; CEBPB is important for gene regulations in the IL-6 signaling pathway
and involved in immune and inflammatory responses as well as embryo
development.
We identified 12 DEGs, with 11 genes more and 1 gene less expressed in
three PBD-ZSD patient iPSCs with PEX1 missense mutations relative to nine
healthy control iPSCs. Two genes with higher expression in patient iPSCs with
homozygous PEX1 missense mutation, BSG and RPS4Y1, are shared in all the
PBD-ZSD patient iPSCs regardless of the mutation genotypes. No DEGs are
shared between patient iPSCs with homozygous PEX1 fs mutations and those
with homozygous PEX1 ms mutations. GO and KEGG analysis failed to identify
any enriched functional categories or pathways among the 12 DEGs. Through
manual search in the David Bioinformatics resource, we identify one DEG,
PRODH, regulates apoptosis in response to oxidative stress.
4.3.3 Differential gene expression among CCALD patient and healthy
control cells
First, we compared the gene expression profiles of CCALD patient and
healthy control fibroblasts. Among the two patient and three control fibroblasts
used for reprogramming, there were no differentially expressed genes (DEGs)
found (>1.2-fold change, FDR<0.1) due to the limited number of samples
77
analyzed. To enhance the statistical power to detect potential DEGs, we
conducted a larger-scale gene expression analysis (>47,000 transcripts) using
cultured skin fibroblasts from five healthy control donors and five CCALD patients.
In these studies, we identified 127 DEGs, including 44 less and 83 more
expressed in patients relative to controls.
Based on GO analysis, we found 4 enriched functional categories (≥4
genes, B-H corrected P<0.05) for DEGs that were more highly expressed in
CCALD patient relative to healthy control fibroblasts. The most specific GO
category for genes more highly expressed in patient fibroblasts was nuclear
lumen genes, including HNRNPA2B1, RCC2, CDKN1C, LMO4, IPO5, SOX9,
MAGOH, TCEA1, RBBP4, MED1, TSEN15, XPOT, and ATF5. KEGG analysis
did not show any enriched categories (≥4 genes, B-H corrected P<0.05) for
DEGs more highly expressed in patient relative to healthy control donor cells. In
contrast DEGs that were more highly expressed in healthy donor relative to
patient fibroblast were enriched for one KEGG category (‘Pathways in Cancer’ as
represented by the SMAD3, LAMA4, PML, and DAPK1 genes), but no GO
categories.
We also used the DAVID Bioinformatics resource to annotate the function
of DEGs and manually searched for genes relevant to peroxisome biology, lipid
metabolism, oxidative stress, and neuroinflammation. The only peroxisomal gene
was AGPS, which is involved in plasmalogen (a type of ether phospholipid)
biosynthesis, and was up-regulated in patient fibroblasts. Two genes encoding
enzymes involved in the Lands cycle, deacylation/reacylation reactions
78
responsible for glycerophospholipid remodeling, were present with the higher
expression of LYPLA1 (deacylation), but lower expression of MBOAT7
(reacylation), in patient relative to control fibroblasts. Two DEGs were involved in
sphingosine metabolism with SMPD1, which converts sphingomyelin to ceramide,
being less highly expressed and SGPL1, which degrades sphingosine-1-
phosphate, being more highly expressed in patient relative to control fibroblasts.
No genes involved in classic oxidative stress responses were found in the list.
The DEGs strongly or primarily related to neuroinflammation included CBLB,
RAB27A, and TCF3, which showed higher expression in patient fibroblasts, as
well as CD81, MASP1, and GRK5, which showed higher expression in healthy
control fibroblasts.
Based on our manual search, we found that 40% of the DEGs (46 genes)
were reported as genes with highest variability in expression based on biopsy
sites, as described in reference (Rinn, Bondre et al. 2006). We also note that the
expression profiles of four out of six DEGs described above as being involved in
neuroinflammation (RAB27A, CD81, MASP1, and GRK5) are also influenced by
the biopsy site. Although all the fibroblasts in our study were obtained from the
upper limbs, the control and patient donor cells were collected and expanded at
different sources (Methods), which could influence their gene expression
signatures.
Then, we studied the gene expression profiles of 5 CCALD iPSCs from 2
CCALD donors and 9 control iPSCs from 3 healthy donors and identified 75
DEGs (22 more and 53 less expressed in patient relative to control iPSCs). There
79
was no overlap with the DEGs uncovered in the cultured skin fibroblasts from the
5 healthy control and 5 CCALD patient donors discussed above. Based on GO
analysis, we found a total of 14 functional categories enriched (≥4 genes, B-H
corrected P<0.05) for DEGs that were more highly expressed in patient donor
cells (Additional file 10). These included blood vessel morphogenesis, regulation
of cellular protein metabolic process, and carboxylic acid metabolic process. In
contrast, GO analysis identified no enriched categories for DEGs that were more
highly expressed in healthy donor cells. KEGG analysis did not show any
enriched pathways (≥4 genes, B-H corrected P<0.05) for DEGs that were either
more highly expressed in the patient or healthy control cells.
Although GO and KEGG analysis did not highlight biological processes
proposed to be relevant to disease, inspection of the DEG functions based on the
DAVID Bioinformatics resource uncovered genes associated with major
hypothesis pertinent to XALD pathogenesis. Amongst the relevant genes with
reduced expression in CCALD patient relative to healthy donor derived iPSCs
were PEX11B (confirmed by qRT-PCR in Figure 4.9A and IMPA1 expression
confirmed in Figure 4.9B) and CD200. The former plays a pivotal role in
peroxisome proliferation and maintenance (Schrader, Reuber et al. 1998; Koch,
Pranjic et al. 2010). Decreased CD200 expression is associated with activation,
accumulation of macrophages, including brain microglia, and causes
inflammatory responses in other systems (Liu, Bando et al. ; Hoek, Ruuls et al.
2000). DEGS more highly expressed in patient relative to control iPSCs were
also related to hypotheses relevant to XALD pathogenesis and lipid metabolism.
80
ULK1 gene is the mammalian homolog of the yeast Atg1 gene, which plays a
critical role in the autophagy-mediated turnover of peroxisomes in yeast (Chen
and White 2011). PLA2G2A is involved in phospholipid turnover (Nakanishi and
Rosenberg 2006). NAAA, THBS1, and BSG all have functions related to
neuroinflammation (Li, Tong et al. ; Stolz, Hammond et al. 1993; Buyse,
Sitaraman et al. 2002; Arora, Gwinn et al. 2005; Tsuboi, Sun et al. 2005; Isenberg,
Jia et al. 2007; Maccarrone, Battista et al. 2007; Martin-Manso, Galli et al. 2008;
Solorzano, Zhu et al. 2009; Tournier JN, Rossi Paccani S et al. 2009). SLC7A8 is
a transporter of thyroid hormones, which can induce peroxisomal biogenesis and
β-oxidation as well as the ABCD2 expression, whose induction can correct
biochemical functions of X-ALD patient fibroblasts (Fourcade, Savary et al. 2003).
(Figure 4.10 and Figure 4.11)
81
Figure 4.3: DEGs with higher expression in all PBD-ZSD patient iPSCs
relative to healthy control iPSCs enriched for mitochondria genes.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
BSG 7.8 8.2 7.9 8.1 7.8 7.8 7.7 7.7 8.1 7.6 7.8 8.2 8.5 9.1 8.5 8.4 8.8 8.2 8.8 8.5 8.9 8.4 8.2
COQ3 6.8 7.4 7.0 6.9 6.9 6.8 7.0 7.3 7.6 7.5 7.1 7.1 7.8 7.8 7.6 7.4 7.2 7.6 7.6 7.4 7.5 7.6 7.8
COQ7 6.0 6.4 6.0 6.0 5.5 5.8 6.1 5.8 5.5 6.1 5.4 5.9 6.5 6.5 6.5 6.3 6.2 6.5 6.3 6.4 6.5 6.4 6.5
CPT1A 6.7 6.5 8.1 6.8 7.2 7.3 7.1 7.9 7.3 7.7 6.3 7.0 7.8 7.9 7.5 7.6 8.0 8.1 7.5 8.2 7.9 7.7 7.8
ACAT1 8.4 8.7 8.6 8.4 8.0 7.9 8.0 9.0 8.4 9.2 8.3 8.4 9.7 9.0 9.2 8.9 8.9 8.9 8.8 9.2 9.0 9.4 9.2
AGK 7.6 7.3 7.5 7.5 7.2 7.6 7.0 7.6 7.6 7.5 7.1 7.4 7.9 7.6 7.7 7.6 8.3 7.8 7.5 7.8 7.6 7.9 7.7
BCL2L2 6.2 6.1 6.6 6.2 5.9 6.4 6.0 6.6 6.5 5.8 6.2 6.5 6.7 6.6 6.4 6.6 7.0 6.1 6.7 6.9 6.7 6.8 6.8
MCL1 10.3 10.6 10.5 10.5 10.1 10.5 10.2 10.3 10.5 10.4 10.3 10.5 10.4 10.4 10.9 11.2 11.0 10.7 10.4 10.8 10.9 10.7 10.8
PTRH2 8.5 8.4 8.4 8.4 8.3 8.4 8.5 8.8 8.7 9.4 8.5 8.8 9.3 8.8 9.1 9.3 9.1 9.2 8.8 8.9 8.8 9.1 9.2
CDK7 8.6 9.0 9.2 8.9 8.6 8.5 8.4 9.7 8.9 9.8 8.8 9.4 10.0 9.6 9.8 9.8 9.1 9.5 9.2 9.8 9.6 9.8 10.0
COX5B 5.4 5.8 5.4 5.4 5.6 5.2 5.6 5.8 5.3 5.7 5.5 5.5 6.0 5.9 6.1 5.9 5.8 5.8 5.8 5.8 6.1 6.2 6.3
ETHE1 5.5 6.0 6.1 5.9 5.5 5.4 5.6 6.8 6.7 6.2 6.2 6.3 6.9 7.2 6.9 6.8 6.6 6.3 6.6 6.9 7.4 6.9 6.9
MRPL48 7.9 8.2 8.1 8.1 7.8 7.8 8.0 7.9 7.8 8.6 7.9 8.1 8.7 8.4 8.5 8.4 8.2 8.3 8.3 8.6 8.4 8.4 8.6
SLC25A36 6.9 7.6 7.8 7.2 7.1 7.1 7.1 8.6 8.1 7.9 6.9 8.4 8.9 7.5 8.8 8.7 9.3 8.7 7.0 8.5 8.5 9.2 9.1
Healthy Control iPSCs PBD-ZSD iPSCs
82
Figure 4.4: DEGs with higher expression in all PBD-ZSD patient iPSCs
relative to healthy control iPSCs enriched for ketone body metabolism.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
ACLY 7.4 7.5 7.9 7.2 7.6 7.6 7.8 7.8 7.9 7.8 7.8 7.9 7.7 7.7 8.2 8.4 8.1 8.0 7.8 8.3 8.2 8.0 8.2
ACLY 7.9 8.3 8.4 7.7 8.1 8.4 8.5 8.1 8.2 8.3 8.4 8.5 8.3 8.3 8.9 8.9 8.7 8.8 8.3 8.7 8.6 8.7 8.6
AKR1A1 9.6 9.7 9.8 9.8 9.6 9.4 9.4 10.3 10.2 10.0 9.8 10.0 10.3 10.2 10.2 10.1 10.0 10.1 10.2 10.4 10.5 10.4 10.2
ATF4 11.1 11.0 10.9 11.1 11.3 11.3 10.9 11.4 11.4 11.8 11.4 11.7 11.6 11.5 12.0 12.0 11.8 12.0 11.7 11.7 11.2 11.7 11.8
BSG 7.8 8.2 7.9 8.1 7.8 7.8 7.7 7.7 8.1 7.6 7.8 8.2 8.5 9.1 8.5 8.4 8.8 8.2 8.8 8.5 8.9 8.4 8.2
CACNA1A 5.1 4.7 6.4 5.0 5.2 5.3 4.6 6.1 6.5 5.7 5.9 6.0 6.3 6.6 6.3 6.3 6.6 6.2 6.8 6.1 6.6 6.2 6.8
COQ3 6.8 7.4 7.0 6.9 6.9 6.8 7.0 7.3 7.6 7.5 7.1 7.1 7.8 7.8 7.6 7.4 7.2 7.6 7.6 7.4 7.5 7.6 7.8
COQ7 6.0 6.4 6.0 6.0 5.5 5.8 6.1 5.8 5.5 6.1 5.4 5.9 6.5 6.5 6.5 6.3 6.2 6.5 6.3 6.4 6.5 6.4 6.5
CPT1A 6.7 6.5 8.1 6.8 7.2 7.3 7.1 7.9 7.3 7.7 6.3 7.0 7.8 7.9 7.5 7.6 8.0 8.1 7.5 8.2 7.9 7.7 7.8
ELOVL5 5.0 5.1 4.7 5.4 4.7 5.0 4.4 5.1 5.0 5.3 4.8 4.7 5.4 5.1 5.4 5.7 5.9 5.3 5.0 5.5 5.6 5.5 5.6
FOLR1 6.9 6.3 6.5 6.8 6.8 6.7 6.8 7.7 6.8 7.3 7.3 6.6 7.8 7.5 7.5 7.4 7.7 7.2 7.6 7.7 7.5 7.6 8.0
GRHPR 8.4 8.5 8.5 8.6 8.3 8.5 8.4 8.7 8.8 8.5 8.6 8.5 8.9 9.0 8.6 8.7 9.0 8.7 9.0 8.9 8.7 8.8 8.7
LIPA 7.1 6.7 7.3 6.8 7.1 7.5 7.2 7.1 6.8 6.1 7.1 7.7 7.8 8.1 7.5 7.2 8.2 8.3 7.9 8.1 7.6 7.9 7.4
Healthy Control iPSCs PBD-ZSD iPSCs
83
Figure 4.5: DEGs with less expression in all PBD-ZSD patient iPSCs relative
to healthy control iPSCs enriched for Wnt signaling pathway.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
WNT3 4.8 5.1 3.9 5.1 5.1 5.6 5.7 4.1 4.5 4.5 4.2 4.1 3.9 4.1 4.1 4.4 4.2 4.0 4.2 4.0 4.1 4.0 3.8
FZD4 5.3 5.7 5.3 6.0 5.4 5.5 5.0 5.3 5.4 5.3 5.4 6.0 5.1 5.2 5.2 5.4 5.0 5.1 5.0 5.2 5.0 5.1 5.1
DKK1 8.0 6.3 5.3 6.3 6.2 8.8 7.8 4.3 5.2 4.9 6.6 6.4 4.2 5.0 5.0 5.9 4.3 4.4 5.6 4.1 3.9 5.1 4.4
PPP2R5E 9.4 9.5 9.3 9.5 9.3 9.4 9.5 9.0 9.2 9.0 9.4 8.8 9.0 9.2 8.8 8.7 8.8 8.8 9.1 9.1 9.3 9.0 9.1
PPP2R1B 9.3 9.0 9.2 9.1 9.4 9.3 9.6 8.2 8.8 8.6 8.9 8.9 8.3 8.4 8.5 8.6 8.1 9.0 8.9 8.5 8.2 8.5 8.5
SOX17 5.9 5.5 4.5 5.7 5.6 6.6 5.2 4.0 4.9 3.9 4.7 5.9 3.9 4.6 4.2 4.7 3.5 3.8 5.3 4.1 4.4 4.0 3.9
Healthy Control iPSCs PBD-ZSD iPSCs
84
Figure 4.6: DEGs in all PBD-ZSD patient iPSCs relative to healthy control
iPSCs enriched for Wnt signaling pathway.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
CRYGD 5.3 5.3 4.9 5.1 5.2 5.1 5.3 4.9 5.2 5.2 5.2 5.2 5.2 5.8 5.9 6.0 5.2 5.9 6.1 5.9 6.0 5.8 5.9
EGR1 7.8 7.9 6.9 6.9 4.9 8.6 5.3 6.2 6.2 8.4 7.4 9.0 8.6 7.8 10.0 10.2 9.4 8.1 8.0 8.6 7.8 9.3 9.3
FOS 7.9 7.8 8.1 7.1 5.1 7.8 6.9 7.3 6.3 8.3 7.2 9.2 9.3 8.1 10.7 10.7 10.1 8.7 9.3 9.3 6.6 9.5 9.5
GLRX 7.1 7.2 6.9 7.6 6.9 6.9 6.8 7.7 7.5 7.4 7.1 7.9 8.2 7.8 7.8 7.5 8.3 7.7 7.6 7.9 7.9 8.4 8.2
MCL1 10.3 10.6 10.5 10.5 10.1 10.5 10.2 10.3 10.5 10.4 10.3 10.5 10.4 10.4 10.9 11.2 11.0 10.7 10.4 10.8 10.9 10.7 10.8
NFE2L1 8.3 8.4 8.8 8.6 8.9 8.7 8.5 9.1 9.1 9.2 9.2 9.1 9.1 9.2 9.1 9.2 9.0 9.3 9.1 9.4 9.4 9.2 9.4
NFE2L1 8.3 8.3 8.8 8.4 8.8 8.7 8.6 9.3 9.1 9.2 9.3 9.3 9.3 9.1 9.6 9.5 9.2 9.7 9.3 9.5 9.3 9.2 9.7
STK24 8.0 7.9 8.1 7.9 7.9 8.0 7.9 7.9 8.1 7.8 8.1 7.9 8.3 8.1 8.4 8.3 8.6 8.2 7.9 8.2 8.3 8.2 8.2
ALDH3A2 8.8 9.2 8.5 9.2 9.3 9.0 9.3 8.4 9.2 8.4 8.9 8.3 7.7 8.5 7.8 7.4 7.7 7.8 8.5 7.5 8.7 7.9 7.6
CLN8 4.4 4.6 4.1 4.5 4.3 4.7 4.3 4.2 4.8 4.3 4.4 4.5 3.6 4.1 3.9 4.0 3.7 4.1 4.3 4.4 4.2 3.9 3.9
DKK1 8.0 6.3 5.3 6.3 6.2 8.8 7.8 4.3 5.2 4.9 6.6 6.4 4.2 5.0 5.0 5.9 4.3 4.4 5.6 4.1 3.9 5.1 4.4
HAS2 7.8 8.1 7.2 8.8 8.4 8.7 8.1 4.4 7.4 4.5 6.7 7.0 4.2 6.2 5.6 5.8 3.9 4.8 6.9 4.8 5.8 5.6 4.4
KPNA1 6.6 6.7 6.3 6.6 6.5 6.7 6.6 5.5 5.9 5.5 5.9 5.6 5.6 5.7 5.3 5.9 6.1 5.9 5.8 5.7 5.5 5.8 5.7
Healthy Control iPSCs PBD-ZSD iPSCs
85
Figure 4.7: DEGs between all PBD-ZSD patient iPSCs and healthy control
iPSCs enriched for inflammation.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
BSG 7.8 8.2 7.9 8.1 7.8 7.8 7.7 7.7 8.1 7.6 7.8 8.2 8.5 9.1 8.5 8.4 8.8 8.2 8.8 8.5 8.9 8.4 8.2
HLA-A 8.9 9.4 10.2 9.3 9.7 9.5 9.4 10.4 10.2 10.0 9.8 9.5 10.5 10.0 10.2 10.1 10.8 10.6 9.9 10.3 10.5 10.6 10.4
HLA-B 7.3 7.2 7.9 7.2 7.5 7.4 7.4 8.7 8.3 8.4 8.1 8.8 8.7 8.1 8.6 8.7 8.3 8.6 8.3 8.7 8.7 9.1 8.6
HLA-E 5.3 5.1 5.8 5.5 5.0 5.0 5.3 6.4 5.6 5.6 5.5 5.9 6.5 6.4 6.1 6.1 6.5 6.0 5.9 6.2 6.3 6.5 5.9
HLA-F 6.2 6.6 6.9 6.3 7.1 6.5 6.4 7.6 7.3 7.5 7.1 7.5 7.7 7.3 7.6 7.5 7.5 7.7 7.3 7.5 7.8 7.8 7.5
HLA-G 6.7 6.8 6.9 6.9 7.0 6.9 7.1 7.3 7.4 7.6 7.3 7.5 7.5 7.1 7.6 7.6 7.6 7.9 7.3 7.4 7.8 7.7 7.5
LIPA 7.1 6.7 7.3 6.8 7.1 7.5 7.2 7.1 6.8 6.1 7.1 7.7 7.8 8.1 7.5 7.2 8.2 8.3 7.9 8.1 7.6 7.9 7.4
M6PR 7.8 7.9 7.6 7.5 7.9 8.1 7.9 7.8 8.0 7.8 7.6 7.9 8.1 8.2 8.2 8.3 8.0 8.6 7.8 8.0 8.2 8.4 8.3
MCAM 5.4 6.0 6.0 6.4 5.3 5.3 4.9 5.8 6.0 5.9 5.8 6.2 5.7 6.7 6.1 6.2 6.5 6.9 6.4 6.1 6.4 6.4 6.3
MCAM 7.0 6.8 7.8 7.1 6.9 6.9 6.6 7.1 7.3 7.3 7.4 7.4 7.1 8.0 7.6 7.6 7.8 7.8 7.8 7.3 7.9 7.6 7.7
MCAM 4.8 5.3 5.5 5.6 5.4 4.9 4.8 5.4 5.3 5.4 5.4 5.6 5.5 5.6 5.9 5.6 5.7 5.8 5.6 5.8 5.6 5.7 5.7
CTSC 10.7 10.6 10.5 10.5 10.7 10.7 10.8 10.7 10.5 10.2 10.5 10.4 10.0 10.0 10.4 10.3 10.2 10.2 10.4 10.2 10.4 10.5 10.4
Healthy Control iPSCs PBD-ZSD iPSCs
86
Figure 4.8: Less expression of PEX1 gene in PBD-ZSD patient iPSCs with
homozygous PEX1 frameshift mutation relative to healthy control iPSCs.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
PEX1 4.8 5.0 5.1 4.9 5.5 4.6 5.2 4.8 5.0 3.5 3.9 3.7 3.9
PBD-PEX1fs iPSCs Healthy Control iPSCs
87
Figure 4.9: qRT-PCR confirmation of differentially expressed genes in
CCALD patient and healthy control iPSCs.
Data from the analysis of (A) PEX11B and (B) IMPA1 gene expression levels are
provided. The error bars represent standard error. In both cases, the expression
was lower in CCALD patient relative to control iPSCs (two-tailed Student’s t-test
p-value <0.5), in accordance with the microarray-based gene expression data.
88
Figure 4.10: DEGs between CCALD patient iPSCs and healthy control
iPSCs involved in peroxisome turnover.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
PEX11B 6.0 5.3 6.2 5.4 5.9 5.6 5.2 5.4 5.6 5.1 4.9 5.1 4.8 4.5
ULK1 6.5 6.2 6.5 6.4 6.0 6.7 5.9 5.9 6.4 5.6 5.9 5.8 5.9 5.8
SLC7A8 3.8 4.4 4.0 4.1 4.5 4.1 4.6 4.1 4.8 5.3 5.3 5.0 4.7 5.0
PLA2G2A 4.7 4.4 4.6 4.5 4.8 4.6 4.8 4.5 4.8 5.2 5.1 4.8 5.2 4.7
Healthy Control iPSCs CCALD iPSCs
89
Figure 4.11: DEGs between CCALD patient iPSCs and healthy control
iPSCs involved in inflammation.
Green, black, and red color blocks represent low, medium, and high gene
expression levels respectively. Numbers in the color blocks represent log
transformed gene expression value.
BSG 7.9 8.2 8.0 7.9 7.7 8.2 7.7 7.9 8.2 8.6 8.7 8.7 8.5 8.6
NAAA 5.1 4.9 5.0 5.0 5.6 5.1 5.0 5.1 4.8 5.8 5.9 5.8 6.0 5.6
THBS1 6.3 7.9 6.0 5.2 5.5 5.8 6.9 6.9 8.8 8.4 8.9 8.9 8.8 9.3
CD200 8.4 7.8 8.5 8.3 8.3 8.2 8.1 8.0 8.3 7.7 7.4 7.9 7.7 7.6
CD200 6.5 6.3 6.5 6.9 6.3 6.4 5.7 5.8 5.9 5.2 5.4 5.5 5.6 5.8
Healthy Control iPSCs CCALD iPSCs
90
4.3.4 Differential DNA methylation among patient and healthy control cells
We didn’t observe any robust differentially methylated loci (DMLs)
between the seven PBD-ZSD patient and three control fibroblasts (Δβ>0.25, B-H
corrected P<0.05). However, we found 49 robust DMLs in comparisons of the
twenty-two PBD-ZSD patient and eight control iPSCs. 13 out of the 19 genes
proximal to these DMLs are poorly expressed. The other 6 expressed genes
show no significant differentially expression between PBD-ZSD patient and
control fibroblasts.
We only found 5 robust DMLs in comparisons of two CCALD patient and
three control fibroblasts and one robust DML in comparisons of five CCALD
patient and nine control iPSCs. All 3 genes proximal to the DMLs in patient and
control fibroblasts were poorly expressed in all these fibroblasts, regardless of
ABCD1-mutation status.
4.4 Discussion
In this chapter, we characterized the cellular lipid metabolism, gene
expression and DNA methylation profiles of PBD patient and CCALD patient iPS
cells.
To begin to address the issue of cellular lipid metabolism in the etiology of
PBD and CCALD, we characterized the lipid profiles of fibroblasts and iPSCs
derived from patients and healthy controls. In keeping with prior reports
(Steinberg, Dodt et al. 2006; Jang, Kang et al. 2011), we found that the PBD and
CCALD patient fibroblasts have elevated VLCFA levels as compared to those
91
from healthy controls. We also found that the PBD patient fibroblasts with PEX10,
PEX12 or PEX26 mutation have reduced plasmalogen levels. However, PBD
patient fibroblasts with homozygous PEX1 fs or ms mutation have similar
plasmalogen levels, as compared to those from healthy controls. This
observation was independent of the growth media tested and occurs despite the
fact that a key enzyme in plasmalogen biosynthesis, AGPS, is 1.7-fold up-
regulated in fibroblasts cultured under normal growth conditions from CCALD
patient relative to healthy controls.
All the PBD iPSCs have slightly lower levels of VLCFA compared to the
healthy control iPSCs, but no significant difference in their plasmalogen contents.
The data indicates that either VLCFAs are not supplemented in the medium or
actively produced in human iPS cells or they are metabolized through functionally
redundant metabolic pathways, for example, β-oxidation of LCFA and MCFA in
mitochondria.
Regardless of their ABCD1 mutation status, the CCALD patient and
healthy control iPSCs showed no differences in their VLCFA or plasmalogen
levels, which were independent of the media tested. A challenge in these
experiments is the relative incompatibility of the growth conditions for iPSCs and
fibroblasts. While fibroblasts can grow in iPSC media, they change their
morphology and have slower growth rates. Meanwhile, iPSCs can rapidly
differentiate in fibroblast media and thus these experiments are not feasible.
Lastly, we cannot rule out the contribution of feeder cells to the lipid profiles of
the iPSCs cells, this most likely reflects previously described differences in iPSC
92
and fibroblast metabolomes (Panopoulos, Yanes et al. 2012). Although ABCD1
mutation carriers show profoundly elevated sVLCFA levels in their blood and
urine and dramatic reductions of VLCFA catabolic activity in their cultured
fibroblasts, their role in disease pathologenesis is still under discussion. In a
subset of asymptomatic patients with normal brain myelination, dietary therapy
with “Lorenzo’s Oil”, a mixture of glycerol trioleate and glyceryl trierucate that
inhibits endogenous sVLCFA synthesis, helps prevent the onset of cerebral
disease in some, but not all individuals (Moser, Raymond et al. 2005; Moser,
Moser et al. 2007; Semmler, Kohler et al. 2008). Likewise, the significance that
decreased plasmalogen levels in the RBCs and white matter of patients has
been the subject of debate (Wilson and Sargent 1993; Moser 1999; Khan, Singh
et al. 2008).
As part of the genomic characterization of our iPS and fibroblast cell
resources, we acquired global gene expression data from patient and healthy
control derived fibroblasts and iPSCs. We identified no DEGs between PBD-ZSD
patient and healthy control fibroblasts, probably due to the limited number of
healthy control samples analyzed. DEGs with higher expression in all PBD-ZSD
patient iPSCs relative to healthy control iPSCs are highly enriched for
mitochondria genes, especially genes involved in mitochondria β-oxidation and
metabolism of β-oxidation by-products, ketone bodies. ACAT1, encoding the
acetyl-Coenzyme A acetyltransferases 1, and CPT1A, encoding the carnitine
palmitoyltransferase 1A, are involved in the mitochondria β-oxidation. Ketone
93
bodies are synthesized with excess acetyl-CoA produced in the β-oxidation as
storage of energy source.
The PEX gene defects in PBD-ZSD patient cells result in reduced or
absence of peroxisome biogenesis and peroxisomal β-oxidation activity.
Correspondingly, we observed increased expression of genes involved in
mitochondria β-oxidation and ketone body metabolism in PBD-ZSD patient
iPSCs. This indicates a rise of active mitochondria lipid metabolism to
compensate the loss of peroxisomal counterparts. This also explains the low
levels of VLCFA in PBD-ZSD patient iPSCs observed in the lipid profiling (Figure
4.1C). Our observations support the current theory of mitochondria and
peroxisome dynamics which consider them as metabolically linked organelles
with cross-talking to keep a balance in lipid homeostasis (Schrader and Yoon
2007). So far, this compensation effect of mitochondria lipid metabolism is only
observed in the induced pluripotent cells but not terminal differentiated fibroblast
cells. Human embryonic, adult and precursor stem cells have mitochondrial
properties different from somatic cell, including their perinuclear localization,
reduced number, oxygen consumption and ATP content (Lonergan, Bavister et al.
2007). It has been validated that human iPSCs have mitochondrial features
similar to those of hESCs (Armstrong, Tilgner et al. 2010). However, the lipid
metabolic status of iPSC/hESC mitochondria has not been evaluated and our
observation of increased gene expression in mitochondrial lipid metabolism may
be associated with the rejuvenated mitochondria energy capacity after induced
pluripotency (Armstrong, Tilgner et al. 2010; Suhr, Chang et al. 2010).
94
Furthermore, mitochondrial abnormalities was described in the original
report of the most severe form of PBD-ZSD, the Zellweger syndrome (cerebro-
hepato-renal syndrome), as a defect in electron transport prior to the
cytochromes (Goldfischer, Moore et al. 1973). Pex5-deficient mice show
embryonic lethality and prominent ultrastructural abnormalities of mitochondria,
changes in the expression and activities of mitochondrial respiratory chain
complexes and increase in heterogeneity of mitochondrial compartment in
various organs and cell types, including liver, heart, kidney, muscle cells and
neutrophils (Baumgart, Vanhorebeek et al. 2001). Even though we haven’t
observed differential expression of mitochondria genes related to the respiratory
chain complexes (other than COX5B), oxidative stress or ATP production in
iPSCs, we need to evaluate these aspects in the differentiated cells derived from
PBD-ZSD patient iPSCs in the future.
In addition to the DEGs involved in mitochondria β-oxidations and ketone
body metabolism, we identified more DEGs involved in lipid metabolism,
including the higher expression of AGK encoded acylglycerol kinase, ELOVL5
encoded ELOVL fatty acid elongase and less expression of ALDH3A2 encoded
aldehyde dehydrogenase in PBD-ZSD patient iPSCs relative to control iPSCs.
Furthermore, we identified DEGs with less expression in PBD-ZSD patient
iPSCs enriched for Wnt signaling pathway, including the secreted signaling
ligand, WNT3, the membrane receptor protein, FZD4, and downstream
regulators, DKK1, SOX17 and PP2A. Overexpression of WNT3 promotes cell
proliferation and increase neuronal differentiation of neural stem cells and neurite
95
outgrowth (Yin, Zhang et al. 2007; David, Canti et al. 2010; Shruster, Eldar-
Finkelman et al. 2011). FZD4 has been shown to be up-regulated during mouse
embryonic stem cell neural differentiation in response to retinoic acid (RA)
(Verani, Cappuccio et al. 2007). Expression of Dkk1 is required for neural
induction in mouse embryonic stem cells (Verani, Cappuccio et al. 2007) and
differentiation into astrocytes, oligodendrocytes and neurons (Ahn, Byun et al.
2008). Down-regulation of Wnt/β-catenin signaling with the addition of Dkk1
during mid and late stages of neurogenesis inhibited neuronal production (Munji,
Choe et al. 2011). PP2A regulates cell growth and division, dorsal development,
midbrain-hindbrain boundary formation and closure of the neural tube (Yang, Wu
et al. 2003). SOX17 regulates oligodendrocyte maturation by promoting the
expression of myelin basic protein (MBP) (Chew, Shen et al. 2011).
All these gene expression changes in Wnt signaling pathway are targeting
towards neural differentiation defects. However, the Wnt signaling pathway is a
complicated and precisely controlled regulatory network. The results of Wnt
signaling gene regulation may be variable corresponding to specific cell types
and timing. Therefore, the effects of these DEGs need to be carefully revisited
and evaluated during the course of differentiation in vitro or development in vivo.
The PBD-ZSD patients with the homozygous PEX1 I700fs framshift mutation, a
change leading to premature truncation and rapid degradation of PEX1 protein,
correlate with the severe Zellweger syndrome phenotypes (Maxwell, Nelson et al.
1999). In patient iPSCs with this homozygous PEX1 frameshift mutation, we
observed significantly less PEX1 gene expression (>2.4-fold, FDR=0.0001),
96
which is consistent with the absence of peroxisomal activities in the patients
(Maxwell, Nelson et al. 1999).
Furthermore, we observed many DEGs between all PBD-ZSD patient and
healthy control iPSCs involved in oxidative stress and inflammations, indicating
the shared pathology of the Zellweger Syndrome spectrum. Many DEGs between
CCALD and control fibroblasts were previously reported to be associated with the
site of biopsy (Rinn, Bondre et al. 2006). This is reasonable given that the patient
and control fibroblasts were acquired from different institutions even though all
biopsies involved the upper limbs of donors. We sought to determine if there was
enrichment for functional categories or biological processes in the DEGs,
keeping in mind the limitations of using cultured cells to study complex diseases
involving interactions been multiple organ systems. Only very broad functional
categories or KEGG pathways were highlighted in these analyses, with none of
them showing a direct relation to disease.
Since there are likely to be gaps in public databases of processes relevant
to peroxisome biology and rare disease pathogenesis, like XALD, we conducted
a manual inspection of gene annotations provided by the DAVID bioinformatics
resource and found one peroxisomal enzyme, alkyl-dihydroxyacetonephosphate-
synthase (AGPS), was up-regulated at expression level in CCALD patient
fibroblasts. AGPS is a rate limiting enzyme in plasmalogen synthesis, which is
imported to peroxisome through PTS2 pathways using the PEX7 shuttle protein
(Grimm, Kuchenbecker et al. 2011). Mutations in AGPS cause cataracts and
male sterility in mice (Liegel, Chang et al. 2011) and RCDP type 2 and 3 in
97
human (Itzkovitz, Jiralerspong et al. 2012). Therefore, the elevated expression of
AGPS is not a direct result of ABCD1 mutation in CCALD patient fibroblast, but
an indirect response to compensate the lack of resource for plasmalogen
synthesis. It also explains the similar plasmalogen levels in both patient and
control fibroblasts.
We also identified two DEGs involved in the Lands’ cycle for
glycerophospholipid remodeling, higher expression of deacylation enzyme
LYPLA1 and lower expression of reacylation enzyme MBOAT7, in CCALD patient
relative to control fibroblasts (Shindou and Shimizu 2009). In addition, two DEGs
were involved in sphingosine metabolism with SMPD1, which converts
sphingomyelin to ceramide, being less expressed and SGPL1, which degrades
sphingosine-1-phosphate, being more expressed, in patient relative to control
fibroblasts (Pyne and Pyne 2000). These can be explained as indirect response
to the accumulation of VLCFA in CCALD patient fibroblasts. However, SMPD1 is
also associated with the site of biopsy.
We also found multiple DEGs involved in immune related processes, but
only two (CBLB and RAB27A) of these genes were not associated with the site of
biopsy. CBLB plays a critical role in antigen-induced immune tolerance and Cblb-
deficient mice immunized with myelin basic protein are more susceptible to
experimental autoimmune encephalomyelitis (EAE), a model for multiple
sclerosis (Chiang, Kole et al. 2000; Gruber, Hermann-Kleiter et al. 2009).
RAB27A mutations can lead to an uncontrolled T lymphocyte and macrophage
activation syndrome in humans with some individuals showing possible leukocyte
98
brain infiltration (Menasche, Pastural et al. 2000). In one Saudi Arabian kindred,
RAB27 mutations were associated with immunodeficiency and progressive
demyelination of brain white matter (de Saint Basile 2007).
The DEGS found in patient and controls iPSCs did not overlap with those
found in fibroblasts and instead were consistent with several leading hypotheses
regarding X-ALD pathogenesis. In particular, we highlight the reduced expression
of PEX11B, a major controller of peroxisome proliferation and neuroinflammatory
genes. Human has 3 PEX11 isoforms, PEX11α, PEX11β, and PEX11γ (Abe and
Fujiki 1998; Schrader, Reuber et al. 1998). PEX11β protein is involved in
peroxisome elongation and constriction elongation (Fagarasanu, Fagarasanu et
al. 2007; Delille, Dodt et al. 2011; Schrader, Bonekamp et al. 2011). There has
been no report of PBD patients with PEX11B mutations until earlier this year that
a novel defect of peroxisome division caused by a homozygous no-sense
mutation in the PEX11B gene in a 26-year old Dutch man (Ebberink, Koster et al.
2012; Thoms and Gartner 2012). This patient presents with atypical symptoms
for PBD, including congenital cataracts at newborn, mild intellectual disability,
progressive hearing loss, sensory nerve involvement, gastrointestinal problems
and recurrent migraine-like episodes. All the standard biochemical peroxisomal
parameters were normal for this individual, except slightly elevated ratio of
C26:0/C22:0 during severe migraine-like episodes. Cultured patient fibroblasts
had peroxisomes variable in size and number and catalase was mis-located in
cytosol instead of peroxisomes in 10% of cultured patient fibroblasts at 37°C and
90% of cells at 40°C (Ebberink, Koster et al. 2012). The identification of the first
99
PEX11β patient extended the spectrum of PBD genotypes and phenotypes.
Pex11B
null mice show several pathologic features, including neuronal migration
defects, enhanced neuronal apoptosis, developmental delay, hypotonia, and
neonatal lethality (Li, Baumgart et al. 2002). Despite these severe phenotypes,
Pex11B null mice displays only mild defects in peroxisomal fatty acid beta-
oxidation and ether lipid biosynthesis (Li, Baumgart et al. 2002). Intriguingly, the
deletion of a single Pex11B allele leads to a slightly increased number of
peroxisomes, levels of oxidative stress in brain tissue, and neuronal cell death in
mice (Ahlemeyer, Gottwald et al. 2012).
ULK1, which plays a critical role in the autophagy-mediated turnover of
peroxisomes in yeast (Chen and White 2011), is lower expressed in CCALD
patient iPSCs. Thus, it is tempting to speculate that if nervous and immune
system cells of ABCD1 mutation carriers indeed have aberrant PEX11B and
ULK1 activity, this could lead to differences in peroxisome abundance that
increase reactive oxygen species (ROS) levels and promote X-ALD
pathogenesis. Of course, it remains to be determined if ABCD1 mutation carriers
have abnormal peroxisome abundance in their cells and tissues. Even, if this is
true, this would not prove that causes increased ROS levels or is the results of
the differential activity of the highlighted genes.
In a similar vein, the increased NAAA, THBS1, BSG (aka CD147 aka
EMMPRIN), and NOTCH1 gene expression in patient relative to control iPSCs is
supportive of hypotheses regarding a predisposition to inflammatory processes in
the brains of XALD patients that can lead to devastating autoimmune responses.
100
NAAA hydrolyzes palmitoylethanolamide (PEA), a naturally occurring lipid amide
that, when administered as a drug, inhibits inflammatory responses (Solorzano,
Zhu et al. 2009). In principle, increasing leukocyte NAAA levels could reduce
PEA levels and promote inflammation. In fact, a chemical inhibitor of NAAA
function attenuates inflammation and tissue damage and improves recovery of
motor function in mice with spinal cord trauma (Solorzano, Zhu et al. 2009).
Intriguingly, CD200 has been proposed to play a role in the immune privileged
status of the CNS with CD200-mediated immune suppression occurs via neuron-
microglial as well as glial-glial interactions in inflammatory conditions (Koning,
Swaab et al. 2009). THBS1 is linked to neuroinflammatory processes involving
astrocyte and microglia through its role in processing and activating the TGF-β
ligand (Lanz, Ding et al. 2010) and is also implicated in responses to oxidative
stress (Ning, Sarracino et al. 2011). Likewise, Notch1 is involved in microglial
associated inflammation (Wei, Chigurupati et al. 2011). Also of relevance are
emerging reports that BSG acts a master regulator of matrix metalloproteinases
(MMP) implicated in most diseases involving neuroinflammation and thus has
been proposed to play a role in the immune-privileged status of the CNS
(Agrawal and Yong 2011).
Furthermore, we noticed that four DEGs, BSG, FOLR1, GRYGD and
RRP7A, are shared among PBD and CCALD patient iPSCs, indicating their
common pathological functions in peroxisomal disorders including inflammation
and oxidative stress.
101
According to the DNA methylation profiles of PBD patient, CCALD patient
and healthy control fibroblasts and iPSCs, we didn’t observe any robust DMLs
between patient and healthy control cells that are associated to differentially
expressed genes. Therefore, we didn’t observe evidence for epigenetic
regulation in the peroxisomal disorders based on our results.
Although all the patient iPSCs are similar to the healthy control iPSCs in
the morphology, stem cell marker expression, epigenetic status and pluripotency,
DEGs revealed transcriptional difference between the patient and control iPSCs.
These variations may be pre-disposed in germ cells and gametes and manifest
later during embryonic development and terminal differentiation. Therefore, we
should monitor and evaluate the differential gene expression and their functional
consequence during the course of development in patient organs and tissues,
especially the differentiation of central nervous system (CNS) cell lineages. This
will provide potential early diagnosis and treatment for patients with peroxisomal
disorders even before they were born. However, this task is hindered by the lack
of suitable experimental models and complex non-autonomous effects in an
organism system. With the derived iPSCs from patient and control somatic cells,
we will be able to perform direct in vitro differentiation and study the defects in a
specific cell types, like the neurons and oligodendrocytes, during the complete
differentiation process.
102
CHAPTER 5
DIFFERENTIATION AND CHARACTERIZATION OF
PBD PATIENT CNS CELLS
5.1 Abstract
Peroxisomes play important roles in central nervous system development
and function. Dysfunction of peroxisomes in brain causes severe myelin
abnormalities, neural inflammation, neuronal migration defects, and neural
differentiation defects. Here, we confirmed that PBD-ZSD patient iPSCs and
healthy control iPSCs differentiate into neural progenitors with variable efficiency
regardless of the mutant genotypes. Patient and control derived neural
progenitors and motor neuron progenitors contain similar levels of VLCFA and
plasmalogen. Patient and control neural progenitors further differentiate into
motor neurons and oligodendrocytes. However, oligodendrocytes derived from
PBD-ZSD patients with PEX1 missense mutations show defective morphologies,
including poor branching.
5.2 Introduction
Peroxisomal disorders exhibit severe neurological dysfunction including
neuronal migration or positioning disturbances, neuronal differentiation defects,
myelin abnormalities, and post-developmental neuronal degeneration. Migration
defects appear in all CNS neuronal classes, but are more prominent in the
103
Purkinje cells and granule cells destined for the outer layer cortex and cerebellar
white matter (Evrard, Caviness et al. 1978; Powers and Moser 1998).
Cultured Pex2
-/-
mouse Purkinje cells appear to have defective
differentiation with poor and irregular dendrite branching (Faust, Banka et al.
2005). Abnormalities in myelin can be divided into three basic types: (1)
hypomyelination, reduction in myelin volume or myelin-staining; (2)
noninflammatory dysmyelination, biochemically abnormal myelin or genetically
defective myelin-forming cells; (3) inflammatory demyelination, inflammatory and
immune-mediated destruction of myelin. All three types of myelin abnormality
have been observed in peroxisomal disorders. Post-developmental neuronal
degeneration has been observed in certain types of peroxisomal disorders,
including the myeloneuropathy of AMN and cerebellar atrophy in RCDP
(Raymond 2009). In the CNS, peroxisomes are more abundant in differentiating
neurons than in mature neurons and are accumulated at axon terminals and
dendrites (McKenna, Arnold et al. 1976; Arnold and Holtzman 1978), indicating
the important role of peroxisomal function in central nervous system development.
Our global gene expression analysis identified less expression of Wnt
signaling genes in the PBD-ZSD patient iPSCs. These down regulated genes in
PBD-ZSD iPSCs are involved in CNS cell proliferation, differentiation and
maturation, indicating pre-disposed abnormalities at the gene expression level at
early stage of embryo development. It may result in defects in CNS cell lineages
derived from patient iPSCs.
104
Animal models of PBD-ZSD have been generated by target inactivation of
Pex2 (Faust and Hatten 1997; Faust, Su et al. 2001), Pex5 (Baes, Gressens et al.
1997; Gressens, Baes et al. 2000) and Pex13 (Maxwell, Bjorkman et al. 2003).
These knockout mouse models have intrauterine growth retardation and severe
hypotonia at birth, and they died within 72 hours. The peroxisome-deficient
fetuses have neuronal migration defects in neocortex, delayed neuronal
maturation, increased apoptosis in the cortical plate, and cerebellar abnormalities
with reduced Purkinje cell development. Pex7 knockout mice (Brites, Motley et al.
2003) have intrauterine growth retardation, severe hypotonia, delayed
ossification of distal bones, dwarfism, and delayed neuronal migration.
Inactivation of Pex11α in mice does not result in noticeable phenotypic
abnormalities (Li, Baumgart et al. 2002). However, complete inactivation of
Pex11β in mice is neonatal lethal, decreases the number of peroxisomes in brain,
impairs neuronal migration, and enhances neuronal apoptosis, developmental
retardation and hypotonia without abrogating peroxisome function (Li, Baumgart
et al. 2002). Furthermore, deletion of a single allele of Pex11β gene is sufficient
to cause oxidative stress and neuronal death, but slightly increases the number
of peroxisomes in brain (Ahlemeyer, Gottwald et al. 2012). Mice carrying null
Abcd2 alleles alone or in concert with Abcd1 null alleles primarily show spinal
axonal degeneration, but limited cerebellar involvement (Pujol, Ferrer et al. 2004;
Ferrer, Kapfhammer et al. 2005). Nevertheless, the neonatal death in many of
these mouse models and the contribution of peroxisomal defects in
105
extraneuronal tissues make it difficult to study the cell autonomous contribution of
peroxisomal dysfunctions in CNS development.
Selective rescue of peroxisomes in liver or brain by tissue-selective
overexpression of Pex5p demonstrates the significance of peroxisomal
metabolism in brain and extraneuronal tissues to the normal development of
mouse neocortex (Janssen, Gressens et al. 2003). Tissue-specific ablation of
functional peroxisomes using Pex5 conditional knockout mice demonstrates that
absence of peroxisomal function in oligodendrocytes but not neurons or
astrocytes has major impact on axonal integrity and neurological functioning
(Kassmann, Lappe-Siefke et al. 2007; Bottelbergs, Verheijden et al. 2010;
Bottelbergs, Verheijden et al. 2012). Peroxisome-deficiency in oligodendrocytes
(Kassmann, Lappe-Siefke et al. 2007), but not the defect in ether lipid synthesis
(Bottelbergs, Verheijden et al. 2012), is causally involved in axonal loss and
neuroinflammation. (Baes and Van Veldhoven 2006; Wanders and Waterham
2006)
In vivo studies in Pex knockout mice demonstrated the important role of
peroxisomes in CNS development, including neuronal migration, proliferation,
differentiation, and survival. However, it is difficult to elucidate the cell
autonomous effect in the CNS cells in a model system with multi-cellular and
multi-organ peroxisome dysfunction. To solve this problem, in vitro culture of
granule neuron and Purkinje cells isolated from Pex2-/- mice have been utilized
as simplified models to study the intrinsic role of CNS peroxisome in neuronal
migration and differentiation. There are several limitations in the current in vitro
106
study of CNS peroxisome. First, the neuronal cells were isolated from animals at
a later point during development, so the pre-disposed non-cell-autonomous
effects cannot be excluded. Second, the isolated neuronal cells are terminal
differentiated cells, which cannot be used to monitor the peroxisome function
during the complete CNS development. Third, the lack of mouse models with
common patient PEX mutations limited the application in drug screening and
testing.
Human ES cells and iPS cells have the potential to generate virtually any
differentiated cell types. With the advance of stem cell technology and knowledge
of embryonic development, hESCs and iPSCs can be efficiently induced into
endoderm, mesoderm, or ectoderm, and further directed to differentiate into
multipotent progenitor cells and unipotent mature cells. It is well established that
hESCs can be induced into expandable, trilineage neural progenitors (Zhang,
Wernig et al. 2001; Xia and Zhang 2009), further directed into motor neuron
progenitors/oligodendrocyte progenitors, and terminally differentiated into mature
motor neurons and oligodendrocytes via transplantation or in vitro culture (Izrael,
Zhang et al. 2007; Hu, Du et al. 2009; Hu and Zhang 2009; Karumbayaram,
Novitch et al. 2009; Sharp, Frame et al. 2010).
With our established PBD and CCALD patient iPSC resources and up-to-
date protocols for in vitro neuronal differentiation of hESCs, we differentiated both
patient and healthy control iPSCs into rosette-structured neural progenitor cells,
further differentiated them into motor neuron progenitor/oligodendrocyte
progenitor cells in the presence of epidermal growth factor (EGF) and retinoic
107
acid (RA), and terminally differentiated them into mature motor neurons and
oligodendrocytes (Figure 5.1). Differentiated cell lineages were characterized by
qRT-PCR, immuno-cytochemistry, and their electrophysiological features. During
in vitro differentiation, we evaluated the morphology of patient and control
derived CNS cells, and enriched neural progenitor and motor neuron progenitor
cells by their unique morphologies (neural rosette and neural sphere) for lipid
profiling. Oligodendrocyte progenitor cells and mature oligodendrocytes can be
purified using surface markers, PDGFRα and O4, for future lipid profiling and
global gene expression analysis.
108
Figure 5.1: Scheme of the neural progenitor (NP), motor neuron (MN) and oligodendrocyte (OD) in vitro
differentiation from human induced pluripotent stem cells (iPSCs).
109
5.3 Results
5.3.1 Variable neural differentiation potency of iPS cells
In the comparative global gene expression analysis of patient and control
iPSCs, we identified lower expression of genes enriched in the Wnt signaling
pathways, including WNT3, FZD4, DKK1, SOX17 and PP2A. WNT3 regulates
cell proliferation and neuronal differentiation of neural stem cells and neurite
outgrowth (Yin, Zhang et al. 2007; David, Canti et al. 2010; Shruster, Eldar-
Finkelman et al. 2011). FZD4 is involved in mouse embryonic stem cell neural
differentiation in response to retinoic acid (RA) (Verani, Cappuccio et al. 2007).
Expression of Dkk1 is required for neural induction in mouse embryonic stem
cells (Verani, Cappuccio et al. 2007) and differentiation into astrocytes,
oligodendrocytes and neurons (Ahn, Byun et al. 2008). PP2A regulates cell
growth and division, dorsal development, midbrain-hindbrain boundary formation
and closure of the neural tube (Yang, Wu et al. 2003). SOX17 regulates
oligodendrocyte maturation by promoting the expression of myelin basic protein
(MBP) (Chew, Shen et al. 2011).
Dysregulation of these Wnt signaling genes might cause neural
differentiation defects. Therefore, it is necessary to test if and how efficiently the
patient and control iPSCs may be differentiated to functional cells of CNS
lineages. In this neuronal functional study, we have focused on the comparison of
PBD-ZSD patient iPSCs and healthy control iPSCs.
First, we tested whether the patient iPSCs can differentiate into neural
progenitor cells as well as the healthy control iPSCs. IPS cells were
110
enzymatically detached and suspended in iPSC medium for 4 days to form EBs
and in neural induction medium (NIM) containing N2 (Invitrogen) for 3 more days
to induce neural differentiation. EBs were plated onto Matrigel® at day 7 and
maintained in the same medium for neural differentiation. Small columnar-like
neural progenitors (NP) started to appear at day 10 and formed neural tube-like
rosettes at day 15. Immunostaining and fluorescence-activated cell sorting
(FACS) indicated that at day 15, >90% of the columnar cells in neural rosettes
uniformly expressed neural progenitor marker, PAX6. Many cells in neural
rosettes are also positive for NESTIN and TuJ1. (Figure 5.2) Each attached EBs
either formed homogeneous neural rosettes or flat non-neural cells.
Neural progenitor inductions were performed in 4 PBD_PEX1fs iPSCs, 7
PBD_PEX1ms iPSCs, 1 PBD_PEX10 iPSC, 3 PBD_PEX12 iPSCs, 2
PBD_PEX26 iPSCs and 6 healthy control iPSCs and repeated at least 3 times in
each iPSC. For each differentiation experiment, the total number of EBs attached
and the number of EBs forming neural rosettes were counted and recorded to
calculate the percentage of EBs forming neural rosettes, which is used to
compare the neural differentiation efficiency. All the PBD-ZSD patient and healthy
control iPSCs formed rosette structured neural progenitors, but with variable
efficiency regardless of the mutant genotypes (Figure 5.3A). It has been noticed
that most healthy control iPSCs maintained consistent neural differentiation
potency after extensive passaging (more than 75 passages). However, the
patient iPSCs carrying homozygous PEX1 I700fs frameshift mutations showed
declined neural differentiation potency in higher passage (Figure 5.3B and C).
111
Whether this is due to the accumulative toxicity caused by peroxisomal defects or
random factors, it will be determined by complementation assays to examine the
neural differential potency in patient iPSCs with PEX1 over-expression or control
iPSCs with PEX1 knockdown.
Neural progenitor cells derived from both patient and control iPSCs were
manually collected based on their distinguished rosette structures and lysed for
lipid profiling analysis. NP cells derived from most PBD-ZSD patient iPSCs have
similarly low VLCFA levels as those derived from healthy control iPSCs, except
NP cells derived from two patient iPSCs with homozygous PEX1 ms mutations
(2.5~3.5-fold increase) and one with PEX10 mutations (1.5-fold increase) (Figure
5.4A). The plasmalogen levels are highly variable among all the NP cells without
significant difference between patient and control cells (Figure 5.4B). Unlike
iPSCs, neural rosettes contain a heterogeneous population of multipotent cells
towards different neural lineages. Therefore, the variance in VLCFA and
plasmalogen levels may be results of in-population difference of cell
compositions. To get an accurate characterization of the patient derived CNS
lineages, it is necessary to obtain a homogeneous population of terminally
differentiated cells, like certain types of neurons and glia cells, for evaluation.
112
Figure 5.2: Induction of neural progenitors from representative PBD-ZSD
iPSCs.
Expression of human neural progenitor (NP) markers, PAX6, NESTIN, and TuJ1.
Red and green colors represent NP marker staining; blue color represents DAPI
nuclear counterstaining.
113
Figure 5.3: Variable neural differentiation potency of PBD-ZSD patient and
healthy control iPSCs.
(A) PBD-ZSD patient and healthy control iPSCs have variable neural
differentiation potency regardless of the mutant genotype. (B) Representative
healthy control iPSC maintain consistent neural differentiation potency along
passaging. (C) Representative PBD-ZSD patient iPSC with PEX1 fs mutation
showed decreased neural differentiation potency along passaging.
A
114
Figure 5.3, Continued
B
C
115
Figure 5.4: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control iPSC derived
neural progenitor cells (NPC).
(A) Relative VLCFA levels in PBD-ZSD patient and healthy control iPSC derived
NP, as represented by the C26:0/C22:0 ratio; (B) Relative plasmalogen levels in
patient and healthy control iPSC derived NP, as represented by the total
phosphatidyl ethanolamine (PE) plasmalogen/C20:0 ratio.
A B
116
5.3.2 Differentiation and characterization of motor neurons
With up-to-date protocols for in vitro neural differentiation, hESC and
iPSCs can be terminally differentiated into motor neurons with high efficiency (up
to 50%) (Hu and Zhang 2009) and enriched after infection with HB-9 promoter
driven-GFP. For motor neuron differentiation, NP cells at day 10 of neural
induction were cultured in NIM medium supplemented with 1µM retinoic acid (RA)
for caudalization for 5 days. Then, rosette structured NP cells were gently blown
off, triturated, and suspended to form motor neuron progenitors (MNP)/neural
spheres (NS) in neural differentiation medium (NDM) containing N2, B27
(Invitrogen) and supplemented with 1µM RA and 100ng/ml sonic hedgehog (SHH)
until day 28. For terminal motor neuron differentiation, MNPs were triturated and
attached to laminin coated cell culture plates and maintained in NDM medium
supplemented with cAMP, ascorbic acid and neurotrophic factors, BDNF, GDNF
and IGF1, for up to 7 weeks (Hu and Zhang 2009; Xia and Zhang 2009).
It was confirmed that both patient and control iPSCs can be differentiated
into OLIG2 and NKX2.2 positive MNPs, and terminally differentiated into HB9
and TuJ1 double positive mature MNs (Figure 5.5). We didn’t observe any
obvious morphological difference between the patient and control derived MNPs
or MNs.
Furthermore, neurons derived from both patient and control iPSCs display
typical neuron electrophysiological characters. Current voltage curve shows that
the tested cell has inward sodium flow and outward potassium flow indicating the
presence of active sodium channels and potassium channels as those on
117
neurons (Figure 5.6A). Action potential curve indicates this cell has sodium and
potassium channels responding to electric stimulation as the response of typical
neurons (Figure 5.6B). However, no spontaneous action potential was detected
in control or patient motor neurons.
MNP cells derived from both patient and control iPSCs were lysed for lipid
profiling analysis and showed similar, but variable levels of VLCFA (Figure 5.7A)
and plasmalogen (Figure 5.7B). Even though, the efficiency of motor neuron
differentiation is high in most iPSCs, the transfection efficiency is too low to
collect enough GFP positive MNs for lipid analysis. We confirmed that PBD-ZSD
patient iPSCs as well as control iPSCs can differentiate into active motor neurons
with prototypical electrophysiological features. However, we found no significant
difference in the VLCFA or plasmalogen levels in the patient and control MNPs.
118
Figure 5.5: Differentiation and maturation of representative PBD-ZSD
patient iPSC-derived motor neuron progenitors (MNP) and motor neurons
(MN).
Expression of MNP markers, OLIG2 and NKX2.2, and MN markers, HB9 and
TuJ1 represented by Red or Green colors. Blue color represents DAPI nuclear
counterstaining.
119
Figure 5.6: Neurons derived from representative PBD-ZSD patient iPSCs
display typical neuron electrophysiological characteristics.
Whole cell patch clamp recording from candidate motor neurons under terminal
motor neuron differentiation condition. (A) Current voltage curves recorded by
voltage clamp; (B) Action potential curves recorded by current clamp.
120
Figure 5.7: Comparison of very long chain fatty acid (VLCFA) and
plasmalogen levels in PBD-ZSD patient and healthy control iPSC derived
motor neuron progenitors (MNP).
(A) Relative VLCFA levels in PBD-ZSD patient and healthy control iPSC derived
MNP, as represented by the C26:0/C22:0 ratio; (B) Relative plasmalogen levels
in patient and healthy control iPSC derived MNP, as represented by the total
phosphatidyl ethanolamine (PE) plasmalogen/C20:0 ratio. The measurements
were performed in multiple iPSCs with the same genetic background.
A B
121
5.3.3 Differentiation and characterization of oligodendrocytes
Peroxisomal biogenesis defects cause accumulation of VLCFA as well as
deficiency of plasmalogen in brain and other tissues, which causes oxidative
stress, inflammation, myelin abnormality and axonopathy. Plasmalogens play
essential roles in maintaining membrane fluidity and protecting neurons from
reactive oxygen species and other damages. In human, up to 70% of myelin
sheath ethanolamine glycerophospholipids are made up of plasmalogens
(Farooqui and Horrocks 2001). Oligodendrocytes are the only myelin-forming
cells in CNS. Peroxisomes in oligodendrocytes provide a critical phospholipid
component of myelin sheath, plasmalogens, and provide protections against
axon degeneration and neuroinflammation. Selective inactivation of peroxisome
biogenesis in oligodendrocytes leads to axonal loss and neuroinflammation and
results in ataxia, tremor and premature death in mutant mice (Kassmann, Lappe-
Siefke et al. 2007). However, selective inactivation of peroxisome function in
neurons or astrocytes doesn’t cause metabolic abnormalities or axonal damage
(Bottelbergs, Verheijden et al. 2010). Therefore, peroxisomal functions in
oligodendrocytes play fundamental protection roles in CNS. Furthermore,
oligodendrocyte peroxisome dysfunction might be the ‘first hit’ in the brain
pathology of peroxisomal biogenesis disorders. An in vitro model of PBD patient
oligodendrocytes will be crucial for understanding the pathomechanisms and
testing new drugs targeting brain defects in PBD.
Additionally, we identified lower expression of SOX17 in PBD-ZSD patient
iPSCs. SOX17 encodes a member of the SOX (SRY-related HMG-box) family of
122
transcription factors involved in the regulation of embryonic development and cell
fate determination. It has been shown that Sox17 regulates oligodendrocyte
maturation by promoting the expression of myelin basic protein (MBP) in rodents
(Chew, Shen et al. 2011). Therefore, the effect of low expression level of SOX17
on oligodendrocyte maturation can be tested by in vitro oligodendrocyte
differentiation of patient and control iPSCs.
For oligodendrocyte differentiation, we focused on PBD-ZSD patient
iPSCs with PEX1 mutations as well as healthy control iPSCs. IPSCs were
enzymatically detached and resuspended in transition medium (TM) containing
50% iPSC medium and 50% glial restrictive medium (GRM) with 5 ng/ml FGF2
and 20 ng/ml EGF in low adherent plates for 2 days. On day 3, EBs were
switched to GRM supplemented with 20 ng/ml EGF and 5 µM RA for 8 days with
medium changed every day. Yellow color neural spheres will start to appear
during the RA treatment. On day 11, the neural spheres are manually selected,
cut to small pieces and maintained in GRM supplemented with 20 ng/ml EGF. On
day 28, the neural spheres were cut to small pieces and attached to Matrigel® in
the same medium. Cells started to migrate out of the clusters. After one week,
the attached cells were dissociated and attached to poly-L-ornithine/fibronectin
double coated plates for oligodendrocyte progenitor expansion. Cells were
maintained in GRM supplemented with N2, FGF2 and EGF. For terminal
oligodendrocyte differentiation, cells were maintained in GRM supplemented with
N2, noggin, and reduced FGF2 and EGF for 2-3 days. Then FGF2 and EGF were
completely withdrawn from the medium with the addition of cAMP, ascorbic acid,
123
and neurotrophic factors, IGF, GDNF and CNTF. (Nistor, Totoiu et al. 2005;
Zhang, Izrael et al. 2006; Izrael, Zhang et al. 2007; Hatch, Nistor et al. 2009;
Sharp, Frame et al. 2010)
Patient iPSCs carrying homozygous PEX1 ms mutations were
differentiated into oligodendrocyte progenitors expressing OP specific markers,
PDGFRα and SOX10, identical to OPs derived from healthy control iPSCs
(Figure 5.8). Patient iPSC derived OPs can be expanded and survive as many
passages as the control iPSC derived OPs.
Two weeks after growth factor withdrawal for terminal OD differentiation,
patient and control cells were positive for O4 by live staining, but morphologically
different. Under brightfield microscope, control ODs present as a homogenous
monolayer of cells. Immunofluorescent staining reveals a large number of highly
branched O4 positive cells. Although patient derived cells show O4 positive
staining, the number of O4 positive cells was very limited and poorly branched.
Shortly after growth factor withdrawal, patient derived cells began to detach from
the cell culture plates. Following a further week of culture in the same condition,
healthy control iPSC derived ODs could be maintained as monolayer cells and
began to express mature oligodendrocyte maker, MBP. However, patient iPSC
derived cells were severely detached and tended to form neural spheres, which
were difficult to fix or stain.
PBD-ZSD patient iPSCs with homozygous PEX1 ms mutations and
healthy control iPSCs can be differentiated into morphologically identical
oligodendrocyte progenitors expressing PDGFRα and SOX10. After growth factor
124
withdrawal, large number of healthy control oligodendrocyte progenitors further
differentiated into highly branched mature oligodendrocytes expressing O4 and
MBP. Under the same terminal differentiation culture condition, patient
oligodendrocyte progenitors also started to become mature as indicated by the
O4 positive staining. However, the number of O4-positive cells was limited in the
patient derived cells and appeared poorly branched. Additionally, the patient
derived cells cannot maintain a monolayer of cells attached to plates; instead
they tend to form neural spheres. Extensive complementation assays,
biochemical and global gene expression analysis are required to test the link
between the observed defects and peroxisomal dysfunction.
125
Figure 5.8: Oligodendrocyte progenitors (OP) derived from representative
PBD-ZSD patient and healthy control iPSCs.
Brightfield pictures of OP derived from PBD_PEX1ms1-iPS1 and control2-iPS3,
and expression of OP markers, PDGFRα and SOX10. Red color represents OP
marker staining; blue color represents DAPI nuclear counterstaining.
126
Figure 5.9: Oligodendrocytes (OD) derived from representative PBD-ZSD
patient and healthy control iPSCs.
Brightfield picture of OD derived from PBD_PEX1ms1-iPS1 and control2-iPS3,
and expression of OD markers, O4 and MBP, at both 10× and 20× objective
magnification. Two weeks after growth factor withdrawal for terminal OD
differentiation, cells are positive for O4 live staining. Following a further week of
culture in the same condition, cells derived from control iPSCs were fixed and
then stained positive for MBP. Cells derived from patient iPSCs were severely
detached and unable to fix or stain for MBP. Red and green colors represent OD
marker staining; blue color represents DAPI nuclear counterstaining.
127
5.4 Discussion
All human diseases caused by peroxisome dysfunction and most mouse
models with loss of peroxisome function are characterized by a variety of
neurological abnormalities during or post brain development, underscoring the
importance of peroxisomes in CNS development and maintenance. However, the
importance of peroxisomes in CNS lineage differentiation hasn’t been well
evaluated. Here, we attempted to utilize our established PBD patient and healthy
donor iPSC resources to model the CNS defects in peroxisomal biogenesis
disorders by differentiating iPSCs into CNS lineages, including neural progenitors,
motor neurons and oligodendrocytes. We characterized the neural induction,
neuron differentiation and oligodendrocyte differentiation of PBD-ZSD patient
derived iPS cells.
First, we tested the neural differentiation potency of 17 PBD-ZSD patient
derived iPSCs and 6 healthy control iPSCs. All 23 iPS cell lines tested can be
induced to form typical neural rosette structures representing neural progenitors.
It indicates that loss of peroxisomal functions does not block the development of
nervous system.
Nevertheless, we observed highly variable differentiation potency among
all iPSCs, regardless of the PEX genotypes or the parental fibroblast origins. This
observation is consistent with an investigation of neural differentiation capacity of
12 human iPSC lines established through different methods, and 6 hESC lines,
which have shown that human iPSCs differentiate into functional neural lineages
using the same transcriptional network over the same time course as hESCs, but
128
with increased variability (Hu, Weick et al. 2010). In addition, this variability of
neural differentiation may reflect the somatic memory remaining due to the
incomplete epigenetic reprogramming observed in some of our iPSCs. Although
the in vitro CNS models we are trying to establish will simplify the systematic
analysis of the effects of peroxisomal inactivity, this variability of neural
differentiation brings new difficulties for evaluating the intrinsic effects of
peroxisomes on CNS cells. Therefore, instead of comparing the overall efficiency
of neural differentiation of patient and control iPSCs, we will focus on
characterization of individual neural lineages, including neural progenitors, and
terminally differentiated motor neurons and oligodendrocytes.
Furthermore, we analyzed the lipid contents of patient and control derived
neural progenitors and motor neuron progenitors. We found similar levels of
VLCFA and plasmalogens among all the iPSC derived neural progenitors. The
lipid contents of cultured cells are influenced by the cultured conditions, as the
fibroblasts cultured in regular fibroblast medium have much higher VLCFA levels
than in iPSC medium shown in Chapter 4. Pex7 knockout mice have normal
phytanic acid level when fed with standard diet, but show drastic accumulation of
phytanic acid when fed with phytol, the precursor of phytanic acid (Brites, Mooyer
et al. 2009). In addition, neural progenitors represent a heterogeneous population
of multipotent neuronal cells, even though they express the same protein
markers and the morphologically indistinguishable (Yaworsky and Kappen 1999).
This also points to the importance of obtaining single populations of terminally
differentiated neural types.
129
Second, we aimed to differentiate patient and control iPSCs towards motor
neuron lineage and to characterize their abnormalities caused by peroxisomal
defects. Both patient and control iPSCs can differentiate into HB9 and TuJ1
double positive motor neurons. Electrophysiology study shows that patient and
control iPSC derived motor neurons developed active sodium and potassium
channels that respond to electric stimulation. Furthermore, patient and control
iPSC derived motor neurons show no distinguishable morphological difference
and were well maintained for up to 4 weeks at in vitro culture condition. It
indicates that under directed differentiation culture conditions with limited non-
cell-autonomous influence of peroxisomal dysfunction, highly enriched motor
neurons with PEX mutations do not have obvious defects. Our in vitro motor
neuron model, as well as the mouse model with neural specific peroxisome
inactivation (Bottelbergs, Verheijden et al. 2010), proves that inactivation of
peroxisomal functions in neurons is not the direct cause of neural abnormality in
peroxisomal disorders. However, in vitro derived patient neural cell lineages
could play a role as effectors in a co-culture system for mechanism study or drug
testing. On the other hand, it is possible that more subtle defects may exist but
were not observed due to technological limitations,
Finally, the patient iPSCs with homozygous PEX1 G843D missense
mutations and control iPSCs were directed into glial lineage. We demonstrated
that both patient and control iPSCs differentiate into oligodendrocyte precursors
with identical morphology and further differentiate into O4-positive
oligodendrocytes. We detected highly branched mature oligodendrocytes
130
expressing MBP in the control iPSC derived oligodendrocyte population. In
contrast, the patient oligodendrocyte with homozygous PEX1 G843D missense
mutations were poorly branched and were unable to remain attached under
terminal differentiation conditions. This may correspond to the reduced SOX17
expression in PBD-ZSD patient iPSCs and needs to be further confirmed after
we have the oligodendrocyte precursor and mature oligodendrocytes purified.
It has been well demonstrated the peroxisome-deficiency in
oligodendrocytes directly causes axonal loss and neuroinflammation (Kassmann,
Lappe-Siefke et al. 2007), which indicates that oligodendrocytes are a good
target for therapy. However, there is no characterization for the peroxisome-
deficient oligodendrocyte reported so far. In vitro oligodendrocytes with
peroxisome dysfunction provide a unique opportunity for understanding the
initiation of neuronal defects in PBD patients and a perfect model for drug testing.
Before we go further into applications, it is crucial to verify the direct
association between the observed morphological abnormalities in patient
oligodendrocytes and their PEX mutations by complementation assays. Then we
will move on to study the biochemical features and global gene expression
profiles of the patient and control oligodendrocytes to identify gene regulation
networks associated with the oligodendrocyte specific peroxisomal defects. By
co-culturing the patient oligodendrocytes with neurons, we can study the direct
influence of oligodendrocyte peroxisomal defects on those neurons.
In our on-going drug testing studies, we have demonstrated that some
nonsense suppressor drugs and proteasome inhibitor drugs are effective in
131
rescuing certain peroxisome functions in patient fibroblasts (Dranchak, Di Pietro
et al. 2010). In long term, in vitro oligodendrocyte models can be applied in
testing the potential drug effects in central nervous system.
In vitro differentiated oligodendrocyte progenitors have been transplanted
into the shiverer model of dysmyelination and injury models for integration,
terminal differentiation into oligodendrocytes and myelin formation (Nistor, Totoiu
et al. 2005; Cloutier, Siegenthaler et al. 2006; Sharp, Frame et al. 2010). In the
future, it is possible to utilize these techniques to build mouse models integrated
with PBD patient oligodendrocytes for personalized disease study and drug
testing.
Overall, peroxisomal disorder patient-specific iPSC and CNS model
systems provide a new platform for recapitulating in detail disease initiation and
progression in early embryogenesis and differentiation, for investigating the cell
autonomous role of peroxisomes in peroxisomal disorder in pathophysiology, and
for screening therapeutic agents tailored to a patient’s genotype in cell
populations most affected by disease.
132
CHAPTER 6
CONCLUSIONS AND PERSPECTIVES
Peroxisomal disorders were caused by peroxisome assembly defects or
single peroxisomal enzyme dysfunction, which affect multiple organs, especially
the central nervous system. Zellweger syndrome (ZS), the most severe type of
peroxisomal disorders, has an incidence of 1 in 40,000 individuals; X-linked
adrenoleukodystrophy (XALD), the most common type of peroxisomal disorders,
has an incidence of 1 in 20,000 individuals. Currently, there is no curative therapy
or long-term effective treatment available for peroxisomal disorders. Ongoing
pathomechanism studies and drug tests on peroxisomal disorders are mainly
established on patient primary fibroblasts, which have limited peroxisomal activity
and knockout mouse models, which do not mirror the exact human mutations or
certain aspects of human symptoms. In this thesis, peroxisomal disorder patient-
specific induced pluripotent stem cell (iPSC) model systems were generated as a
new platform for investigating detail disease initiation and progression in early
embryogenesis and differentiation and for screening therapeutic agents tailored
to a patient’s genotype in cell populations most affected by disease.
Characterizations of these patient-specific iPSCs and iPSC derived CNS cell
lineages, including neural progenitors, motor neurons, and oligodendrocytes,
provide a novel perspective that supports leading hypothesis regarding disease
pathogenesis. Oligodendrocyte, the sole source of CNS myelination and a direct
cause of peroxisomal disorder pathogenesis in brain, will be an ideal target for
133
peroxisomal disorder therapy; Patient-specific oligodendrocytes in culture dish,
will be a unique platform for peroxisomal disorder drug screening and testing.
6.1 Reprogramming, pluripotency, and peroxisomal disorders
We have generated iPSC resources for long-term purpose of developing
novel tissue culture models for elucidating the pathogenesis of peroxisomal
disorders and screening for more effective customized drug therapies. Skin
fibroblasts from 9 PBD patients, 2 CCALD patients and 3 healthy donors were
successfully reprogrammed to form iPSCs with the hallmark molecular properties
of pluripotency including the expression of pluripotency genes, the achievement
of global epigenetic reprogramming, and the capability of differentiation into cell
types of all three germ layers. Most patient and control iPSCs can be stably
maintained for extensive passages without losing pluripotency or genome
integrity.
Unlike the parental fibroblasts of peroxisomal disorder patients, which
show elevated VLCFA levels and/or decreased plasmalogen levels, all patient
iPSCs and healthy control iPSCs show no significant difference in their lipid
contents. We postulate that either VLCFAs may not be supplied in the medium or
actively synthesized in human iPSCs, or VLCFAs may be metabolized through
functionally redundant metabolic pathways, for example, β-oxidation of LCFA and
MCFA in mitochondria.
Although peroxisomal disorder patients show elevated VLCFA levels in
their blood and urine and reduced VLCFA catabolic activity in their cultured
134
fibroblasts, their role in disease pathogenesis is still under discussion. Likewise,
the significance of decreased plasmalogen levels in the brain white matter of
CCALD patients also is unclear (Wilson and Sargent 1993; Khan, Singh et al.
2008). Previous research demonstrated that defects in VLCFA catabolic activity
(Heinzer, Watkins et al. 2003) or ether lipid synthesis (Bottelbergs, Verheijden et
al. 2012) were not the causes of common severe syndromes of peroxisomal
disorders, like the activation of the innate immune system and axonal loss in the
central nervous system.
The successful generation of iPSCs from multiple types of peroxisomal
disorder patients carrying a variety of peroxisomal gene mutations indicates that
normal peroxisomal functions are not necessary for reprogramming skin
fibroblasts into iPSCs and maintaining pluripotency. Our peroxisomal disorder
patient and healthy donor iPSC resources provide a novel in vitro model for the
study of detail disease initiation and progression in early embryogenesis and
differentiation, especially the central nervous system development.
6.2 Differential gene expression in PBD and CCALD iPS cells
We studied the global gene expression profiles of PBD-ZSD, CCALD
patient and healthy control iPSCs and identified differentially expressed genes
(DEGs) between patient and control cells. Mitochondria genes involved in β-
oxidation, β-oxidation by-product metabolism and mitochondrial respiratory chain
have higher expression in all PBD-ZSD patient iPSCs relative to healthy control
iPSCs, which could be in response to the loss of β-oxidation activity in
135
peroxisomes and explain the normal VLCFA level in PBD-ZSD patient iPSCs.
Mitochondrial abnormalities and respiratory chain defects have been reported in
PBD-ZSD patients and mouse models (Goldfischer, Moore et al. 1973; Baumgart,
Vanhorebeek et al. 2001). Furthermore, this result supports the existence of
crosstalk between these metabolically linked organelles for keeping the cellular
lipid homeostasis (Schrader and Yoon 2007). It may also indicate a unique
property of mitochondria in pluripotent stem cells.
Wnt signaling pathway genes involved in promoting cell proliferation,
neuronal differentiation, neurite outgrowth and oligodendrocyte maturation have
lower expression in all PBD-ZSD patient iPSCs relative to healthy control iPSCs,
which indicates potential problems in central nervous system differentiation from
patient iPSCs and was further investigated.
We also observed two PEX genes differentially expressed between patient
and control iPSCs. In keeping with prior report (Maxwell, Allen et al. 2002), PEX1
expression is significantly lower in PBD-ZSD patient iPSCs carrying PEX1
frameshift mutation which causes premature truncation and rapid degradation of
PEX1 protein. A novel discovery is the reduced expression of PEX11B, a major
controller of peroxisome proliferation and neuroinflammatory genes, in CCALD
patient iPSCs relative to healthy control iPSCs. PEX11β protein functions in
peroxisome elongation and constriction elongation (Fagarasanu, Fagarasanu et
al. 2007; Delille, Dodt et al. 2011; Schrader, Bonekamp et al. 2011). These
results indicate that single peroxisomal enzyme deficiency in CCALD may be
accompanied by additional defects in peroxisomal proliferation and assembly.
136
Defects in PEX11B gene haven’t been identified in any peroxisomal disorder
patients until earlier this year. A novel defect of peroxisome division due to a
homozygous non-sense mutation in PEX11B was identified in a 26-year old man
with atypical symptoms for a PBD (Ebberink, Koster et al. 2012; Thoms and
Gartner 2012). PEX11β patient has symptoms shown in most PBD patients,
including mild intellectual disability, congenital cataracts at birth and progressive
sensory neuronal hearing loss, as well as atypical symptoms, like the recurrent
migraine-like episodes. All the standard biochemical peroxisomal parameters
were normal in this patient expect slightly elevated VLCFA levels during
migraine-like episodes. Cultured patient fibroblasts have peroxisomes variable in
size and number, indicating defects in peroxisomal division and proliferation. Mis-
located catalase in cytosol was observed in cultured patient fibroblasts with more
mis-localization at higher temperature, indicating defects in peroxisome assembly.
Phenotypes observed in Pex11B knockout mouse models resemble patient
symptoms. Pex11B
null mice show several pathologic features, including
neuronal migration defects, enhanced neuronal apoptosis, developmental delay,
hypotonia, and neonatal lethality (Li, Baumgart et al. 2002). Despite these severe
phenotypes, Pex11B null mice display only mild defects in peroxisomal fatty acid
beta-oxidation and ether lipid biosynthesis (Li, Baumgart et al. 2002). Intriguingly,
the deletion of a single Pex11B allele leads to a slightly increased number of
peroxisomes, levels of oxidative stress in brain tissue, and neuronal cell death in
mice (Ahlemeyer, Gottwald et al. 2012).
137
To determine the influence of reduced PEX11B expression in CCALD, we
need to first confirm the deficiency of PEX11β protein and its involvement in
peroxisomal number regulation in CCALD patient iPSCs and differentiated cell
types. Then we can study the functional involvement of PEX11β in peroxisomal
activity by recovering PEX11B expression in CCALD iPSCs and differentiated
cells or in iPSCs derived from PEX11B patient. Furthermore, we can validate our
results in CCALD mouse models or characterize the Abcd1-Pex11B double-
knockout model.
PBD-ZSD and CCALD patient iPSCs also have DEGs relevant to
neuroinflammation and oxidative stress, which is consistent with major
hypotheses regarding peroxisomal disorder pathogenesis. Furthermore, the
DEGs between patient and control iPSCs were independent of cellular VLCFA
levels, which did not vary significantly according to mutation status.
The highlighted genes provide new leads for pathogenic mechanisms that
can be explored in animal models and human tissue specimens. We suggest that
our iPSC resources will have multiple applications that include assisting efforts to
identify genetic and environmental modifiers and screening for therapeutic
interventions tailored towards affected cell populations and genotypes.
6.3 Influence of copy number variance and epigenetic variance in iPS cells
We have identified de novo copy number variances (CNV) in some patient
and control iPSCs. The observed genomic deletions always affect only one allele
and genomic amplifications always results in duplication in copy number. We
138
have detected reoccurrence of regional duplications of chromosome 12q11.21
and chromosome 20p11.21, which encompass genes involved in embryogenesis,
cellular growth and division and may benefit the growth selection of these iPSCs.
Microarray data shows no corresponding gain of gene expression or change of
DNA methylation in iPSCs with duplications in these two regions. The influence
of CNVs in iPSC differentiation needs to be monitored and evaluated to separate
them from the influence caused by peroxisomal gene mutations.
We extended the multi-platform analysis of CNV regions to all fibroblasts
and iPSCs with global gene expression and DNA methylation assays. In almost
all circumstances, the expression levels of genes with CNV regions were within
range of diploid samples. Although there are several possible explanations for
these observations, this could reflect the effects of selection whereby CNVs are
only tolerated or counter benefit in iPSCs if they involve genomic regions that do
not influence the initiation of reprogramming or maintenance of pluripotency. It is
also possible that epistatic interactions reduce the effects of CNVs on the gene
expression profiles of iPSCs.
Global DNA methylation analyses confirmed epigenetic reprogramming
featuring iPSC-specific methylation and demethylation as compared to parental
fibroblasts. However, this process is not uniformly completed in all iPS cell lines
as incomplete demethylation was observed in several patient and control iPS cell
lines, which reflects retained epigenetic memory of somatic cells (Ohi, Qin et al.
2011). Several genes proximal to the incomplete demethylation regions are
involved in transcription regulation and early development. Even though none of
139
these genes were differentially expressed between iPSCs with complete and
incomplete demethylation, the differential methylation among iPSCs may result in
variance in differentiation potency. Therefore, extra caution and confirmation
steps need to be taken for characterization of disease relevant features of
differentiated cells.
Variances in the copy numbers of specific genomic segments and
epigenetic modifications exist among iPSCs regardless of the genotypes of
peroxisomal genes. This provides an added level of complexity that needs to be
considered when comparing the phenotypes of patient and donor-derived iPS
cells. Some variances were caused by technique limitations. Reprogramming
induced by retrovirus mediated ectopic expression of pluripotency genes results
in retroviral vector integration which increases the genome instability and gene
dysregulation. Chromosome aberrations were also frequently acquired in culture
(Ben-David, Mayshar et al. 2011).
In principle genomic and epigenomic changes that result from the
regoramming process can be minimized by improving reprogramming methods.
For example, alternative gene delivery methods, like non-integrative viruses
(Okita, Nakagawa et al. 2008), plasmids (Yu, Hu et al. 2009), RNA (Warren,
Manos et al. 2010), or microRNA (Anokye-Danso, Trivedi et al. 2011), will avoid
the genomic integration of virus. Furthermore, the inclusion of appropriate small
molecules, like histone deacetylase (HDAC) inhibitors, greatly improves
reprogramming efficiency and facilitates the epigenetic remodeling (Huangfu,
Osafune et al. 2008). Ultimately, strictly chemical-based methods would be ideal
140
as long as they do not involve agents that enhance the mutation rate of the
genome.
For identification of differences truly associated with peroxisomal
mutations, we need to overcome or minimize the random variations among all
iPSCs. Increasing the number of iPSCs for each individual and the number of
parental fibroblasts will increase the statistical power of analysis. Any
phenotypical difference observed between patient and control cells requires
verification in multiple iPSCs with the same genetic background. Furthermore,
the change has to be verified by complementation assays or drug recovery
assays.
6.4 Defects in PBD iPS cell derived CNS cells
After confirmation and characterization of peroxisomal disorder patient
and healthy control iPSCs, we extended our study to the central nervous system
cell lineages. We confirmed that all PBD patient and control iPSCs can
differentiate into neural progenitors, though the differentiation potency is variable
among all iPSCs. The variance in neural differentiation potency has been
observed in iPSCs as well as hESCs, but with increased variability in iPSCs (Hu,
Weick et al. 2010), which may be associated with the memory of somatic cells.
Motor neuron differentiation was also achieved in PBD patient and healthy
control iPSCs. PBD patient and healthy control iPSC derived motor neurons
express motor neuron specific markers and have typical neural electrophysiology.
PBD patient iPSC derived motor neurons have no obvious morphological
141
difference compared to healthy control iPSC derived motor neurons. The in vitro
motor neuron model, as well as the mouse model with neural specific
peroxisome inactivation (Bottelbergs, Verheijden et al. 2010), proves that the
neural abnormalities in peroxisomal disorders are not direct consequence of
peroxisomal defects in neurons, but triggered by the accumulation of toxicity and
inflammation due to peroxisomal defects in oligodendrocytes and other organs or
tissues, especially liver.
We demonstrated that patient iPSCs with homozygous PEX1 G843D
missense mutations and control iPSCs differentiate into oligodendrocyte
precursors with identical morphology and further differentiate into O4 positive
oligodendrocytes. We detected highly branched mature oligodendrocytes
expressing MBP in the control iPSC derived oligodendrocyte population. In
contrast, the patient oligodendrocytes with homozygous PEX1 G843D missense
mutations were poorly branched and were unable to remain attached under
terminal differentiation conditions, which may be directly associated with reduced
peroxisomal activity in these cell and may be caused by the loss of SOX17 gene
regulations towards oligodendrocyte maturation as identified in PBD patient
iPSCs.
Peroxisome-deficiency in oligodendrocytes directly causes axonal loss
and neuroinflammation (Kassmann, Lappe-Siefke et al. 2007), which indicates
that oligodendrocytes are a good target for therapy. In vitro oligodendrocytes with
peroxisome dysfunction provide a unique opportunity for understanding the
initiation of neuronal defects in PBD patients and a perfect model for drug testing.
142
Before we move into the application of our iPS and CNS cell models, it is
crucial to vigorously validate that the observed morphological abnormalities in
patient oligodendrocytes are truly caused by the PEX gene defects. It is also
critical to understand the mechanisms underlying the cellular defects and to
characterize and validate specific markers or features for the development of
drug screening platform. This will be further discussed in the next section.
6.5 Future directions
We have managed to differentiate PBD patient iPSCs into putative
oligodendrocytes expressing protein makers of mature oligodendrocytes. In vitro
derived patient oligodendrocyte culture is a good candidate for studying
peroxisomal disorder pathomechanism in brain and screening therapeutic agents
targeting brain defects in peroxisomal disorders. Although highly enriched, the in
vitro differentiation culture system contains a heterogeneous population of
oligodendrocyte progenitor and mature cells, which increases variances and
difficulties for data interpretation. Therefore, it is crucial to obtain a homogeneous
population of oligodendrocyte progenitors or mature cells. To achieve this goal,
on one hand, the in vitro oligodendrocyte differentiation protocol will be optimized
to increase the differentiation efficiency; on the other hand, oligodendrocyte
progenitors and mature oligodendrocytes will be bound separately to microbeads
conjugated to antibodies against oligodendrocyte progenitor surface marker,
PDGFRα, or mature oligodendrocyte marker, purified using the MSCS® cell
separation system (Miltenyi Biotec), and confirmed by FACS assays.
143
Standard biochemical lipid profile and global gene expression profile
analysis will be performed on purified oligodendrocyte progenitors and mature
oligodendrocytes to identify differences in lipid content and gene expression
between patient and control cells.
Phenotypic differencs between patient and control cells will be validated
by overexpression of PEX genes in patient cells and knock-down of PEX genes
in control cells. Several in vitro and in vivo systems will be established for the
investigation of the functional consequence of oligodendrocyte peroxisomal
defects in CNS, including the in vitro neuron-oligodendrocyte co-culture system,
oligodendrocyte progenitor maturation in brain culture system, and shiverer mice
brain and spinal cord transplantation of in vitro differentiated oligodendrocyte
progenitors (Nistor, Totoiu et al. 2005; Cloutier, Siegenthaler et al. 2006; Sharp,
Frame et al. 2010). Patient iPSCs and oligodendrocytes will be genetically
corrected using zinc-finger nucleases (Urnov, Miller et al. 2005; Lombardo,
Genovese et al. 2007). Their functional corrections will be evaluated in the in
vitro co-culture systems and in vivo transplantation systems, which will be the
first step in the development of gene therapy for correcting neurological defects
of peroxisomal disorders.
In parallel to the study of peroxisomal functions and dysfunctions in the
central nervous system, my colleague, Win-Yan Yik, is making an effort in in vitro
differentiation of mature hepatocytes using our PBD patient and healthy control
iPSC resources and investigation of peroxisomal functions and dysfunctions in
144
liver, another critical organ system in lipid metabolism through peroxisomal
functions and peroxisomal disorder pathology.
We are also engaged in perfecting the iPSC resources by using non-
integration delivery reprogramming methods, including RNA mediated
reprogramming and microRNA mediated reprogramming, and by reprogramming
patient and control fibroblasts expressing green fluorescent protein harboring a
PTS1 targeting sequence (GFP-PTS1 fusion proteins) for more convenient live
cell monitoring of peroxisomal matrix protein import.
With thorough characterization of peroxisomal disorder patient iPSCs and
differentiated cell lineages and the improvement of iPSC resources, the iPS and
CNS cell models of peroxisomal disorders will provide in-depth knowledge for
normal peroxisomal function in early embryogenesis and peroxisomal disorder in
early in-utero onset pathology, which will facilitate the prenatal diagnosis and
treatment of peroxisomal disorders, and provide a unique platform for
personalized therapy development.
145
BIBLIOGRAPHY
Aasen, T., A. Raya, et al. (2008). "Efficient and rapid generation of induced
pluripotent stem cells from human keratinocytes." Nat Biotechnol 26(11):
1276-1284.
Abe, I. and Y. Fujiki (1998). "cDNA cloning and characterization of a constitutively
expressed isoform of the human peroxin Pex11p." Biochem Biophys Res
Commun 252(2): 529-533.
Agrawal, S. M. and V. W. Yong (2011). "The many faces of EMMPRIN - roles in
neuroinflammation." Biochim Biophys Acta 1812(2): 213-219.
Ahlemeyer, B., M. Gottwald, et al. (2012). "Deletion of a single allele of the
Pex11beta gene is sufficient to cause oxidative stress, delayed
differentiation and neuronal death in mouse brain." Dis Model Mech 5(1):
125-140.
Ahn, S. M., K. Byun, et al. (2008). "Olig2-induced neural stem cell differentiation
involves downregulation of Wnt signaling and induction of Dickkopf-1
expression." PLoS One 3(12): e3917.
Ambrosini, A., S. Paul, et al. (2007). "Human subtelomeric duplicon structure and
organization." Genome Biol 8(7): R151.
Amps, K., P. W. Andrews, et al. (2011). "Screening ethnically diverse human
embryonic stem cells identifies a chromosome 20 minimal amplicon
conferring growth advantage." Nat Biotechnol 29(12): 1132-1144.
Anokye-Danso, F., C. M. Trivedi, et al. (2011). "Highly efficient miRNA-mediated
reprogramming of mouse and human somatic cells to pluripotency." Cell
Stem Cell 8(4): 376-388.
Armstrong, L., K. Tilgner, et al. (2010). "Human induced pluripotent stem cell
lines show stress defense mechanisms and mitochondrial regulation
similar to those of human embryonic stem cells." Stem Cells 28(4): 661-
673.
Arnold, G. and E. Holtzman (1978). "Microperoxisomes in the central nervous
system of the postnatal rat." Brain Res 155(1): 1-17.
Arora, K., W. M. Gwinn, et al. (2005). "Extracellular cyclophilins contribute to the
regulation of inflammatory responses." J Immunol 175(1): 517-522.
Asheuer, M., I. Bieche, et al. (2005). "Decreased expression of ABCD4 and BG1
genes early in the pathogenesis of X-linked adrenoleukodystrophy." Hum
Mol Genet 14(10): 1293-1303.
Baes, M., P. Gressens, et al. (1997). "A mouse model for Zellweger syndrome."
Nat Genet 17(1): 49-57.
Baes, M. and P. P. Van Veldhoven (2006). "Generalised and conditional
inactivation of Pex genes in mice." Biochim Biophys Acta 1763(12): 1785-
1793.
Barth, P. G., C. B. Majoie, et al. (2004). "Neuroimaging of peroxisome biogenesis
disorders (Zellweger spectrum) with prolonged survival." Neurology 62(3):
439-444.
146
Baumgart, E., I. Vanhorebeek, et al. (2001). "Mitochondrial alterations caused by
defective peroxisomal biogenesis in a mouse model for Zellweger
syndrome (PEX5 knockout mouse)." Am J Pathol 159(4): 1477-1494.
Ben-David, U., Y. Mayshar, et al. (2011). "Large-scale analysis reveals acquisition
of lineage-specific chromosomal aberrations in human adult stem cells."
Cell Stem Cell 9(2): 97-102.
Berger, J. and J. Gartner (2006). "X-linked adrenoleukodystrophy: clinical,
biochemical and pathogenetic aspects." Biochim Biophys Acta 1763(12):
1721-1732.
Berger, J., A. Pujol, et al. (2010). "Current and future pharmacological treatment
strategies in X-linked adrenoleukodystrophy." Brain Pathol 20(4): 845-856.
Bjorkhem, I., L. Sisfontes, et al. (1986). "Simple diagnosis of the Zellweger
syndrome by gas-liquid chromatography of dimethylacetals." J Lipid Res
27(7): 786-791.
Bosnali, M. and F. Edenhofer (2008). "Generation of transducible versions of
transcription factors Oct4 and Sox2." Biol Chem 389(7): 851-861.
Bottelbergs, A., S. Verheijden, et al. (2010). "Axonal integrity in the absence of
functional peroxisomes from projection neurons and astrocytes." Glia
58(13): 1532-1543.
Bottelbergs, A., S. Verheijden, et al. (2012). "Peroxisome deficiency but not the
defect in ether lipid synthesis causes activation of the innate immune
system and axonal loss in the central nervous system." J
Neuroinflammation 9(1): 61.
Brites, P., P. A. Mooyer, et al. (2009). "Plasmalogens participate in very-long-
chain fatty acid-induced pathology." Brain 132(Pt 2): 482-492.
Brites, P., A. M. Motley, et al. (2003). "Impaired neuronal migration and
endochondral ossification in Pex7 knockout mice: a model for rhizomelic
chondrodysplasia punctata." Hum Mol Genet 12(18): 2255-2267.
Buyse, M., S. V. Sitaraman, et al. (2002). "Luminal leptin enhances CD147/MCT-
1-mediated uptake of butyrate in the human intestinal cell line Caco2-
BBE." J Biol Chem 277(31): 28182-28190.
Carey, B. W., S. Markoulaki, et al. (2009). "Reprogramming of murine and human
somatic cells using a single polycistronic vector." Proc Natl Acad Sci U S A
106(1): 157-162.
Cartier, N. and P. Aubourg (2010). "Hematopoietic stem cell transplantation and
hematopoietic stem cell gene therapy in X-linked adrenoleukodystrophy."
Brain Pathol 20(4): 857-862.
Chen, H. Y. and E. White (2011). "Role of autophagy in cancer prevention."
Cancer Prev Res (Phila) 4(7): 973-983.
Chevillard, G., M. C. Clemencet, et al. (2004). "Targeted disruption of the
peroxisomal thiolase B gene in mouse: a new model to study disorders
related to peroxisomal lipid metabolism." Biochimie 86(11): 849-856.
Chew, L. J., W. Shen, et al. (2011). "SRY-box containing gene 17 regulates the
Wnt/beta-catenin signaling pathway in oligodendrocyte progenitor cells." J
Neurosci 31(39): 13921-13935.
147
Chiang, Y. J., H. K. Kole, et al. (2000). "Cbl-b regulates the CD28 dependence of
T-cell activation." Nature 403(6766): 216-220.
Cloutier, F., M. M. Siegenthaler, et al. (2006). "Transplantation of human
embryonic stem cell-derived oligodendrocyte progenitors into rat spinal
cord injuries does not cause harm." Regen Med 1(4): 469-479.
Corzo, D., W. Gibson, et al. (2002). "Contiguous deletion of the X-linked
adrenoleukodystrophy gene (ABCD1) and DXS1357E: a novel neonatal
phenotype similar to peroxisomal biogenesis disorders." Am J Hum Genet
70(6): 1520-1531.
David, M. D., C. Canti, et al. (2010). "Wnt-3a and Wnt-3 differently stimulate
proliferation and neurogenesis of spinal neural precursors and promote
neurite outgrowth by canonical signaling." J Neurosci Res 88(14): 3011-
3023.
De Duve, C. and P. Baudhuin (1966). "Peroxisomes (microbodies and related
particles)." Physiol Rev 46(2): 323-357.
de Saint Basile, G. (2007). Inherited Hemophagocytic Syndromes. Primary
Immunodeficiency Diseases: A Molecular and Genetic Approach. H. D. S.
Ochs, C. I. E.; Puck, J. M. . New York, Oxford Univ. Press: 578-588.
Dean, M. and T. Annilo (2005). "Evolution of the ATP-binding cassette (ABC)
transporter superfamily in vertebrates." Annu Rev Genomics Hum Genet 6:
123-142.
Delille, H. K., G. Dodt, et al. (2011). "Pex11pbeta-mediated maturation of
peroxisomes." Commun Integr Biol 4(1): 51-54.
Dimos, J. T., K. T. Rodolfa, et al. (2008). "Induced pluripotent stem cells
generated from patients with ALS can be differentiated into motor
neurons." Science 321(5893): 1218-1221.
Dranchak, P. K., E. Di Pietro, et al. (2010). "Nonsense suppressor therapies
rescue peroxisome lipid metabolism and assembly in cells from patients
with specific PEX gene mutations." Journal of Cellular Biochemistry.
Duinsbergen, D., M. Eriksson, et al. (2008). "Induced pluripotency with
endogenous and inducible genes." Exp Cell Res 314(17): 3255-3263.
Ebberink, M. S., J. Koster, et al. (2012). "A novel defect of peroxisome division
due to a homozygous non-sense mutation in the PEX11beta gene." J Med
Genet 49(5): 307-313.
Ebert, A. D., J. Yu, et al. (2009). "Induced pluripotent stem cells from a spinal
muscular atrophy patient." Nature 457(7227): 277-280.
Eminli, S., J. Utikal, et al. (2008). "Reprogramming of neural progenitor cells into
induced pluripotent stem cells in the absence of exogenous Sox2
expression." Stem Cells 26(10): 2467-2474.
Evrard, P., V. S. Caviness, Jr., et al. (1978). "The mechanism of arrest of
neuronal migration in the Zellweger malformation: an hypothesis bases
upon cytoarchitectonic analysis." Acta Neuropathol 41(2): 109-117.
Fagarasanu, A., M. Fagarasanu, et al. (2007). "Maintaining peroxisome
populations: a story of division and inheritance." Annu Rev Cell Dev Biol
23: 321-344.
148
Farooqui, A. A. and L. A. Horrocks (2001). "Plasmalogens: workhorse lipids of
membranes in normal and injured neurons and glia." Neuroscientist 7(3):
232-245.
Faust, P. L., D. Banka, et al. (2005). "Peroxisome biogenesis disorders: the role
of peroxisomes and metabolic dysfunction in developing brain." J Inherit
Metab Dis 28(3): 369-383.
Faust, P. L. and M. E. Hatten (1997). "Targeted deletion of the PEX2 peroxisome
assembly gene in mice provides a model for Zellweger syndrome, a
human neuronal migration disorder." J Cell Biol 139(5): 1293-1305.
Faust, P. L., H. M. Su, et al. (2001). "The peroxisome deficient PEX2 Zellweger
mouse: pathologic and biochemical correlates of lipid dysfunction." J Mol
Neurosci 16(2-3): 289-297; discussion 317-221.
Feeney, K. M., A. Saade, et al. (2011). "In vivo analysis of the cell cycle
dependent association of the bovine papillomavirus E2 protein and
ChlR1." Virology 414(1): 1-9.
Ferrer, I., J. P. Kapfhammer, et al. (2005). "Inactivation of the peroxisomal
ABCD2 transporter in the mouse leads to late-onset ataxia involving
mitochondria, Golgi and endoplasmic reticulum damage." Hum Mol Genet
14(23): 3565-3577.
Forss-Petter, S., H. Werner, et al. (1997). "Targeted inactivation of the X-linked
adrenoleukodystrophy gene in mice." J Neurosci Res 50(5): 829-843.
Fourcade, S., J. Lopez-Erauskin, et al. (2008). "Early oxidative damage
underlying neurodegeneration in X-adrenoleukodystrophy." Hum Mol
Genet 17(12): 1762-1773.
Fourcade, S., M. Ruiz, et al. (2010). "Valproic acid induces antioxidant effects in
X-linked adrenoleukodystrophy." Hum Mol Genet 19(10): 2005-2014.
Fourcade, S., S. Savary, et al. (2003). "Thyroid hormone induction of the
adrenoleukodystrophy-related gene (ABCD2)." Mol Pharmacol 63(6):
1296-1303.
Freberg, C. T., J. A. Dahl, et al. (2007). "Epigenetic reprogramming of OCT4 and
NANOG regulatory regions by embryonal carcinoma cell extract." Mol Biol
Cell 18(5): 1543-1553.
Garashi, M., D. Belchis, et al. (1976). "Brain gangliosides in
adrenoleukodystrophy." J Neurochem 27(1): 327-328.
Gilg, A. G., A. K. Singh, et al. (2000). "Inducible nitric oxide synthase in the
central nervous system of patients with X-adrenoleukodystrophy." J
Neuropathol Exp Neurol 59(12): 1063-1069.
Goldfischer, S., C. L. Moore, et al. (1973). "Peroxisomal and mitochondrial
defects in the cerebro-hepato-renal syndrome." Science 182(4107): 62-64.
Gressens, P., M. Baes, et al. (2000). "Neuronal migration disorder in Zellweger
mice is secondary to glutamate receptor dysfunction." Ann Neurol 48(3):
336-343.
149
Grimm, M. O., J. Kuchenbecker, et al. (2011). "Plasmalogen synthesis is
regulated via alkyl-dihydroxyacetonephosphate-synthase by amyloid
precursor protein processing and is affected in Alzheimer's disease." J
Neurochem 116(5): 916-925.
Grinnell, K. L., B. Yang, et al. (2007). "De-differentiation of mouse interfollicular
keratinocytes by the embryonic transcription factor Oct-4." J Invest
Dermatol 127(2): 372-380.
Gruber, T., N. Hermann-Kleiter, et al. (2009). "PKC-theta modulates the strength
of T cell responses by targeting Cbl-b for ubiquitination and degradation."
Sci Signal 2(76): ra30.
Gump, J. M. and S. F. Dowdy (2007). "TAT transduction: the molecular
mechanism and therapeutic prospects." Trends Mol Med 13(10): 443-448.
Hatch, M. N., G. Nistor, et al. (2009). "Derivation of high-purity oligodendroglial
progenitors." Methods Mol Biol 549: 59-75.
Heinzer, A. K., P. A. Watkins, et al. (2003). "A very long-chain acyl-CoA
synthetase-deficient mouse and its relevance to X-linked
adrenoleukodystrophy." Hum Mol Genet 12(10): 1145-1154.
Hernandez, M. A., R. Schulz, et al. (2010). "The diagnosis of inherited metabolic
diseases by microarray gene expression profiling." Orphanet J Rare Dis 5:
34.
Ho, J. K., H. Moser, et al. (1995). "Interactions of a very long chain fatty acid with
model membranes and serum albumin. Implications for the pathogenesis
of adrenoleukodystrophy." J Clin Invest 96(3): 1455-1463.
Ho, Y. S., Y. Xiong, et al. (2004). "Mice lacking catalase develop normally but
show differential sensitivity to oxidant tissue injury." J Biol Chem 279(31):
32804-32812.
Hoek, R. M., S. R. Ruuls, et al. (2000). "Down-regulation of the macrophage
lineage through interaction with OX2 (CD200)." Science 290(5497): 1768-
1771.
Hruban, Z., M. Gotoh, et al. (1974). "Structure of hepatic microbodies in rats
treated with acetylsalicylic acid, clofibrate, and dimethrin." Lab Invest 30(1):
64-75.
Hsu, W., L. Zeng, et al. (1999). "Identification of a domain of Axin that binds to
the serine/threonine protein phosphatase 2A and a self-binding domain." J
Biol Chem 274(6): 3439-3445.
Hu, B. Y., Z. W. Du, et al. (2009). "Differentiation of human oligodendrocytes from
pluripotent stem cells." Nat Protoc 4(11): 1614-1622.
Hu, B. Y., J. P. Weick, et al. (2010). "Neural differentiation of human induced
pluripotent stem cells follows developmental principles but with variable
potency." Proc Natl Acad Sci U S A 107(9): 4335-4340.
Hu, B. Y. and S. C. Zhang (2009). "Differentiation of spinal motor neurons from
pluripotent human stem cells." Nat Protoc 4(9): 1295-1304.
Huang da, W., B. T. Sherman, et al. (2009). "Systematic and integrative analysis
of large gene lists using DAVID bioinformatics resources." Nat Protoc 4(1):
44-57.
150
Huangfu, D., K. Osafune, et al. (2008). "Induction of pluripotent stem cells from
primary human fibroblasts with only Oct4 and Sox2." Nat Biotechnol
26(11): 1269-1275.
Hudson, C., D. Clements, et al. (1997). "Xsox17alpha and -beta mediate
endoderm formation in Xenopus." Cell 91(3): 397-405.
Inoue, A., T. Li, et al. (2007). "Loss of ChlR1 helicase in mouse causes lethality
due to the accumulation of aneuploid cells generated by cohesion defects
and placental malformation." Cell Cycle 6(13): 1646-1654.
Isenberg, J. S., Y. Jia, et al. (2007). "Thrombospondin-1 inhibits nitric oxide
signaling via CD36 by inhibiting myristic acid uptake." J Biol Chem 282(21):
15404-15415.
Itskovitz-Eldor, J., M. Schuldiner, et al. (2000). "Differentiation of human
embryonic stem cells into embryoid bodies compromising the three
embryonic germ layers." Mol Med 6(2): 88-95.
Itzkovitz, B., S. Jiralerspong, et al. (2012). "Functional characterization of novel
mutations in GNPAT and AGPS, causing rhizomelic chondrodysplasia
punctata (RCDP) types 2 and 3." Hum Mutat 33(1): 189-197.
Izrael, M., P. Zhang, et al. (2007). "Human oligodendrocytes derived from
embryonic stem cells: Effect of noggin on phenotypic differentiation in vitro
and on myelination in vivo." Mol Cell Neurosci 34(3): 310-323.
Jang, J., H. C. Kang, et al. (2011). "Induced pluripotent stem cell models from X-
linked adrenoleukodystrophy patients." Ann Neurol 70(3): 402-409.
Janssen, A., P. Gressens, et al. (2003). "Neuronal migration depends on intact
peroxisomal function in brain and in extraneuronal tissues." J Neurosci
23(30): 9732-9741.
Jiang, L., X. Luo, et al. (2011). "PDRG1, a novel tumor marker for multiple
malignancies that is selectively regulated by genotoxic stress." Cancer
Biol Ther 11(6): 567-573.
Karaman, M. W., M. L. Houck, et al. (2003). "Comparative analysis of gene-
expression patterns in human and african great ape cultured fibroblasts."
Genome Res 13(7): 1619-1630.
Karumbayaram, S., B. G. Novitch, et al. (2009). "Directed Differentiation of
Human-Induced Pluripotent Stem Cells Generates Active Motor Neurons."
Stem Cells 27(4): 806-811.
Kassmann, C. M., C. Lappe-Siefke, et al. (2007). "Axonal loss and
neuroinflammation caused by peroxisome-deficient oligodendrocytes." Nat
Genet 39(8): 969-976.
Kemp, S. and R. Wanders (2010). "Biochemical aspects of X-linked
adrenoleukodystrophy." Brain Pathol 20(4): 831-837.
Khan, I. F., R. K. Hirata, et al. (2010). "Engineering of human pluripotent stem
cells by AAV-mediated gene targeting." Mol Ther 18(6): 1192-1199.
Khan, M., J. Singh, et al. (2008). "Plasmalogen deficiency in cerebral
adrenoleukodystrophy and its modulation by lovastatin." J Neurochem
106(4): 1766-1779.
Kim, J. B., V. Sebastiano, et al. (2009). "Oct4-induced pluripotency in adult neural
stem cells." Cell 136(3): 411-419.
151
Kim, J. B., H. Zaehres, et al. (2008). "Pluripotent stem cells induced from adult
neural stem cells by reprogramming with two factors." Nature 454(7204):
646-650.
Kirov, S. A., B. Zhang, et al. (2007). "Association analysis for large-scale gene
set data." Methods Mol Biol 408: 19-33.
Klimanskaya, I., Y. Chung, et al. (2007). "Derivation of human embryonic stem
cells from single blastomeres." Nat Protoc 2(8): 1963-1972.
Kobayashi, T., N. Shinnoh, et al. (1997). "Adrenoleukodystrophy protein-deficient
mice represent abnormality of very long chain fatty acid metabolism."
Biochem Biophys Res Commun 232(3): 631-636.
Koch, J., K. Pranjic, et al. (2010). "PEX11 family members are membrane
elongation factors that coordinate peroxisome proliferation and
maintenance." J Cell Sci 123(Pt 19): 3389-3400.
Koning, N., D. F. Swaab, et al. (2009). "Distribution of the immune inhibitory
molecules CD200 and CD200R in the normal central nervous system and
multiple sclerosis lesions suggests neuron-glia and glia-glia interactions."
J Neuropathol Exp Neurol 68(2): 159-167.
Laffe, L. (1999). "Ketone Bodies: a Review of Physiology, Pathophysiology and
Application of Monitoring to Diabetes." Diabetes Metab Res Rev 15: 412-
426.
Lanz, T. V., Z. Ding, et al. (2010). "Angiotensin II sustains brain inflammation in
mice via TGF-beta." J Clin Invest 120(8): 2782-2794.
Laurent, L. C., I. Ulitsky, et al. (2011). "Dynamic changes in the copy number of
pluripotency and cell proliferation genes in human ESCs and iPSCs during
reprogramming and time in culture." Cell Stem Cell 8(1): 106-118.
Lazarow, P. B. and Y. Fujiki (1985). "Biogenesis of peroxisomes." Annu Rev Cell
Biol 1: 489-530.
Lazarow, P. B., M. Robbi, et al. (1982). "Biogenesis of peroxisomal proteins in
vivo and in vitro." Ann N Y Acad Sci 386: 285-300.
Lengerke, C. and G. Q. Daley (2009). "Disease models from pluripotent stem
cells." Ann N Y Acad Sci 1176: 191-196.
Leonard, A. E., E. G. Bobik, et al. (2000). "Cloning of a human cDNA encoding a
novel enzyme involved in the elongation of long-chain polyunsaturated
fatty acids." Biochem J 350 Pt 3: 765-770.
Lerou, P. H., A. Yabuuchi, et al. (2008). "Derivation and maintenance of human
embryonic stem cells from poor-quality in vitro fertilization embryos." Nat
Protoc 3(5): 923-933.
Li, W., W. Wei, et al. (2009). "Generation of rat and human induced pluripotent
stem cells by combining genetic reprogramming and chemical inhibitors."
Cell Stem Cell 4(1): 16-19.
Li, X., E. Baumgart, et al. (2002). "PEX11alpha is required for peroxisome
proliferation in response to 4-phenylbutyrate but is dispensable for
peroxisome proliferator-activated receptor alpha-mediated peroxisome
proliferation." Mol Cell Biol 22(23): 8226-8240.
152
Li, X., E. Baumgart, et al. (2002). "PEX11 beta deficiency is lethal and impairs
neuronal migration but does not abrogate peroxisome function." Mol Cell
Biol 22(12): 4358-4365.
Li, Y., X. Tong, et al. "Thrombospondin1 deficiency reduces obesity-associated
inflammation and improves insulin sensitivity in a diet-induced obese
mouse model." PLoS One 6(10): e26656.
Liao, J., C. Cui, et al. (2009). "Generation of induced pluripotent stem cell lines
from adult rat cells." Cell Stem Cell 4(1): 11-15.
Liegel, R., B. Chang, et al. (2011). "Blind sterile 2 (bs2), a hypomorphic mutation
in Agps, results in cataracts and male sterility in mice." Mol Genet Metab
103(1): 51-59.
Liu, G. H., B. Z. Barkho, et al. (2011). "Recapitulation of premature ageing with
iPSCs from Hutchinson-Gilford progeria syndrome." Nature 472(7342):
221-225.
Liu, G. H., K. Suzuki, et al. (2011). "Targeted gene correction of laminopathy-
associated LMNA mutations in patient-specific iPSCs." Cell Stem Cell 8(6):
688-694.
Liu, H., F. Zhu, et al. (2008). "Generation of induced pluripotent stem cells from
adult rhesus monkey fibroblasts." Cell Stem Cell 3(6): 587-590.
Liu, Y., Y. Bando, et al. "CD200R1 agonist attenuates mechanisms of chronic
disease in a murine model of multiple sclerosis." J Neurosci 30(6): 2025-
2038.
Lombardo, A., P. Genovese, et al. (2007). "Gene editing in human stem cells
using zinc finger nucleases and integrase-defective lentiviral vector
delivery." Nat Biotechnol 25(11): 1298-1306.
Lonergan, T., B. Bavister, et al. (2007). "Mitochondria in stem cells."
Mitochondrion 7(5): 289-296.
Lowry, W. E., L. Richter, et al. (2008). "Generation of human induced pluripotent
stem cells from dermal fibroblasts." Proc Natl Acad Sci U S A 105(8):
2883-2888.
Lu, J. F., E. Barron-Casella, et al. (2007). "The role of peroxisomal ABC
transporters in the mouse adrenal gland: the loss of Abcd2 (ALDR), Not
Abcd1 (ALD), causes oxidative damage." Lab Invest 87(3): 261-272.
Lu, J. F., A. M. Lawler, et al. (1997). "A mouse model for X-linked
adrenoleukodystrophy." Proc Natl Acad Sci U S A 94(17): 9366-9371.
Maccarrone, M., N. Battista, et al. (2007). "The endocannabinoid pathway in
Huntington's disease: a comparison with other neurodegenerative
diseases." Prog Neurobiol 81(5-6): 349-379.
Maeba, R. and N. Ueta (2003). "Ethanolamine plasmalogen and cholesterol
reduce the total membrane oxidizability measured by the oxygen uptake
method." Biochem Biophys Res Commun 302(2): 265-270.
Maeba, R. and N. Ueta (2003). "Ethanolamine plasmalogens prevent the
oxidation of cholesterol by reducing the oxidizability of cholesterol in
phospholipid bilayers." J Lipid Res 44(1): 164-171.
153
Marchetto, M. C., C. Carromeu, et al. (2010). "A model for neural development
and treatment of Rett syndrome using human induced pluripotent stem
cells." Cell 143(4): 527-539.
Martin-Manso, G., S. Galli, et al. (2008). "Thrombospondin 1 promotes tumor
macrophage recruitment and enhances tumor cell cytotoxicity of
differentiated U937 cells." Cancer Res 68(17): 7090-7099.
Mast, F. D., J. Li, et al. (2011). "A Drosophila model for the Zellweger spectrum of
peroxisome biogenesis disorders." Dis Model Mech 4(5): 659-672.
Maxwell, M., J. Bjorkman, et al. (2003). "Pex13 inactivation in the mouse disrupts
peroxisome biogenesis and leads to a Zellweger syndrome phenotype."
Mol Cell Biol 23(16): 5947-5957.
Maxwell, M. A., T. Allen, et al. (2002). "Novel PEX1 mutations and genotype-
phenotype correlations in Australasian peroxisome biogenesis disorder
patients." Hum Mutat 20(5): 342-351.
Maxwell, M. A., P. V. Nelson, et al. (1999). "A common PEX1 frameshift mutation
in patients with disorders of peroxisome biogenesis correlates with the
severe Zellweger syndrome phenotype." Hum Genet 105(1-2): 38-44.
McKenna, O., G. Arnold, et al. (1976). "Microperoxisome distribution in the
central nervous system of the rat." Brain Res 117(2): 181-194.
Menasche, G., E. Pastural, et al. (2000). "Mutations in RAB27A cause Griscelli
syndrome associated with haemophagocytic syndrome." Nat Genet 25(2):
173-176.
Moretti, A., M. Bellin, et al. (2010). "Patient-specific induced pluripotent stem-cell
models for long-QT syndrome." N Engl J Med 363(15): 1397-1409.
Moser, A. B., S. J. Steinberg, et al. (2011). "Human and great ape red blood cells
differ in plasmalogen levels and composition." Lipids Health Dis 10: 101.
Moser, H. W. (1991). "Peroxisomal disorders." Clin Biochem 24(4): 343-351.
Moser, H. W. (1999). "Genotype-phenotype correlations in disorders of
peroxisome biogenesis." Mol Genet Metab 68(2): 316-327.
Moser, H. W., A. Mahmood, et al. (2007). "X-linked adrenoleukodystrophy." Nat
Clin Pract Neurol 3(3): 140-151.
Moser, H. W. and A. B. Moser (1991). " Measurement of saturated very long
chain fatty acids in plasma. In: Hommes, F. A. (ed)." Techniques in
diagnostic human biochemical genetics: a laboratory manual, Wiley-Liss,
New York.
Moser, H. W., A. B. Moser, et al. (2007). ""Lorenzo's oil" therapy for X-linked
adrenoleukodystrophy: rationale and current assessment of efficacy." J
Mol Neurosci 33(1): 105-113.
Moser, H. W., G. V. Raymond, et al. (2005). "Follow-up of 89 asymptomatic
patients with adrenoleukodystrophy treated with Lorenzo's oil." Arch
Neurol 62(7): 1073-1080.
Munji, R. N., Y. Choe, et al. (2011). "Wnt signaling regulates neuronal
differentiation of cortical intermediate progenitors." J Neurosci 31(5): 1676-
1687.
Nagan, N. and R. A. Zoeller (2001). "Plasmalogens: biosynthesis and functions."
Prog Lipid Res 40(3): 199-229.
154
Nakanishi, M. and D. W. Rosenberg (2006). "Roles of cPLA2alpha and
arachidonic acid in cancer." Biochim Biophys Acta 1761(11): 1335-1343.
Ning, M., D. A. Sarracino, et al. (2011). "Proteomic temporal profile of human
brain endothelium after oxidative stress." Stroke 42(1): 37-43.
Nishikawa, S., R. A. Goldstein, et al. (2008). "The promise of human induced
pluripotent stem cells for research and therapy." Nat Rev Mol Cell Biol 9(9):
725-729.
Nishino, K., M. Toyoda, et al. (2011). "DNA methylation dynamics in human
induced pluripotent stem cells over time." PLoS Genet 7(5): e1002085.
Nistor, G. I., M. O. Totoiu, et al. (2005). "Human embryonic stem cells
differentiate into oligodendrocytes in high purity and myelinate after spinal
cord transplantation." Glia 49(3): 385-396.
Novikoff, A. B. and W. Y. Shin (1964). "The endoplasmatic reticulum in the golgi
zone and ist relations to microbodies, golgiapparatus and autophagic
vacuoles in rat liver cells." J. Mikros. Oxford 3: 187-206.
Ohi, Y., H. Qin, et al. (2011). "Incomplete DNA methylation underlies a
transcriptional memory of somatic cells in human iPS cells." Nat Cell Biol
13(5): 541-549.
Okita, K., M. Nakagawa, et al. (2008). "Generation of mouse induced pluripotent
stem cells without viral vectors." Science 322(5903): 949-953.
Panopoulos, A. D., O. Yanes, et al. (2012). "The metabolome of induced
pluripotent stem cells reveals metabolic changes occurring in somatic cell
reprogramming." Cell Res 22(1): 168-177.
Park, I. H., N. Arora, et al. (2008). "Disease-specific induced pluripotent stem
cells." Cell 134(5): 877-886.
Park, I. H., P. H. Lerou, et al. (2008). "Generation of human-induced pluripotent
stem cells." Nat Protoc 3(7): 1180-1186.
Pasca, S. P., T. Portmann, et al. (2011). "Using iPSC-derived neurons to uncover
cellular phenotypes associated with Timothy syndrome." Nat Med 17(12):
1657-1662.
Patel, M. and S. Yang (2010). "Advances in reprogramming somatic cells to
induced pluripotent stem cells." Stem Cell Rev 6(3): 367-380.
Pettus, B. J., M. Baes, et al. (2004). "Mass spectrometric analysis of ceramide
perturbations in brain and fibroblasts of mice and human patients with
peroxisomal disorders." Rapid Commun Mass Spectrom 18(14): 1569-
1574.
Pike, B. L., T. C. Greiner, et al. (2008). "DNA methylation profiles in diffuse large
B-cell lymphoma and their relationship to gene expression status."
Leukemia 22(5): 1035-1043.
Platta, H. W., F. El Magraoui, et al. (2009). "Pex2 and pex12 function as protein-
ubiquitin ligases in peroxisomal protein import." Mol Cell Biol 29(20): 5505-
5516.
Platta, H. W., F. El Magraoui, et al. (2007). "Ubiquitination of the peroxisomal
import receptor Pex5p is required for its recycling." J Cell Biol 177(2): 197-
204.
155
Platta, H. W. and R. Erdmann (2007). "Peroxisomal dynamics." Trends Cell Biol
17(10): 474-484.
Platta, H. W. and R. Erdmann (2007). "The peroxisomal protein import
machinery." FEBS Lett 581(15): 2811-2819.
Powers, J. M. and H. W. Moser (1998). "Peroxisomal disorders: genotype,
phenotype, major neuropathologic lesions, and pathogenesis." Brain
Pathol 8(1): 101-120.
Powers, J. M., Z. Pei, et al. (2005). "Adreno-leukodystrophy: oxidative stress of
mice and men." J Neuropathol Exp Neurol 64(12): 1067-1079.
Prestele, J., G. Hierl, et al. (2010). "Different functions of the C3HC4 zinc RING
finger peroxins PEX10, PEX2, and PEX12 in peroxisome formation and
matrix protein import." Proc Natl Acad Sci U S A 107(33): 14915-14920.
Pujol, A., I. Ferrer, et al. (2004). "Functional overlap between ABCD1 (ALD) and
ABCD2 (ALDR) transporters: a therapeutic target for X-
adrenoleukodystrophy." Hum Mol Genet 13(23): 2997-3006.
Pujol, A., C. Hindelang, et al. (2002). "Late onset neurological phenotype of the
X-ALD gene inactivation in mice: a mouse model for
adrenomyeloneuropathy." Hum Mol Genet 11(5): 499-505.
Pyne, S. and N. J. Pyne (2000). "Sphingosine 1-phosphate signalling in
mammalian cells." Biochem J 349(Pt 2): 385-402.
Qi, C., Y. Zhu, et al. (1999). "Absence of spontaneous peroxisome proliferation in
enoyl-CoA Hydratase/L-3-hydroxyacyl-CoA dehydrogenase-deficient
mouse liver. Further support for the role of fatty acyl CoA oxidase in
PPARalpha ligand metabolism." J Biol Chem 274(22): 15775-15780.
Raymond, G. V. (2009). Peroxisomal Disorders. Neural Lipids. G. Tettamanti,
Goracci, G., Springer Science+Business Media: 632-669.
Rhodin, J. (1954). Correlation of ultrastructural organization and function in
normal and experimentally changed proximal tubule cells of the mouse
kidney. . Stockholm, Sweden:, Karolinska Instituet. PhD.
Rinaldo, P., D. Matern, et al. (2002). "Fatty acid oxidation disorders." Annu Rev
Physiol 64: 477-502.
Rinn, J. L., C. Bondre, et al. (2006). "Anatomic demarcation by positional
variation in fibroblast gene expression programs." PLoS Genet 2(7): e119.
Ruf, I. K., P. W. Rhyne, et al. (2001). "EBV regulates c-MYC, apoptosis, and
tumorigenicity in Burkitt's lymphoma." Curr Top Microbiol Immunol 258:
153-160.
Schrader, M., N. A. Bonekamp, et al. (2011). "Fission and proliferation of
peroxisomes." Biochim Biophys Acta.
Schrader, M., B. E. Reuber, et al. (1998). "Expression of PEX11beta mediates
peroxisome proliferation in the absence of extracellular stimuli." J Biol
Chem 273(45): 29607-29614.
Schrader, M. and Y. Yoon (2007). "Mitochondria and peroxisomes: are the 'big
brother' and the 'little sister' closer than assumed?" Bioessays 29(11):
1105-1114.
156
Semmler, A., W. Kohler, et al. (2008). "Therapy of X-linked
adrenoleukodystrophy." Expert Rev Neurother 8(9): 1367-1379.
Setchell, K. D., P. Bragetti, et al. (1992). "Oral bile acid treatment and the patient
with Zellweger syndrome." Hepatology 15(2): 198-207.
Sharp, J., J. Frame, et al. (2010). "Human embryonic stem cell-derived
oligodendrocyte progenitor cell transplants improve recovery after cervical
spinal cord injury." Stem Cells 28(1): 152-163.
She, X., J. E. Horvath, et al. (2004). "The structure and evolution of centromeric
transition regions within the human genome." Nature 430(7002): 857-864.
Shindou, H. and T. Shimizu (2009). "Acyl-CoA:lysophospholipid
acyltransferases." J Biol Chem 284(1): 1-5.
Shruster, A., H. Eldar-Finkelman, et al. (2011). "Wnt signaling pathway
overcomes the disruption of neuronal differentiation of neural progenitor
cells induced by oligomeric amyloid beta-peptide." J Neurochem 116(4):
522-529.
Singh, H., M. Brogan, et al. (1992). "Peroxisomal beta-oxidation of branched
chain fatty acids in human skin fibroblasts." J Lipid Res 33(11): 1597-1605.
Singh, I. (1996). "Mammalian peroxisomes: metabolism of oxygen and reactive
oxygen species." Ann N Y Acad Sci 804: 612-627.
Singh, I. and A. Pujol (2010). "Pathomechanisms underlying X-
adrenoleukodystrophy: a three-hit hypothesis." Brain Pathol 20(4): 838-
844.
Singh, J., M. Khan, et al. (2011). "HDAC inhibitor SAHA normalizes the levels of
VLCFAs in human skin fibroblasts from X-ALD patients and downregulates
the expression of proinflammatory cytokines in Abcd1/2-silenced mouse
astrocytes." J Lipid Res 52(11): 2056-2069.
Sinner, D., S. Rankin, et al. (2004). "Sox17 and beta-catenin cooperate to
regulate the transcription of endodermal genes." Development 131(13):
3069-3080.
Smith, J. J., M. Marelli, et al. (2002). "Transcriptome profiling to identify genes
involved in peroxisome assembly and function." J Cell Biol 158(2): 259-
271.
Soldner, F., D. Hockemeyer, et al. (2009). "Parkinson's disease patient-derived
induced pluripotent stem cells free of viral reprogramming factors." Cell
136(5): 964-977.
Solorzano, C., C. Zhu, et al. (2009). "Selective N-acylethanolamine-hydrolyzing
acid amidase inhibition reveals a key role for endogenous
palmitoylethanolamide in inflammation." Proc Natl Acad Sci U S A 106(49):
20966-20971.
Stadtfeld, M., M. Nagaya, et al. (2008). "Induced pluripotent stem cells generated
without viral integration." Science 322(5903): 945-949.
Steinberg, S., L. Chen, et al. (2004). "The PEX Gene Screen: molecular
diagnosis of peroxisome biogenesis disorders in the Zellweger syndrome
spectrum." Mol Genet Metab 83(3): 252-263.
Steinberg, S., R. Jones, et al. (2008). "Investigational methods for peroxisomal
disorders." Curr Protoc Hum Genet Chapter 17: Unit 17 16.
157
Steinberg, S. J., G. Dodt, et al. (2006). "Peroxisome biogenesis disorders."
Biochim Biophys Acta 1763(12): 1733-1748.
Steinberg, S. J., G. V. Raymond, et al. (2003-2012). Peroxisome Biogenesis
Disorders, Zellweger Syndrome Spectrum. GeneReviews [internet].
Steinberg, S. J., A. Snowden, et al. (2009). "A PEX10 defect in a patient with no
detectable defect in peroxisome assembly or metabolism in cultured
fibroblasts." J Inherit Metab Dis 32(1): 109-119.
Stolz, A., L. Hammond, et al. (1993). "cDNA cloning and expression of the human
hepatic bile acid-binding protein. A member of the monomeric reductase
gene family." J Biol Chem 268(14): 10448-10457.
Suhr, S. T., E. A. Chang, et al. (2010). "Mitochondrial rejuvenation after induced
pluripotency." PLoS One 5(11): e14095.
Sun, N., N. J. Panetta, et al. (2009). "Feeder-free derivation of induced
pluripotent stem cells from adult human adipose stem cells." Proc Natl
Acad Sci U S A 106(37): 15720-15725.
Takahashi, K., K. Okita, et al. (2007). "Induction of pluripotent stem cells from
fibroblast cultures." Nat Protoc 2(12): 3081-3089.
Takahashi, K., K. Tanabe, et al. (2007). "Induction of pluripotent stem cells from
adult human fibroblasts by defined factors." Cell 131(5): 861-872.
Takahashi, K. and S. Yamanaka (2006). "Induction of pluripotent stem cells from
mouse embryonic and adult fibroblast cultures by defined factors." Cell
126(4): 663-676.
Tamura, S., S. Yasutake, et al. (2006). "Dynamic and functional assembly of the
AAA peroxins, Pex1p and Pex6p, and their membrane receptor Pex26p." J
Biol Chem 281(38): 27693-27704.
Tazi, K. A., J. J. Quioc, et al. (2009). "Protein array technology to investigate
cytokine production by monocytes from patients with advanced alcoholic
cirrhosis: An ex vivo pilot study." Hepatol Res 39(7): 706-715.
Theda, C., A. B. Moser, et al. (1992). "Phospholipids in X-linked
adrenoleukodystrophy white matter: fatty acid abnormalities before the
onset of demyelination." J Neurol Sci 110(1-2): 195-204.
Thoms, S. and R. Erdmann (2006). "Peroxisomal matrix protein receptor
ubiquitination and recycling." Biochim Biophys Acta 1763(12): 1620-1628.
Thoms, S. and J. Gartner (2012). "First PEX11beta patient extends spectrum of
peroxisomal biogenesis disorder phenotypes." J Med Genet 49(5): 314-
316.
Tiscornia, G., O. Singer, et al. (2006). "Production and purification of lentiviral
vectors." Nat Protoc 1(1): 241-245.
Tournier JN, Rossi Paccani S, et al. (2009). "Anthrax toxins: a weapon to
systematically dismantle the host immune defenses." Mol Aspects Med.
30(6): 456-466.
Tsuboi, K., Y. X. Sun, et al. (2005). "Molecular characterization of N-
acylethanolamine-hydrolyzing acid amidase, a novel member of the
choloylglycine hydrolase family with structural and functional similarity to
acid ceramidase." J Biol Chem 280(12): 11082-11092.
158
Urnov, F. D., J. C. Miller, et al. (2005). "Highly efficient endogenous human gene
correction using designed zinc-finger nucleases." Nature 435(7042): 646-
651.
Vargas, C. R., M. Wajner, et al. (2004). "Evidence that oxidative stress is
increased in patients with X-linked adrenoleukodystrophy." Biochim
Biophys Acta 1688(1): 26-32.
Verani, R., I. Cappuccio, et al. (2007). "Expression of the Wnt inhibitor Dickkopf-1
is required for the induction of neural markers in mouse embryonic stem
cells differentiating in response to retinoic acid." J Neurochem 100(1):
242-250.
Wanders, R. J. (2004). "Metabolic and molecular basis of peroxisomal disorders:
a review." Am J Med Genet A 126A(4): 355-375.
Wanders, R. J. (2004). "Peroxisomes, lipid metabolism, and peroxisomal
disorders." Mol Genet Metab 83(1-2): 16-27.
Wanders, R. J. and H. R. Waterham (2006). "Biochemistry of mammalian
peroxisomes revisited." Annu Rev Biochem 75: 295-332.
Wang, X. M., T. C. Greiner, et al. (2010). "Identification and functional relevance
of de novo DNA methylation in cancerous B-cell populations." J Cell
Biochem 109(4): 818-827.
Wang, Y., M. McClelland, et al. (2009). "Analyzing microarray data using
WebArray." Cold Spring Harb Protoc 2009(8): pdb prot5260.
Warren, L., P. D. Manos, et al. (2010). "Highly efficient reprogramming to
pluripotency and directed differentiation of human cells with synthetic
modified mRNA." Cell Stem Cell 7(5): 618-630.
Watkins, P. A., A. B. Moser, et al. (2010). "Identification of differences in human
and great ape phytanic acid metabolism that could influence gene
expression profiles and physiological functions." BMC Physiol 10: 19.
Wei, Z., S. Chigurupati, et al. (2011). "Notch activation enhances the microglia-
mediated inflammatory response associated with focal cerebral ischemia."
Stroke 42(9): 2589-2594.
Wernig, M., C. J. Lengner, et al. (2008). "A drug-inducible transgenic system for
direct reprogramming of multiple somatic cell types." Nat Biotechnol 26(8):
916-924.
Wernig, M., A. Meissner, et al. (2008). "c-Myc is dispensable for direct
reprogramming of mouse fibroblasts." Cell Stem Cell 2(1): 10-12.
Wernig, M., J. P. Zhao, et al. (2008). "Neurons derived from reprogrammed
fibroblasts functionally integrate into the fetal brain and improve symptoms
of rats with Parkinson's disease." Proc Natl Acad Sci U S A 105(15): 5856-
5861.
Wilson, R. and J. R. Sargent (1993). "Lipid and fatty acid composition of brain
tissue from adrenoleukodystrophy patients." J Neurochem 61(1): 290-297.
Woltjen, K., I. P. Michael, et al. (2009). "piggyBac transposition reprograms
fibroblasts to induced pluripotent stem cells." Nature 458(7239): 766-770.
Xia, X., M. McClelland, et al. (2005). "WebArray: an online platform for
microarray data analysis." BMC Bioinformatics 6: 306.
159
Xia, X. and S. C. Zhang (2009). "Differentiation of neuroepithelia from human
embryonic stem cells." Methods Mol Biol 549: 51-58.
Yang, J., J. Wu, et al. (2003). "PP2A:B56epsilon is required for Wnt/beta-catenin
signaling during embryonic development." Development 130(23): 5569-
5578.
Yaworsky, P. J. and C. Kappen (1999). "Heterogeneity of neural progenitor cells
revealed by enhancers in the nestin gene." Dev Biol 205(2): 309-321.
Yazawa, M., B. Hsueh, et al. (2011). "Using induced pluripotent stem cells to
investigate cardiac phenotypes in Timothy syndrome." Nature 471(7337):
230-234.
Ye, Z., H. Zhan, et al. (2009). "Human-induced pluripotent stem cells from blood
cells of healthy donors and patients with acquired blood disorders." Blood
114(27): 5473-5480.
Yik, W. Y., S. J. Steinberg, et al. (2009). "Identification of novel mutations and
sequence variation in the Zellweger syndrome spectrum of peroxisome
biogenesis disorders." Hum Mutat 30(3): E467-480.
Yin, Z. S., H. Zhang, et al. (2007). "Wnt-3a protein promote neuronal
differentiation of neural stem cells derived from adult mouse spinal cord."
Neurol Res 29(8): 847-854.
Yu, J., K. Hu, et al. (2009). "Human induced pluripotent stem cells free of vector
and transgene sequences." Science 324(5928): 797-801.
Yu, J., M. A. Vodyanik, et al. (2007). "Induced pluripotent stem cell lines derived
from human somatic cells." Science 318(5858): 1917-1920.
Yusa, K., S. T. Rashid, et al. (2011). "Targeted gene correction of alpha1-
antitrypsin deficiency in induced pluripotent stem cells." Nature 478(7369):
391-394.
Zhang, B., S. Kirov, et al. (2005). "WebGestalt: an integrated system for exploring
gene sets in various biological contexts." Nucleic Acids Res 33(Web
Server issue): W741-748.
Zhang, P. L., M. Izrael, et al. (2006). "Increased myelinating capacity of
embryonic stem cell derived oligodendrocyte precursors after treatment by
interleukin-6/soluble interleukin-6 receptor fusion protein." Mol Cell
Neurosci 31(3): 387-398.
Zhang, S. C., M. Wernig, et al. (2001). "In vitro differentiation of transplantable
neural precursors from human embryonic stem cells." Nat Biotechnol
19(12): 1129-1133.
Zou, J., M. L. Maeder, et al. (2009). "Gene targeting of a disease-related gene in
human induced pluripotent stem and embryonic stem cells." Cell Stem
Cell 5(1): 97-110.
Zou, J., P. Mali, et al. (2011). "Site-specific gene correction of a point mutation in
human iPS cells derived from an adult patient with sickle cell disease."
Blood 118(17): 4599-4608.
Abstract (if available)
Abstract
Peroxisomal disorders are a group of genetically heterogeneous metabolic diseases caused by defects in peroxins, proteins encoded by PEX genes that function in peroxisome biogenesis, or in a single peroxisomal protein that has a more targeted effect on specific peroxisome functions. In general, peroxisome disorders can affect almost every organ system, with especially devastating effects on the nervous, hepatic, and adrenocortical systems. ❧ Currently, there is no curative therapy or long-term effective treatment available for peroxisomal disorders. Ongoing pathomechanism studies, diagnostics, and drug testing are mainly established on patient-derived primary fibroblasts and Pex gene knockout mouse models, which do not represent the exact human mutations and most clinical aspects of the human disease. ❧ In this thesis, I describe a new model system which we established for studying the pathology of peroxisomal disorders and testing new therapeutic agents. We generated induced pluripotent stem cells (iPSCs) from primary skin fibroblasts of multiple healthy controls and patients with peroxisomal biogenesis disorders (PBD), caused by genetic defects in PEX genes, or the childhood cerebral form of X-linked adrenoleukodystrophy (CCALD), caused by genetic defects in the ABCD1 gene that encodes a peroxisome membrane protein involved in very long chain fatty acid (VLCFA) metabolism. Candidate iPSCs were subject to global expression, DNA methylation, and genotyping analysis and tested for pluripotency through in vitro embryoid body differentiation and in vivo teratoma formation. We characterized the gene expression and biochemical profiles of these patient-specific iPSCs and further differentiated these iPSCs into pathologically related central nervous system cell (CNSC) lineages, including neural progenitors, motor neurons, and oligodendrocytes. ❧ Our molecular characterization of iPSCs and CNSCs provided a novel perspective into disease mechanisms that supports leading hypotheses regarding disease pathogenesis including the pivotal roles of neuroinflammation, lipid metabolism, and aberrant mitochondrial function. Our novel resources also provide a first step required for the development and interpretation of patient-specific model systems that investigate non-cell autonomous processes relevant to the etiology of peroxisomal disorders. These iPSC and CNS cell resources could also have applications for high content screening (HCS) of chemical libraries for candidate drugs that directly address the cell type specificity of disease and the nature of the mutations found in the patient population.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Development of targeted therapies for peroxisome biogenesis disorders
PDF
Developing novel in vitro model systems to investigate therapeutic hypotheses for peroxisome biogenesis disorders
PDF
The mechanisms of somatic cell reprogramming
PDF
Estimation of carrier frequencies of peroxisomal disorders and deficiencies in global populations
PDF
Characterization of mouse models of peroxisome biogenesis disorders: a study of dental enamel phenotypes
PDF
Development of computational tools to assist high content screening to identify drug therapies for peroxisome biogenesis disorders
PDF
Exploring stem cell pluripotency through long range chromosome interactions
PDF
Derivation and characterization of human embryonic stem (hES) cells and human induced pluripotent stem (hiPS) cells in clinical grade conditions
PDF
Interaction of epigenetics and SMAD signaling in stem cells and diseases
PDF
The splicing error of FOXP1 in type I myotonic dystrophy
PDF
Derivation, expansion and characterization of human hippocampal primordial cells from normal and diseased iPSCs
PDF
Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis
PDF
Craniofacial skull joint and temporomandibular joint (TMJ) in homeostasis and disease
PDF
Genetics and the environment: evaluating the role of noncoding RNA in autism spectrum disorder
PDF
Epigenetic plasticity of cultured female human embryonic stem cells and regulation of gene expression and chromatin by PR-SET7 mediated H4K20me1
PDF
An iPSC-based biomarker strategy to identify neuroregenerative responders to allopregnanolone
PDF
Lung mesenchyme cell biology
PDF
Pleotropic potential of Stat3 in determining self-renewal, apoptosis, and differentiation in mouse embryonic stem cells
PDF
Identifying allele-specific DNA methylation in mammalian genomes
PDF
Glucocorticoids and Runx2 synergistically stimulate Wif1 and compromise pre-osteoblats in vitro
Asset Metadata
Creator
Wang, Xiaoming (author)
Core Title
IPS and CNS cell models of peroxisomal disorders
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Molecular Biology
Publication Date
07/30/2014
Defense Date
06/19/2012
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
adrenoleukodystrophy,central nervous system,induced pluripotent stem cells,OAI-PMH Harvest,peroxisome,peroxisome biogenesis disorders,reprogramming
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Hacia, Joseph G. (
committee chair
), Smith, Andrew D. (
committee member
), Stallcup, Michael R. (
committee member
)
Creator Email
wang14@usc.edu,yyxmwang@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-75961
Unique identifier
UC11289531
Identifier
usctheses-c3-75961 (legacy record id)
Legacy Identifier
etd-WangXiaomi-1064.pdf
Dmrecord
75961
Document Type
Dissertation
Rights
Wang, Xiaoming
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
Tags
adrenoleukodystrophy
central nervous system
induced pluripotent stem cells
peroxisome
peroxisome biogenesis disorders
reprogramming