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

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

CHARACTERIZATION OF HUMAN EMBRYONIC STEM CELL DERIVED
RETINAL PIGMENT EPITHELIAL CELLS FOR AGE-RELATED MACULAR
DEGENERATION

by

Jamie Hsiung

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
(GENETIC, MOLECULAR AND CELLULAR BIOLOGY)

December 2014
i

Dedication

I dedicate this thesis to my mom and dad for their never ending encouragement and support.

ii

Acknowledgements

I would like to thank my PI, Dr. David Hinton for giving me the opportunity to work in
his laboratory. Despite how hectic your schedule usually is, I appreciate how you are always
willing to meet with me to discuss my projects. I am grateful for all of your patience and advice
in supporting my experiments, writing, and presentations, especially when it came down to the
final few months before my dissertation defense date.
I would like to thank my mentor Dr. Danhong Zhu for going beyond simply teaching me
cell culture. You have provided me great feedback over the years on the direction of my project
and offered vital constructive criticism when I really needed it. You have been essential to the
success of my project.
I would like to thank my committee members, Dr. Cheryl Craft and Dr. Wange Lu for
providing me wonderful suggestions and comments in regards to my projects, my thesis, and life
after graduation.
I would like to thank current and former members of the Hinton lab, particularly
Christine Spee, Dr. Ram Kannan, Dr. Shikun He, Dr. Takeshi Yoshida, Dr. Nymph Chan, and
Dr. Kumar Parameswaranpg for not only being a knowledgeable network for troubleshooting
experiments and discussing my projects, but also being the components of a wonderful, caring
laboratory to work in. I thoroughly enjoyed my time here because of you all.
I would like to thank Ernesto Barron, Eric Barron, and Anthony Rodriguez for their help
with microscopy images and data quantification.
iii

Finally, I would like to thank my parents Bill and Judy Hsiung, and my husband Tri Au.
I am so grateful to have such a loving family who is always willing to make sacrifices and put
my needs ahead of theirs.

iv

Table of Contents
Dedication……………………………………………………………………………………........i
Acknowledgements………………………………………………………………………………ii
List of Tables……………………………………………………………………………………vii
List of Figures………………………………………………………………………………….viii

Chapter 1: Introduction: Retinal Pigment Epithelial Cells in Age Related Macular
Degeneration
1.1: Introduction to Retinal Pigment Epithelial (RPE) Cells…………………………..………….1
1.2: Properties of Polarized RPE ……………………………………………………..……..........2
1.3: RPE Developmental Process…………………………………………………….…………..3
1.4: RPE Cells in Age-related Macular Degeneration (AMD)……………………………............5
1.5: Role of Oxidative Stress in AMD Pathogenesis………………………………………...........7
1.6: Other RPE Degenerative Diseases……………………………………………………..........10
1.7: Therapeutic Options to Restore Vision………………………………………………...........10
1.8: Introduction to Embryonic Stem Cells………………………………………...……………11
1.9: Differentiating RPEs from hESCs……………………………………………………..........13
1.10: hESC-RPE in Clinical Work……………………………………………………………….14
1.11: Introduction to Thesis Projects…………………………………………………………….16

Chapter 2: Human Embyronic Stem Cell Derived Polarized Retinal Pigment Epithelial
Cells have Higher Resistance to Oxidative Stress-Induced Cell Death than Non-Polarized
Cultures
2.1: Abstract……………………………………………………………………………….……..18
2.2: Introduction………………………………………………………………………………….20
2.3: Materials and Methods………………………………………………………………............23
2.4: Results……………………………………………………………………………………….28
v

2.4.1 Polarized RPE are more resistant to H
2
O
2
-mediated apoptosis…….....…...............28
2.4.2 Polarized RPE TER drop corresponds to sudden increase in H
2
O
2
cell
death……………….……………………………………………………………………..35
2.4.3 Non-polarized RPE have higher levels of pro-apoptotic signaling
pathways………………………..………………………………………………………..36
2.4.4 Polarized RPE have constitutively higher levels of cell survival
signaling.............................................................................................................................45
2.4.5 Polarized RPE express constitutively higher levels of antioxidants…………….....50
2.5: Discussion……………………………………………………………………………...........54

Chapter 3: Retinal Pigment Epithelial Cell Conditioned Medium Contains Factors
Responsible for Increased RPE Cell Differentiation from Human Embryonic Stem Cells
3.1: Abstract………………………………………………………………………….…………..62
3.2: Introduction……………..…………………………………………………………………...64
3.3: Materials and Methods...…………………………………………………………………….68
3.4: Results…………………………………………………………………………….................74
3.4.1 hESC grown in CM resulted in greater amount of pigmented cells……………….74
3.4.2 CM induced higher number of RPE differentiation from hESC…………………...77
3.4.3 Growth factor array reveals potential factors in CM and
differentiating hESC responsible for increasing amount of RPE yield………………….83
3.4.4 Addition of NT3/NT4 and PDGF AA can increase RPE yield
during hESC differentiation……………………………………………………………...90
3.4.5 Neither CM nor GF addition can increase the speed of RPE
differentiation……………………………………………………….…………………....95
3.4.6 CM and GF addition can increase RPE proliferation………………,..…………..101
3.4.7 CM and NT3/NT4 addition can increase number of RPE colonies during
differentiation……………………………………………………………….…………..102
3.5: Discussion…………………………………………………………………………...……..105

Chapter 4: Summary and Future Directions
4.1: Summary…………………………………………………………………………………...113
vi

4.2. Future Directions…………………………………………………………………………..120

References……………………………………………………………………………………...123

vii

List of Tables

3.1: Growth factor array performed on media of differentiating H9 cells and
polarized H9-RPE………………..................................................................................................84
3.2: The listed growth factor combinations were screened to see if they can
increase the amount of differentiated RPE cells relative to regular medium…………………...90

viii

List of Figures

2.1 H9-RPE cells plated at different levels of confluency had varying
responses to H
2
O
2
treatment……………………………………………………………………29
2.2 Polarized H9-RPE cells have highest resistance to H
2
O
2
treatment……..…………………32
2.3 Highest cell death in H
2
O
2
treated polarized H9-RPE corresponds to TER drop…………..36
2.4 Treated non-polarized RPE have higher expression of pro-apoptotic p-p38
and p-JNK relative to polarized RPE……………………………………………………………38
2.5 Treated non-polarized RPE have significantly higher expression of pro-apoptotic
p53 and p21 relative to polarized RPE…………………………………………………………..41
2.6 Treated sub-confluent RPE have highest pro-apoptotic Bax expression levels……..………44
2.7 Less cell death in polarized H9-RPE can be attributed toward higher activation
of cell survival signaling, such as p-Akt and p-PTEN…………………………………………...47
2.8 Polarized RPE constitutively have higher levels of anti-apoptotic Bcl-2…………………....49
2.9 There were significantly higher levels of antioxidant proteins SOD1 and
catalase expressed in polarized RPE relative to non-polarized RPE…………………………….52
3.1 hESC-RPE cells grown in RPE-CM yield higher amount of pigmented cells
than regular media………………………………………………………………………………..76
3.2 Pigmented patches in H9 and hES3 cell cultures immunostain positive for
Bestrophin, an RPE marker………………………………………………………………………78
3.3 hESC-RPE cells cultured in CM express higher RPE marker genes than when
cultured in regular medium………………………………………………………………………80
3.4 CM increases the number of RPE cells, not the level of RPE65 or Bestrophin
being expressed in individual cells………………………………………………………………82
3.5 The following growth factors from the array were selected and tested to see
ix

if they had a role in increasing RPE differentiation……………………………………………...87
3.6 H9-ESC differentiated with addition of NT3/NT4 resulted in higher amount
of pigmented cells relative to cells grown in regular medium…………………………………...92
3.7 Q-RT PCR show that addition of PDGF AA and NT3/NT4 results in an increase
of RPE-specific marker genes relative to regular media at 8 weeks of differentiation………….94
3.8 There was no difference in the levels of early expressing genes of the
RPE differentiation pathway between regular and CM in hES3 and H9 cells…………………..96
3.9 Q-RT PCR shows CM did not increase the rate of RPE differentiation from hESC…………98
3.10 Q-RT PCR shows that the addition of PDGF AA and NT3/NT4 also did
not increase the rate of RPE differentiation from H9 ESC……………………..……………....100
3.11 RPE cells grown in CM and media supplemented with PDGF AA and
NT3/NT4 helped increase RPE cell proliferation relative to regular medium………………….102
3.12 H9 cells differentiated in CM and media supplemented with NT3/NT4
had significantly higher numbers of differentiated RPE colonies, but cells
differentiated in media supplemented with PDGF AA did not…………………………………103
1

Chapter 1: Introduction

1.1: Introduction to Retinal Pigment Epithelial (RPE) Cells
Retinal pigment epithelial (RPE) cells are a monolayer of polarized cells that comprise
the outer blood retinal barrier
1
. Located between the retina and the choroidal blood vessels, the
apical side of the RPE faces the photoreceptors while the basal side is adjacent to the Bruch’s
membrane, which separates the RPE from the choriocapillaris
2
. Polarized RPE are characterized
by their hexagonal shape with tight cell-cell junctions, apical microvilli to aid in rod outer
segment phagocytosis, and localization of the Na/K-ATPase pump to the apical side, which helps
maintain proper ionic concentrations for phototransduction as well as the energy for
transepithelial transport
3
. RPE cells are pigmented cells capable of absorbing stray light,
phagocytosis of fatty acid-rich photoreceptor outer rod segments, as well as transporting water,
ions, and metabolic waste products to the basal blood vessel side, and in turn carrying glucose,
retinol and fatty acids back to the apical photoreceptor side
4
. Additionally, RPE cells are
essential for maintaining the visual cycle. Because photoreceptors are unable to re-isomerize all-
trans-retinal into all-cis retinal after photon absorption, all-trans-retinal is transported into RPE
cells, converted to all-cis retinal and then shuffled back into the photoreceptors again
2
. In
addition to secreting various immunosuppressive factors since it is found in an immune-
privileged site
5
, polarized RPE cells also preferentially secrete pigment epithelial derived factor
(PEDF) and vascular endothelial growth factor (VEGF) into the apical and basal sides
respectively in order to help maintain proper functioning of the photoreceptors and choroidal
blood vessels
6-8
.

2

1.2: Properties of Polarized RPE
In order to function properly, RPE cells must be polarized so that they are properly
oriented with their apical side facing the retina and their basal side facing the Bruch’s membrane.
Although RPE cells are polarized in vivo, polarization is lost during passaging in in vitro cell
cultures; the cells lose their pigmentation, and their original hexagonal shapes become more
elongated in the non-polarized state. RPE cells need to first reach confluency before polarization
develops, which can take up to 4 weeks. Polarization can be assessed visually by looking for the
formation of pigmented, hexagonal shapes, but the level of their polarization can be measured
via their transepithelial resistance (TER) using a voltometer when grown on Transwell chamber
inserts. In vitro studies of polarized fetal RPE and human embryonic stem cell derived RPE
(hESC-RPE) cells show that polarized RPE cells secreted the neurotrophic, anti-angiogenic
growth factor pigment epithelial derived factor (PEDF) about 80 times higher than non-polarized
RPE cells
9
. PEDF has also been shown to play an important role in retina differentiation.
Furthermore the polarized apical and basal sides of the RPE layer secrete different levels of
growth factors. Specifically, in fetal RPE, PEDF was found to be secreted higher on the apical
side while the proangiogenic vascular endothelial growth factor (VEGF) was found to be
secreted higher on the basal side
6
. Neuroprotectin D1 (NPD1), which is regulated by
neurotrophins, helps prevent oxidative stress-related apoptosis triggered by A2E, a toxic
byproduct of phototransduction. Interestingly, NPD1 has also been shown to be secreted on the
apical side of polarized RPE cells
10
. These studies therefore indicate that polarized RPE cells
and adjacent photoreceptors may be better protected from oxidative stress than non-polarized
RPE due to its secretion of neuroprotective factors. PEDF likely plays a role in differentiation
of retinal cell types as well as retinal neurosphere formation
11
. Because the neural retina
3

differentiation pathway is closely tied to RPE differentiation, it is highly likely that polarized
conditioned medium, which contains a high quantity of PEDF, could also play a role in RPE
differentiation.
1.3: RPE Developmental Process
The RPE is closely linked to not only the health and proper functioning of the retina, but
developmentally as well in a process that is heavily regulated by specific transcription factors.
During RPE development, the neuroectoderm can form the eye field (a derivative of the neural
tube), at which point several early stage neural differentiation transcription factors such as Pax6
and other proteins such as Beta 3 Tubulin are expressed. Both the RPE and retina form from the
anterior part of the neural tube of the neuroectoderm called the diencephalon. The main first step
is when the eye field laterally splits into two optic vesicles forming the left and right sides. The
cells that constitute the optic vesicle are indistinguishable from each other, both molecularly and
morphologically, and express a multitude of transcription factors including Six3, Pax6, Rx1, and
Otx2, which are essential in initiation of eye development
12
. At this point, these cells are
capable of differentiating into the RPE, neural retina, or optic stalk, with the final determination
based on the activation or inhibition of different signaling molecules which regulate the
expression of transcription factors that drive the identity and differentiation of these cells
13
.
Each optic vesicle has a lens placode, made up of the surface ectoderm, occurring at E9.5 in mice
and 28 days of gestation in humans
14
. The lens placode has been shown to express fibroblast
growth factor (FGF) 1 and FGF2
15
while parts of the optic vesicles express FGF8, FGF9, FGF15
and the FGF receptors
13
. As development progresses, the lens placode invaginates upon itself,
forming the optic cup at E11.5 in mice and 31-37 days of gestation in humans, and eventually the
pit closes to form the lens vesicle. Retinal progenitor cells that develop into the retina or RPE
4

cells depending on which genes are expressed are derived from the optic cup. At this stage of
development, FGF signaling, which has been found to be activated by the Ras-Raf-MEK-ERK
signaling pathway, represses RPE differentiation and promotes neural retina differentiation.
When FGF2 signaling is inhibited, neural retinal differentiation has been found to be prevented,
and when transgenic mice were generated that express FGF9 in the presumptive RPE, the RPE is
transformed into an additional neural retina
16
. In contrast, RPE-inducing signals appear to be
coming from the surrounding extraocular mesenchyme, which also expresses activin, a member
of the transforming growth factor beta (TGF-β) family while the activin receptors are located in
the optic vesicle; when chick optic vesicles were cultured without the presence of extraocular
mesenchyme, RPE marker expression levels dropped while neural retina marker expression
levels increased. However, when activin was re-added into the culture system, RPE marker
expression levels were increased while neural retina-specific genes were downregulated or
repressed
17
. While signaling molecules from the surrounding cells guide the optic vesicle cells
toward an RPE fate by activating or inhibiting transcriptional regulators within the prospective
RPE, there are only a select group of transcriptional regulators that have been shown to be
required for specific RPE differentiation: Pax6, Mitf, and Otx1/2
17
. Mitf, which has been
shown to be responsible for the melanin-producing cells, including the RPE and melanocytes,
has been found to be expressed in the developing RPE. When Mitf was inactivated in mice and
quail, the presumptive RPE development was disabled and resulted in hyperproliferative
unpigmented cells that developed into the neural retina
18-20
. Otx1/2, which are initially
expressed in the optic vesicle have been found to interact with Mitf in the RPE nuclei to increase
melanocyte gene activation. Otx1/2 are transcription factors which play an important role in
anterior head formation, and whose expression becomes limited to the region of only the
5

presumptive RPE during the optic cup development, but remain active in adult RPE. Pax6
expression is found throughout the optic vesicle and in the developing optic cup, but is lost at the
later optic cup stage. It is hypothesized that Otx1/2 and Pax6 initiates specification of the RPE,
along with Wnt signaling, activating Mitf expression. Then, as Pax6 expression is decreased
during RPE development, Mitf with Otx proteins, could drive RPE differentiation without the
need for Pax6.
In addition to these 3 core genes, there are other molecules involved in subsequent steps
to further RPE differentiation, including cell cycle regulators which inhibit RPE proliferation, the
bone morphogenic proteins (BMPs) and sonic hedgehog (SHH). Although BMP2, BMP4, and
BMP7 are found in the ocular regions, their receptors are found mainly in the ventral part of the
optic cup
21
. When Noggin, a BMP antagonist is overexpressed, RPE marker expression levels
decreased, and instead, markers for the optic stalk were expressed, indicating that BMP signaling
helps organize the region where the neural retina, RPE, and optic stalk meet
22
. SHH signaling
plays a role in early eye field specification, as well as optic cup development
23
. When SHH
signaling was disturbed, it was found to disturb Otx2 expression and as well as pigment
development in the ventral RPE. When SHH was misexpressed, Otx2 was found to be
ectopically activated and resulted in the development of pigmented cells in the neural retina
24
.
1.4: RPE Cells in Age-Related Macular Degeneration (AMD)
The macula is the central part of the retina which contains more than one layer of
ganglion cells. Approximately 6 mm in diameter, the macula is dominated by rods and its center
foveola is comprised exclusively of cones, which is responsible for its high central visual acuity
25
. The accumulation of extracellular lipid and protein deposits called drusen is one of the main
6

features of age-related macular degeneration (AMD), one of the leading types of visual
impairment in the developed world. Drusen are concentrated in the macula and may be
associated with dysfunctional or death of RPE
26
. There are 2 late forms of AMD. The early,
dry form of AMD, characterized by the formation of drusen in which vision is not yet affected,
can promote the advanced forms of AMD which include geographic atrophy (GA) and choroidal
neovascularization. GA, the advanced dry form of AMD, is characterized by large regions of
RPE loss and secondary photoreceptor degeneration, particularly concentrated in the macula
region, causing blurred vision and central vision loss
27
. CNV, or the wet form of AMD, is
characterized by new, leaky blood vessels growing from the choroid through a break in the
Bruch’s membrane and under the retina causing photoreceptor loss
28
. CNV is also associated
with excessive vascular endothelial growth factor (VEGF) expression in the RPE, and treatments
for it include targeting VEGFA. In 2010, there were over 2 million patients over the age of 50
with late stage AMD, and the number of patients is expected to double by 2050 (National Eye
Institute).
AMD is a complex disease in which both genetic and environmental factors contribute to
its development. Some unavoidable risk factors include aging, family history, and gender, while
environmental factors include higher body mass index, smoking, cardiovascular disease, sunlight
exposure, hypertension, increased serum cholesterol, and nutrition
25
, with smoking as the
biggest non-genetic risk for AMD
29
. Studies have indicated that there is a strong correlation
between smoking and all forms of AMD, with the finding that smoking resulted in RPE
abnormalities
30-32
, likely stemming from the high amount of oxidant compounds found in the
cigarette smoke
33, 34
. There have also been findings of some genetic variants that have been tied
to AMD risk. A single nucleotide polymorphism (SNP) in complement factor (CF) H has been
7

found to be associated with advanced AMD
35, 36
. Other variants in the genes involved in the
complement cascade have been found to affect AMD risk including CFB, complement
component 3, and CFI
30
. Studies have also indicated that inflammation in response to the
development of drusen could be a part of AMD development. Genome wide association studies
have also shown there to be genetic variances in genes involved in lipid metabolism including
ABCA1, LIPC, CETP, and LPL
37, 38
. Notably, genetic variances have been found in genes that
are associated with oxidative stress, including a polymorphism in superoxide dismutase 2 as well
as mitochondrial DNA
39, 40
. Additionally, high levels of oxidative stress-induced damage have
been found in the retinas of patients with GA
41
, indicating that oxidative stress is a major
contributor to the loss of RPE.
1.5: Role of Oxidative Stress in AMD Pathogenesis
Oxidative stress has been attributed as one of the major factors in the development of
AMD
42, 43
. Susceptibility increases with age since there is an accumulation of damaging
reactive oxygen species (ROS) and a decline in antioxidants. ROS include free radicals which
contain one or more unpaired electrons in their outer orbits like superoxide anion (O
2
-
∙) and
hydroxyl ion (OH∙), which aim to reach a stable, non-reactive state by extracting electrons from
other molecules, thus rendering those molecules reactive, and causing an oxidative chain
reaction. ROS also include non-radicals like hydrogen peroxide (H
2
O
2
) and singlet oxygen (O
2
)
which contain their full set of electrons but are still in an unstable state
44
. These ROS can be
generated from various cellular sources such as the mitochondrial electron transport chain, which
accounts for 90% of our total oxygen consumption, the microsomal electron transport chain,
NAD(P)H oxidases, and lipid peroxidation
45
. Additionally, ROS production can be caused by
environmental factors like air pollutants, cigarette smoke, irradiation, and reperfusion injury.
8

The retina is a hotspot for the formation of ROS for several reasons. The retina is
exposed to high levels of cumulative irradiation, and it, as well as the adjacent RPE, contains
abundant photosensitizers which can add to accumulation of reactive oxygen intermediates
(ROI)
44
. Additionally, not only is the retina the highest consumer of oxygen relative to any
other tissue in the body, but the photoreceptor outer segment (POS) membranes are rich in
polyunsaturated fatty acids, which can oxidize easily, and is speculated to be one of the driving
factors of AMD development. One of the many functions of the RPE is that they phagocytose
and digest these diurnally shed POS, and it has been shown to be correlated to a nine-fold
increase in extracellular H
2
O
2
. However, as aging progresses, lipofuscinogenesis occurs when
the RPE cell’s lysosomal ablility is impaired. Lipofuscin consists of granular, dark yellow
pigments of lipids containing residues of lysosomal digestion of the POS. It accumulates in the
lysosomes which in turn increases the accumulation of ROS and free radicals
46, 47
, thus further
contributing to RPE destabilization. Lipofuscin, which also accumulates in the granules of other
post-mitotic cell types, consists of free radical-damaged fat and 2% protein as well as the bis-
retinoid fluorophore A2E, which originated from the release of all-trans-retinal when rhodopsin
is photoactivated
48
. A2E, a major component of lipofuscin, has been thought to play a role in
delaying RPE phagocytosis of POS, photoxidation, RPE cell damage, and inflammation
49
. Age
and lipofuscin have been linked to AMD; it was shown that 19% of RPE intracellular space is
occupied by lipofuscin in 81-90 year olds, as opposed to 1% in 10 year olds. Furthermore,
substantial amounts of lipofuscin have been found in the RPE cells of AMD eyes
50
.
Antioxidants such as catalase also decrease with age, but superoxide dismutase activity remained
unchanged (Liles et al., 1991).
9

Lipofuscin components are deposited from the RPE into the Bruch’s membrane, and
contributes to drusen formation as well as thickening of the Bruch’s membrane
49
. Proteins
frequently found in drusen from donor eyes of AMD patients consist of: αB-crystallin, βB1-
crystallin, complement component 9 (CC9), tissue inhibitor of metalloprotease 3 (TIMP3), βB2-
crystallin, vitronectin, Histone H2A2 (H2AE), clusterin (APOJ), βA3/βA4-crystallins, βS-
crystallin, and annexin 2
51

In addition to drusen formation, other key characteristics of AMD include basal laminar
deposits (BlamD) consisting of membranous material and collagen between the RPE and its
basement membrane as well as basal linear deposits (BlinD) within the inner collagenous layer
of the Bruch’s membrane. BlamD are initially thin and continuous in early AMD, but
accumulates as the disease progresses to include large drusen (>125 μm in diameter) as well as
BlinD, and eventually causes RPE atrophy
25
. The proteins associated with these various types
of deposits include vitronectin, clusterin, serum amyloid P, amyloid-β, complement
components, apolipoprotein E (ApoE), and have also been found deposited in other age-related
diseases like atherosclerosis and Alzheimer’s Disease
52
. The formation of drusen is a key
characteristic of the early dry form of AMD, which is also accompanied by thickening of the
Bruch’s membrane, but vision is not affected yet. As the amount and size of drusen increases
and inflammation progresses as a result of dysregulation of several complement regulatory
proteins (CRPs), including CFH, CFB, CF2, and CF3 (Bok, 1993), there is higher risk of the
disease progressing into advanced AMD or geographic atrophy. While there have been findings
that a diet containing lutein with zeaxanthin and omega 3 fatty acids could be partially protective
against AMD progression,
53, 54
, there is no treatment for the advanced form of dry AMD
55
.
10

In AMD, the affected RPE is the primary site responsible for both wet and dry forms of
the disease due to accumulation of drusen and lipofuscin in the macula, and damaging effects of
oxidative stress and inflammation. RPE degeneration and death lead to geographic atrophy
while RPE activation and secretion of VEGF lead to CNV.
1.6: Other RPE Degenerative Diseases
Similar to AMD in which there is central vision loss, Stargardt macular degeneration
(SMD) is characterized by having mutations in the ABCA4 gene, a gene encoding ATP binding
cassette that transports molecules across photoreceptors
56
. The mutation causes fluorescent
lipofuscin granules to be deposited in the RPE and leads to an increase in oxidative stress and
complement activation, thus causing RPE degeneration
57
. The failure of RPE cells to
phagocytose photoreceptor outer segments was found not only to cause drusen formation in
AMD, but it is also known to cause the autosomal recessive form of retinitis pigmentosa
58
.
Other examples of vision loss diseases due to dysfunctional, degenerative or RPE cell loss
include Best vitelliform macula dystrophy (BVMD), caused by a mutation in the Best1 gene
which is hypothesized to encode a Ca2+ activated Cl- channel membrane protein in the RPE cell.
BVMD is characterized by the deposition of lipofuscin-like material below and within the RPE
and can cause atrophy of the macula and/or choroidal neovascular
56, 59
.
1.7: Therapeutic Options to Restore Vision
Although there are some therapeutic approaches to alleviate the progression of these
diseases, many patients will eventually lose their vision; therefore, it is crucial to find ways to
regenerate RPE cells to replenish the degenerated/disrupted RPE cells, which may stop the
disease progression
60
. Animal models with retinal degeneration have shown successful
photoreceptor rescue and vision restoration after RPE cells were transplanted into the subretinal
space
59
. For example, in the Royal College of Surgeons (RCS) rat, there is a mutation in the
11

receptor tyrosine kinase gene MERTK, causing defective phagocytosis of the shed outer
segments of the rod cells by the RPE cells and resulting in retinal degeneration; however, vision
was shown to be rescued through transplantation of ARPE-19 cells (immortalized human RPE
cell line) into the subretinal space of the RCS rat. Autologous RPE can also be transplanted from
the periphery to the central retina and has been shown to partially restore vision in people with
AMD
61
. Although several studies have used ARPE-19 and fetal RPE cells in cellular therapy to
treat retinal diseases, immortalized cell lines do not fully differentiate into polarized RPE, are not
feasible for clinical work and fetal RPE cells are limited. Another potential source for RPE cells
is human induced pluripotent stem cells (iPSC) which, by forced expression of Oct4, Sox2,
Nanog and Lin28
62
or Oct4, Sox2, cMyc, Klf4
63
in somatic cells, can produce pluripotent cells
that can self-renew and generate cells of all three germ layers. Additionally, it has the advantage
of producing cells that are patient-specific and immune-matched
64
, but also the disadvantage of
carrying the genes that were defective from the start. If iPSC were to be used therapeutically,
then a new cell bank would need to be developed for each patient. Recent studies have also
shown that iPSC may retain some epigenetic markers of the original cell type (Kim et al., 2010),
and there have also been concerns about certain reprogramming methods which could cause
patient immune rejection
65
.
1.8: Introduction to Embryonic Stem Cells
Human embryonic stem cells (hESC) are a promising source for generating unlimited
amounts of RPE cells. Fertilization results in the formation of a zygote, which, after a series of
mitotic divisions, forms a solid mass of cell called the morula. Five days after fertilization, the
morula develops into the blastocyst which has an outer layer of cells called the trophoblast as
well as an inner cell mass, from which ESC are derived from. ESC possess some key properties
12

that make them promising for treating degenerative diseases or injuries. First, they are
pluripotent cells that are able to differentiate into any of the three germ layers, the endoderm,
mesdoderm, and ectoderm, which can give rise to almost any adult or fetal cell type. For
example, cells from the stomach or intestine originally came from the endoderm, while the
mesoderm can give rise to other cell types such as bone or connective tissue. The ectoderm,
which forms the outer layer of the embryo, develops into the surface ectoderm (epidermis, hair,
nails, etc.) and the neuroectoderm which is made up of the neural crest (peripheral nervous
system, melanocytes, etc.) and the neural tube (brain, retina, spinal cord, etc.). These
differentiated derivatives could be used in cell replacement therapy in various conditions
including Parkinson’s disease, leukemia, juvenile-onset diabetes mellitus, and cardiac infarcts
66,
67
. Second, ESC are capable of self-renewal, meaning that they are able to maintain the
undifferentiated state in vitro indefinitely. Studies have shown that they can be maintained in
vitro for at least one year, yielding around 250 population doublings, and still remain
phenotypically and karyotypically stable
68
. This is advantageous because in order to use these
hESC derivatives for experiments as well as for cell therapy, the ESC as the original source must
be scaled up. However, because ESC are predisposed to initiate differentiation, steps must be
taken to help maintain their undifferentiated state while they are expanded. Usually they are
cultured as adherent cells with mouse embryonic fibroblast (MEF) feeder cells which help
support hESC attachment, and act as a stem cell niche through supporting hESC growth and
survival by secretion of growth factors. Because using feeder methods is more technically
challenging, Matrigel, which is composed of mainly heparan sulfate proteoglycan, collagen IV,
and laminin can also be used to help hESC attach. When cells were grown on Matrigel-coated
13

plates in conditioned medium from MEFs, they were able to maintain their undifferentiated state
for over 130 population doublings
69
.
1.9: Differentiating and Expanding RPE from hESC
Several hESC lines, including H1, H7, H9 and hES3, have been shown to be able to
differentiate into RPE cells when basic fibroblast growth factor is removed from the medium.
hESC have the tendency to spontaneously follow the neural differentiation pathway when there
are no other types of inducing factors, thus spontaneously differentiating into RPE cells later on
as well. It can take up to 4-6 weeks for pigmented cells to first appear in the differentiating
cultures, and an additional 2 weeks for RPE cells to appear, which is evident by their hexagonal,
darkly pigmented shapes appearing in the mix with other cell types. The newly-formed RPE
cells make up only a small percentage of cells differentiated from hESC. Thus, many groups
have been studying ways in which they can speed up the rate of RPE differentiation as well as
increase the yield of RPE cells. RPE cells are usually harvested from hESC after 10 weeks of
differentiation, in order to allow time for the colonies to proliferate. Depending on the growth
pattern of the RPE colonies, they can be harvested manually by cutting if the colonies are more
spread out in a fan-like appearance, or enzymatically if the cells are grown in tighter colonies.
The growth pattern of RPE cells during the differentiation process is dependent on the hESC
line. Once separated from other differentiating cell types, the RPE cells are allowed to re-
develop their polarity (requiring approximately 4 weeks) before passaging again. Although there
are initially other differentiated cell types mixed in with the RPE, the purity of RPE cells
increases during passaging for several reasons: (A). other differentiated cells typically have
shorter lifespans than RPE and eventually detach from culture, (B). RPE cells are more adherent
to the culture plates; thus, the cells are enzymatically treated twice during passaging—once to
14

remove contaminating cells and a second time to detach RPE cells, (C). any remaining
differentiated cells are eventually outcompeted by the surrounding RPE cells. Ideally,
experiments utilize hESC-RPE from Passage 3-4. Thus, not only does it require over 2 weeks to
differentiate RPE cells ready to harvest, but it can also take at least another 3 months to utilize.
1.10: hESC-RPE in Clinical Work
There have been several studies performed on transplantation of hESC-derived RPE cells
to replace dying or degenerate RPE cells using animal models. The two main methods of hESC-
RPE cell delivery into the sub-retinal space are injection of RPE cells in suspension and
implantation of pre-polarized RPE cells grown on a special membrane which mimics the Bruch’s
membrane. Schraermeyer et al. in 2001 was the first group to show that transplantation of
murine ESCs into the subretinal space of RCS rats by suspension injection was able to rescue
photoreceptor degeneration, while years later, another group was able to show consistent
generation of RPE from hESC
70
. Soon, other groups demonstrated that hESC-RPE cells were
able to rescue photoreceptor degeneration in RCS rats when transplanted into the rats at 6 weeks
of age with the photoreceptor outer nuclear layer (ONL) still intact despite diminished ONL in
the control rats
71, 72
. The rescue effect was demonstrated with injection of 20,000 cells, and the
effect was even more evident at higher number injections of 50 – 100,000. They were also able
to verify the identity of these cells by positive staining for RPE marker genes RPE65 and
Bestrophin, demonstrate that the cells no longer proliferate, show that these cells remain viable
for over 100 days in vivo, that these cells settled into the subretinal space, and did not show any
signs of teratoma formation. These cells have also been used to rescue photoreceptor
degeneration in mice with the ELOVL4 mutation, a model for SMD, without any tumor
formation
71
.
15

Although the suspension-injection method of RPE transplantation is a surgically simpler
technique, it is difficult to assess where the injected cells integrate, or account for the cells that
are lost during attachment. Also, because the cells are initially not polarized, it is uncertain
whether they are fully capable of performing typical RPE tasks, or whether they are even able to
polarize. The implantation of pre-polarized RPE cells grown on a carrier substrate can help
alleviate some of the drawbacks of suspension-injection of RPE cells. Some proposed carrier
substrates include amniotic membranes, which have been shown to be an acceptable growth
substrate for RPE in vivo and in vitro by strong expression of RPE marker genes, cross-linked
collagen, and artificial membranes like poly-L-lactic-acid/poly-lactic-co-glycolicacid
(PLLA/PLGA) and surface-modified poly(hydroxybutyrate-
co-hydroxyvalerate) thin films
73
. Parylene, an inert membrane that is being studied by various
groups, is a mimic of the Bruch’s membrane with the proper porosity that allows the diffusion of
certain size molecules and the right surface which allows for proper RPE attachment. hESC-
RPE has been shown to properly attach, proliferate, and polarize on parylene substrate. Diniz et
al., showed improved survival of hESC-RPE cells when implanted as a monolayer on parylene
into the subretinal space in immunocompromised rats relative to suspension-injection RPE cells.
There were also no reports of teratoma formation
74
. The drawbacks to this method include the
introduction of a foreign material into the eye, as well as a relatively more invasive surgery
required for the transplantation.
FDA-approved clinical trials to replace RPE cells in retinal degenerative diseases using
hESC have also begun within the past few years. Schwartz et al., in 2012 was the first group to
show transplantation of hESC-RPE cells into human patients, one with dry AMD (Clinical Trial
Number NCT01344993), and another with SMD (Clinical Trial Number NCT01469832) in
16

ongoing Phase I/II clinical studies. Using hESC-RPE that were made under Good
Manufacturing Practices, the cells were xeno-free, pathogen-free, and clean of contaminating
pluripotent cells. The patients were injected with a cell suspension of 50,000 RPE cells and after
4 months, there were no signs of adverse effects such as abnormal growth, immune rejection or
teratoma formation in either patient. While most clinical studies have focused on transplantation
through suspension-injection of RPE cells, a previous clinical study indicated that transplanting
sheets of adult human allogeneic RPE cells into patients with exudative/wet AMD failed to
improve visual function
75
. Treatments for wet AMD typically targets VEGF or consists of
surgical ablation of the neovascular membranes, but another group will be starting a Phase 1
clinical trial later this year for hESC-RPE transplantation in patients with acute wet AMD and
rapid vision loss (Clinical Trial Number NCT01691261).
1.11: Introduction to Thesis Projects
My thesis was based on the idea that understanding the biology of RPE differentiation
and maturation would lead to improved methods to treat AMD patients with hESC-derived RPE.
For the first project in this thesis, I compared the resistance to oxidative stress of hESC-derived
polarized RPE to non-polarized RPE. I hypothesized that polarized RPE cells will not succumb
to oxidative stress-induced apoptosis as readily as non-polarized RPE. The results of this project
will be useful in clinical studies of hESC-RPE implantation because it will provide us with the
knowledge of whether suspension-injection of RPE cells is more sensitive to oxidative stress—
since non-polarized RPE serves as a model for freshly attached RPE cells following injection
into the subretinal space of AMD patients (a highly oxidative stressful environment)—relative to
polarized RPE grown and implanted on a membrane. If true, it would suggest that strategies
using sheets of polarized RE should show improved survival of the transplanted cells.
17

For my second project, I compared the effects of polarized, hESC-RPE-derived
conditioned medium (CM) on the differentiation of RPE cells from hESC. I hypothesized that
CM will increase the amount of RPE differentiation in hESC when compared to hESC
differentiated in regular, control medium. Then, I identified the growth factors secreted in CM
that play a role in augmenting RPE differentiation, as well as the mechanisms behind CM and
growth factors expanding the amount of differentiated RPE cells. These results will also be
significant in clinical research because it is a way to improve manufacturing by increasing the
amount of RPE differentiation from hESC, and it can also provide us some insight on the various
growth factors being secreted when polarized hESC-RPE are transplanted into the subretinal
space, as well as in in vitro cell cultures.

18

Chapter 2: Human Embyronic Stem Cell Derived Polarized Retinal Pigment Epithelial
Cells have Higher Resistance to Oxidative Stress-Induced Cell Death than Non-Polarized
Cultures
Abstract
Purpose: Oxidative stress mediated injury to the retinal pigment epithelium (RPE) is a major
factor involved in the pathogenesis of age-related macular degeneration (AMD), the leading
cause of blindness in the elderly. We hypothesize that polarized monolayers of human embryonic
stem cell (hESC) derived RPE are more resistant to oxidative stress-induced cell death relative to
non-polarized RPE cells. This work has clinical relevance since both cell suspensions and
polarized monolayers of hESC-RPE are being evaluated for their potential for cellular therapy in
patients with AMD.
Methods: RPE cells were differentiated from the hESC cell line H9. Polarized, non-
polarized/confluent, non-polarized/sub-confluent hESC H9-RPE cells were treated with H
2
O
2

and then analyzed for apoptotic cell death using TUNEL stain and cleaved caspase 3 levels, as
well as apoptotic signaling proteins phosphorylated (p)-p38, p-JNK, p53, Bcl-2, Bax. Proteins
involved in cell survival Akt signaling activation and antioxidant proteins superoxide dismutase
(SOD) 1, SOD2, and catalase were also studied.
Results: TUNEL stains revealed highest amount of cell death in non-polarized/sub-confluent H9
RPE and little cell death in polarized H9-RPE cells when all samples were treated with 600 μM
of H
2
O
2
. There were higher levels of p-p38 and p-JNK in treated non-polarized RPE relative to
polarized cells. Q-RT PCR and western blot results indicated higher expression levels of the
anti-apoptotic Bcl-2 in untreated polarized RPE relative to untreated non-polarized RPE, while
19

highest levels of pro-apoptotic Bax and cleaved caspase 3 fragments were found in treated sub-
confluent cells. Polarized RPE also had constitutively higher levels of Akt signaling, SOD1 and
catalase.
Conclusions: These results indicate that non-polarized RPE, especially sub-confluent cells, are
most sensitive to oxidative stress-induced apoptosis. Polarized hESC-RPE cells have higher
tolerance to oxidative stress relative to non-polarized cells, most likely due to their constitutively
higher pro-survival levels of Bcl-2 and p-Akt as well as SOD1 and catalase. These results
suggest that implantation of polarized hESC-RPE for treating patients with geographic atrophy
AMD should have better survival than injections of hESC-RPE in suspension.

20

Introduction
Retinal pigment epithelial (RPE) cells are found in vivo as a polarized monolayer of
highly pigmented neuroepithelial cells with an apical side facing the neural retina and a basal
side facing the Bruch’s membrane overlying the blood vessel-rich choroid with tight cell-to-cell
junctions. They supply the retina with nutrients, form the outer the blood retinal barrier, secrete
growth factors, absorb stray light, and phagocytose shed photoreceptor outer segments. Polarized
RPE also have several other defining characteristics, including the ability to transport substances
between the photoreceptors and the blood vessels
2
, apically localized microvilli and Na/K
ATPase, high melanocyte pigmentation, and well-defined tight junctions
6
.
When RPE cells are damaged or lost, photoreceptors also become impaired, resulting in
vision loss, as occurring in age-related macular degeneration (AMD), the leading cause of
blindness among the elderly worldwide
27
. There are two forms of late AMD: the advanced dry
form known as geographic atrophy, and the wet form known as choroidal neovascularization
(CNV). In the early dry form of AMD, there is an accumulation of extracellular lipid and protein
deposits called drusen between the basal side of the RPE and Bruch’s membrane. As the amount
and size of drusen increases, it promotes oxidative stress and inflammation, and the patient may
progress to one of the late, blinding forms of the disease. CNV is characterized by the formation
of new, leaky blood vessels which disrupts the Bruch’s membrane and RPE and grows into the
subretinal space
28
, which could also cause disruption of the macula and loss of central vision.
Oxidative stress is thought to contribute to the onset and progression of AMD. Aging
results in an increase of cellular exposure to oxidative stress, in which there is a build-up of
damaging reactive oxygen species (ROS) and a decline in antioxidant enzymes. ROS, which
include free radicals like superoxide anion (O
2
-
∙) and hydroxyl ion (OH∙), and non-radicals like
21

hydrogen peroxide (H
2
O
2
) and singlet oxygen (O
2
), can be generated in vivo from the
mitochondrial electron transport chain, lipid peroxidation, or NADPH oxidase as well as from
the outside environment through smoking, irradiation, or air pollution
76
. The RPE helps protect
itself from oxidative damage through antioxidant enzymes like superoxide dismutase (SOD) and
catalase, which are found in the RPE as well as the photoreceptors. SOD, which catalyzes the
quenching of the superoxide anion to produce H
2
O
2
and oxygen, exists in 3 isoforms co-factored
with a metal: Cu/Zn SOD1 is found in the cytoplasm, Fe/Mn SOD2 is located in the
mitochondria, and Nickel SOD3 is extracellular. It was found that in SOD1-deficient mice, high
levels of oxidative stress caused the formation of drusen, Bruch’s membrane thickening, as well
as CNV
77
. Catalase, an iron-dependent enzyme that scavenges H
2
O
2
has been found to be active
in the neurosensory retina as well as the RPE, but its activity decreases in eyes with AMD, and
further drops during aging
78
.
Because many AMD patients will eventually lose their vision, it is crucial to find ways to
replace lost RPE cells which may slow the disease progression
60
. hESC are a good potential
source for generating RPE cells for the treatment of many retinal diseases because they can be
indefinitely self-renewed and expanded to generate an almost unlimited source of young RPE
cells. There are currently two common methods for the delivery of ESC-derived RPE into the
sub-retinal space: injection of RPE cells in suspension and the insertion of a monolayer of pre-
polarized RPE cells. Suspension-injection of RPE cells into patients with dry AMD and
Stargardt’s macular dystrophy (SMD) is currently in Phase I/II clinical trials (Clinical Trial
Numbers: dry AMD, NCT01344993; SMD, NCT01691261) in studies sponsored by Advanced
Cell Technology
79
. Both methods have their advantages and disadvantages. While the RPE
suspension injection method is quicker and less invasive, there is no way of controlling where or
22

if they attach, or whether they can survive onto the damaged Bruch’s membrane once they do
attach. Also, there is the possibility of cell clumps forming that could damage the neural retina.
The other method consists of surgically inserting a patch of polarized RPE grown on a special
membrane designed to mimic the Bruch’s membrane into the subretinal space. While this
method is surgically more invasive and introduces a foreign material (the membrane) into the
eye, the main advantage is that the new RPE cells, which are replacing the damaged cells, are
already polarized
74
.
Although the RPE monolayer is polarized in vivo, in vitro RPE cell culture models need
time to develop their polarity. When hESC-derived RPE cells are seeded or passaged, they
initially lose their polarity and require at least 4 weeks to become fully polarized again.
Polarized cells have their own unique properties including tight cell-to-cell junctions, the
presence of pigmented melanosomes, which have been shown to protect against oxidative stress
80
, as well as the ability to secrete higher amounts of neurotrophic factors like pigment epithelial
derived factor (PEDF) relative to non-polarized RPE
9
. Thus, we hypothesize that they may be
more resistant to oxidative stress relative to cells that have not yet polarized.
We aim to determine whether polarized hESC-RPE cells are more resistant to cell death
relative to non-polarized hESC-RPE, and then compare the levels of activated apoptotic and cell
survival signaling pathways and antioxidants expressed between polarized and non-polarized
RPE. This project is significant because it can help determine whether it is more advantageous
to deliver pre-polarized RPE cells into the subretinal space of AMD patients relative to RPE
suspension injected cells—which must first adhere to the Bruch’s membrane and initially remain
in a non-polarized state—based on their respective abilities to resist oxidative stress.
23

Materials and Methods
H9 ESC Stem Cell Culture and Differentiation
H9 ESC cells (WiCell, Madison, WI) were passaged every 5 days. Stem cell colonies
were manually cut with the StemPro EZ Passage tool (Invitrogen, Carlsbad, CA), split 1:6 and
cultured in mTesR 1 medium (Stem Cell Technologies, Vancouver, BC, Canada) on 6-well
plates coated with hESC-qualified, lactate dehydrogenase elevating virus (LDEV)-free Matrigel
(BD Biosciences, San Jose, CA). Medium was changed daily with mTeSR 1 and spontaneous
differentiation was initiated after 7 days of culturing, which consisted of changing the mTesR 1
medium to 50/50 DMEM:F12, L-Glutamine, 15 mM HEPES (Corning Life Sciences,
Tewksbury, MA) supplemented with KOSR (Knockout Serum Replacement) (Life
Technologies, Carlsbad, CA). Medium was changed every 3-4 days during the 10 week
differentiation period.
RPE Cell Purification and Culture
After 10 weeks of differentiation, RPE cells were purified from the other differentiating
cell types and we referred to these newly purified RPE cells as “Passage 0” cells. To make
“Passage 0”, the differentiating cells that include RPE cells were washed with Dulbecco’s
phosphate-buffered saline (DPBS) without Ca
2++
or Mg
2++
(Corning Life Sciences, Tewksbury,
MA), and then incubated with 0.05% Trypsin, 0.1% EDTA in HBSS (Corning Life Sciences,
Tewksbury, MA) at 37
o
C for 5 minutes. Because RPE cells are more adherent to the plate, non-
RPE cells that detached first were pipetted out and remaining cells were washed with DPBS.
The second incubation with Trypsin-EDTA at 37
o
C for 5 minutes was performed in order to
detach RPE cells. Next, cells were counted and plated at a density of 1.3x10
5
cells/cm
2
onto
24

plates coated with phenol-red free, growth factor reduced, LDEV-free Matrigel (BD Biosciences,
San Jose, CA). Passage 0 cells continue to be cultured in the same differentiation medium as
described above. Within 4 weeks of culturing, the RPE cells develop polarity and remaining
contaminating cell types detach as they usually do not survive for such a long time period, and/or
are outcompeted by the abundance of RPE cells, resulting in purer cultures of RPE cells,
evidenced by an even layer of hexagonal-shaped, pigmented cells. These 4 week old cells were
then split into Passage 1, following the same protocol as described. Cells were passaged every 4
weeks, and were utilized for experiments up through Passage 4. Based on positive and negative
selection markers, the purity at Passage 4 is thought to be >95% (results not shown).
Plating Polarized and Non-Polarized/Confluent, Non-Polarized/Sub-Confluent RPE Cells
Polarized RPE cells from Passage 3 or 4 were grown on Matrigel-coated Transwell plates
with 0.4 µm pore size (Corning Life Sciences, Tewksbury, MA) until TER reached about 350
Ω∙cm
2
(approximately 4 weeks after seeding). TER of polarized RPE was measured using an
epithelial voltometer (EVOM) (World Precision Instruments, Sarasota, FL). The reading for the
“blank” was obtained by measuring an empty well containing only medium and Matrigel. Non-
polarized/confluent RPE cells were plated at a density of 1.3x10
5
cells/cm
2
in order to reach
confluency the following day, and non-polarized/sub-confluent RPE cells were plated at 5.0x10
4

cells/cm
2
to reach approximately 70% confluence the following day. Both confluent and sub-
confluent cells were also plated on Matrigel-coated plates and treated with H
2
O
2
the day after
plating.

25

H
2
O
2
Cell Treatment
Non-polarized confluent and sub-confluent cells were initially treated for 24 hours with a
range of H
2
O
2
concentrations from 0-1000 µM while polarized RPE were also initially treated
with a range of 0-1400 µM H
2
O
2
to determine which dosage was associated with initiation of
cell death. Non-polarized RPE were treated with 0 and 600 µM H
2
O
2
while polarized RPE were
treated with 0, 600, 1000 µM H
2
O
2
in serum-free media for most of the experiments. Depending
on the experiment, the length of treatment ranged from 15 minutes, 8 hours, or 24 hours. After
treatment, cells were harvested immediately for RNA for Q-RT PCR and protein for western
blot.
Quantitative RT-PCR
Post-H
2
O
2
treatment, polarized and non-polarized RPE cells were harvested with Buffer RLT
containing β-mercaptoethanol using the RNeasy kit (Qiagen, Valencia, CA), and homogenized
with the Qiashredder (Qiagen, Valencia, CA). RNA extraction was performed following the
manufacturer’s instructions, and reconstituted in 30 µl nuclease-free water. The quality of the
RNA was assessed by running a 1% agarose-ethidium bromide gel and checking for the strong
presence of the 28S and 18S ribosomal RNA bands, with no smearing and no contaminating
genomic DNA. One µl of RNA was then converted to cDNA following the instructions of the
ImPromII kit (Promega, Madison, WI). Q-RT PCR was performed in duplicate following the
instructions of the Light Cycler 480 SYBR Green I Master (Roche, Indianapolis, IN). CT values
were obtained for the genes of interest and GAPDH housekeeping gene. Fold changes were
calculated based on the average of 3 different biological samples. * indicates statistical
26

significance with p<0.05, and ** indicates p<0.01. The primer sequences of the genes of interest
are as follows:
SOD1: Forward: TCCATGTTCATGAGTTTGGAGAT, Reverse:
TCTGGATAGAGGATTAAAGTGAGGA
SOD2: Forward: AATCAGGATCCACTGCAAGG, Reverse: TAAGCGTGCTCCCACACAT
Bcl-2: Forward: AGTACCTGAACCGGCACCT, Reverse: GCCGTACAGTTCCACAAAGG
Bax: Forward: AGCAAACTGGTGCTCAAGG, Reverse: TCTTGGATCCAGCCCAAC
p21: Forward: CCGAGGCACTCAGAGGAG, Reverse: AGCTGCTCGCTGTCCACT
Western Blot
Post- H
2
O
2
treatment, cells were harvested with 1:1 Mammalian Protein Extraction Reagent
(Thermo Scientific, Waltham, MA) containing 1:100 addition of Protease Inhibitor Cocktail
(Sigma Aldrich, St. Louis, MO) and Laemmli buffer (Biorad, Hercules, CA) and 1:20 addition of
β-mercaptoethanol. Samples were boiled for 5 minutes and 10 μg of sample were loaded into
10% or 15% Tris-HCl gels (Biorad, Hercules, CA), depending on the protein size. The gel was
transferred onto PVDF membrane (Millipore, Temecula, CA) and incubated with the following
antibodies at 4
o
C overnight at the specified dilution in 5% non-fat milk with TBS and 0.1%
Tween buffer (TBST): p38 (Cell Signaling, Beverly, MA) (1/1000), p-p38 (Cell Signaling,
Beverly, MA) (1/1000), JNK (Cell Signaling, Beverly, MA) (1/1000), p-JNK (Cell Signaling,
Beverly, MA) (1/1000), Bcl-2 (Cell Signaling, Beverly, MA) (1/500), Bax (Cell Signaling,
Beverly, MA) (1/1000), Caspase 3 (Cell Signaling, Beverly, MA) (1/1000), SOD1 (Abcam,
Cambridge, MA) (1/1000), SOD2 (Abcam, Cambridge, MA) (1/1000), p53 (Millipore,
27

Temecula, CA) (1/1000), Akt (Cell Signaling, Beverly, MA) (1/1000), p-Akt (Cell Signaling,
Beverly, MA) (1/1000), p-PTEN (Cell Signaling, Beverly, MA) (1/1000). The following day,
the membranes were incubated for 1 hour with corresponding secondary antibodies anti-rabbit
peroxidase or anti-mouse peroxidase (Vector Laboratories, Burlingame, CA) at 1/5000 dilution
in 5% milk in TBST buffer before subjecting to ECL substrate (Thermo Scientific, Waltham,
MA) followed by exposure to film. Densitometry analysis of Western Blots was performed
using Image J software with N=3. Student T test was used to determine statistical significance; *
indicates p value < 0.05, ** indicates p<0.01.
TUNEL stain
Post-H
2
O
2
treatment, cells were fixed in 4% paraformaldehyde for 30 minutes, then washed with
PBS and permeabilized by incubating with 2% Triton-X100 in PBS for 15 minutes at room
temperature. Cells were washed 3x’s with PBS and then TUNEL reaction mix was applied to
the cells following the instructions of the DeadEnd TUNEL stain kit (Promega, Madison, WI).
Positive control for TUNEL staining consisted of RPE cells incubated with 2 units/ul of DNAse I
for 10 minutes after permeabilization with Triton-X100. Samples were mounted using
Vectashield Mounting Medium with DAPI (Vector Laboratories, Burlingame, CA). Images
were taken at 3 random overlapping fields using the 10x objective and the number of positively
stained green cells was counted relative to the total number of DAPI-stained nuclei in order to
obtain the percentage of positively stained cells in each field. The percentages were averaged
and student T-test was performed for statistical significance; * indicates p<0.05, ** indicates
p<0.01.

28

Results
2.4.1 Polarized RPE are more resistant to H
2
O
2
-mediated apoptosis
The various confluencies of H9-RPE cells were plated and then treated with increasing
dosages ranging from 200-1000 μM H
2
O
2
for 24 hours in order to gauge the ideal concentration
to analyze cell death. At 200 μM H
2
O
2
treatment, both non-polarized and polarized cells
appeared normal, and at 400 μM H
2
O
2,
sub-confluent cells appeared to be slightly stressed with
shrunken morphology, while confluent and polarized cells appeared to be unaffected (data not
shown). The result indicated that 600 µM H
2
O
2
demonstrated best differential amounts of cell
death with 1000 μM H
2
O
2
showing substantial cell death in polarized RPE. At 600 µM H
2
O
2
as
shown in Fig. 2.1, sub-confluent H9-RPE cells started to show rounding up of cells and cell
detachment while confluent cells showed focal cell detachment; however, the polarized H9-RPE
did not appear to be affected by the treatment. At 800 µM treatment, all sub-confluent cells, as
well as the majority of confluent cells, detached, while polarized cells showed focal alterations in
cell shape (data not shown), and at 1000 µM treatment, all non-polarized RPE detached while
polarized RPE began to show some detachment.
29

We wanted to determine next whether the increased amount of morphologically injured cells in
non-polarized cells was evidence of apoptosis. The non-polarized and polarized H9-RPE cells
treated at 600 μM H
2
O
2
were analyzed for cell death using TUNEL staining. For non-polarized
RPE in Fig. 2.2A, at 600 μM H
2
O
2
, despite many cells detaching from the plate, almost 100% of
all remaining sub-confluent cells stained positive for TUNEL, in contrast to a smaller number of
TUNEL-positive stained confluent cells. There were no TUNEL-positive cells detected in
30

treated polarized cells at 600 μM, but when the dosage was increased to 1000 μM H
2
O
2
, there
were many positive-stained cells (Fig. 2.2B). Cells were treated with DNAse I for the positive
control TUNEL stain, while the negative control excluded background fluorescence (Fig. 2.2C).
Quantitation of the data is shown in Fig. 2.2D. Results showed that there was a significantly
higher percentage of cell death resulting from oxidative stress in sub-confluent cells relative to
confluent cells (p<0.05) at 400 and 600 μM H
2
O
2
and relative to polarized cells (p<0.01) at 400,
600, and 800 μM H
2
O
2
as well as at 1000 μM H
2
O
2
(p<0.05). There was also a significantly
higher percentage of positive stained cells in confluent cells relative to polarized cells at 400 and
600 μM H
2
O
2
(p<0.05). At 800 μM H
2
O
2
, all sub-confluent cells detached and the remaining
confluent cells were 100% positively stained, but there was significantly less positive stained
polarized cells (p<0.01). At 1000 μM treatment, whereas all non-polarized cells detached, the
number of positive stained cells in polarized cultures began to increase. Caspase 3 is a major
regulator of cell death which, upon activation, executes apoptosis by catalyzing the cleavage of
certain cellular proteins at specific amino acid sequences. In Fig. 2.2E, we looked at protein
levels of cleaved caspase 3, which has a 19/17 kDA and 12 kDA band. Cultures were treated for
8 hours and harvested for protein. Among untreated cells, western blot showed polarized RPE
had significantly lower levels of cleaved caspase 3 relative to non-polarized cells while sub-
confluent RPE had the highest (p<0.05). Following treatment, there was a significant increase in
cleaved caspase 3 levels in non-polarized RPE relative to respective untreated controls (p<0.05).
Comparing among treated cells, polarized RPE at both treatments had lowest levels of cleaved
caspase 3 (p<0.01), while between treated non-polarized RPE sub-confluent RPE had
significantly higher amount of caspase 3 fragments (p<0.05). These results indicate that
31

polarized RPE are more resistant to oxidative stress-induced apoptosis relative to non-polarized
cells.
32

33

34

35

2.4.2 Polarized RPE TER drop corresponds to sudden increase in H
2
O
2
cell death
Additionally, we looked at the transepithelial resistance (TER) of the polarized cells after
treatment with increasing concentrations of H
2
O
2
. The TER was relatively stable at
approximately 350 Ω·cm
2
when treated with 400 and 600 μM H
2
O
2
, decreased slightly at 800
μM, and then dropped significantly at 1000 μM H
2
O
2
(p<0.05) (Fig.2.3A). When the decline in
TER was plotted against the percentage of cell death, the TER remained constant at
approximately 350 Ω·cm
2
until a sharp increase in cell death, which corresponded to a TER drop
to 150 Ω·cm
2
(Fig.2.3B). This suggests that the abrupt increase of cell death at 1000 μM H
2
O
2
,
which corresponded to the sudden drop in TER, was associated with the detachment of cells
from the monolayer. These results show that non-polarized cells are more susceptible to cell
death from H
2
O
2
stress relative to polarized cells, and the level of confluence plays a significant
role in the degree of cell death from oxidative stress. Also, in polarized cells, it appeared to
require nearly double the concentration of H
2
O
2
used on non-polarized cells to cause any
significant cell death.
36

2.4.3 Non-polarized RPE have higher levels of pro-apoptotic signaling pathways
We looked at apoptotic signaling pathways which were likely to play a role in non-
polarized RPE’s sensitivity to oxidative stress. We focused on the c-Jun NH-2 terminal kinase
37

(JNK) and p-38 mitogen-activated protein kinase (MAPK) signaling pathways which have been
found to be essential to mediate cell stress, including oxidative stress-induced apoptosis
81
. Fig.
2.4A indicates baseline levels of p-p38 in untreated polarized RPE were undetectable while there
were significantly higher levels of p-p38 in untreated non-polarized cells (p<0.05). In treated
non-polarized cells,

there was a significant increase in p-p38 expression relative to respective
controls (p<0.05), while there was no change in polarized RPE with either treatment. When
compared among treated samples, p-p38 levels in sub-confluent RPE were highest compared to
confluent (p<0.05) and polarized RPE at 600 and 1000 μM H
2
O
2
(p<0.01); polarized RPE at
both dosages had the least amount of p-p38. p-JNK followed a similar pattern to p-p38 levels in
treated RPE cells. Fig. 2.4B shows among the untreated control cells, p-JNK levels in polarized
RPE were undetectable, while sub-confluent cells had the highest p-JNK levels. When polarized
RPE were treated with 600 and 1000 μM H
2
O
2
, there was no change in p-JNK levels relative to
the polarized control. However, when non-polarized RPE were treated with H
2
O
2,
there were
significantly higher increases in p-JNK levels relative to respective untreated cultures (confluent,
p<0.05, sub-confluent, p<0.01)
.
Among treated RPE cultures, sub-confluent RPE had
significantly highest levels of p-JNK relative to polarized RPE at 600 and 1000 μM H
2
O
2
(p<0.01). These results show that the p38 and JNK apoptotic signaling pathways are more
highly activated in non-polarized RPE than polarized RPE when under oxidative stress.
38

39

JNK and p38 have previously been shown to phosphorylate and activate the tumor
suppressor factor p53. p53 is a transcription factor which can regulate apoptosis in response to
oxidative stress via members of the Bcl-2 family of proteins in the mitochondrial apoptotic
signaling pathway. Also, p53 is known to activate Cdk inhibitor from the Cip/Kip family,
p21
Waf1/Cip1
which is pro-apoptotic and responsible for cell cycle arrest. Interestingly, Fig. 2.5A
shows that even before H
2
O
2
treatment, among the untreated cultures, polarized RPE had the
lowest levels of p53 while untreated sub-confluent RPE had the highest expression (p<0.05).
Following treatment, when p53 levels in non-polarized cells were compared to untreated cells in
each corresponding group, we see p53 levels significantly increase (confluent, p<0.05, sub-
confluent p<0.01), but there was no change in polarized cells. Among the treated groups, sub-
confluent RPE had the most significant rise in p53 expression relative to polarized RPE at both
treatments (p<0.01). These results indicate that p53 expression is activated by H
2
O
2
stress in all
RPE types, but the activation in sub-confluent cells is strongest. This is significant because p53
is a key activator of the apoptotic process as well as cell cycle arrest via p21.
Treated RPE cells show expression levels of p21 to be similar to p53. Western blot
quantification indicated baseline levels of p21 in untreated non-polarized RPE were significantly
higher than untreated polarized RPE (confluent, p<0.05, sub-confluent, p<0.01) (Fig. 2.5B). Q-
RT PCR and western blot indicate non-polarized RPE had significantly higher p21 expression
(Q-RT PCR, p<0.05, protein quantification, p<0.01) after treatment relative to corresponding
untreated cultures while p21 in polarized RPE increased in 1000 μM-treated cultures (p<0.05) on
the protein level only. Comparing between treated cells, sub-confluent RPE showed
significantly highest increase in p21 expression after treatment relative to polarized RPE at both
dosages on the protein level (p<0.01) and only at 600 μM on the mRNA level (p<0.05).
40

Confluent cells had significantly higher p21 expression than 600 μM H
2
O
2
treated polarized RPE
(Q-RT PCR, p<0.05, protein quantification, p<0.01). These results indicate that non-polarized
RPE, particularly sub-confluent cells, are most susceptible to cell cycle arrest via p53-mediated
activation of p21. This data also shows that the stronger p38 and JNK activation signaling in
non-polarized RPE resulting from H
2
O
2
insult likely caused an increase in p53 expression.
41

42

The mitochondrial outer membrane permeability (MOMP), significant for the onset of
apoptosis via caspase activation, is highly regulated by a group of proteins in the B-cell
lymphoma (Bcl-2) family. Some of the pro-apoptotic proteins in the Bcl-2 family include BH3-
only domain proteins Bcl-2 associated X protein (Bax) and Bcl-2-associated killer (BAK),
which, when activated by intrinsic or extrinsic stimuli, can disrupt the MOMP and initiate
apoptosis. Anti-apoptotic protein Bcl-2 prevents MOMP by binding to Bax and Bak. The
MOMP is dependent on the activation of upstream signaling pathways. Other groups have found
evidence of JNK and p38 kinase directly phosphorylating and activating Bax and consequently
apoptosis
82
, in addition to direct activation of p53, which itself has been shown to
transcriptionally induce the expression of Bax, thus inhibiting Bcl-2’s anti-apoptotic ability
83
.
p53 may also inhibit Bcl-2 by transactivation of Cdc42, which activates a signaling pathway to
phosphorylate and inactivate Bcl-2 (Thomas et al., 2000).
Fig. 2.6 compares Bax mRNA and protein expression levels in treated and untreated
polarized and non-polarized RPE. Q-RT PCR results indicate that after 600 µM H
2
O
2
treatment
for 24 hours, Bax expression levels approximately doubled (p<0.05) relative to respective
untreated cultures in non-polarized cultures, while in polarized RPE, there was no significant
change in Bax expression levels at either treatment relative to untreated RPE. When comparing
treated cells among different cultures, Bax levels in non-polarized cells were significantly greater
than polarized cells at 600 µM H
2
O
2
(p<0.05). In contrast to the Q-RT PCR results, protein
expression levels of Bax were unchanged throughout the various samples with the exception of a
significant drop in Bax levels in sub-confluent cells after treatment despite the high number of
dying cells. This could be attributed to the majority of treated sub-confluent cells having already
passed the beginning stage of apoptosis since many cells have detached. Q-RT PCR, a more
43

sensitive assay, was likely able to detect the few remaining cells that still express Bax. To
determine why Bax levels actually dropped in the treated sub-confluent cultures, cells were
treated for 8 instead of 24 hours. Western blot indicated no change in Bax levels in polarized
RPE or confluent RPE after 8 hours of H
2
O
2
treatment. However, in sub-confluent cells, there
was a significant (p<0.05) increase in Bax expression levels after 600 µM H
2
O
2
treatment
relative to untreated sub-confluent cells, as well as relative to treated confluent RPE (p<0.01) and
polarized RPE at both H
2
O
2
dosages.
44

45

Taken together, these results suggest that non-polarized RPE are more sensitive to
oxidative stress-induced apoptosis resulting from higher activation levels of JNK and p-38
signaling pathways, which in turn could either directly initiate the MOMP through Bax or via
p53 activation of Bax in order to activate caspase 3.
2.4.4 Polarized RPE have constitutively higher levels of cell survival signaling
Next, we looked at possible cell survival signaling mechanisms responsible for polarized
RPE’s higher resistance to oxidative stress. We decided to look at the phosphoinositide 3-kinase
(PI3K)/Akt cell survival signaling pathway, since it has been shown to be activated in the RPE
following exposure to H
2
O
2
(Yang et al., 2006) to drive cell survival as well as increase SOD1
expression (Rojo et al., 2004). Thus, we expect polarized RPE to have higher levels of Akt
activation relative to non-polarized cells. RPE cells were treated with H
2
O
2
for 15 minutes in
order to best capture the PI3K/Akt signaling activation and harvested for western blot. Fig. 2.7A
shows that there was no significant difference in Akt levels between polarized and non-polarized
RPE, while phosphorylated Akt (p-Akt) was significantly higher in polarized relative to non-
polarized RPE (p<0.05) in baseline untreated samples. After H
2
O
2
treatment, all cell groups
experienced significant p-Akt increase relative to corresponding untreated cultures (confluent,
p<0.01, sub-confluent and polarized, p<0.05). The relative increase in p-Akt among the treated
polarized and non-polarized samples was similar to the baseline p-Akt found highest in polarized
RPE. These findings indicate that polarized RPE normally have a higher active Akt survival
pathway, relative to non-polarized RPE.
We also looked at phosphatase and tensin homolog (PTEN), a negative regulator for the
PI3K/Akt signaling pathway. While levels of PTEN inversely correlate with P- Akt,
46

phosphorylation of PTEN causes PTEN inactivation. Fig. 2.7B indicates that oxidative stress
treatment does not increase p-PTEN expression in any of the RPE culture types; however, there
are significantly higher baseline p-PTEN levels (p<0.05) in polarized RPE relative to non-
polarized RPE in untreated cells. These results indicate that polarized RPE are constitutively
more inclined to survive relative to non-polarized RPE, since polarized RPE had higher levels of
p-PTEN relative to confluent and sub-confluent cells.
47

48

Activation of the PI3/ Akt pathway has been shown to upregulate Bcl-2 expression
through the cAMP-response element-binding protein on its promoter (Pugazhenthi, 2000). Fig.
2.8A shows baseline levels of Bcl-2 protein expression to be significantly higher in polarized
RPE relative to confluent and sub-confluent cells. Comparing within cell types, Q-RT PCR
shows Bcl-2 expression levels significantly dropped by half (p<0.05) in treated sub-confluent
cells while western blot quantification confirmed a significant decrease in Bcl-2 expression
(p<0.05) Bcl-2 levels remained unchanged in confluent and polarized cells after treatment.
When compared between the three treated groups, the drop in Bcl-2 expression was significant
(p<0.05) between polarized and confluent RPE as well as polarized and sub-confluent RPE at
600 and 1000 µM H
2
O
2.
Fig. 2.8B indicates in polarized RPE, there was a higher Bcl-2:Bax
ratio relative to non-polarized cells. These results, which correlate with our p-Akt findings,
indicate that polarized RPE are protected from induction of apoptosis compared to non-polarized
cells due to their higher anti-apoptotic Bcl-2 levels likely resulting from higher Akt
phosphorylation.
49

50

2.4.5 Polarized RPE express constitutively higher levels of antioxidants
We wanted to see whether there were any differences in antioxidant expression levels
between polarized and non-polarized RPE cells. We were interested in looking at SOD levels,
since the PI3/Akt pathway has been shown to regulate SOD1 expression. SOD, one of the
antioxidant enzymes found in the RPE that convert superoxide radicals to molecular oxygen and
hydrogen peroxide, has two forms: the copper-zinc containing SOD (SOD1) located in the
cytosol, and manganese-containing SOD (SOD2) located in the mitochondria. We looked at
mRNA and protein levels of both forms. Q-RT PCR and western blot quantification indicate that
following 600 μM H
2
O
2
treatment, SOD1 levels dropped significantly (p<0.05) in sub-confluent
cells relative to untreated control, decreased insignificantly in confluent cells, and stayed
approximately the same in polarized cells at both dosages (Fig. 2.9A). Interestingly, SOD1
Western blots show significantly higher levels of SOD1 in untreated polarized RPE relative to
untreated confluent and sub-confluent cells (p<0.05), which had the lowest SOD1 expression.
Q-RT PCR and western blot quantification also indicated that among treated cells, sub-confluent
had the largest decrease in SOD1 expression levels (p<0.05) relative to confluent and polarized
RPE at both dosages. Treated confluent RPE also had significantly (p<0.05) lower SOD1
expression levels relative to polarized RPE treated at both dosages on the protein level but was
insignificant on the mRNA level.
In contrast, Q-RT PCR results showed that SOD2 expression in non-polarized cells
increased significantly (p<0.05) after 600 μM H
2
O
2
treatment relative to the same dosage in
polarized cells. Treatment at 1000 μM H
2
O
2
in polarized cells resulted in unchanged SOD2
expression compared to untreated control (Fig. 2.9B). However, unlike Q-RT PCR, SOD2
51

protein expression levels did not change in any of the cells post-treatment compared to
corresponding control cells.
Another antioxidant factor, catalase, was found to be significantly naturally higher in
polarized RPE relative to non-polarized RPE (p<0.05) (Fig. 2.9C). Following treatment, catalase
expression in polarized RPE remained high and unchanged while sub-confluent RPE
experienced a significant drop (p<0.05) relative to corresponding untreated cultures; the decrease
in catalase levels in treated confluent RPE was insignificant. Among treated cells, non-polarized
RPE expressed the lowest catalase levels relative to polarized RPE at 600 μM H
2
O
2
(p<0.05).
These results indicate that polarized RPE are strongly protected from oxidative stress which may
be a result of their constitutively higher levels of SOD1 and catalase. Additionally, SOD1 and
catalase expression levels remained high, even after higher dosages of H
2
O
2
, whereas they
decreased in non-polarized cells.
52

53

54

Discussion
Oxidative stress is thought to be one of the major factors contributing to the death of
RPE in AMD. Transplantation of hESC-derived RPE cells have the potential to replenish and
replace damaged or lost RPE cells by suspension injection of cells into the subretinal space or
placement of a pre-polarized RPE monolayer onto the Bruch’s membrane. Here we wanted to
compare the resistance to oxidative stress of hESC-derived polarized relative to non-polarized
confluent and sub-confluent RPE, since it will give us some indication of how the transplanted
RPE cells will react in the eye of an AMD patient—a highly oxidative stressed in vivo
environment—when they are pre-polarized relative to non-polarized, suspension-cultured
injected cells. The non-polarized confluent and sub-confluent cell cultures act as a model of RPE
cell behavior when cells are suspension injected into the sub-retinal space in that once the cells
adhere, they may end up sparsely populating the area (sub-confluent culture), or end up densely
populated and adjacent to one another (confluent culture). Our results here show clearly that
polarized RPE cells are more resistant to H
2
O
2
-mediated apoptosis than non-polarized confluent
and sub-confluent cultures.
These findings are highly relevant for RPE cell replacement therapy. hESC-derived RPE
cells are currently in Phase I/II clinical trials to replace damaged/degenerated cells in patients
with geographic atrophy (Clinical Trial Number NCT01344993), and SMD (Clinical Trial
Number NCT01469832) and another Phase I clinical trial is planned to begin in 2014 to study
hESC-RPE monolayer implantation in patients with wet AMD (Clinical Trial Number
NCT01691261). Schwartz et al., published their preliminary results in 2012 in which they
injected hESC-RPE in suspension into the subretinal space of one eye of one patient with AMD
and one with SMD. At the 4 month follow-up, they reported no sign of teratoma formation in
55

either patient as well as no abnormal proliferation, immune rejection of the cells, tumorigenicity,
or post-operative inflammation, but it is uncertain whether there will be immune rejection in the
long term
84
. The method of injecting RPE cells in suspension also has other caveats: it is
unknown how many cells, if any, can adhere to the Bruch’s membrane, the cells have a greater
chance of forming clumps, there is no control of where the cells end up, and it is uncertain
whether they can form a polarized monolayer. In order to minimize these concerns, another
method of RPE transplantation involves implantation of pre-polarized RPE cells grown on a
biodegradable or non-biodegradable substrate which mimics the Bruch’s membrane. Diniz et al.
in 2013 found that placement of pre-polarized RPE grown on a non-biodegradable,
biocompatible parylene membrane into the sub-retinal space of immunocompromised nude rats
resulted in greater RPE survival relative to cell suspension transplantation, indicating that the
state of RPE cells during surgery could play a role in determining the fate of the RPE after
implantation. Although surgery for monolayer implantation is more invasive than RPE cell
injection, the advantages of this method are that the cells are non-proliferative, already in an
intact monolayer, and already polarized, expressing key RPE markers. Our results indicate that
sub-confluent cells, which are most likely the initial state that the RPE cells find themselves in
vivo post-injection, are highly sensitive to oxidative stress, and if the surviving cells are able to
grow to confluency, they are still more likely to begin apoptosis relative to polarized RPE, since
the in vivo environment of an AMD eye is highly concentrated with reactive oxygen
intermediates. Thus, another advantage of using pre-polarized hESC-RPE cells during
transplantation would be that they are more resistant to oxidative stress.
H
2
O
2
was selected to induce oxidative stress in our study for several reasons. H
2
O
2
, in
addition to being highly permeable to the cell membrane and invoking high cytotoxicity, is one
56

of the major reactive oxygen species, and a precursor to tissue-damaging free radicals. Unlike
tert-butyl hydroperoxide (tBH), another reagent used to generate oxidative stress, H
2
O
2
is also
naturally produced in vivo. In normal human eyes, H
2
O
2
levels in the aqueous humor have been
reported to be in the range of 14-31 μM
85
while under pathologic conditions such as cataracts,
the H
2
O
2
levels can range from 33-500 μM in humans
86, 87
. The dosage concentrations we used
for H
2
O
2
were similar to pathological conditions in vivo. When we treated the H9-RPE cells for
24 hours at 400 μM, the sub-confluent cells appeared round and shrunken while confluent and
polarized RPE cells appeared to be normal; 600 μM treatment for 24 hours resulted in visible cell
death in confluent cells and cell detachment from sub-confluent RPE, while polarized RPE were
morphologically unaffected. H
2
O
2
was shown to induce necrosis at higher (700-1000 μM )
concentrations in ARPE-19 cells, while lower concentrations were able to show RPE apoptosis
88
. Our studies focused on cell death through apoptosis, since apoptosis is thought to be the
cause of RPE cell death in AMD. This is based on TUNEL stains of RPE from post-mortem
human eyes from AMD patients as well as increased caspase activation levels in the RPE
cultured from post-mortem older human eyes (>81 years old) compared to post-mortem younger
eyes (<41 years old)
89, 90
. Several other groups have also shown H
2
O
2
stimulating apoptosis in
RPE cells
81, 88, 91
.
Our results indicate that H
2
O
2
-treated polarized RPE were most resistant to oxidative
stress while sub-confluent RPE were most sensitive. Because cleaved caspase 3 (data not
shown) and Bax levels decreased after 24 hour H
2
O
2
treatment in sub-confluent cells, we treated
all cell types for a shorter timer period of 8 hours. We proposed that the drop in Bax and cleaved
caspase 3 levels following 24 hour treatment was attributed toward sub-confluent cells’ higher
sensitivity to apoptosis, so that during cell harvesting, the treated sub-confluent cells were in a
57

later stage of apoptosis. The results of the 8 hour treatment appeared to confirm this since there
was an increase in Bax levels in sub-confluent cells, but there was not much change in treated
confluent and polarized RPE relative to each corresponding untreated control. The high presence
of lower molecular weight products in treated sub-confluent RPE suggests further degradation of
caspase 3, indicating their higher sensitivity to oxidative stress. These results indicate that within
non-polarized cells, the degree of confluency plays a role in their resistance to oxidative stress.
One way anti-apoptotic Bcl-2 prevents the permeabilization of the mitochondrial outer
membrane from initiating caspase activation is by binding to the pro-apoptotic family member
Bax, which thwarts it from binding to the mitochondrial outer membrane and preventing the
release of cyctochrome C into the cytoplasm
92, 93
. If cytochrome C were released from the
mitochondria, however, it would eventually drive apoptosis by binding to the apoptotic protease-
activating factor 1 (APAF 1), thus inducing its conformational change into another structure
called the apoptosome, which cleaves and activates the effector caspases and ultimately
apoptosis
92, 93
. We showed polarized RPE having constitutively higher expression levels of Bcl-
2 that remained unchanged after oxidative stress treatment, while non-polarized RPE
constitutively had lower levels of Bcl-2 that dropped in confluent and (significantly in) sub-
confluent cultures after oxidative stress treatment. Polarized RPE possess some unique
properties that could account for their naturally higher Bcl-2 expression. For example, both
polarized fetal and hESC-RPE have been shown to secrete extremely high levels of PEDF, a
serpin family member, relative to non-polarized fetal and hESC RPE cells
9
. Mukherjee et al.,
found in 2007 that PEDF in conjunction with docosahexaenoic acid, a component of the
phospholipids in the photoreceptor outer segment, was able to stimulate Bcl-2 expression during
oxidative stress while decreasing pro-apoptotic Bax and caspase 3 activation. Another group
91

58

also noticed a decline in Bcl-2 expression in 200 μM H
2
O
2
treated fetal RPE cells. They found
the decline in Bcl-2 levels to be time-dependent and almost undetectable by western blot after 7
hours of treatment, but they did not specify whether these cells were polarized. The decline in
Bcl-2 in treated non-polarized cells could be another indication of increased apoptosis because
Bcl-2 is known to be digested by caspase 3, with the digested fragments traveling to the
mitochondria and releasing additional cytochrome c into the cytoplasm for more caspase
activation
94
. Conversely, upon oxidative stress treatment, Bax levels were shown to increase in
confluent and sub-confluent cells whereas Bax levels in polarized RPE cells remained
unchanged. These results show that non-polarized H9 RPE are more sensitive to cell death
relative to polarized RPE via the intrinsic apoptotic pathway.
After determining that polarized RPE are more resistant to oxidative stress and seeing
that it constitutively has higher levels of Bcl-2 which helps prevent MOMP and instigation of
apoptosis, we wanted to see if this could be attributed to any differences between polarized and
non-polarized RPE in cell survival pathways upstream of the Bcl-2 family. H
2
O
2
has previously
been shown to induce the reactive oxygen intermediate (ROI)-mediated PI3K which activates the
Akt/protein kinase B cell survival signaling pathway
95, 96
, and in doing so, it has also been
shown to protect ARPE-19 cells from oxidant-induced cell death
97
. The Akt pathway has been
found to be important in regulating various signaling pathways including cell growth,
proliferation, glycogen metabolism and cell survival
98
. Although there was an increase in p-Akt
in all treated cell types, we observed higher levels of p-Akt in untreated polarized RPE relative to
non-polarized RPE cells, indicating that polarized RPE are innately more protected.
Additionally, Akt activation has been shown to phosphorylate and inactivate pro-apoptotic
molecules including caspase 9, BAD, forkhead in rhabdomyosarcoma, and glycogen synthase
59

kinase
99-101
, as well as transcriptionally increase Bcl-2 expression
102
, which correlate with our
findings of naturally higher p-Akt and Bcl-2 levels in polarized RPE compared to non-polarized
cultures.
Activation of the JNK and p38 MAPK pathways upstream of the Bcl-2 family has
typically been shown to play a significant role in mediating apoptotic initiation resulting from
cellular stress. Our results showed, among treated cultures, highest p-p38 and p-JNK levels in
sub-confluent RPE, while levels in polarized RPE remained low. p-JNK has been shown to
inactivate Bcl-2 and Bcl-xl and induce apoptosis by releasing Bax from the protein 14-3-3 in the
cytoplasm after phosphorylation of 14-3-3
103
. Likewise, p-p38 MAPK was also shown to induce
Bax translocation into the mitochondria to induce apoptosis in nitric-oxide mediated cell death in
neurons
104
. Similarly, both JNK and p38 were shown to be activated as a result of H
2
O
2
-
treatment, and be responsible for Bax translocation into the mitochondria to induce apoptosis in
RPE cells
81
. p-JNK and p-p38 has also been shown to activate p53, which in turn,
transcriptionally activates Bax as well as other pro-apoptotic transcription factors like p21, which
is also responsible for cell cycle arrest. We found p53 and p21 expression levels were similar to
p-JNK and p-p38 levels: among treated cells, sub-confluent RPE expressed highest p53 and p21
levels whereas polarized RPE had lowest expression levels. This indicates that sub-confluent
cells, in addition to being most sensitive to oxidative-stress induced apoptosis, were also most
sensitive to cell cycle arrest mediated through p53. Taken together, these results indicate that
non-polarized RPE, and particularly sub-confluent cells, experienced higher levels of p53
activated, Bax-mediated apoptosis associated with activation of the JNK and p38 MAPK
pathway.
60

In all cells, there is a complex balance between pro-apopototic and anti-apoptotic factors
in deciding whether a particular cellular insult activates apoptosis or cell survival signaling. The
overexpression of anti-apoptotic Bcl-2 has been shown to inactivate JNK signaling and decrease
apoptosis caused by various reagents
105
. However, in our non-polarized RPE cells, despite the
increased p-Akt in treated cells which was expected to increase Bcl-2 levels, Bcl-2 levels
actually decreased and resulted in a rise in pro-apoptotic Bax instead. This could be attributed to
our findings that there is an increase in pro-apoptotic phosphorylation of p38 and JNK in
confluent RPE and an even larger elevation in sub-confluent RPE cells, whereas polarized RPE
cells did not experience significant increase in these factors after treatment.
We found that another reason why polarized cells are more resistant to oxidative stress-
mediated cell death is that they constitutively have higher levels of antioxidant proteins SOD1
and catalase. SOD1 expression, the copper/zinc form of superoxide dismutase, which catalyzes
the reaction of superoxide anion to H
2
O
2
and oxygen, was found by others to be increased as a
result of PI3K/Akt signaling through NF kappa B in PC12 pheochromocytoma cells
106
.
Additionally, SOD1 has also been shown to play a role in activating the Akt cell survival
pathway after transient focal cerebral ischemia in mice
107
, indicating here that the high levels of
SOD1 could correlate with the naturally high levels of p-Akt in polarized RPE. Various groups
have also found correlations between Bcl-2 and SOD. Overexpressing Bcl-2 was shown to
increase SOD activity in astrocytes,
108
and SOD knockout mice also had the same phenotype as
bcl-2 knockout mice
109
. While SOD2 levels were shown to increase slightly in all cells after
treatment, the level of increase appeared to be similar throughout all the samples, indicating that
SOD1 most likely plays a greater role in preventing oxidative stress-mediated apoptosis in RPE
cells. Our results demonstrated decreased catalase levels in confluent RPE and even more so in
61

sub-confluent RPE after H
2
O
2
treatment, which correlates to findings that H
2
O
2
decreases
catalase levels in via PI3K/Akt signaling
110
; however, our polarized RPE results showed that
catalase levels remain high after oxidative stress treatment, in spite of the Akt activation. These
results could mean that there are other factors specific to polarized RPE cells that help increase
catalase expression, such as Bcl-2 which was expressed highly in polarized RPE and has been
shown by others to increase catalase activity in PC12 cells
111
.
We found here that polarized RPE were most resistant to H
2
O
2
mediated cell death based
on having constitutively higher levels of cell survival signaling, anti-apoptotic protein Bcl-2
expression, and antioxidants compared to non-polarized RPE. Based on these results, we
conclude that it would be most beneficial to the survival of the transplanted cells as well as the
affected photoreceptors if the implanted RPE cells were pre-polarized before transplantation in
clinical studies.
62

Chapter 3: Retinal Pigment Epithelial Cell Conditioned Medium Contains Factors
Responsible for Increased RPE Cell Differentiation from Human Embryonic Stem Cells
Abstract
Purpose: Geographic atrophy (GA) in patients with Age-related macular degeneration (AMD) is
characterized by the loss of retinal pigment epithelial (RPE) cells in the macular region. Human
embryonic stem cells (hESC)-derived RPE have shown promise as a cellular therapy for GA;
however, the differentiation from hESC is a very lengthy process with an initially low RPE yield.
The purpose of this study was to determine whether conditioned medium (CM) collected from
polarized hESC-RPE promoted differentiation of hESC into RPE and to identify the putative
growth factors involved.
Methods: CM was collected from polarized hESC-RPE. Two hESC lines, hES3 and H9, were
differentiated into RPE by culturing in CM, growth factor (GF)-supplemented, or regular
medium for up to 8 weeks. Cells were harvested bi-weekly and Q-RT PCR was performed for
RPE-specific marker genes. A protein array was performed on CM as well as the media from
differentiating H9 cells to screen for possible GFs responsible for greater RPE cell appearance.
Results: Quantification of the pigmented areas in H9 and hES3 cells grown in CM relative to
regular medium indicated substantially higher amount of pigmented cells in CM. Additionally,
Q-RT PCR results showed significantly higher expression of RPE-specific genes in the H9 and
hES3 cells cultured in CM relative to the regular medium at 8 weeks of differentiation. A
protein array analyzing the conditioned media of differentiating cells as well as polarized RPE
indicated the GFs, neurotrophin 3 (NT3), neurotrophin 4 (NT4), and platelet-derived growth
factor AA (PDGF AA) to be possible candidates that increase the amount of differentiated RPE.
63

When cells were cultured in CM, PDGF AA, or a combination of NT3/NT4, there was
significantly higher expression of RPE-specific genes relative to the H9 cells regular medium at
8 weeks of differentiation. hES3 cells cultured in CM also showed higher RPE marker genes
than H9 cells differentiated in regular medium. We found CM, PDGF AA, or NT3/NT4
additions were able to increase RPE cell proliferation, while only CM and NT3/NT4 could
increase differentiation of RPE colonies from H9 cells.
Conclusions: These results indicate that polarized RPE cell CM can increase the amount of RPE
cells differentiated from H9 and hES3 lines via RPE proliferation and differentiation. CM, as
well as the media from differentiating H9 cells, contains certain secreted factors (NT3, NT4, and
PDGF AA) that can also cause an increase in the amount of differentiated RPE cells.

64

Introduction
Age-related macular degeneration (AMD) is the leading cause of blindness among the
elderly in the developed world, affecting over 30 million people world-wide
27
. The macula is
the photoreceptor-dense, central part of the retina which is responsible for clear, high acuity
central vision required for everyday tasks such as driving or reading. In geographic atrophy
(GA), a blinding complication of AMD, retinal pigment epithelial (RPE) cells, which are
essential to the health of the retina, become dysfunctional and die, thus resulting in degeneration
of retinal photoreceptors
112
. RPE cells are a polarized monolayer of highly pigmented cells
capable of absorbing stray light from the retina, have tight cell-to-cell junctions, and form the
outer blood retinal barrier. The apical side contains microvilli which faces the photoreceptors
and aids with phagocytosis their shed outer segments, while the basal side is lined by the Bruch’s
membrane and faces the choroidal blood vessels
2
. In addition to being able to transport water,
ions, and metabolic waste from the subretinal space into the blood vessels, RPE cells can deliver
nutrients such as glucose and fatty acids from the blood to the photoreceptors
1
. Also, RPE cells
are capable of secreting various types of growth factors which help maintain proper functioning
of the photoreceptors as well as the blood vessels.
The formation of drusen, extracellular lipid and protein deposits, between the RPE and
Bruch’s membrane is a key characteristic of the early dry form of AMD. As the amount and size
of drusen increases, eliciting inflammation, there is higher risk of the disease progressing into
advanced AMD or geographic atrophy, which is characterized by large regions of dying RPE and
photoreceptors. Another type of advanced AMD is choroidal neovascularization (CNV), or the
wet form of AMD, which is characterized by new, leaky blood vessels growing through the
Bruch’s membrane adjacent to the RPE layer
28
. While CNV is caused by excessive vascular
65

endothelial growth factor (VEGF) expression in the RPE, and treatments for it include targeting
VEGFA, there is no treatment for the advanced form of dry AMD
55
. Human embryonic stem
cells (hESC) are a potential source to generate RPE cells for cell replacement therapy (Haruta et
al., 2005). Early stage clinical trial evaluating the safety and efficacy of such therapies are
currently underway (Clinical Trial Numbers NCT01344993, NCT01691261). hESC are
pluripotent, can be indefinitely self-renewed
68
, and expanded to generated almost an unlimited
source of RPE cells.
There are two basic methods of differentiating RPE cells from hESC: spontaneous
differentiation and embryoid body (EB) suspension culture. For spontaneous differentiation,
differentiation initiates when basic fibroblast growth factor (bFGF) is removed from the medium,
since bFGF to support hESC renewal
113
. Dark brown pigmented spots that can develop into
RPE spontaneously appear approximately 4-8 weeks after removal of bFGF from the medium.
Another way of hESC differentiation into RPE cells is through embryoid bodies (EB). With this
method, the hESC colonies are cut into cell clusters and the EB are then suspension cultured for
a period of time from a few days up to a few weeks, followed by seeding onto an adherent
substrate such as laminin-coated plates, and then further cultured for RPE cell differentiation
which can take up to 8 weeks to appear
114
. Although these are the two main RPE differentiation
methods, many groups have devised variations of these methods in hopes of generating faster
and/or more RPE cells, such as usage of the addition of nicotinamide and activin, which has
neuroprotective cell survival qualities and helps direct RPE differentiation (Idelson et al, 2009).
In vivo, RPE cells are already known to release several trophic factors in order to
maintain proper retina integrity, such as angiogenic regulatory factors like VEGF or the
neuroprotective pigment epithelium-derived factor (PEDF). Additionally, RPE cells are known
66

to release platelet-derived growth factor (PDGF) for cell differentiation/maturation regulation, as
well as provide trophic support for photoreceptor cells and the choroid via secretion of PEDF,
ciliary neurotrophic factor (CNTF), and neurotrophin-3 (NT-3) among others
115
. It was
previously observed in our lab that when hESC were cultured in conditioned medium (CM),
there appeared to be a greater amount of RPE cells relative to hESC cultured in regular medium.
Thus, we hypothesize that the CM contains factors that are responsible for increasing the amount
of hESC differentiated RPE cells. This project aims to confirm that CM is responsible for
increasing the amount of RPE cells differentiated from hESC, determine which RPE cell-
released factors found in the CM are responsible for this increase in RPE differentiation, and
then analyze the mechanism as to how CM is able to increase RPE yield.
The derivation of RPE cells from hESC is a lengthy and inefficient process. With
spontaneous differentiation, it can take at least 4 weeks for the first pigmented cells to appear
and at least an additional 2-3 weeks for the first RPE cell patches to appear. Additionally,
previously studies reported only about 1% of the differentiated cells are RPE after 4-8 weeks of
culture with some variability between cell lines
70
. From a laboratory perspective, not only is the
waiting period very time consuming, but resource consuming as well, since during these weeks
of culture, large amounts of medium need to be changed, and the long time period between hESC
culture and the first appearance of RPE cells also runs a greater risk of cell contamination. From
a clinical trial manufacturing perspective, it is also not feasible for RPE cells to take so long to
appear because following RPE cell differentiation, the cells still need to be passaged and further
expanded into cell banks which require even more time. Therefore, this project is important
because it aims to help improve RPE cell differentiation methods and to better understand the
role of particular growth factors in RPE differentiation. Furthermore, by understanding the
67

components of the RPE conditioned medium, we will better understand the microenvironment of
the subretinal space after cell therapy.

68

Materials and Methods
CM Obtainment
Conditioned medium (CM) was collected twice a week from polarized H9-RPE cultured in either
DMEM:F12 medium supplemented with 10% KOSR, 1 mM glutamine (Corning Life Sciences,
Tewksbury, MA). The CM of cells from various passages (1-4) was pooled together. CM was
obtained from RPE cells no older than 2 months old per passage.
H9 Cell Culture and Differentiation
H9 ESC cells (WiCell, Madison WI) were passaged every 5 days. Stem cell colonies
were manually cut with the StemPro EZ Passage tool (Invitrogen, Carlsbad, CA), split 1:6 and
cultured in mTesR 1 medium (Stem Cell Technologies, Vancouver, BC, Canada) on 6-well
plates coated with hESC-qualified, lactate dehydrogenase elevating virus (LDEV)-free Matrigel
(BD Biosciences, San Jose, CA). Medium was changed daily with mTeSR 1 and spontaneous
differentiation was initiated after 7 days of culturing, which consisted of changing the mTesR 1
media to “regular” medium (DMEM:F12, L-Glutamine, 15 mM HEPES (Corning Life Sciences,
Tewksbury, MA) supplemented with KOSR (Knockout Serum Replacement) (Life
Technologies, Carlsbad, CA), CM (50/50 mixture of regular medium and CM), or regular
medium + growth factor (GF). GF’s were added at the following concentrations: NT3
(Millipore, Temecula, CA) (100 pg/ml), NT4 (Abcam, Cambridge, MA) (25 pg/ml), BDNF
(Abcam, Cambridge, MA) (15 pg/ml), KGF (ProSpec, East Brunswick, NJ) (450 pg/ml), BMP7
(ProSpec, East Brunswick, NJ) (230 pg/ml), PDGF AA (130 pg/ml) (Millpore, Temecula, CA).
Media was changed every 3-4 days during the 10 week differentiation period. Pigmented cells
69

began spontaneously appearing at around 4 weeks and usually developed into RPE cells at
approximately 6 weeks.
hES3 Cell Differentiation
Human embryonic stem 3 (hES3) cells were grown on mouse feeder cells in 35 mm dishes, and
were obtained from the USC Stem Cell Core Facility. The cells were mechanically cut into
small cell clusters with a 25-gauge ocular microsurgical knife and suspension cultured for 3 days
in neuronal differentiation medium (DMEM:F12 1:1 with 10% KOSR, 1xN2, 1xB27, 2 mM
glutamine, 1 ng/ml Noggin, 1 ng/ml DKK1, 5 ng/ml IGF) to form EB to guide the cells toward a
neuronal differentiation state
9
. DKK1 is a Wnt antagonist, which helps with the forebrain
development and Noggin is a BMP signaling antagonist which is involved in dorsal development
and head formation
116
. Following the 3 day suspension culture, the EBs were plated on to
Matrigel coated plates and cultured in the same differentiation media as H9 cells described
above.
RNA extraction
Differentiating hES3 and H9 cells were harvested weekly up to 10 weeks post-differentiation.
hESC were cultured in DMEM:F12 and CM:DMEM:F12, as well as in GF-supplemented
medium. Cells were lysed with Buffer RLT containing β-mercaptoethanol using the RNeasy kit
(Qiagen, Valencia, CA), and homogenized with the Qiashredder (Qiagen, Valencia, CA). RNA
extraction was performed following the manufacturer’s instructions, and reconstituted in 30 µl
nuclease-free water. The quality of the RNA was assessed by running a 1% agarose-ethidium
bromide gel and checking for the strong presence of the 28S and 18S ribosomal RNA bands,
with no smearing and little to no genomic DNA bands near the wells of the gel. One µl of RNA
70

was then converted to cDNA following the instructions of the ImPromII kit (Promega, Madison,
WI).
Q-RT PCR
Q-RT PCR was performed in duplicate following the instructions of the Light Cycler 480 SYBR
Green I Master (Roche, Indianapolis, IN). CT values were obtained for the genes of interest and
GAPDH housekeeping gene. Fold changes were calculated based on the average of 5 different
biological samples. * indicates statistical significance with p<0.05. The primer sequences of the
genes of interest are as follows. The expression levels of hESCs grown in untreated, CM, and
GF-treated was measured relative to undifferentiated hES3 or H9 cells. The primer sequences
for the genes studied are listed as follows:
PAX6: Forward: CGCTAATGGGCCAGTGAGGAGC, Reverse:
AAGGCCTCACACATCTGCGCG
PMEL: Forward: GGGCATCTTGCTGGTGTT, Reverse:
GAGAAGTCTTGCTTCATAAGTCTGC
RPE65: Reverse: CCAGATGCCTTGGAAGAAGA, Reverse
CTTGGCATTCAGAATCAGGAG
Bestrophin: Forward: TTCTATGTTGGCTGGCTGAA, Reverse: GCAGGTCCTGGTGCATCT
Beta 3 Tubulin: Forward: GCAACTACGTGGGCG ACT, Reverse:
CGAGGCACGTACTTGTGAGA
Vimentin: Forward: GACCAGCTAACCAACGACAAAA, Reverse:
GAAGCATCTCCTCCTGCAAT
71

PEDF: Forward: ACGCTATGGCTTGGATTCAG, Reverse:
GGTCAAATTCTGGGTCACTTTT
PDGF AA: Forward: CAGTCAGATCCACAGCATCC, Reverse:
CAGGCTGGTGTCCAAAGAAT
NT3: Forward: ACACGGATGCCATGGTTACT, Reverse: GAGGATCCTGAGGTCTGCAC,
NT4: Forward: GGAGGAGGTGCTGACAGG, Reverse: TGGGACTCAATTGGCACAC
BDNF: Forward: AACTTTGGGAAATGCAAGTGTT, Reverse:
AAGGATGGTCATCACTCTTCTCA
BMP7: Forward: TCCAAGACGCCCAAGAAC, Reverse: ACAGCTCGTGCTTCTTACAGG
FGF7: Forward: GCAAAGTAAAAGGGACCCAAG, Reverse:
TCACTTTCCACCCCTTTGAT
GAPDH: Forward: CCCCGGTTTCTATAAATTGAGC, Reverse:
CACCTTCCCCATGGTGTCT
Ki67: Forward: GAGAGTAACGCGGAGTGTCA, Reverse: TCACTGTCCCTATGACTTCTG
Protein Array
A quantitative protein array (Human Growth Factor Array Q1) from Ray Biotech which tested
for 40 growth factors was used on H9 cells. Unfrozen, undiluted medium samples were obtained
from the following cultures: undifferentiated H9-ESC, 2, 4, 6, and 8-week old differentiating H9
cells, Passage 3 polarized H9-RPE, polarized fetal RPE, and HEK 293 cells. Except for H9-ESC
medium which was collected after 1 day of culture, media was collected after 3 days of culturing
in all other samples. Arrays were performed following the instructions from Ray Biotech.

72

Cell Viability Studies
RPE cells were plated at 5x10
4
cells/cm
2
with regular medium, CM, addition of NT3/NT4 or
PDGF AA. Cells were harvested 2 days after plating for Q-RT PCR experiments. For
PrestoBlue (Life Technologies, Carlsbad, CA) analysis, the reagent was added to the cultures 2
days after plating according to manufacturer’s instructions, and incubated at 37
o
C for 2 hours
before assaying for resazurin to resofurin reduction. Assay was performed obtaining the
absorbance of the samples at 570 nm and 600 nm wavelengths using the Benchmark Plus
Microplate Spectrophotometer (Biorad, Hercules, CA), and normalizing the 570 nm absorbance
values to the 600 nm values. Percent reduction of PrestoBlue Reagent was calculated using the
following equation provided by Life Technologies: [((O2xA1)-(O1xA2))/((R1xN2)-
(R2xN1))]x100, where O=molar extinction coefficient of oxidized PrestoBlue reagent at 570 nm
(80586), O2=molar extinction coefficient of oxidized PrestoBlue reagent at 600 nm (117216),
R1=molar extinction coefficient of reduced PrestoBlue reagent at 570 nm (155677), R2=molar
extinction coefficient of reduced PrestoBlue reagent at 600 nm (14652), A1=absorbance of test
wells at 570 nm, A2=absorbance of test wells at 600, N1=absorbance of media only wells at 570
nm, N2=absorbance of media only wells at 600 nm.
Pigment Area Quantification
Pigmented Area Percentage: the percentage of pigmented area was calculated by dividing the
pigmented area by the total colony area. The colony area was determined by drawing a close-
fitting rectangle around each individual colony using Powerpoint, which also indicated the length
and width for area calculation. Six samples for each regular medium and CM were quantified
for differentiating H9 and hES3 cells.
73

Pigmented Colony Quantification: the number of pigmented colonies per well of a 6 well plate
were counted with N=6.
Immunostain
hES3 and H9 cells were fixed in 4% paraformaldehyde for 30 minutes, followed by PBS
washing. Bestrophin primary antibody (Abcam, Cambridge, MA) was added at dilution of 1:500
and incubated overnight at 4
o
C. The next day, cells were washed with PBS, and secondary
antibody was added at a dilution of 1/2000 and incubated at room temperature for 1 hour. Cells
were washed again with PBS and mounted using Vectashield Mounting Medium containing
DAPI (Vector Laboratories, Burlingame, CA).

74

Results
3.4.1 hESC grown in CM resulted in greater amount of pigmented cells
We tested the effect of CM on RPE differentiation in two different types of hESC lines,
H9 and hES3. Fig. 3.1A shows that at 8 and 10 weeks after differentiation, CM-grown H9 cells
had more pigmented cells relative to regular DMEM:F12-grown H9 cells. Similarly, hES3 cells
differentiated in CM also had higher amount of pigmented cells relative to hES3 cells grown in
regular medium. For both H9 and hES3 cell lines, it appeared that CM yielded not only a greater
number of pigmented colonies, but the size of the dark colonies was also larger than the ones
cultured in regular medium. Interestingly, differentiating hES3-RPE cell colonies had a more
fan-like appearance and appeared to be less pigmented than differentiating H9-RPE colonies
which looked like darker rods. The total pigmented areas were quantified in H9 and hES3 cell
lines (Fig. 3.1B). We tested the effect of CM on RPE differentiation in two different types of
hESC lines, H9 and hES3. Here CM referred to the collected CM diluted with DMEM:F12
medium at a 1:1 ratio. Fig. 1A shows that at 8 and 10 weeks of differentiation, CM-grown H9
and hES3 cells had more pigmented cells relative to cells grown in regular DMEM:F12 medium.
For both cell lines, it appeared that CM yielded a greater number of pigmented colonies and the
size of individual dark colonies was also larger. Interestingly, differentiating hES3-RPE cell
colonies had a more fan-like appearance and appeared less pigmented than the darker, rod-like
differentiating H9-RPE. The pigmented colony percentages were quantified in both cell lines
(Fig. 3.1B). H9 cells grown in CM resulted in 80% of the colony area being pigmented
compared to 25% when cultured in regular medium. For hES3 cells grown in CM,
approximately 40% of the colony area consisted of pigmented cells while regular medium
yielded 5%. These results indicate that although H9 cells overall differentiated 5-6 times the
75

amount of dark colonies relative to hES3, both lines cultured in CM still had higher amount of
pigmented area compared to its corresponding regular medium.
76

77

3.4.2 CM induced higher number of RPE differentiation from hESC
Next, we wanted to confirm that these pigmented cells are RPE cells. Bestrophin is an
RPE-specific calcium-activated anion channel found in the cell membrane, and here we used it
as a protein marker for mature RPE cells. Eight week old differentiating H9 and hES3 cells
cultured in regular and CM immunostained positive for Bestrophin in the cell membrane.
Additionally, we did not find any difference in the intensity level of Bestrophin expression
between samples cultured in regular media or CM in H9 or hES3 cells (Fig. 3.2).
78

79

The pigmented cells were also confirmed to be RPE cells by measuring the levels of
RPE-specific marker genes in 8 week old differentiating H9 and hES3 cell cultures. RNA was
harvested from 8 week old differentiating H9 and hES3 cells grown in regular medium and CM.
Q-RT PCR was performed on selected RPE-specific marker genes. In addition to Bestrophin,
these included premelanosome protein (PMEL) which is essential for pigmentation
117
and is
expressed in middle-stage to mature RPE cells, as well as the mature RPE marker RPE65 which
is a protein integral for regenerating photoreceptor visual pigment in the visual cycle. Fig. 3.3
indicates statistically significantly higher expression in PMEL, Bestrophin and RPE65 in hES
cells differentiated in CM relative to regular medium (p<0.05). These results confirm that the
mix of differentiating cell types grown in CM contain higher amounts of RPE marker-expressing
cells relative to the various cell types differentiated in regular media in both H9 and hES3 cells.
80

81

Q-RT PCR was used to measure gene marker expression levels under the assumption that
gene expression levels were an indication of the relative amount of differentiated RPE cells.
However, the increase in RPE marker gene levels of the hESCs cultured in CM could have come
from increased expression of those genes in each individual cell, and not necessarily the result of
an increase in total number of RPE cells. To test this, we plated equal numbers of RPE cells in
regular medium and CM and cultured them for one month until they were confluent and
polarized, as evidenced by their pigmented, hexagonal shapes. We harvested the cells for RNA
and compared RPE65 and Bestrophin expression levels between the RPE grown in CM and
regular medium using Q-RT PCR. If the increase in these RPE marker genes was due to an
increase in cell number, because the sub-cultured RPE cell populations were initially evenly
distributed between regular medium and CM, the gene expression levels should be the same
between the two media. However, if there was an increase in RPE65 or Bestrophin levels in
CM-cultured RPE cells, then we would conclude that CM causes an increase in RPE65 and
Bestrophin expression in each individual cell. Fig. 3.4 indicates that there is no difference in
RPE65 and Bestrophin expression levels between RPE cells cultured in CM and regular media,
indicating that the increase in RPE-specific genes in differentiating cells correlates to the total
amount of RPE cells in the plate and not to the individual cell.
82

We have shown that there is a higher amount of differentiated RPE cells in H9 and hES3
cells cultured in CM relative to regular medium. Next, we wanted to determine which proteins
secreted by RPE cells were responsible for the increased amount of RPE cells. At this point, we
decided to focus on the H9 cell line because it appeared to yield a higher amount of RPE cells
than the hES3 line, and because H9-RPE cells are being used by our group for a proposed
clinical trial.
83

3.4.3 Growth factor array reveals potential factors in CM and differentiating hESC
responsible for increasing amount of RPE yield
We performed a quantitative growth factor (GF) array (Quantibody Human Growth
Factor Array 1, Ray Biotech, Norcross, GA) on the polarized H9-RPE CM as well as media
obtained from various time points of H9 cell differentiation. Although the data from
differentiating H9 cell media may not necessarily reveal the relevant GFs in RPE CM that
increases the amount of RPE differentiation, it can let us know of potential growth factors that
could guide the process of eye field development and early RPE differentiation. Specifically,
this array was performed on the media secreted from the following H9 cells: undifferentiated H9-
ESC, differentiating H9-ESC cell cultures from weeks 2, 4, 6, 8, polarized H9-RPE, polarized
fetal RPE, and HEK293 cells. Polarized fetal RPE conditioned medium was used as a positive
control comparison relative to polarized H9-RPE, since we expect similar levels of GF secretion
between the two cells while human embryonic kidney (HEK) 293 cell conditioned medium was
used to further narrow down potential growth factors that may be RPE-specific. Finally, media
from 2, 4, 6, and 8 weeks of H9 ESC differentiation as well as polarized H9-RPE CM was used
in order to assess the different types of GFs (mainly neurotrophic factors) that are secreted as
some ESC cells undergo differentiation into RPE cells. Table 3.1 shows the results of the growth
factor array screen.
84

85

Out of the 40 GFs that the array screened for, we performed another screen to further
narrow down which GFs may play a role in increasing RPE differentiation based on their
secretion pattern during the various differentiation time periods from 2 to 8 weeks and in RPE-
CM (Fig. 3.5). We verified the GF secretion levels in differentiating cells and in RPE-CM by
using Q-RT PCR on each cell types’ harvested RNA with the understanding that there should be
a correlation between RNA levels and protein levels secreted into the media. The left panel
represents the results of the array for the selected GF, and the right panel represents the Q-RT
PCR results of expression levels of the selected gene in differentiating H9 ESC’s and polarized
RPE relative to H9 stem cells. From the GFs we screened, it appeared that gene expression
levels correlated with protein secretion levels. Bone morphogenic protein 7 (BMP7) and
keratinocyte growth factor (KGF)/fibroblast growth factor (FGF) 7 were secreted or expressed at
relatively low levels during the H9 cell differentiation process up until week 8, which happens to
be when RPE cells are most prevalent in the mix of various cell types. Furthermore, BMP7 and
KGF secretion or expression levels are highest in RPE-CM (Fig. 3.5A-B, 3.5C-D). We also
looked at secretion and expression levels of platelet derived growth factor AA (PDGF AA), the
homodimer of the PDGF A isoform (Fig. 3.5E-F), which was secreted highly throughout RPE
differentiation as well as in polarized RPE cultures.
In addition to the above-mentioned growth factors, we focused on neurotrophic growth
factors because they have been shown to induce neuroprotective signaling in RPE (Bazan et al.,
2008), and enhance RPE cell survival (Mukherjee, 2007). Fig. 3.5G-H, I-J show that
neurotrophin factor 3 and 4 (NT3 and NT4) gradually increased and peaked at 8 weeks post-
differentiation, which also coincides with the path of RPE differentiation in both growth factor
secretion levels and gene expression levels relative to H9 ESCs. Interestingly, NT3 secretion
86

was about 10 times more than NT4 during cell differentiation. However, NT4 secretion and
expression levels dropped in H9-RPE CM relative to differentiating H9 cells, while NT3
secretion and expression levels were not detected in H9-RPE CM. This indicates that although
NT3 most likely was not one of the growth factors in RPE CM responsible for increasing the
amount of differentiated RPE cells, it may still play a role in directing H9-ESC differentiation
into RPE, since it has also been found to play a role in RPE development
118
. Another
neurotrophic factor, brain-derived neurotrophic factor (BDNF), was not secreted as strongly
throughout the ESC differentiation process relative to NT3 and NT4, and was not found to be
secreted at 8 weeks post-differentiation or in RPE CM (Fig.5K-L). However, we decided to test
it as well as NT4 because together they have been shown to induce BMP7 expression in
embryonic neurons
119
.
87

88

89

We decided to culture H9 cells with various combinations of NT3 (200 pg/ml), NT4 (25
pg/ml), BDNF (15pg/ml) as well as with the addition of BMP7 (1000 pg/ml) and KGF (130
pg/ml) at 7 weeks of differentiation, and PDGF AA (130 pg/ml). These concentrations, with the
exception of PDGF AA, were determined based on the results of the quantitative growth factor
array. The concentration of PDGF AA tested was obtained from the results of an ELISA assay
that was previously performed. The growth factor combinations tested are listed in Table 3.2.
Each growth factor condition was cultured in triplicate in 24 well plates. While most of the
additions did not appear to increase the amount of differentiated RPE cells (data not shown),
there were some candidates that appeared to increase the number of pigmented cells beginning at
6 to 8 weeks of differentiation. These candidates initially included NT4/BDNF, NT3, NT3/NT4,
and PDGF AA but were narrowed down to just NT3/NT4 and PDGF AA after Q-RT PCR did
not show increased expression of RPE 65 and Bestrophin in the others.
90

3.4.4 Addition of NT3/NT4 and PDGF AA can increase RPE yield during hESC
differentiation
Fig. 3.6 shows higher amount of pigmented cells in cells cultured with NT3/NT4 and
PDGF AA after 6 and 8 weeks of culture relative to untreated cells. Although individual
additions of NT3 and NT4 did not appear to affect the amount of RPE differentiation relative to
the untreated medium (data not shown), it appears that NT3/NT4 treated cells, like CM, has more
RPE colonies as well as larger colonies compared to regular-media grown cells. PDGF AA
treated cells appear to only have larger colonies relative to H9 cells grown in regular media (Fig.
3.6A). The pigmented areas were again quantified (Fig. 3.6B) and the results indicated that the
91

CM (p<0.05), PDGF AA (p<0.01) and NT3/NT4 (p<0.05) addition produced significantly higher
pigmented area than regular medium, with CM and NT3/NT4 addition yielding the highest
percentage of pigmented cells.
92

93

Fig. 3.7 shows the Q-RT PCR gene expression results of PMEL, RPE65, Bestrophin, and
PEDF in H9 cells at 8 weeks of differentiation grown in media supplemented with NT3/NT4 and
PDGF AA. It appears that addition of PDGF AA did not increase PMEL gene expression while
cells grown with NT3/NT4 insignificantly increased PMEL expression (Fig. 3.7A). It appears
that NT3/NT4 and PDGF AA treated cells, like CM, had approximately 2 fold higher
Bestrophin, RPE65, and PEDF expression levels (p<0.05) relative to untreated cells (Fig. 3.7B-
D). Although PDGF AA treatment also expressed significantly higher levels of RPE marker
genes Bestrophin, RPE65, and PEDF, it seemed slightly less efficient increasing the amount of
RPE than CM or NT3/NT4 supplementation with approximately a 1.5 to 2 fold change in these
genes.
94

95

The next question that needs to be addressed is how CM and the addition of NT3/NT4
and PDGF AA are causing the increase of RPE cells in differentiating H9 ESC. We decided to
look at the following possible mechanisms of whether CM and GF additions help increase: (A)
the speed of RPE differentiation, (B) proliferation of the differentiated RPE cells, and (C) the
number of RPE colonies differentiated from hESCs.
3.4.5 Neither CM nor GF addition can increase the speed of RPE differentiation
To determine whether CM increases the speed of RPE differentiation relative to regular
media, we harvested RNA from differentiating hESC at 2 and 4 weeks post-differentiation, as
well as undifferentiated H9 and hES3 stem cells as our “Week 0”, to use as our reference. We
measured levels of Beta 3 tubulin and Pax6 which are expressed in the early stages of the eye
field development, as well as vimentin, an intermediate filament which is also expressed in
neural progenitors, and found no difference in expression levels between cells cultured in CM
and regular media as differentiation progressed in H9 and hES3 cells (Fig. 3.8). These results
indicate that CM does not appear to guide the differentiation process in the RPE cell direction
any more than regular media.
96

97

Next, we wanted to determine if RPE cells appear sooner when H9 and hES3 cells are
differentiated in CM vs. regular medium. As hESC differentiation progressed, RNA was also
harvested from 4, 6, and 8 weeks post-differentiation. In addition to PMEL, which is expressed
beginning at the middle stage of RPE differentiation, we also looked at mature RPE marker
genes, RPE65, Bestrophin as well PEDF gene expression. These later expressing genes showed
significantly higher expression levels of PMEL (p<0.05) at 8 weeks post differentiation in hESC
(Fig. 3.9A-B). Although Bestrophin and RPE65 first appeared in 4 week old cells in H9 cells,
there was no significant difference between the two culture media until weeks 6 (p<0.05) and 8
(p<0.01), in which H9 cells cultured in CM had significantly higher Bestrophin expression
relative to regular media (Fig. 3.9C, 3.9E). In hES3 cells, Bestrophin-expressing RPE cells do
not appear until 6 weeks (with RPE65 expressed slightly earlier at 4 weeks) and although there is
higher expression in cells grown in CM, the difference is not significant until 8 weeks (p<0.01)
(Fig. 3.9D, 3.9F). H9 differentiating cells experienced significantly higher PEDF expression
levels at 8 weeks in CM-differentiated cells (p<0.05) (Fig. 9D), but the increase in PEDF
expression was not statistically significant in CM-grown cells at 4, 6, and 8 weeks of
differentiation in hES3 (Fig. 3.9G-H). These results indicate CM may not speed up the
appearance of RPE cells, but may play a role in increasing the amount of RPE in later stages of
RPE differentiation
98

99

Because NT3/NT4 and PDGF AA addition increased the number of RPE cells like CM,
we also looked at whether these GF’s could cause RPE cells to appear sooner than regular media.
For PMEL, RPE65, Bestrophin, and PEDF, the GF additions did not significantly change their
expression levels relative to the regular media until 6 or 8 weeks of differentiation (Fig. 3.10).
PMEL expression levels did not vary greatly among the different GF treatments until 8 weeks,
when there was an insignificant increase in NT3/NT4 treated cells relative to regular medium
(Fig. 3.10A). RPE65 and Bestrophin expression first appear at 4 weeks with insignificant
differences throughout cells cultured in regular media, CM, NT3/NT4, and PDGF AA (Fig.
3.10B-C). However, there were significant increases in RPE 65 expression at 6 and 8 weeks in
CM and GF cultured cells (p<0.05). The expression levels of PEDF also show similar
expression levels between the four culture conditions at 2 and 4 weeks of differentiation, with
treated cells beginning to show higher PEDF expression levels at 6 weeks. At 8 weeks of
differentiation, all treated culture conditions show significantly higher PEDF expression relative
to cells grown in regular media (p<0.05) (Fig. 3.10D). These results indicate that neither of the
added GFs could speed up the rate of RPE differentiation relative to regular media, but were also
able to increase the amount of RPE cells.
100

101

3.4.6 CM and GF addition can increase RPE proliferation
Next we wanted to see whether the addition of CM, NT3/4, and PDGF AA helps increase
RPE cell proliferation. To do this, we plated RPE cells at a low density of 5x10
4
cells/cm
2
and
cultured them for 2 days in regular media, CM, NT3/NT4-, and PDGF AA- supplemented media.
Fig. 3.11A shows CM having the highest density of RPE cells relative to growth factor-treated
cells and regular media, and NT3/N4 and PDGF AA supplemented media also had higher
number of cells relative the regular media. We measured the gene expression level of cell
proliferation marker Ki67 in the 2 day old cells using Q-RT PCR and found the results correlated
with Fig. 3.11A; CM, PDGF AA, and NT3/4 supplemented media all had significantly higher
expression levels of Ki67 relative to untreated cells (CM, p<0.01, PDGF AA, NT3/NT4, p<0.05)
(Fig. 3.11B). Next, we looked at the reduction levels of resazurin to resorufin using the
PrestoBlue cell viability reagent as a measure of cell proliferation. As the reagent enters a living
cell, the resazurin (blue) is reduced to resofurin, which is red and fluorescent. Cells plated in the
4 different media conditions were grown for 2 days, incubated with PrestoBlue reagent for 30
minutes at 37
o
C and absorbance was measured to detect resofurin reduction. Fig. 3.11C
indicates cells grown in CM and with NT3/NT4 and PDGF AA additions had significantly
higher percentage reduction of resazurin than the regular media (p<0.01). Interestingly,
comparison to PDGF AA and NT3/4-supplemented RPE cells, CM had significantly higher
reduction of resazurin (p<0.05). These results indicate that one of the ways CM as well as
NT3/NT4 and PDGF AA increases the amount of RPE in differentiating cells could be by
increasing RPE cell proliferation.
102

3.4.7 CM and NT3/NT4 addition can increase number of RPE colonies during
differentiation
In addition to CM and NT3/NT4 playing a role in RPE cell proliferation, it appears that
CM and NT3/4 can also increase the number of differentiated RPE colonies relative to untreated
103

cells (Fig. 3.12). The number of RPE colonies per well were counted in the 4 different media
conditions, with N=6 for each group. Individual pigmented bodies were counted as single
colonies, regardless of the size. CM and NT3/NT4 cultured cells yielded more colonies per well
(averaging approximately 47 and 43 respectively) relative to regular medium (p<0.05) (29
colonies) and to PDGF AA-treated medium (24 colonies) and the pigmented areas appeared
slightly larger as well, indicating that they could play dual roles in increasing the number of
differentiated RPE colonies as well as proliferation of those colonies. Interestingly, although
PDGF AA has been found earlier to significantly increase RPE cell proliferation, it did not
increase the number of RPE colonies relative to cells differentiated in regular media with an
average of 24 colonies per well; in this case, the individual RPE colonies appeared larger only.

104

These results show that CM appears to play a role in the later differentiation stages in
both hESC cell lines; however, it, along with NT3/NT4 and PDGF AA additions did not appear
to speed up the rate of RPE differentiation. Instead, it seems that CM and NT3/NT4 are able to
increase the differentiation and proliferation of RPE cells while PDGF AA appears to mainly
play a role in RPE proliferation.
105

Discussion
RPE cells naturally secrete a large variety of growth factors in order to maintain proper
retinal and choroidal morphology and function, as well as their own integrity. By collecting the
medium from polarized H9-RPE cells and using that to differentiate hESC, we were able to show
that CM has the ability to increase RPE differentiation from hESC. Although we did not
measure growth factor secretion differences between the different passages or between the age of
the cells, another group
61, 115
has shown that in fetal RPE, there are significantly different trophic
factor secretion levels between passages as well as between the age of the cells, namely in
secretion of bFGF, VEGF-A, and PEDF.
Two different lines of hESC cultured in CM, hES3 and H9, were able to show a higher
amount of pigmented cells compared to cells grown in regular medium. These pigmented cells
were identified to be RPE cells by expressing key RPE markers such as Bestrophin, RPE65,
PMEL, and PEDF, with hESC cultured in CM expressing higher levels of these markers, which
was indicative of the greater amount of RPE cells. While both cell lines experienced an increase
in RPE cells when grown in CM, H9 cells were constitutively able to differentiate a greater
amount of RPE than hES3 cells, even when hES3 cells were grown in CM, illustrating the
differences between hESC lines. H9, as well as the H1 cell line, have been shown to be some of
the most efficient lines at differentiating RPE cells
61
. hES3 and H9 cells also followed different
differentiation protocols. hES3 cells formed EB in a neuronal medium supplemented with
(Noggin, DKK-1, IGF-1) for 3 days to direct the cells to a neural fate, modeled after Lamba et
al.,
120
and then switched to our differentiation medium, while H9 cells were directly switched to
differentiation medium at 7 days old. From our experience, this appeared to decrease the high
amount of cell death that hES3 cells experienced if they were directly switched to differentiation
106

medium. These findings illustrate the varying capability levels that different hESC lines have in
terms of RPE differentiation as well as their individual responses to various means of
differentiation.
It is known that hESC will spontaneously differentiate into RPE cells on their own, but
RPE differentiation of RPE cells can be very time-consuming and result in a low yield of RPE
cells. After 4-8 weeks of culture, the efficiency of pigmented cell formation consisted of ~1%
70
. Thus, many groups have looked into various ways of differentiating them from hESC in
hopes of speeding up the process and/or increasing the yield of RPE. Wnt and Nodal antagonists
which induced retinal progenitors from which RPE cells derive was shown to result in 38% of
hESC being pigmented after 8 weeks of culture
121
. Another group showed the addition of
nicotinamide and Activin A resulting in 70% of ESC clusters to be pigmented at 8 weeks
59
.
With RPE CM, we measured the pigmented area in each dish, and determined it was about 4-6
times the size of the cells grown in regular medium in H9 and hES3 cells.
The pigmented area of H9 cells grown in CM was measured at 8 weeks to be
approximately 4 times higher than cells grown in regular medium, which was in contrast to gene
expression level fold changes. RPE65, Bestrophin, and PEDF levels were approximately twice
as high in CM cultured H9 cells compared to regular medium at 8 weeks. There is a similar
occurrence in hES3 cells. While the pigmented area of hES3 cells cultured in CM was
approximately 6 times greater than cells grown in regular medium, PMEL, Bestrophin, and
RPE65 expression levels were about 4 times greater in CM. These results suggest that not all
cells in the outlined pigmented areas are RPE and that contaminating cells differentiating down
other pathways are also included; it may be possible that there are more contaminating cells in
the larger pigmented areas.
107

From the results of the growth factor array, non-neurotrophic factors BMP7, KGF and
PDGF AA initially appeared to have potential in increasing RPE differentiation. Both BMP7
and KGF, despite minimal secretions during RPE differentiation, experienced a spike in H9-RPE
CM and fetal RPE CM while remaining relatively low in HEK293 cells. BMP7 has also been
found to be expressed in the surface ectoderm over the optic vesicle, the mesenchyme, as well as
the presumptive RPE during eye development while, Geiger et al., in 2005 showed KGF having
a cell proliferative effect on ARPE-19 cells via the MAPK p42/p44 pathway at 50 ng/ml
treatment. However, when we added KGF to H9 cells that had been differentiating for 7 weeks,
there did not appear to be an increase in RPE cells relative to regular medium, albeit our dosage
was lower at 400 pg/ml, in accordance with our growth factor array results. Although PDGF AA
appeared to be highly expressed throughout the differentiation process as well as in RPE CM, it
was selected because it has been shown to play a role in RPE cell proliferation (Andrews, 1999)
and to be an autocrine growth regulator in RPE cells in that it secretes PDGF and also has PDGF
receptors (Campochiaro et al., 2004). We determined out of these non-neurotrophic growth
factors, only PDGF AA was capable of increasing the amount of differentiated RPE cells.
NT3, NT4, along with BDNF and nerve growth factor (NGF) belong to a group known as
neurotrophins, a subset of neurotrophic factors that activate signaling pathways via the Trk
family of receptor tyrosine kinases
122
and are known to have survival and differentiation-
inducing effects in epithelial and neuronal populations. BDNF along with its receptor TrkB has
been shown to promote differentiation of the retina as well as differentiation and survival of RPE
in Xenopus
123
while NT3 has been shown to promote neuroepithelial differentiation as well as
survival of retinal ganglion and amacrine cells
124
. Although NT4 has been shown to bind to the
TrkB receptor and NT3 has its preferred TrkC receptor to initiate signaling (and does bind to
108

TrkB at a lower affinity) for neuron survival
125

126
there have not been any reports of the two
neurotrophins working synergistically to promote differentiation, as was shown here. Despite
previous groups showing the expression of TrkB in RPE cells
127
, surprisingly, addition of BDNF
did not appear to increase RPE differentiation, not even in conjunction with NT4.

After determining that CM, along with GF PDGF AA and NT3/NT4 were capable of
increasing the amount of RPE during differentiation, we were interested in determining the
mechanism by examining whether these factors increased (A) the speed of RPE differentiation,
(B) RPE proliferation, or (C) the amount of RPE colonies.
Neither CM or our tested GFs gave any indication of speeding up the rate of RPE
differentiation from hES3 or H9 cells with RPE cells appearing at approximately 6 weeks of
differentiation in both CM and regular cultures, but with significantly higher levels of RPE cells
in CM cultures. Several groups have been able to find ways to differentiate hESC so that RPE
cells appear earlier than 6 weeks. One way was by first generating spherical neural masses
(SNM) which develop neural tube-like structures during passaging from which RPE cells
develop; these SNMs are expandable without the loss of RPE differentiation ability, and are
easily frozen and thawed
128
. RPE generation from SNMs requires approximately 2 weeks,
128

thus speeding up the typical 6 week differentiation process from hESCs. Perhaps the most time
and yield efficient RPE differentiation was reported recently through using retinal-inducing
factors IGF1, Noggin, DKK1 and bFGF, as well as the addition of Activin A, nicotinamide and
SU5402, and vasoactive intestinal peptide at specific points of differentiation , with an efficiency
of approximately 80%, and the ability to isolate RPE cells at 14 days
129
. It would be interesting
to see whether the quantity of these quick-differentiating cells with the above-mentioned factors
could be further increased in conjunction with the CM components we identified as well.
109

We showed that one of the ways that H9-RPE CM increased the amount of differentiating
RPE cells was through increasing RPE proliferation. Similarly, RPE CM from post-mortem
human eyes have also been shown to stimulate cell proliferation in human eyes, astrocytes, and
corneal fibroblasts
130
. In our studies we found CM’s proliferative ability to be attributed to one
of its components, PDGF AA, which was found to be expressed highly throughout the 8 week
differentiation process as well as in CM, and was previously found to increase RPE proliferation
as well act as an autocrine stimulator for RPE cells
131, 132
. Leschey et al., also found that the
following growth factors, insulin, PDGF, basic and acidic FGF, epidermal growth factor (EGF),
insulin like growth factor 1 (IGF-1) were all potent stimulators of RPE proliferation by
themselves as well as in any combination with each other, indicating that they work
synergistically with each other. Interestingly, our growth factor array did not pick up on
secretion of bFGF or IGF-1 in RPE CM, and EGF was barely expressed in H9-RPE and fetal
RPE CM.
CM’s ability to increase RPE proliferation is also advantageous because it could shorten
the amount of time needed in RPE polarization during RPE expansion. During passaging, RPE
polarity is initially lost. It can take up to a month for RPE cells to polarize, requiring RPE cells
to first become highly confluent before polarization begins. Therefore, by increasing the rate of
cell proliferation, RPE cells grown with CM can reach confluency quicker than those grown in
untreated media, thus lowering the amount of time required for RPE polarization to occur.
We determined that another way CM was able to increase the amount of RPE was by
increasing RPE differentiation from H9 cells. This was indicated by an increase in pigmented
colonies in 8 week old H9 cells relative to regular medium. In contrast, the addition of solely
PDGF AA showed no increase of RPE colonies, but larger individual colonies, indicating PDGF
110

AA’s bigger role in RPE proliferation than RPE differentiation. The particular growth factors
responsible for CM’s ability to increase RPE differentiation are still unclear. NT4, whose levels
peaked at 8 weeks of H9 differentiation, was secreted at relatively low levels in H9-RPE CM,
and when added to the regular medium, did not appear to increase the amount of differentiated
RPE cells. However, while NT3 (which was not detected in H9-RPE CM) by itself also did not
appear to increase the amount of pigmented cells during H9 differentiation, a combination of
NT3/NT4 significantly increased the amount of RPE cells by increasing the number of
pigmented colonies during H9 differentiation. In E11.5 mice, NT3 was found to be expressed in
the developing RPE, but expression gradually decreased as pigmentation developed, which
correlates to our lack of NT3 expression in the CM of polarized H9-RPE
118
. These findings
indicate that NT3 may play a role in inducing early RPE differentiation, as indicated by an
increase in the amount of differentiated RPE from H9 cells, but once the RPE cells are
differentiated, its expression wanes. Because the combination of NT3 with NT4 appeared to
result in high numbers of large, differentiated RPE colonies, it is likely that they work in
conjunction to drive RPE differentiation and proliferation.
In addition to increasing RPE differentiation from hESCs, H9-RPE conditioned medium
may also have the potential to be protective of RPE and retinal cell types. We did not find KGF
able to increase RPE differentiation from H9 cells, but it may play a more direct role in
protecting RPE from cell stress. KGF is a mitogenic factor that has been shown to thwart
permeability increases in epithelial
133
and endothelial
134
barriers in a variety of cell types
resulting from oxidative stress. In the RPE cell line ARPE-19, KGF has been shown to alleviate
DNA strand breaks and encourage DNA repair as a result of H
2
O
2
treatment
135
, indicating that it
111

could be an important component of RPE CM that helps alleviate the cellular effects of oxidative
stress.
PEDF, which has been shown to be protective of photoreceptors, and to be antisenescent
of premature RPE, has also been shown to be secreted in extremely high concentrations in
polarized hESC-RPE cells
9
and it is likely one of the factors in CM that helped increase the
amount of RPE differentiation. The trophic factors found in RPE CM were also shown to
decrease retinal apoptosis and retinal cytotoxicity in degenerating porcine retina as well as
human retina from AMD and non-AMD patients. When one of the trophic factors, PEDF, was
neutralized in RPE-CM, its effectiveness was lowered in reducing retinal apoptosis; thus, the
group concluded that PEDF, which was found to be extremely highly secreted in fetal RPE,
could be one of the factors in fetal RPE-CM that helps preserve the retina
115, 136
. We tried
adding PEDF recombinant protein at a concentration of 100 ng/ml to differentiating hES3 cells
for 8 weeks, but there did not appear to be any difference in pigmented cell levels relative to
regular medium (data not shown). Since we have shown that PEDF is secreted at lower levels in
non-polarized RPE and only secreted in larger amounts in polarized RPE, it is a late factor which
may play an important trophic role in RPE that has already been differentiated.
While there are many methods currently available that increase RPE yield during hESC
differentiation, CM appears to most convenient and versatile. In labs that already work with
polarized RPE cells, CM is easily obtainable and economical with no additional growth factors
needed, has previously been shown to be heat and pH stable
130
, has the ability to spontaneously
increase RPE differentiation and RPE proliferation, and has been found to contain components
that protect the RPE as well. Additionally, with its ability to enhance RPE proliferation, it can
help with RPE cell expansion during passaging by cutting down the time needed for RPE
112

polarization. The components may hold promise in improving the manufacturing process of
RPE differentiation from hESC.

113

Chapter 4: Summary and Future Directions
Summary
Polarized RPE are responsible for many activities required for visual function. Since
photoreceptors do not have a direct blood supply, the RPE regulates the flow of molecules and
nutrients from the choroidal blood vessels into the retina and the waste products from the retina
into the blood vessels
1
. Phagocytosis of shed outer receptor segments is performed by RPE
interaction of the apical microvilli with the photoreceptors, and requires the proper distribution
of proteins involved in the phagocytic process, such as αvβ5 integrin, to be expressed on the
apical side
137
. RPE cells also play an important role in re-isomerizing 11-trans retinol deposited
into the subretinal space into 11-cis retinal via the RPE65 protein, and re-transporting it back into
the photoreceptors through its apical side. Additionally, Na/K ATPase pumps are found on the
apical side to help maintain the high Na+ environment required for the process of visual signal
transduction
3
.
The accumulation of drusen between the RPE and the Bruch’s membrane is one of the
main features of AMD. Oxidative stress as a result of aging is thought to be one of the main
factors in contributing to the cause of AMD. As the amount of drusen increases in size and
number, early stage AMD progresses into GA, in which the RPE cells degenerate and die,
leading to photoreceptor loss and the loss of central vision
25
Although there is no cure for dry
AMD yet, or other similar diseases like SMD where the RPE cells also degenerate or die, RPE
replacement strategies offer a potential solution. Fetal and adult RPE have been delivered in
suspension in patients with atrophic or exudative AMD, as well as the RPE cell line ARPE-19.
However, these methods are disadvantageous because of limited supply, especially with fetal
tissue, as well as the possibility of immune rejection
65, 138, 139
. With ARPE-19, it has been
114

shown that they can still dedifferentiate into other cell types after multiple passages, lose their
original morphology, and downregulate important RPE proteins while upregulating neuronal cell
marker expression.
65, 138, 139
Autologous transplantation is when a small section of healthy
peripheral RPE is removed and placed under the macula to replace the dysfunctional RPE.
Although immune rejection is less likely in autologous transplantation, the viability of the cells is
a concerning factor because these cells are senescent and most likely carry the same genetic
factor information as the degenerate RPE cells. Another challenge in these grafts is the
attachment of the RPE cells onto Bruch’s membrane
73
. Other types of surgeries include macular
translocation in which the neural retina is detached and rotated so that the macula is turned away
from diseased RPE and faces portions of healthy RPE instead. Although visual acuity was
maintained for up to 5 years with this procedure, there could be high complications that develop
as a result of the difficult surgery
140, 141
.
Because AMD affects such a large population of people, and there is a high risk of
complications that could develop from the complex surgical methods described, hESC is a
promising source of cells to replace lost RPE. Since hESC are able to self-renew and remain in
an undifferentiated state indefinitely, hESC-RPE cells are more readily available than fetal RPE
or RPE cell lines, and can also be expanded into large quantities to accommodate the large
number of patients with AMD. hESC-RPE cells are not senescent and can also be screened for
AMD-associated genes, but there are still concerns about immune rejection. While iPSC-RPE
cells were initially thought to resolve the problem of patient immune rejection, there have still
been concerns with this matter because of the integrative virus methods used to reprogram for
pluripotency. However, to combat this, researchers have been working on improved non-
integrative reprogramming methods as well as direct mRNA or protein transduction
65
. Though
115

patient-derived iPSC-RPE are currently being studied as replacement for degenerating RPE in
AMD, there are still concerns with the iPSC-RPE retaining memory of their previous cells.
Because iPSC-RPE cells are specific to each patient, cell supply will be limited since the
patient’s somatic cells must be first reprogrammed into iPSC cells before allowed to differentiate
into RPE cells. In contrast, hESC-RPE cells would be more readily available for large
populations of patients.
hESC-RPE can be transplanted into the subretinal space by subretinal injection of RPE
cells, or as sheet of polarized RPE cells grown on membrane
65
. While the implantation of pre-
polarized RPE may seem more advantageous, the surgery is more invasive than if the cells were
suspension injected. Additionally, the surgical placement of the membrane, regardless of
whether it is biodegradable or non-biodegradable, nonetheless introduces foreign material into
the eye. However, suspension-injection of hESC-RPE also has its drawbacks. Although
Schwartz et al., demonstrated no adverse effect or teratoma formation in suspension injected
RPE cells into an AMD patient, questions still arise if these cells will attach correctly on the
Bruch’s membrane and form a polarized monolayer.
Studies have already illustrated the important functions of polarized RPE in vivo on the
health of the retina through its ability to transport substances and molecules between the blood
vessels and photoreceptors, its role in the visual cycle, its secretion of trophic and angiogenic
factors localized to the apical or basal sides, and its ability to phagocytose photoreceptor outer
segments. My two projects further depict the unique properties of polarized hESC-RPE cells
with respect to being more resistant to oxidative stress and also of being capable of secreting
factors that help increase RPE differentiation from hESC.
116

My first project showed polarized hESC-RPE cells were more resistant to oxidative
stress-mediated apoptosis relative to non-polarized RPE which included confluent and sub-
confluent cells. Sub-confluent cells had the highest percentage of dead cells when treated with
600 μM H
2
O
2
followed by a significantly smaller percentage of dead cells in confluent RPE,
while polarized RPE cells were seemingly unaffected. Instead, polarized RPE began dying at
nearly double the concentration of H
2
O
2
at 1000 μM. At this dosage, non-polarized RPE had
already completely detached from the plate. Furthermore, it was found that sub-confluent RPE
underwent apoptosis resulting from a shorter H
2
O
2
dosage period than confluent and polarized
RPE; while treated, confluent and polarized RPE did not show any changes in cleaved caspase 3
levels following 8 hours of H
2
O
2
treatment + medium change, there was a notable increase in
treated sub-confluent cells. Instead, confluent RPE showed a significant increase in cleaved
caspase 3 levels while polarized RPE showed no change at 24 hours of treatment relative to
corresponding untreated cultures.
We looked at some signaling pathways upstream of the Bcl-2 family of proteins
responsible for maintaining the MOMP and regulating apoptosis. We found that untreated
polarized RPE cells naturally expressed higher levels of p-Akt, which could play a role in their
higher resistance to oxidative stress, relative to non-polarized RPE. The Akt signaling pathway
has been shown to be important in cell survival, activating Bcl-2
102
to prevent Bax from
permeabilizing the mitochondrial membrane and initiating apoptosis. In contrast, non-polarized
cells naturally had higher activation of pro-apoptotic signaling pathways through p-JNK and p-
p38. Within non-polarized cells, sub-confluent cells had greater expression of p-JNK and p-p38
than confluent cultures in both treated and untreated cells, which likely resulted in their lower
levels of Bcl-2 and higher Bax levels relative to polarized RPE.
117

In addition to polarized RPE being predisposed to having higher Akt and Bcl-2 levels,
polarized RPE were also found to have constitutively higher levels of antioxidants SOD1 and
catalase which likely protected it against oxidative stress. After H
2
O
2
treatments of 600 and
1000 μM, SOD1 and catalase expression levels did not decrease at all in polarized RPE; treated
non-polarized RPE cells experienced a drop in both SOD1 and catalase. Within non-polarized
RPE, the drop was most evident in sub-confluent cells relative to confluent RPE, illustrating that
sub-confluent RPE were the least capable of protecting themselves against oxidative stress.
The results of this first project have clinical significance because it showed that the
method of transplanting pre-polarized RPE cells into the sub-retinal space may be more
advantageous than suspension injection of RPE cells. Non-polarized RPE cells used here
represented an in vitro model of RPE cells in the sub-retinal space following suspension
injection. Like non-polarized RPE cell cultured in vitro, the injected cells require time to
establish their polarity. When initially injected, it is uncertain how many cells can attach onto
the Bruch’s membrane, and in our in vitro cultures, we observed that in order for cells to
polarize, they need to be in close proximity to one another. It is likely during this time period
that the injected RPE cells are far apart from each other, analogous to our sub-confluent cultures,
and they would need to proliferate enough so that they are adjacent to each other before they
become polarized. However, because our results indicated that sub-confluent H9-RPE are most
susceptible to oxidative stress-mediated apoptosis and cell cycle arrest, the cells, which are in a
high oxidative stress environment of an AMD patient, may not get the chance to proliferate
enough to polarize. Furthermore, because polarized cells showed that they continue to express
high levels of SOD1 and catalase when exposed to oxidative stress and minimal cell death,
despite nearly doubling our dosage to 1000 μM, a concentration that went beyond the H
2
O
2

118

levels found in the aqueous humor of the eye under pathologic conditions stress, we propose that
transplantation with polarized RPE will yield the most effective results in replacing damaged
RPE cells in AMD patients.
While hESC-generated RPE cells can be an unlimited source of cells for therapeutic use,
the differentiation time period from hESC to the appearance of RPE cells can take up to 8 weeks.
Furthermore, RPE cells must be purified from surrounding cells, and cultured for 1 month
followed by additional passages before they can be rid of other contaminating cell types. In the
second part of my project, we looked at the effect of polarized-RPE CM on hESC differentiation
in two types of hESC cell lines, H9 and hES3. CM was shown to increase the amount of RPE
cell differentiated from both hES3 and H9 cells. However, CM-treated hES3 cells resulted in a
lower RPE yield relative to H9 cells cultured in regular medium, which was one of the reasons
why H9 cells was used for the bulk of the project; the H9 line is also one of the lines currently
used in clinical research. We also performed a growth factor array to screen for potential
proteins in CM that could be responsible for increasing the amount of RPE during
differentiation; additionally, we screened the media of differentiating H9 cells at various weeks
leading up to the appearance of RPE cells. Based on the array, a combination of NT3/NT4 and
PDGF AA added back to the regular medium resulted in an increase in the amount of
differentiated RPE cells. While the addition of PDGF AA increased the amount of RPE cells
relative to regular medium, they were nonetheless noticeably less than the H9 cells cultured with
CM or the addition of NT3/NT4. Although CM and the GF additions we tried were able to
increase the amount of differentiated RPE cells, neither CM or GF additions were able to
increase the speed of RPE differentiation by having RPE cells appear sooner than 6 weeks.
Instead, we found that they all increased the RPE yield in differentiating H9-ESC’s by increasing
119

RPE proliferation. We noticed that cells cultured in PDGF AA produced RPE colonies that were
larger than H9 cells cultured in regular medium; however the number of colonies was relatively
small. In contrast, we found that addition of NT3/NT4, like CM, significantly increased the
number of differentiated RPE colonies as well as the size of the colonies relative to cells cultured
in regular medium. These results indicated that CM and the addition of NT3/NT4 increased the
amount of RPE cells by inducing differentiation of RPE colonies and likely expanded the size of
those colonies by their ability to induce RPE proliferation. On the other hand, PDGF AA was
found to only encourage RPE proliferation.
This project showed that CM was able to increase RPE differentiation from 2 different
hESC lines by producing more RPE colonies and then likely increasing the proliferation of those
RPE cells. Because the differentiation process yields a low amount of RPE cells, and it requires
a long time to differentiate and purify before they are usable, these findings will help improve
hESC differentiation and hESC-RPE culture techniques. Although other groups have
demonstrated increased RPE cell differentiation by addition of other factors, CM used here also
has its advantages. CM is versatile, commonly available, able to differentiate RPE and retinal
progenitor cells, protect retinal cells and likely RPE cells as well. This project also shows
clinical significance in the future transplantation polarized RPE into the subretinal space because
now we have a better understanding of the RPE microenvironment. Based on the results of the
growth factor screen, we are more aware of the growth factors that polarized RPE cells can
secrete and how this can affect the surrounding photoreceptors and choroidal blood vessels
within AMD patients. Transplantation of polarized RPE cells will result in cell secretion of
factors that will help with retinal differentiation and proliferation as well.

120

Future Directions
We determined that polarized RPE constitutively have higher levels of Bcl-2, p-Akt and
antioxidants SOD1 and catalase relative to non-polarized RPE. It will be interesting to
determine what transcriptional networks are regulating this group of anti-apoptotic and anti-
oxidant genes. These networks might also regulate the development of components of the
differentiated phenotype such as the presence of tight junctions and presence of mature
melanosomes. Other groups have indicated that the melanosomes in RPE act as antioxidants in
protecting the RPE against non-photic oxidative stress
80
. Non-polarized cells initially lose their
pigmentation during culturing, and pigmentation slowly begins to return when cells become
tightly confluent with full development of pigmentation when they are polarized. Perhaps some
of the components (ie. premelanosome, tyronsinase, tyrosinase-related protein-1) that regulate
melanosome formation in polarized RPE could also play a double role in regulation of SOD1 and
catalase expression.
Polarized RPE is also unique from non-polarized RPE in its high secretion levels of
PEDF into the extracellular space; however, little work has been done on the signaling pathways
that PEDF can activate. One group has shown that PEDF can block p38 MAPK signaling in
VEGF-induced vascular permeability
142
. Similarly, we can also look more in-depth at PEDF’s
signaling mechanism to see if it is responsible for inhibiting p38 and JNK phosphorylation, as we
found in our polarized RPE. The PEDF receptor (PEDF-R) was recently identified and is
expressed on the surface of both retinal and RPE cells. It is a member of the patatin-like
phospholipase domain-containing 2 (PNPLA2) family, a transmembrane phospholipase that has
been found to exhibit strong phospholipase A2 activity upon PEDF binding
143
that liberates fatty
acids from phospholipid substrates. PEDF-R was also recently found to be responsible for
121

retinal cell survival
144
. While there has been little work done on phospholipase A2 relative to
PI3/Akt cell survival signaling, it will be interesting to see if PEDF-R plays a role in being
directly responsible for polarized RPE cells’ constitutively higher levels of p-Akt.
In the second project, out of the many growth factor combinations we tested, only
NT3/NT4 and PDGF AA appeared to affect the amount of differentiated RPE cells; however,
there were other possible candidates that the growth factor array identified that we could also try,
and these were neurotrophic factors that appear to be geared toward RPE protection. One of
these is hepatocyte growth factor (HGF) which is secreted extremely highly in Week 8 of
differentiation and has been shown to defend against RPE and photoreceptor degeneration, and
also protects RPE from a drop in glutathione levels
145
. Another neurotrophic factor that was not
included in the growth factor array but appears promising in RPE differentiation is ciliary
neurotrophic factor (CNTF) and leukemia inhibitory factor (LIF), which has also been shown to
play a role in RPE survivability. CNTF has been shown to increase RPE survivability and NT3
secretion to the apical side
146
, which we have already shown to be significant in RPE
differentiation and proliferation.
It would be interesting to see which signaling pathways are being activated or enhanced
when RPE cells are differentiated from hESCs in CM and NT3/NT4 relative to regular medium.
Some pathways to look at could include SHH signaling pathway, since it has been shown to play
a major role in RPE and retinal development in vivo, or the TGF-β signaling pathway because it
has been shown to prevent signaling for the specification of the neural retina during development
of the optic cup. We suspect that CM could protect against oxidative stress based on the results
of the growth factor array, and the knowledge that high levels of PEDF were found to be secreted
in CM. We could also test the abilities of CM in protecting non-polarized RPE from oxidative
122

stress by culturing those cells in CM and treating with H
2
O
2
to determine if there is less cell
death. This might also help answer the question of whether the increased amount of RPE cells
differentiated in CM can be attributed to CM helping to decrease RPE cell die off from stress
during the cell culture process. This project also briefly showed the potential of CM to cut the
time required for RPE polarization by increasing the proliferation of non-polarized cells, which
first need to reach confluency before they polarize. It will be interesting to see if culturing non-
polarized RPE with CM is also able to increase the degree of polarity by measuring their TER.
These non-polarized cells would need to be cultured in CM for only approximately 2 weeks,
since once the cells polarize, they will begin to secrete CM on their own.
Both hESC and iPSC are promising sources in the future for generating youthful RPE
cells for transplantation in patients with degenerating and lost RPE. Depending on the patient
applicability, hESC could be the ideal source for generating vast amounts of RPE cells for the
general patient population, while iPSC derived RPE, could be better suited for individuals who
have adverse immune reactions to hESC-RPE, since iPSC could be originated from their own
somatic cells. Regardless of whether RPE cells are originated from hESC or iPSC, the
differentiation and expansion process of RPE cells is still a lengthy process. Therefore, knowing
that polarized RPE CM has the ability to increase hESC differentiation as well as RPE
proliferation could likely apply to iPSC as well, in increasing the differentiation abilities of iPSC
into RPE, and possibly shortening its expansion time, thus offering a patient-specific option to
generate high quantities of RPE cells for RPE cell therapy.

123

References

1. Strauss O: The retinal pigment epithelium in visual function, Physiol Rev 2005, 85:845-881
2. Marmorstein AD: The polarity of the retinal pigment epithelium, Traffic 2001, 2:867-872
3. Dornonville de la Cour M: Ion transport in the retinal pigment epithelium. A study with double
barrelled ion-selective microelectrodes, Acta Ophthalmol Suppl 1993, 1-32
4. Bok D: The retinal pigment epithelium: a versatile partner in vision, J Cell Sci Suppl 1993, 17:189-
195
5. Streilein JW, Ma N, Wenkel H, Ng TF, Zamiri P: Immunobiology and privilege of neuronal retina
and pigment epithelium transplants, Vision Res 2002, 42:487-495
6. Sonoda S, Sreekumar PG, Kase S, Spee C, Ryan SJ, Kannan R, Hinton DR: Attainment of polarity
promotes growth factor secretion by retinal pigment epithelial cells: relevance to age-related macular
degeneration, Aging (Albany NY) 2010, 2:28-42
7. Steele FR, Chader GJ, Johnson LV, Tombran-Tink J: Pigment epithelium-derived factor:
neurotrophic activity and identification as a member of the serine protease inhibitor gene family, Proc
Natl Acad Sci U S A 1993, 90:1526-1530
8. Witmer AN, Vrensen GF, Van Noorden CJ, Schlingemann RO: Vascular endothelial growth factors
and angiogenesis in eye disease, Prog Retin Eye Res 2003, 22:1-29
9. Zhu D, Deng X, Spee C, Sonoda S, Hsieh CL, Barron E, Pera M, Hinton DR: Polarized secretion of
PEDF from human embryonic stem cell-derived RPE promotes retinal progenitor cell survival, Invest
Ophthalmol Vis Sci 2011, 52:1573-1585
10. Bazan NG: Neurotrophins induce neuroprotective signaling in the retinal pigment epithelial cell
by activating the synthesis of the anti-inflammatory and anti-apoptotic neuroprotectin D1, Adv Exp Med
Biol 2008, 613:39-44
11. De Marzo A, Aruta C, Marigo V: PEDF promotes retinal neurosphere formation and expansion in
vitro, Adv Exp Med Biol 2010, 664:621-630
12. Zuber ME, Gestri G, Viczian AS, Barsacchi G, Harris WA: Specification of the vertebrate eye by a
network of eye field transcription factors, Development 2003, 130:5155-5167
13. Martinez-Morales JR, Rodrigo I, Bovolenta P: Eye development: a view from the retina
pigmented epithelium, Bioessays 2004, 26:766-777
14. Graw J: The genetic and molecular basis of congenital eye defects, Nat Rev Genet 2003, 4:876-
888
15. Pittack C, Grunwald GB, Reh TA: Fibroblast growth factors are necessary for neural retina but
not pigmented epithelium differentiation in chick embryos, Development 1997, 124:805-816
16. Zhao S, Overbeek PA: Tyrosinase-related protein 2 promoter targets transgene expression to
ocular and neural crest-derived tissues, Dev Biol 1999, 216:154-163
17. Fuhrmann S, Levine EM, Reh TA: Extraocular mesenchyme patterns the optic vesicle during early
eye development in the embryonic chick, Development 2000, 127:4599-4609
18. Mochii M, Ono T, Matsubara Y, Eguchi G: Spontaneous transdifferentiation of quail pigmented
epithelial cell is accompanied by a mutation in the Mitf gene, Dev Biol 1998, 196:145-159
19. Nakayama A, Nguyen MT, Chen CC, Opdecamp K, Hodgkinson CA, Arnheiter H: Mutations in
microphthalmia, the mouse homolog of the human deafness gene MITF, affect neuroepithelial and
neural crest-derived melanocytes differently, Mech Dev 1998, 70:155-166
20. Nguyen M, Arnheiter H: Signaling and transcriptional regulation in early mammalian eye
development: a link between FGF and MITF, Development 2000, 127:3581-3591
124

21. Belecky-Adams T, Adler R: Developmental expression patterns of bone morphogenetic proteins,
receptors, and binding proteins in the chick retina, J Comp Neurol 2001, 430:562-572
22. Adler R, Belecky-Adams TL: The role of bone morphogenetic proteins in the differentiation of
the ventral optic cup, Development 2002, 129:3161-3171
23. Perron M, Boy S, Amato MA, Viczian A, Koebernick K, Pieler T, Harris WA: A novel function for
Hedgehog signalling in retinal pigment epithelium differentiation, Development 2003, 130:1565-1577
24. Zhang XM, Yang XJ: Temporal and spatial effects of Sonic hedgehog signaling in chick eye
morphogenesis, Dev Biol 2001, 233:271-290
25. Handa JT: How does the macula protect itself from oxidative stress?, Mol Aspects Med 2012,
33:418-435
26. Hageman GS, Luthert PJ, Victor Chong NH, Johnson LV, Anderson DH, Mullins RF: An integrated
hypothesis that considers drusen as biomarkers of immune-mediated processes at the RPE-Bruch's
membrane interface in aging and age-related macular degeneration, Prog Retin Eye Res 2001, 20:705-
732
27. Gehrs KM, Anderson DH, Johnson LV, Hageman GS: Age-related macular degeneration--
emerging pathogenetic and therapeutic concepts, Ann Med 2006, 38:450-471
28. Bressler NM, Bressler SB, Fine SL: Age-related macular degeneration, Surv Ophthalmol 1988,
32:375-413
29. Seddon JM: Genetic and environmental underpinnings to age-related ocular diseases, Invest
Ophthalmol Vis Sci 2013, 54:ORSF28-30
30. Sobrin L, Seddon JM: Nature and nurture- genes and environment- predict onset and
progression of macular degeneration, Prog Retin Eye Res 2013,
31. Mitchell P, Wang JJ, Smith W, Leeder SR: Smoking and the 5-year incidence of age-related
maculopathy: the Blue Mountains Eye Study, Arch Ophthalmol 2002, 120:1357-1363
32. Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FL, 3rd: Risk factors for the incidence of
Advanced Age-Related Macular Degeneration in the Age-Related Eye Disease Study (AREDS) AREDS
report no. 19, Ophthalmology 2005, 112:533-539
33. Velilla S, Garcia-Medina JJ, Garcia-Layana A, Dolz-Marco R, Pons-Vazquez S, Pinazo-Duran MD,
Gomez-Ulla F, Arevalo JF, Diaz-Llopis M, Gallego-Pinazo R: Smoking and Age-Related Macular
Degeneration: Review and Update, J Ophthalmol 2013, 2013:895147
34. Sin HP, Liu DT, Lam DS: Lifestyle modification, nutritional and vitamins supplements for age-
related macular degeneration, Acta Ophthalmol 2013, 91:6-11
35. Edwards AO, Ritter R, 3rd, Abel KJ, Manning A, Panhuysen C, Farrer LA: Complement factor H
polymorphism and age-related macular degeneration, Science 2005, 308:421-424
36. Haines JL, Hauser MA, Schmidt S, Scott WK, Olson LM, Gallins P, Spencer KL, Kwan SY,
Noureddine M, Gilbert JR, Schnetz-Boutaud N, Agarwal A, Postel EA, Pericak-Vance MA: Complement
factor H variant increases the risk of age-related macular degeneration, Science 2005, 308:419-421
37. Chen W, Stambolian D, Edwards AO, Branham KE, Othman M, Jakobsdottir J, Tosakulwong N,
Pericak-Vance MA, Campochiaro PA, Klein ML, Tan PL, Conley YP, Kanda A, Kopplin L, Li Y, Augustaitis KJ,
Karoukis AJ, Scott WK, Agarwal A, Kovach JL, Schwartz SG, Postel EA, Brooks M, Baratz KH, Brown WL,
Brucker AJ, Orlin A, Brown G, Ho A, Regillo C, Donoso L, Tian L, Kaderli B, Hadley D, Hagstrom SA,
Peachey NS, Klein R, Klein BE, Gotoh N, Yamashiro K, Ferris Iii F, Fagerness JA, Reynolds R, Farrer LA, Kim
IK, Miller JW, Corton M, Carracedo A, Sanchez-Salorio M, Pugh EW, Doheny KF, Brion M, Deangelis MM,
Weeks DE, Zack DJ, Chew EY, Heckenlively JR, Yoshimura N, Iyengar SK, Francis PJ, Katsanis N, Seddon
JM, Haines JL, Gorin MB, Abecasis GR, Swaroop A: Genetic variants near TIMP3 and high-density
lipoprotein-associated loci influence susceptibility to age-related macular degeneration, Proc Natl Acad
Sci U S A 2010, 107:7401-7406
125

38. Neale BM, Fagerness J, Reynolds R, Sobrin L, Parker M, Raychaudhuri S, Tan PL, Oh EC, Merriam
JE, Souied E, Bernstein PS, Li B, Frederick JM, Zhang K, Brantley MA, Jr., Lee AY, Zack DJ, Campochiaro B,
Campochiaro P, Ripke S, Smith RT, Barile GR, Katsanis N, Allikmets R, Daly MJ, Seddon JM: Genome-wide
association study of advanced age-related macular degeneration identifies a role of the hepatic lipase
gene (LIPC), Proc Natl Acad Sci U S A 2010, 107:7395-7400
39. Kimura K, Isashiki Y, Sonoda S, Kakiuchi-Matsumoto T, Ohba N: Genetic association of
manganese superoxide dismutase with exudative age-related macular degeneration, Am J Ophthalmol
2000, 130:769-773
40. Canter JA, Olson LM, Spencer K, Schnetz-Boutaud N, Anderson B, Hauser MA, Schmidt S, Postel
EA, Agarwal A, Pericak-Vance MA, Sternberg P, Jr., Haines JL: Mitochondrial DNA polymorphism A4917G
is independently associated with age-related macular degeneration, PLoS One 2008, 3:e2091
41. Shen JK, Dong A, Hackett SF, Bell WR, Green WR, Campochiaro PA: Oxidative damage in age-
related macular degeneration, Histol Histopathol 2007, 22:1301-1308
42. Boulton M: Ageing of the retinal pigment epithelium: implications for transplantation, Graefes
Arch Clin Exp Ophthalmol 2004, 242:76-84
43. Jarrett SG, Boulton ME: Consequences of oxidative stress in age-related macular degeneration,
Mol Aspects Med 2012, 33:399-417
44. Beatty S, Koh H, Phil M, Henson D, Boulton M: The role of oxidative stress in the pathogenesis of
age-related macular degeneration, Surv Ophthalmol 2000, 45:115-134
45. Gutteridge JM, Halliwell B: Free radicals and antioxidants in the year 2000. A historical look to
the future, Ann N Y Acad Sci 2000, 899:136-147
46. Delori FC, Goger DG, Dorey CK: Age-related accumulation and spatial distribution of lipofuscin in
RPE of normal subjects, Invest Ophthalmol Vis Sci 2001, 42:1855-1866
47. Weiter JJ, Delori FC, Wing GL, Fitch KA: Retinal pigment epithelial lipofuscin and melanin and
choroidal melanin in human eyes, Invest Ophthalmol Vis Sci 1986, 27:145-152
48. Boyer NP, Higbee D, Currin MB, Blakeley LR, Chen C, Ablonczy Z, Crouch RK, Koutalos Y:
Lipofuscin and N-retinylidene-N-retinylethanolamine (A2E) accumulate in retinal pigment epithelium in
absence of light exposure: their origin is 11-cis-retinal, J Biol Chem 2012, 287:22276-22286
49. Mazzitello KI, Arizmendi CM, Family F, Grossniklaus HE: Formation and growth of lipofuscin in
the retinal pigment epithelium cells, Phys Rev E Stat Nonlin Soft Matter Phys 2009, 80:051908
50. Feeney-Burns L, Hilderbrand ES, Eldridge S: Aging human RPE: morphometric analysis of
macular, equatorial, and peripheral cells, Invest Ophthalmol Vis Sci 1984, 25:195-200
51. Rabin DM, Rabin RL, Blenkinsop TA, Temple S, Stern JH: Chronic oxidative stress upregulates
Drusen-related protein expression in adult human RPE stem cell-derived RPE cells: a novel culture model
for dry AMD, Aging (Albany NY) 2013, 5:51-66
52. Bowes Rickman C, Farsiu S, Toth CA, Klingeborn M: Dry age-related macular degeneration:
mechanisms, therapeutic targets, and imaging, Invest Ophthalmol Vis Sci 2013, 54:ORSF68-80
53. Moeller SM, Voland R, Tinker L, Blodi BA, Klein ML, Gehrs KM, Johnson EJ, Snodderly DM,
Wallace RB, Chappell RJ, Parekh N, Ritenbaugh C, Mares JA: Associations between age-related nuclear
cataract and lutein and zeaxanthin in the diet and serum in the Carotenoids in the Age-Related Eye
Disease Study, an Ancillary Study of the Women's Health Initiative, Arch Ophthalmol 2008, 126:354-364
54. Tan JS, Wang JJ, Flood V, Rochtchina E, Smith W, Mitchell P: Dietary antioxidants and the long-
term incidence of age-related macular degeneration: the Blue Mountains Eye Study, Ophthalmology
2008, 115:334-341
55. Campa C, Harding SP: Anti-VEGF compounds in the treatment of neovascular age related
macular degeneration, Curr Drug Targets 2011, 12:173-181
56. Allikmets R: A photoreceptor cell-specific ATP-binding transporter gene (ABCR) is mutated in
recessive Stargardt macular dystrophy, Nat Genet 1997, 17:122
126

57. Radu RA, Yuan Q, Hu J, Peng JH, Lloyd M, Nusinowitz S, Bok D, Travis GH: Accelerated
accumulation of lipofuscin pigments in the RPE of a mouse model for ABCA4-mediated retinal
dystrophies following Vitamin A supplementation, Invest Ophthalmol Vis Sci 2008, 49:3821-3829
58. Duncan JL, LaVail MM, Yasumura D, Matthes MT, Yang H, Trautmann N, Chappelow AV, Feng W,
Earp HS, Matsushima GK, Vollrath D: An RCS-like retinal dystrophy phenotype in mer knockout mice,
Invest Ophthalmol Vis Sci 2003, 44:826-838
59. Idelson M, Alper R, Obolensky A, Ben-Shushan E, Hemo I, Yachimovich-Cohen N, Khaner H,
Smith Y, Wiser O, Gropp M, Cohen MA, Even-Ram S, Berman-Zaken Y, Matzrafi L, Rechavi G, Banin E,
Reubinoff B: Directed differentiation of human embryonic stem cells into functional retinal pigment
epithelium cells, Cell Stem Cell 2009, 5:396-408
60. da Cruz L, Chen FK, Ahmado A, Greenwood J, Coffey P: RPE transplantation and its role in retinal
disease, Prog Retin Eye Res 2007, 26:598-635
61. Liao JL, Yu J, Huang K, Hu J, Diemer T, Ma Z, Dvash T, Yang XJ, Travis GH, Williams DS, Bok D, Fan
G: Molecular signature of primary retinal pigment epithelium and stem-cell-derived RPE cells, Hum Mol
Genet 2010, 19:4229-4238
62. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, Nie J, Jonsdottir GA,
Ruotti V, Stewart R, Slukvin, II, Thomson JA: Induced pluripotent stem cell lines derived from human
somatic cells, Science 2007, 318:1917-1920
63. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S: Induction of
pluripotent stem cells from adult human fibroblasts by defined factors, Cell 2007, 131:861-872
64. Bharti K, Miller SS, Arnheiter H: The new paradigm: retinal pigment epithelium cells generated
from embryonic or induced pluripotent stem cells, Pigment Cell Melanoma Res 2011, 24:21-34
65. Carr AJ, Smart MJ, Ramsden CM, Powner MB, da Cruz L, Coffey PJ: Development of human
embryonic stem cell therapies for age-related macular degeneration, Trends Neurosci 2013, 36:385-395
66. Gearhart J: New potential for human embryonic stem cells, Science 1998, 282:1061-1062
67. Rossant J, Nagy A: In search of the tabula rasa of human cells, Nat Biotechnol 1999, 17:23-24
68. Amit M, Carpenter MK, Inokuma MS, Chiu CP, Harris CP, Waknitz MA, Itskovitz-Eldor J, Thomson
JA: Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential
for prolonged periods of culture, Dev Biol 2000, 227:271-278
69. Xu C, Inokuma MS, Denham J, Golds K, Kundu P, Gold JD, Carpenter MK: Feeder-free growth of
undifferentiated human embryonic stem cells, Nat Biotechnol 2001, 19:971-974
70. Klimanskaya I, Hipp J, Rezai KA, West M, Atala A, Lanza R: Derivation and comparative
assessment of retinal pigment epithelium from human embryonic stem cells using transcriptomics,
Cloning Stem Cells 2004, 6:217-245
71. Lu B, Malcuit C, Wang S, Girman S, Francis P, Lemieux L, Lanza R, Lund R: Long-term safety and
function of RPE from human embryonic stem cells in preclinical models of macular degeneration, Stem
Cells 2009, 27:2126-2135
72. Lund RD, Wang S, Klimanskaya I, Holmes T, Ramos-Kelsey R, Lu B, Girman S, Bischoff N, Sauve Y,
Lanza R: Human embryonic stem cell-derived cells rescue visual function in dystrophic RCS rats, Cloning
Stem Cells 2006, 8:189-199
73. Pan CK, Heilweil G, Lanza R, Schwartz SD: Embryonic stem cells as a treatment for macular
degeneration, Expert Opin Biol Ther 2013, 13:1125-1133
74. Diniz B, Thomas P, Thomas B, Ribeiro R, Hu Y, Brant R, Ahuja A, Zhu D, Liu L, Koss M, Maia M,
Chader G, Hinton DR, Humayun MS: Subretinal implantation of retinal pigment epithelial cells derived
from human embryonic stem cells: improved survival when implanted as a monolayer, Invest
Ophthalmol Vis Sci 2013, 54:5087-5096
75. Tezel TH, Del Priore LV, Berger AS, Kaplan HJ: Adult retinal pigment epithelial transplantation in
exudative age-related macular degeneration, Am J Ophthalmol 2007, 143:584-595
127

76. Cai J, Nelson KC, Wu M, Sternberg P, Jr., Jones DP: Oxidative damage and protection of the RPE,
Prog Retin Eye Res 2000, 19:205-221
77. Imamura Y, Noda S, Hashizume K, Shinoda K, Yamaguchi M, Uchiyama S, Shimizu T, Mizushima
Y, Shirasawa T, Tsubota K: Drusen, choroidal neovascularization, and retinal pigment epithelium
dysfunction in SOD1-deficient mice: a model of age-related macular degeneration, Proc Natl Acad Sci U S
A 2006, 103:11282-11287
78. Liles MR, Newsome DA, Oliver PD: Antioxidant enzymes in the aging human retinal pigment
epithelium, Arch Ophthalmol 1991, 109:1285-1288
79. Schwartz SD, Hubschman JP, Heilwell G, Franco-Cardenas V, Pan CK, Ostrick RM, Mickunas E,
Gay R, Klimanskaya I, Lanza R: Embryonic stem cell trials for macular degeneration: a preliminary report,
Lancet 379:713-720
80. Burke JM, Kaczara P, Skumatz CM, Zareba M, Raciti MW, Sarna T: Dynamic analyses reveal
cytoprotection by RPE melanosomes against non-photic stress, Mol Vis 2011, 17:2864-2877
81. Ho TC, Yang YC, Cheng HC, Wu AC, Chen SL, Chen HK, Tsao YP: Activation of mitogen-activated
protein kinases is essential for hydrogen peroxide -induced apoptosis in retinal pigment epithelial cells,
Apoptosis 2006, 11:1899-1908
82. Kim BJ, Ryu SW, Song BJ: JNK- and p38 kinase-mediated phosphorylation of Bax leads to its
activation and mitochondrial translocation and to apoptosis of human hepatoma HepG2 cells, J Biol
Chem 2006, 281:21256-21265
83. Miyashita T, Reed JC: Tumor suppressor p53 is a direct transcriptional activator of the human
bax gene, Cell 1995, 80:293-299
84. Schwartz SD, Hubschman JP, Heilwell G, Franco-Cardenas V, Pan CK, Ostrick RM, Mickunas E,
Gay R, Klimanskaya I, Lanza R: Embryonic stem cell trials for macular degeneration: a preliminary report,
Lancet 2012, 379:713-720
85. Spector A, Ma W, Wang RR: The aqueous humor is capable of generating and degrading H2O2,
Invest Ophthalmol Vis Sci 1998, 39:1188-1197
86. Ramachandran S, Morris SM, Devamanoharan P, Henein M, Varma SD: Radio-isotopic
determination of hydrogen peroxide in aqueous humor and urine, Exp Eye Res 1991, 53:503-506
87. Richer SP, Rose RC: Water soluble antioxidants in mammalian aqueous humor: interaction with
UV B and hydrogen peroxide, Vision Res 1998, 38:2881-2888
88. Kim MH, Chung J, Yang JW, Chung SM, Kwag NH, Yoo JS: Hydrogen peroxide-induced cell death
in a human retinal pigment epithelial cell line, ARPE-19, Korean J Ophthalmol 2003, 17:19-28
89. Dunaief JL, Dentchev T, Ying GS, Milam AH: The role of apoptosis in age-related macular
degeneration, Arch Ophthalmol 2002, 120:1435-1442
90. Bhattacharya S, Chaum E, Johnson DA, Johnson LR: Age-related susceptibility to apoptosis in
human retinal pigment epithelial cells is triggered by disruption of p53-Mdm2 association, Invest
Ophthalmol Vis Sci 2012, 53:8350-8366
91. Jin GF, Hurst JS, Godley BF: Hydrogen peroxide stimulates apoptosis in cultured human retinal
pigment epithelial cells, Curr Eye Res 2001, 22:165-173
92. Tait SW, Green DR: Mitochondria and cell death: outer membrane permeabilization and beyond,
Nat Rev Mol Cell Biol 2010, 11:621-632
93. Aouacheria A, Cibiel A, Guillemin Y, Gillet G, Lalle P: Modulating mitochondria-mediated
apoptotic cell death through targeting of Bcl-2 family proteins, Recent Pat DNA Gene Seq 2007, 1:43-61
94. Clem RJ, Cheng EH, Karp CL, Kirsch DG, Ueno K, Takahashi A, Kastan MB, Griffin DE, Earnshaw
WC, Veliuona MA, Hardwick JM: Modulation of cell death by Bcl-XL through caspase interaction, Proc
Natl Acad Sci U S A 1998, 95:554-559
95. Wang X, McCullough KD, Franke TF, Holbrook NJ: Epidermal growth factor receptor-dependent
Akt activation by oxidative stress enhances cell survival, J Biol Chem 2000, 275:14624-14631
128

96. Shaw M, Cohen P, Alessi DR: The activation of protein kinase B by H2O2 or heat shock is
mediated by phosphoinositide 3-kinase and not by mitogen-activated protein kinase-activated protein
kinase-2, Biochem J 1998, 336 ( Pt 1):241-246
97. Yang P, Peairs JJ, Tano R, Jaffe GJ: Oxidant-mediated Akt activation in human RPE cells, Invest
Ophthalmol Vis Sci 2006, 47:4598-4606
98. Song G, Ouyang G, Bao S: The activation of Akt/PKB signaling pathway and cell survival, J Cell
Mol Med 2005, 9:59-71
99. Hanada M, Feng J, Hemmings BA: Structure, regulation and function of PKB/AKT--a major
therapeutic target, Biochim Biophys Acta 2004, 1697:3-16
100. Li X, Bijur GN, Jope RS: Glycogen synthase kinase-3beta, mood stabilizers, and neuroprotection,
Bipolar Disord 2002, 4:137-144
101. Franke TF, Hornik CP, Segev L, Shostak GA, Sugimoto C: PI3K/Akt and apoptosis: size matters,
Oncogene 2003, 22:8983-8998
102. Pugazhenthi S, Nesterova A, Sable C, Heidenreich KA, Boxer LM, Heasley LE, Reusch JE:
Akt/protein kinase B up-regulates Bcl-2 expression through cAMP-response element-binding protein, J
Biol Chem 2000, 275:10761-10766
103. Tsuruta F, Sunayama J, Mori Y, Hattori S, Shimizu S, Tsujimoto Y, Yoshioka K, Masuyama N,
Gotoh Y: JNK promotes Bax translocation to mitochondria through phosphorylation of 14-3-3 proteins,
Embo J 2004, 23:1889-1899
104. Ghatan S, Larner S, Kinoshita Y, Hetman M, Patel L, Xia Z, Youle RJ, Morrison RS: p38 MAP kinase
mediates bax translocation in nitric oxide-induced apoptosis in neurons, J Cell Biol 2000, 150:335-347
105. Srivastava RK, Sollott SJ, Khan L, Hansford R, Lakatta EG, Longo DL: Bcl-2 and Bcl-X(L) block
thapsigargin-induced nitric oxide generation, c-Jun NH(2)-terminal kinase activity, and apoptosis, Mol
Cell Biol 1999, 19:5659-5674
106. Rojo AI, Salinas M, Martin D, Perona R, Cuadrado A: Regulation of Cu/Zn-superoxide dismutase
expression via the phosphatidylinositol 3 kinase/Akt pathway and nuclear factor-kappaB, J Neurosci
2004, 24:7324-7334
107. Noshita N, Sugawara T, Lewen A, Hayashi T, Chan PH: Copper-zinc superoxide dismutase affects
Akt activation after transient focal cerebral ischemia in mice, Stroke 2003, 34:1513-1518
108. Papadopoulos MC, Koumenis IL, Xu L, Giffard RG: Potentiation of murine astrocyte antioxidant
defence by bcl-2: protection in part reflects elevated glutathione levels, Eur J Neurosci 1998, 10:1252-
1260
109. Jang JH, Surh YJ: Potentiation of cellular antioxidant capacity by Bcl-2: implications for its
antiapoptotic function, Biochem Pharmacol 2003, 66:1371-1379
110. Venkatesan B, Mahimainathan L, Das F, Ghosh-Choudhury N, Ghosh Choudhury G:
Downregulation of catalase by reactive oxygen species via PI 3 kinase/Akt signaling in mesangial cells, J
Cell Physiol 2007, 211:457-467
111. Ellerby LM, Ellerby HM, Park SM, Holleran AL, Murphy AN, Fiskum G, Kane DJ, Testa MP, Kayalar
C, Bredesen DE: Shift of the cellular oxidation-reduction potential in neural cells expressing Bcl-2, J
Neurochem 1996, 67:1259-1267
112. Vugler A, Lawrence J, Walsh J, Carr A, Gias C, Semo M, Ahmado A, da Cruz L, Andrews P, Coffey
P: Embryonic stem cells and retinal repair, Mech Dev 2007, 124:807-829
113. Levenstein ME, Ludwig TE, Xu RH, Llanas RA, VanDenHeuvel-Kramer K, Manning D, Thomson JA:
Basic fibroblast growth factor support of human embryonic stem cell self-renewal, Stem Cells 2006,
24:568-574
114. Klimanskaya I: Retinal pigment epithelium, Methods Enzymol 2006, 418:169-194
115. Kolomeyer AM, Sugino IK, Zarbin MA: Characterization of conditioned media collected from
cultured adult versus fetal retinal pigment epithelial cells, Invest Ophthalmol Vis Sci 2011, 52:5973-5986
129

116. Dhara SK, Stice SL: Neural differentiation of human embryonic stem cells, J Cell Biochem 2008,
105:633-640
117. Hellstrom AR, Watt B, Fard SS, Tenza D, Mannstrom P, Narfstrom K, Ekesten B, Ito S, Wakamatsu
K, Larsson J, Ulfendahl M, Kullander K, Raposo G, Kerje S, Hallbook F, Marks MS, Andersson L:
Inactivation of Pmel alters melanosome shape but has only a subtle effect on visible pigmentation, PLoS
Genet 2011, 7:e1002285
118. Bennett JL, Zeiler SR, Jones KR: Patterned expression of BDNF and NT-3 in the retina and
anterior segment of the developing mammalian eye, Invest Ophthalmol Vis Sci 1999, 40:2996-3005
119. Ortega JA, Alcantara S: BDNF/MAPK/ERK-induced BMP7 expression in the developing cerebral
cortex induces premature radial glia differentiation and impairs neuronal migration, Cereb Cortex 2010,
20:2132-2144
120. Lamba DA, Karl MO, Ware CB, Reh TA: Efficient generation of retinal progenitor cells from
human embryonic stem cells, Proc Natl Acad Sci U S A 2006, 103:12769-12774
121. Osakada F, Ikeda H, Mandai M, Wataya T, Watanabe K, Yoshimura N, Akaike A, Sasai Y,
Takahashi M: Toward the generation of rod and cone photoreceptors from mouse, monkey and human
embryonic stem cells, Nat Biotechnol 2008, 26:215-224
122. Lewin GR, Barde YA: Physiology of the neurotrophins, Annu Rev Neurosci 1996, 19:289-317
123. Liu ZZ, Zhu LQ, Eide FF: Critical role of TrkB and brain-derived neurotrophic factor in the
differentiation and survival of retinal pigment epithelium, J Neurosci 1997, 17:8749-8755
124. de la Rosa EJ, Arribas A, Frade JM, Rodriguez-Tebar A: Role of neurotrophins in the control of
neural development: neurotrophin-3 promotes both neuron differentiation and survival of cultured
chick retinal cells, Neuroscience 1994, 58:347-352
125. Minichiello L, Casagranda F, Tatche RS, Stucky CL, Postigo A, Lewin GR, Davies AM, Klein R: Point
mutation in trkB causes loss of NT4-dependent neurons without major effects on diverse BDNF
responses, Neuron 1998, 21:335-345
126. Coppola V, Kucera J, Palko ME, Martinez-De Velasco J, Lyons WE, Fritzsch B, Tessarollo L:
Dissection of NT3 functions in vivo by gene replacement strategy, Development 2001, 128:4315-4327
127. Hackett SF, Friedman Z, Freund J, Schoenfeld C, Curtis R, DiStefano PS, Campochiaro PA: A splice
variant of trkB and brain-derived neurotrophic factor are co-expressed in retinal pigmented epithelial
cells and promote differentiated characteristics, Brain Res 1998, 789:201-212
128. Cho MS, Kim SJ, Ku SY, Park JH, Lee H, Yoo DH, Park UC, Song SA, Choi YM, Yu HG: Generation of
retinal pigment epithelial cells from human embryonic stem cell-derived spherical neural masses, Stem
Cell Res 2012, 9:101-109
129. Buchholz DE, Pennington BO, Croze RH, Hinman CR, Coffey PJ, Clegg DO: Rapid and efficient
directed differentiation of human pluripotent stem cells into retinal pigmented epithelium, Stem Cells
Transl Med 2013, 2:384-393
130. Bryan JA, 3rd, Campochiaro PA: A retinal pigment epithelial cell-derived growth factor(s), Arch
Ophthalmol 1986, 104:422-425
131. Leschey KH, Hackett SF, Singer JH, Campochiaro PA: Growth factor responsiveness of human
retinal pigment epithelial cells, Invest Ophthalmol Vis Sci 1990, 31:839-846
132. Campochiaro PA, Hackett SF, Vinores SA, Freund J, Csaky C, LaRochelle W, Henderer J, Johnson
M, Rodriguez IR, Friedman Z, et al.: Platelet-derived growth factor is an autocrine growth stimulator in
retinal pigmented epithelial cells, J Cell Sci 1994, 107 ( Pt 9):2459-2469
133. Waters CM, Savla U, Panos RJ: KGF prevents hydrogen peroxide-induced increases in airway
epithelial cell permeability, Am J Physiol 1997, 272:L681-689
134. Gillis P, Savla U, Volpert OV, Jimenez B, Waters CM, Panos RJ, Bouck NP: Keratinocyte growth
factor induces angiogenesis and protects endothelial barrier function, J Cell Sci 1999, 112 ( Pt 12):2049-
2057
130

135. Geiger RC, Waters CM, Kamp DW, Glucksberg MR: KGF prevents oxygen-mediated damage in
ARPE-19 cells, Invest Ophthalmol Vis Sci 2005, 46:3435-3442
136. Kolomeyer AM, Sugino IK, Zarbin MA: Characterization of the effects of retinal pigment
epithelium-conditioned media on porcine and aged human retina, Graefes Arch Clin Exp Ophthalmol
2013, 251:1515-1528
137. Nandrot EF, Anand M, Sircar M, Finnemann SC: Novel role for alphavbeta5-integrin in retinal
adhesion and its diurnal peak, Am J Physiol Cell Physiol 2006, 290:C1256-1262
138. Dunn KC, Aotaki-Keen AE, Putkey FR, Hjelmeland LM: ARPE-19, a human retinal pigment
epithelial cell line with differentiated properties, Exp Eye Res 1996, 62:155-169
139. Alge CS, Suppmann S, Priglinger SG, Neubauer AS, May CA, Hauck S, Welge-Lussen U, Ueffing M,
Kampik A: Comparative proteome analysis of native differentiated and cultured dedifferentiated human
RPE cells, Invest Ophthalmol Vis Sci 2003, 44:3629-3641
140. Chen FK, Patel PJ, Uppal GS, Tufail A, Coffey PJ, Da Cruz L: Long-term outcomes following full
macular translocation surgery in neovascular age-related macular degeneration, Br J Ophthalmol 2010,
94:1337-1343
141. Takeuchi K, Kachi S, Iwata E, Ishikawa K, Terasaki H: Visual function 5 years or more after
macular translocation surgery for myopic choroidal neovascularisation and age-related macular
degeneration, Eye (Lond) 2012, 26:51-60
142. Yang J, Duh EJ, Caldwell RB, Behzadian MA: Antipermeability function of PEDF involves blockade
of the MAP kinase/GSK/beta-catenin signaling pathway and uPAR expression, Invest Ophthalmol Vis Sci
2010, 51:3273-3280
143. Subramanian P, Notario PM, Becerra SP: Pigment epithelium-derived factor receptor (PEDF-R): a
plasma membrane-linked phospholipase with PEDF binding affinity, Adv Exp Med Biol 2010, 664:29-37
144. Subramanian P, Locatelli-Hoops S, Kenealey J, DesJardin J, Notari L, Becerra SP: Pigment
epithelium-derived factor (PEDF) prevents retinal cell death via PEDF Receptor (PEDF-R): identification of
a functional ligand binding site, J Biol Chem 2013, 288:23928-23942
145. Jin M, Yaung J, Kannan R, He S, Ryan SJ, Hinton DR: Hepatocyte growth factor protects RPE cells
from apoptosis induced by glutathione depletion, Invest Ophthalmol Vis Sci 2005, 46:4311-4319
146. Li R, Wen R, Banzon T, Maminishkis A, Miller SS: CNTF mediates neurotrophic factor secretion
and fluid absorption in human retinal pigment epithelium, PLoS One 2011, 6:e23148

Asset Metadata

Creator Hsiung, Jamie (author)

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

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 09/10/2014

Defense Date 04/16/2014

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

Tag apoptosis,differentiation,OAI-PMH Harvest,oxidative stress,retinal pigment epithelial cell,stem cell

Format application/pdf (imt)

Language English

Advisor Hinton, David R. (committee chair), Craft, Cheryl M. (committee member), Lu, Wange (committee member), Zhu, Danhong (committee member)

Creator Email jamie1210@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-473693

Unique identifier UC11286825

Identifier etd-HsiungJami-2903.pdf (filename),usctheses-c3-473693 (legacy record id)

Legacy Identifier etd-HsiungJami-2903.pdf

Dmrecord 473693

Document Type Dissertation

Format application/pdf (imt)

Rights Hsiung, Jamie

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

Abstract (if available)

Abstract Retinal pigment epithelial (RPE) cells are a monolayer of polarized cells that are located between the retina and the choroidal blood vessels. The apical side of the RPE faces the photoreceptors while the basal side is adjacent to the Bruch’s membrane, which separates the RPE from the choriocapillaris. Polarized RPE are characterized by their hexagonal shape with tight cell-cell junctions and apical microvilli to aid in rod outer segment phagocytosis. RPE cells are pigmented cells capable of absorbing stray light, phagocytosis of fatty acid-rich photoreceptor outer rod segments, as well as transporting water, ions, and metabolic waste products to the basal blood vessel side, and in turn carrying glucose, retinol and fatty acids back to the apical photoreceptor side. Degenerate or dying RPE can cause secondary loss of photoreceptors, which can lead to age-related macular degeneration (AMD), the leading cause of blindness in the elderly in developed countries. Oxidative stress mediated injury to the RPE is thought to be a major factor involved in the pathogenesis of AMD. Human embryonic stem cell (hESC) derived RPE can serve as a potential source of young, viable RPE cells to replace dying RPE cells in patients with AMD. ❧ For our first project, we hypothesize that polarized monolayers of hESC-RPE are more resistant to oxidative stress-induced cell death relative to non-polarized RPE cells. This work has clinical relevance since both cell suspensions and polarized monolayers of hESC-RPE are being evaluated for their potential for cellular therapy in patients with AMD. We found highest amount of cell death in non-polarized/sub-confluent hESC-RPE and little cell death in polarized hESC-RPE cells when all samples were treated with the same dosage of H₂O₂. There were higher levels of pro-apoptotic p-p38, p-JNK and cleaved caspase 3 in treated non-polarized RPE relative to polarized cells. On the other hand, anti-apoptotic Bcl-2 levels were highest in polarized RPE relative to non-polarized RPE. Polarized RPE also had constitutively higher levels of Akt signaling, superoxide dismutase 1 (SOD 1) and catalase. These results indicate that non-polarized RPE, especially sub-confluent cells, are most sensitive to oxidative stress-induced apoptosis. Polarized hESC-RPE cells have higher tolerance to oxidative stress relative to non-polarized cells, most likely due to their constitutively higher pro-survival levels of Bcl-2 and p-Akt as well as SOD1 and catalase. These results suggest that implantation of polarized hESC-RPE for treating patients with geographic atrophy AMD should have better survival than injections of hESC-RPE in suspension. ❧ Human embryonic stem cells (hESC)-derived RPE have shown promise as a cellular therapy for AMD