Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Morphological and functional evaluation of hESC-RPE cell transplantation in RCS rats
(USC Thesis Other)
Morphological and functional evaluation of hESC-RPE cell transplantation in RCS rats
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MORPHOLOGICAL AND FUNCTIONAL EVALUATION OF HESC-RPE CELL
TRANSPLANTATION IN RCS RATS
by
Laura Liu
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
(CELL AND NEUROBIOLOGY)
August 2013
ii
Table of Contents
List of Tables iv
List of Figures v
Abbreviations vi
Abstract vii
Chapter 1 9
INTRODUCTION 9
Chapter 1 References 25
Chapter 2 34
In Vivo Functional Comparison of Polarized and Non-polarized hESC-RPE Cells
Transplantation in RCS Rats 34
ABSTRACT 34
MATERIALS AND METHODS 37
RESULTS 43
DISCUSSION 50
Chapter 2 References 54
Chapter 3 58
Histologic Comparison of Polarized and Non-polarized hESC-RPE Cells Transplantation in
RCS Rats 58
ABSTRACT 58
INTRODUCTION 59
MATERIALS AND METHODS 60
RESULTS 64
DISCUSSION 71
Chapter 3 References 74
Chapter 4 78
Long-term Efficacy and Safety of RPE Monolayer Derived From Human Embryonic Stem
Cell 78
ABSTRACT 78
INTRODUCTION 79
METHODS 80
RESULTS 85
DISCUSSION 89
Chapter 4 References 93
Chapter 5 98
iii
Using optical coherence tomography as a proxy for evaluation of RPE implants in the RCS
rat. 98
ABSTRACT 98
INTRODUCTION 99
MATERIALS AND METHODS 100
RESULTS 102
DISCUSSION 108
Chapter 5 References 111
Chapter 6 112
Histological correlates of Changes in the Fundus Autofluorescence Pattern in a Rat Retinal
Degeneration Model 112
ABSTRACT 112
INTRODUCTION 114
MATERIALS AND METHODS 117
RESULTS 120
DISCUSSION 126
Chapter 6 References 134
Chapter 7 140
CONCLUSION 140
Chapter 7 References 145
iv
List of Tables
Chapter 2
Table 1
Chapter 4
Table 1. Survival of hESC-RPE after transplantation into subretinal space of RCS rat
48
88
v
List of Figures
Chapter 1
Figure 1. Anatomy of the retina
Figure 2. Functions of retinal pigment epithelium
Figure 3. Confocal immunofluorescence, scanning electron microscopic, and transmission electron
microscopic images of polarized hESC-RPE
Figure 4. Design of parylene substrate and substrate with confluent hESC-RPE
Figure 5. Fundus picture of a rat eye immediately after parylene transplantation
Chapter 2
Figure 1. Surface anatomy of the superior colliculus
Figure 2. Threshold contour map of the superior colliculus
Figure 3. Visual threshold area versus visual filed area
Figure 4. Optokinetic head tracking duration in transplanted RCS rats
Figure 5. Mean maximum b-wave amplitude of RCS rat after transplantation
Figure 6. Infrared fundus photos and OCT scans of retina of RCS rat after surgery
Chapter 3
Figure 1. Light microscopic and confocal images of a retinal section of a transplanted RCS rat
Figure 2. Light microscopic transmission and confocal images of retina sections of hESC-RPE
suspension transplanted RCS rats
Figure 3. The outer nuclear layer cell count for controls and implanted animals
Figure 4. Confocal images of rhodopsin-positive phagosomes
Figure 5. Phagosomes count within the RPE
Figure 6. Confocal images of the CD68 and GFAP immunofluorescence in transplanted RCS rats
Figure 7. CD68 and GFAP expression in transplanted RCS rats
Chapter 4
Figure 1. Infrared and OCT image demonstrating the subretinal implant of the parylene membrane.
Figure 2. Light microscopic images and confocal images of hESC-RPE/parylene transplanted retinas
of RCS rat
Figure 3. Light microscopic images and confocal image of retina section of hESC-RPE suspension
transplanted RCS rats.
Figure 4. Optokinetic head tracking durations at after surgery
Chapter 5
10
11
19
21
23
43
43
45
46
47
49
64
66
67
68
69
70
71
83
86
87
89
Figure 1. OCT scan and histology comparison of normal Copenhagen rat
Figure 2. The OCT scans of the RCS rat
Figure 3. Histologic section of RCS rats
Figure 4. Thickness measure of the retinal layers
Figure 5. Infrared fundus photo and OCT scan of transplanted RCS rats
Chapter 6
Figure 1. Fundus autofluorescence image taken at different age in RCS rats
Figure 2. Progressive loss of the different layers of retina at different quadrant in RCS
Figure 3. Debris zone measured from the optic nerve and correlation with the histology
Figure 4. Fundus autofluorescence image an RCS rat that received subretinal injection
Figure 5. CD68 and GFAP expression in RCS rats
Figure 6. Spectral image of autofluorescence
103
103
105
106
107
121
122
124
124
125
126
vi
Abbreviations
AMD Age-related macular degeneration
BSS
Balanced salt solution
CFH Complement factor H
ECM
Extracellular matrix
ERG
Electroretinography
FAF
Fundus autofluorescence
GCL Ganglion cell layer
GFAP Glial fibrillary acidic protein
H&E Hematoxylin and eosin
hESC Human embryonic stem cells
hESC-RPE Human embryonic stem cells derived Retinal Pigment Epithelium
INL Inner nuclear layer
IPL Inner plexiform layer
MerTK Mer tyrosine kinase gene
NFL Nerve fiber layer
OCT Ocular coherence tomography
OHT Optokinetic head tracking
ONL Outer nuclear layer
OPL Outer plexiform layer
PEDF Pigment epithelium-derived factor
POS Photoreceptor outer segments
PUFA Polyunsaturated fatty acids
RCS Royal College of Surgeon
RD Retinal degeneration
RPE Retinal pigment epithelium
RT-PCR Reverse transcription polymerase chain reaction
SD-OCT spectral-domain optical coherence tomography
vii
Abstract
Age-related Macular Degeneration (AMD) is the leading cause of blindness among the
elderly in United States. It is categorized into two types: neovascular (wet) AMD, caused by
abnormal blood vessel growth in the choriocapillaris, and atrophic (dry) AMD, which is
characterized by the extensive loss of retinal pigment epithelium, followed by the loss of
photoreceptors. While several treatments are available for wet AMD, there is no effective way
to address dry AMD. Although tablet formulation of select vitamins and minerals has been
shown to slow the progression of dry AMD in some patients, there are currently no medical or
surgical treatments. An effective therapy for dry AMD is needed.
The purpose of this study is to evaluate the feasibility of using a cell based therapy to replace the
dysfunctional retinal pigment epithelium (RPE), which is one of the contributing factors of
AMD. The scientific evidence supporting such treatment comes from clinical studies using a
surgical approach called macular translocation, in which the area with the diseased RPE is
moved to an area with healthy RPE support. However, the surgery is complicated and carries a
large risk of ocular complications, so it is rarely performed nowadays. The experiments outlined
in this dissertation examined the feasibility of using an alternative approach by in vitro expansion
of an RPE cell line derived from human embryonic stem cells.
The specific aim of the study was to examine the functional outcome of transplantation of RPE
cells in the dystrophic RCS rat by performing both in vivo functional examination and
histological evaluation. The study also compared the outcome difference of cells delivered in
different formats, either as a monolayer or a cell suspension. We evaluated the correlation of
several non-invasive examinations and functional outcomes to see if these non-invasive
viii
techniques have the potential for predicting the treatment effects once the study moves to the
human clinical trial.
Our results showed that RPE transplantation as a monolayer had better visual function
preservation compared to cell suspension injection and sham therapy. Histologic evaluation
showed better preservation of retinal structure in animals transplanted with RPE monolayer. In
vivo imaging showed correlation with the implant that may be translatable for future clinical
evaluation of human clinical trials. Future studies will continue to develop the surgical
technique in larger mammals for future application in human patients.
9
Chapter 1
INTRODUCTION
Age-related Macular Degeneration (AMD)
AMD is the leading cause of blindness among the elderly in the United States. It affects
approximately 1.47% of the US population 40 years and older, or 1.75 million people (Friedman,
O'Colmain et al. 2004), a number that is estimated to double to 2.95 million in 2020 due to the
aging of the population. AMD is categorized into two types: neovascular (wet) AMD and
atrophic (dry) AMD (Ambati, Ambati et al. 2003). Several treatment modalities are available for
wet AMD (Gonzales 2005; Takeda, Colquitt et al. 2007; Brown, Michels et al. 2009). For
example, intravitreal anti-vascular endothelial growth factor (VEGF) injection targeting the
neovascular lesions has become the mainstay in wet AMD treatment recently (Brown, Michels et
al. 2009). However, there is no effective treatment available for dry AMD.
Atrophic AMD is characterized by the extensive loss of the retinal pigment epithelium
(RPE), followed by the loss of photoreceptors. While tablet formulation of select vitamins and
minerals has been shown to slow disease progression in some patients, there are no medical or
surgical treatments currently available for advanced dry AMD (Olson, Erie et al. 2011). An
effective therapy for dry AMD is needed. Advances in tissue engineering techniques make it
possible to develop cell-based therapy for treatment of such disease.
Retina and Retinal Pigment Epithelium
The retina is a light sensitive neuronal cell layer at the back of the eye.
The rods and cones are the photoreceptors, or the light sensing element of the retina (Figure 1).
10
They transform light into electrical signals through phototransduction cascade. The signals then
relay first to the bipolar cell, and then the ganglion cell through synaptic connection. The
ganglion cell sends its axons to the brain through the optic nerve.
Figure 1. Photoreceptors in the retina convert light signals into electrical signals and send them to an intermediate
layer, bipolar cells, then to the retinal ganglion cells which send its axons to the brain through the optic nerve.
(Image: Courtesy of Pearson Education, Inc.)
In a normal retina, the RPE forms a polarized monolayer between photoreceptors and the
choroid. It has several critical functions that support and help maintain the various activities of
the photoreceptors (Strauss 2005). (Figure 2) [Adapted from (Strauss 2005)]. First, the RPE
phagocytizes waste material, photoreceptor outer segments (POS) which are shed diurnally.
Second, the RPE re-isomerizes phototransduction end product all-trans- retinol back to 11-cis-
retinal for the next phototransduction cascade. Third, the RPE maintains the outer blood–retinal
barrier and controls the exchange of fluid and molecules between the choroid and the outer
11
retina. The RPE also supports the choriocapillaris through the secretion of VEGF, which is
expressed at the basolateral surface (Strauss 2005). In the case where the RPE is damaged, the
choriocapillaris loses its support and becomes atrophic (Del Priore, Kaplan et al. 1996). Adult
RPE does not have proliferative capacity; thus, in retinal diseases such as dry AMD, a secondary
degeneration of the adjacent tissue, including retina and choriocapillaris, will occur due to the
loss of support from RPE.
Figure 1. Summary of the functions of retinal pigment epithelium. VEGF, vascular endothelial growth factor; PEDF,
pigment epithelium-derived factor; Epithel, epithelium.
Pathogenesis of Dry AMD
The pathogenesis of AMD is not completely understood. Several mechanisms have been
implicated as the pathogenesis of AMD. Oxidative stress and inflammation are the two major
events thought to be related to the development of the disease (Nowak 2006; Kinnunen,
Petrovski et al. 2012). Although recent studies revealed that genetic factor, smoking, aging, and
body weight all significant risk factors for AMD, the direct mechanism is still not well
12
understood (Schaumberg, Hankinson et al. 2007; Seddon, Francis et al. 2007). The pathogenesis
is likely multifactorial.
Oxidative Stress
Oxidative stress refers to the progressive cellular damage that is caused by reactive
oxygen species (ROS) and contributes to protein misfolding and evoking functional
abnormalities during senescence (Kinnunen, Petrovski et al. 2012). Although pathologic changes
are observed in four functionally related tissues in the outer retina: photoreceptors, RPE, Bruch’s
membrane and choriocapillaris, the degeneration of the RPE have been researched the most
extensively. Researchers consider the degeneration of the RPE to be the event that initiates the
molecular pathways leading to AMD (Nowak 2006).
The outer retina is more susceptible to oxidative stress due to its unique environment
which is rich in polyunsaturated fatty acids (PUFA) from the molecules of retinoic acid cycle
and the metabolic waste in photoreceptors and RPE (Winkler, Boulton et al. 1999). PUFA are
susceptible to oxidation in the presence of oxygen and the outer retina happens to be high in
oxygen content due to the high oxygen demand of photoreceptors which aggravates the oxidative
stress. One of the critical functions of RPE is phagocytosis of photoreceptor outer segments
(POS) which are rich in PUFA. The phagocytosis of POS by RPE cells generates oxidative stress
caused by ROS. In normal conditions, there are several mechanisms for RPE to remove the
waste material inside the cell. The heat-shock protein repairs damaged protein or misfolded
protein (Hartl and Hayer-Hartl 2002). Damaged/ misfolded protein that failed to repair is
processed by proteasomal protein degradation pathways (Kopito 2000; Wojcik 2002).
13
Autophagy is another protein clearance system in the RPE. It is a specific lysosomal
clearance system that targets intracellular protein aggregates. It removes cytoplasmic deposits
by forming double-membrane-bound autophagosomes which undergo fusion with endosomes
and lysosomes for degradation (Eskelinen and Saftig 2009). In conditions such as aging or retinal
disease, these protein clearance systems can fail, causing lipofuscin, a degradation product of
POS, to accumulate in the lysosomes and cytoplasm of the RPE (Nowak 2006). Lipofuscin is
known to be a photoinducible free radical generator (Boulton, Dontsov et al. 1993). Schutt et al
reported that RPE loaded with lipofuscin showed a significant higher rate of apoptosis after
exposure to blue light compared to the RPE cells not loaded with lipofuscin (Schutt, Davies et al.
2000). The eye is an organ that functions as a light detector; light exposure over time inevitably
aggravates the oxidative stress.
Inflammation
Drusen are yellow-white deposits between the RPE and the inner collagenous zone of
Bruch’s membrane, comprised of the mostly non-degraded material extruded from the RPE
(Bressler, Silva et al. 1994; Fine, Berger et al. 2000). There are two types of drusen defined
clinically by their appearance, “hard” and “soft.” The presence of large numbers of soft drusen in
the macula is a risk factor for developing advanced AMD. Proteomic and histopathologic studies
revealed that the drusen contain many immune related components such as dendritic cell
processes, immunoglobulins, class II antigens, and components of the complement cascade such
as complement factor H (CFH), membrane attack complex (C5b-9) (Anderson, Mullins et al.
2002; Kijlstra, La Heij et al. 2005). It is possible that the RPE remnants triggered the local
inflammatory response and complements system. The local reactions cause subsequent damage
to the adjacent retinal tissue.
14
In the past few years, human genome-wide association studies have identified various
polymorphisms that support a role of inflammation in the pathogenesis with AMD (Edwards,
Ritter et al. 2005; Hageman, Anderson et al. 2005; Haines, Hauser et al. 2005; Klein, Zeiss et al.
2005; Rivera, Fisher et al. 2005; Yang, Stratton et al. 2008). A common single nucleotide
polymorphism was found to be strongly associated with the development of AMD; CFH
(Jakobsdottir, Conley et al. 2005; Rivera, Fisher et al. 2005). The gene coding for CFH is
located at CFH at 1q32 (Edwards, Ritter et al. 2005; Haines, Hauser et al. 2005). A T to C
substitution in at exon 9 of CFH representing a tyrosine-histidine change at amino acid 402
(Y402H) was found to be strongly associated with AMD (Zareparsi, Branham et al. 2005).
Prospective studies of CFH revealed that for individuals who were homozygous for the high risk
allele(HH), the risk of developing AMD increased by a factor of 3.92 compared to homozygous
for low risk allele (YY) (Schaumberg, Hankinson et al. 2007). CFH regulates the compliment
system by accelerating the dissociation of a product of compliment cascade, C3Bb (Tuo, Grob et
al. 2012). Thus changes in the CFH can lead to increased inflammatory response. This is
evidence that inflammatory reaction plays a role in AMD.
Innate Immunity
The innate immune system of the eye is composed of inflammatory cells such as
macrophages that respond to pathogens via pattern recognition receptors which sense pathogen-
associated molecular patterns and endogenous danger signals released by injured cells (Bianchi
2007). NOD-like receptors (NLRs) sense danger signals and trigger the assembly of large
cytoplasmic complexes called inflammasomes (Petrilli, Dostert et al. 2007). Dying RPE cells can
activate the inflammasome through NALP3 (a protein encoded by a member of NOD-like
receptor family, and subsequently stimulating the production of IL-1b and other pro-
15
inflammatory cytokines (Petrovski, Berenyi et al. 2011). Pathologic findings of AMD
demonstrate RPE autophagy and inflammatory cells involvement, which also supports the theory
that innate immune system plays a role in the pathogenesis of AMD (Killingsworth, Sarks et al.
1990). However, the role of innate immunity in the pathogenesis of AMD is still controversial.
The Multifactorial Pathogenesis of AMD
Prospective human studies examined the risk factors of AMD reveal that besides CFH,
single-nucleotide polymorphisms in the genes LOC387715/ARMS2 is also strongly correlated
with the progression of AMD and related visual loss (Schaumberg, Hankinson et al. 2007;
Seddon, Francis et al. 2007). The LOC387715/ARMS2 gene encodes a mitochondrial protein of
unknown function (Barot, Gokulgandhi et al. 2011). The prospective human studies also found
that smoking and obesity, other known risk factors for AMD, multiply the risk of developing the
disease. The fact that these risk factors are all independently correlated with the development of
AMD indicates that the pathogenesis of AMD may be multifactorial.
Treatment -Clinical Studies of RPE Transplantation
Currently, there is no effective treatment available for dry AMD due to the complex
pathogenesis of the disease. However, many different surgical approaches have been developed
for wet AMD in order to replace the dysfunctional RPE with healthy RPE in situ to prevent
further retinal degeneration. Macular translocation is a surgery that used the concept of RPE
transplantation, but instead of transplanting RPE, it moved the fovea photoreceptor to an area
with healthy RPE support by surgically rotating the retina (Toth, Lapolice et al. 2004). The
results of the surgery demonstrated significant improvement in visual function (Toth, Lapolice et
al. 2004). However, the surgery is complicated and carries a large risk of ocular complications so
16
it is rarely performed nowadays. The results of the surgery imply that placing healthy RPE in the
degenerating retina can rescue the remaining photoreceptors. Other surgical approaches that use
the concept of autologous RPE patch transplant also involve complicated procedures. Moreover,
those procedures requires sacrificing part of the peripheral vision from where the graft is taken
from because the adult RPE has no proliferation capacity in vivo, and the donor site will remain
without vision (Binder, Stanzel et al. 2007). Despite the significant drawbacks of these
approaches, recent studies have shown that restoring a proper niche for photoreceptors with RPE
transplantation can help slow the degeneration process, which might be helpful in the case of dry
AMD, where the known pathogenesis is mostly related to the dysfunctional RPE (da Cruz, Chen
et al. 2007). However, a technique for in vitro expanded RPE graft is needed in order to avoid
the necessity of taking host tissue and performing high risk surgery.
Cells for Transplantation
A variety of cell lines have been tested as candidates to restore the function of the
photoreceptors. The cells that are considered for transplant must be non-toxic to the host tissue
and have no tumorigenicity. The preparations that show potential for visual functional
improvement and cellular integrations are fetal retinal sheets (Aramant and Seiler 2004), retinal
progenitor cells (Tucker, Redenti et al. 2010), and RPE (Little, Castillo et al. 1996; Seiler and
Aramant 1998; Lund, Kwan et al. 2001; Aramant and Seiler 2004; Binder, Stanzel et al. 2007). It
would be ideal if implants that have the potential to regenerate or replace the photoreceptor cells
such as fetal retinal sheets or retinal progenitor cells can be transplanted. However, in order for
the transplanted photoreceptors to form neural connections with the host retina, the
transplantation needs to be performed at early stage of development (Aramant and Seiler 2004).
The other drawback of using fetal retinal sheets or retinal progenitor cells is that these cells are
17
obtained from the retinas of embryos which will remain an ethical issue when moving to human
clinical trials.
On the other hand, the RPE connects with the photoreceptors through microvilli on its
apical surface and does not require a neuronal connection. Thus it would be more feasible to use
the RPE for transplantation.
RPE Delivered as a Cell Suspension May Not Function as Normal RPE
Several RPE transplantation studies have been conducted in animal models using
different cell preparations, e.g. human fetal RPE, ARPE19 (a spontaneously immortalized human
RPE cell line), genetically engineered human RPE cell lines, human embryonic stem cells
(hESC) derived RPE, human iPS-derived RPE, Schwann cells (Little, Cox et al. 1998; Lund,
Adamson et al. 2001; Lund, Kwan et al. 2001; Carr, Vugler et al. 2009; Lu, Malcuit et al. 2009;
Pinilla, Cuenca et al. 2009). The majority of these studies demonstrated some rescue of
photoreceptors following subretinal injection of RPE cell suspension. Sauve et al used ARPE-19,
a spontaneous immortalized human RPE cell line, for subretinal injection into the RCS rat's eye
and demonstrated the preservation of electrophysiological response (electroretinography , ERG)
until postnatal age day (P) 60(Sauve, Pinilla et al. 2006). During long-term observation (P120),
the amplitude of the ERG responses considerably decreased, suggesting a neurotrophic effect
related to the cell treatment itself or the surgical procedure causing a transient up-regulation of
trophic factors. Idelson et al delivered hESC-derived RPE (hESC-RPE) into the subretinal space
of RCS rats and observed only transient preservation of the retinal function as evidenced by ERG
recordings (Idelson, Alper et al. 2009). In the majority of the investigations, the injected cells
remained as cell clumps without polarity and failed to demonstrate faithful integration with the
18
host retina (Little, Cox et al. 1998; Lund, Adamson et al. 2001; Lund, Kwan et al. 2001; Pinilla,
Cuenca et al. 2009).
The other potential problem with RPE cell suspension therapy for AMD lies in the fact
that the RPE is an anchorage dependent tissue that needs to survive on a sheet (Gilmore 2005). In
the normal condition, it attaches to the Bruch’s membrane. However, in the case of AMD, the
Bruch’s membrane has been damaged by the disease or aging process and may not be able to
provide the niche for RPE to survive (Moore and Clover 2001; Gullapalli, Sugino et al. 2005).
In vitro Evidence Showed That Polarized RPE Behave Closer to Physiological Condition
Cell polarity refers to spatial differences in the shape, structure of the organelle that
enables development of specific function (Bergstralh and St Johnston 2012). Epithelial cells are
one example of cells that exert functional difference at different side of the cells by having
apical-basal polarity (Bergstralh and St Johnston 2012). Observations based on in vitro analysis
suggested that, compared to the non-polarized RPE, the polarized RPE behave more close to
normal or innate RPE (Zhu, Deng et al. 2011). This is based on the morphology of the polarized
RPE that shows presence of microvilli on the apical surface and formation of basal lamina on the
basal surface (Figure 3B, C) [Adapted from (Zhu, Deng et al. 2011)]. In the innate RPE, such an
organization is required for interdigitation with photoreceptors on one side and with the Bruch’s
membrane on the other side (Zhu, Deng et al. 2011). Another significant feature of the polarized
RPE monolayer is the presence of interconnecting tight junctions that are critical in the
maintenance of blood-retinal barrier in the retina (Figure 3A, C). Polarized RPE also
demonstrates polarity in the expression of trophic factors, a phenomenon observed in innate RPE.
In innate RPE, the apical surface expresses more pigment epithelium-derived factor (PEDF) to
19
support photoreceptors and the basal side more VEGF to support the metabolic exchange
between the RPE and choroidal vasculature.
Figure 3. Confocal immunofluorescence, scanning electron microscopic (SEM), and transmission electron
microscopic (TEM) images of polarized hESC-RPE. (A) The highly polarized hESC-RPE cells exhibited positive
immunofluorescent staining of the tight junction proteins ZO-1 (green). (B) The polarized sheet of hESC-RPE
showed apical localization of microvilli on scanning electron microscopic image. (C) Transmission electron
microscopic image of two cells in a polarized hESC-RPE sheet shows apical microvilli, melanin granules in apical
cytoplasm (black arrows) and apical tight junctions (white arrow) between the two cells.
Other RPE specific genes and their protein products were examined in hESC-RPE,
including RPE65 and bestrophin, cellular retinaldehyde-binding protein and PEDF (Krill, Morse
et al. 1966; Bunt-Milam and Saari 1983; Steele, Chader et al. 1993; Katz, Wendt et al. 2005;
Pang, Chang et al. 2005). Reverse transcription polymerase chain reaction (RT-PCR),
immunofluorescent staining and Western blot analysis were used to determine whether the RPE
cells derived from hESC cells can express these RPE-specific or related genes in cultured hESC-
RPE cells. By RT-PCR, Western blot, and immunofluorescent staining, RPE65 was expressed in
polarized hESC-RPE cultures, but not in non-polarized hESC-RPE cultures. By RT-PCR, VMD2
also showed higher expression in polarized hESC-RPE when compared with non-polarized
hESC-RPE. Polarized hESC-RPE also expressed PEDF at a higher level compared to non-
polarized hESC-RPE. The above in vitro assays suggest that polarized hESC-RPE behave close
to normal innate RPE. We posit that RPE cells injected as a suspension do not survive in the
20
subretinal space long term because the cells did not regain polarity after implantation and thus do
not have the molecular and functional attributes of normal RPE.
Material of Substrate for Monolayer RPE Culture and Transplantation
The material used for RPE culture and transplantation should carry several properties to
be ideal (Binder, Stanzel et al. 2007). First, it needs to support the growth and maintenance of
RPE phenotype (Binder, Stanzel et al. 2007). Secondly, it needs to allow molecule transport
because a nonpermeable material can block the metabolic support from the choriocapillaris
(Binder, Stanzel et al. 2007). Thirdly, it must be biocompatible to avoid host reaction (Binder,
Stanzel et al. 2007). Fourthly, it needs to have a physical property that allows for surgical
manipulation but also remains relatively thin in order to fit into the subretinal space without
displacing the retina (Binder, Stanzel et al. 2007).
Many non-biodegradable materials have been tested on animal surgical model, for
example, hydrogel (Singh, Woerly et al. 2001), polyurethane (Williams, Krishna et al. 2005),
polyester (Stanzel, Liu et al. 2012). Biodegradable substrate have been tested as well, for
example, PHBV8 film (Tezcaner, Bugra et al. 2003), Poly-L-lactic acid (Hadlock, Singh et al.
1999), collagen (Thumann, Viethen et al. 2009). Binder et al reviewed the benefits and negatives
of each substrate and found that all materials have some drawbacks (Binder, Stanzel et al. 2007).
The biodegradable materials are generally thicker with limited permeability, while the non-
biodegradable material can be fabricated to a very thin layer with pores that facilitate molecular
diffusion. Tissue reaction was also observed in many studies using biodegradable material
(Binder, Stanzel et al. 2007). The other potential issue with biodegradable substrate is that the
RPE cells are anchorage dependent cells and may not be able to survive when the substrate starts
21
to degrade. Based on this information, using a non-biodegradable substrate seems to be a more
feasible option.
Our study used a non-biodegradable parylene-C membrane to grow hESC-RPE and
subsequent implantation in the rat’s subretinal space. It is a USP class VI biocompatible polymer
which is used in the manufacturing of medical devices, e.g. the retinal prosthesis (Rodger, Fong
et al. 2008; Li, Rodger et al. 2011). Our implant is fabricated with additional ultrathin areas (0.3
µm) which are permeable to molecules up to 1008 kD (Lu, Liu et al. 2011) (Figure 4]. This
enables the passage of nutrients and waste materials between the outer retina and the choroid
comparable to the RPE basement membrane and Bruch’s membrane complex (Lu, Liu et al.
2011).
Figure 4. (A) The design of the parylene substrate used for culturing hESC-RPE. The substrate is 6.5 um thick
containing ultrathin areas. The ultrathin areas (thickness 0.3µm, diameter 35µm) are represented by circles. (B) A
rat parylene implant seeded with confluent hESC-RPE cultured on top of it.
Animal Model
There is no ideal animal model of dry AMD due to the complexity of disease. Because
researchers hypothesized that the dysfunctional RPE is the key element of pathogenesis in dry
AMD, it is reasonable that many groups used animal models of RPE dysfunction for
transplantation study of potential therapy for AMD. The Royal College of Surgeon (RCS) rat is
A B
22
widely used an animal model of RPE dysfunction for transplantation studies (Li and Turner 1991;
Sauve, Klassen et al. 1998; Pardue, Phillips et al. 2005; Lu, Malcuit et al. 2009). It has a
naturally occurring mutation of the Mer tyrosine kinase gene (MerTK) (Nandrot and Dufour
2010). This mutation causes a defect in the RPE phagocytosis of POS. Previous studies showed
that the degeneration of the retina in the RCS rat begins at approximately P20 (Valter, Maslim et
al. 1998) and reaches 20% of the outer nuclear layer loss by P32 (our unpublished observations).
This causes accumulation of toxic debris in the subretinal space and secondary retinal
degeneration (Matthes and LaVail 1989). The other RPE dysfunction model includes
intraperitoneally administration of sodium iodate, which causes both RPE and photoreceptor loss
that limit it use for studies looking at rescue effect of the retina (Kiuchi, Yoshizawa et al. 2002).
The other RPE dysfunction model uses mechanical debridement of Bruch’s membrane during
surgery (Phillips, Sadda et al. 2003).
Time of intervention is critical for studies looking at the therapeutic effects in animal
models of degenerative disease (Wang, Lu et al. 2008). Sheedlo et al examined the photoreceptor
cells in RCS rats transplanted with RPE cells at different ages (P10, 17, and 26) and found that
the rats transplanted at P17 showed more preservation of photoreceptors at 2 months of age
(Sheedlo, Li et al. 1991). Most researchers choose to implant the cells before P21, where the
degeneration is at the very early stage, in order to obtain maximum therapeutic effect in
preventing the degeneration. However, in clinical practice, it will be unlikely to treat patients
with dry AMD at a very early stage because the visual impairment would not be to an extent that
the patient would seek surgical therapy. In our study, we chose to intervene at P27 to P29, where
the retina is at the early-mid stage of degeneration, so as to better resemble the clinical situation
where the patient receiving treatment would have some extent of symptoms. The other reason for
23
the treatment for late intervention is because the rat eye is still developing before P21.
Transplantation surgery would be more difficult to perform in smaller eyes and carries more risk
of displacing the implant into other layers when the rat eye grows.
Transplantation of hESC-RPE Cultured on Parylene Membrane
The surgical approach for rodents differs from human surgery due to several anatomic
differences. Rodents have a small eye, with their crystalline lens taking up to 90% of the space in
the vitreous cavity. It is not feasible to perform surgery through an internal approach as done in
human subjects where a vitrectomy is performed. Transplantation techniques for rats were
modified from Pardue et al using a transcleral approach. (Pardue, Phillips et al. 2005) We were
able to transplant hESC-RPE coated on parylene substrates into the rat’s subretinal space (Figure
5). Possible surgical sheer force effect on the implanted cells was evaluated based on
microscopic evaluation of the hESC-RPE over the implant. The results showed that surgery
caused only less than 2% cell loss that not may not be critical considering its influence on
therapeutic benefits (Yuntao Hu et al, article in press). Our preliminary data demonstrates the
feasibility of this approach for subretinal placement of the hESC-RPE grafts in rats.
Figure 5. Fundus examination of the implanted eye performed
immediately after surgery. Because the retinal vessels are found over the
implant, the fundus image suggests successful subretinal placement.
24
Structure of Thesis
This thesis examines how the polarized RPE will survive and function when implanted as
a monolayer into a diseased retina. Chapter 2 presents the short term in vivo visual functional
evaluation focusing on superior colliculus recording, whereas chapter 3 discusses the histologic
analysis of the function and survival of transplanted polarized RPE and the preservation of ONL.
In chapter 4, we report the long-term results of the study, and in chapters 4 and 5, we focus on
the development of in vivo examination methods that can be clinically applicable for evaluation
of results of transplantation.
25
CHAPTER 1 REFERENCES
Ambati, J., B. K. Ambati, et al. (2003). "Age-related macular degeneration: Etiology,
pathogenesis, and therapeutic strategies." Survey of Ophthalmology 48(3): 257-293.
Anderson, D. H., R. F. Mullins, et al. (2002). "A role for local inflammation in the formation of
drusen in the aging eye." Am J Ophthalmol 134(3): 411-431.
Aramant, R. B. and M. J. Seiler (2004). "Progress in retinal sheet transplantation." Prog Retin
Eye Res 23(5): 475-494.
Barot, M., M. R. Gokulgandhi, et al. (2011). "Mitochondrial dysfunction in retinal diseases."
Curr Eye Res 36(12): 1069-1077.
Bergstralh, D. T. and D. St Johnston (2012). "Epithelial cell polarity: what flies can teach us
about cancer." Essays Biochem 53: 129-140.
Bianchi, M. E. (2007). "DAMPs, PAMPs and alarmins: all we need to know about danger." J
Leukoc Biol 81(1): 1-5.
Binder, S., B. V. Stanzel, et al. (2007). "Transplantation of the RPE in AMD." Prog Retin Eye
Res 26(5): 516-554.
Boulton, M., A. Dontsov, et al. (1993). "Lipofuscin is a photoinducible free radical generator." J
Photochem Photobiol B 19(3): 201-204.
Bressler, N. M., J. C. Silva, et al. (1994). "Clinicopathologic correlation of drusen and retinal
pigment epithelial abnormalities in age-related macular degeneration." Retina 14(2): 130-
142.
Brown, D. M., M. Michels, et al. (2009). "Ranibizumab versus verteporfin photodynamic
therapy for neovascular age-related macular degeneration: Two-year results of the
ANCHOR study." Ophthalmology 116(1): 57-65 e55.
26
Bunt-Milam, A. H. and J. C. Saari (1983). "Immunocytochemical localization of two retinoid-
binding proteins in vertebrate retina." J Cell Biol 97(3): 703-712.
Carr, A. J., A. A. Vugler, et al. (2009). "Protective effects of human iPS-derived retinal pigment
epithelium cell transplantation in the retinal dystrophic rat." PLoS One 4(12): e8152.
da Cruz, L., F. K. Chen, et al. (2007). "RPE transplantation and its role in retinal disease." Prog
Retin Eye Res 26(6): 598-635.
Del Priore, L. V., H. J. Kaplan, et al. (1996). "Retinal pigment epithelial debridement as a model
for the pathogenesis and treatment of macular degeneration." Am J Ophthalmol 122(5):
629-643.
Edwards, A. O., R. Ritter, 3rd, et al. (2005). "Complement factor H polymorphism and age-
related macular degeneration." Science 308(5720): 421-424.
Eskelinen, E. L. and P. Saftig (2009). "Autophagy: a lysosomal degradation pathway with a
central role in health and disease." Biochim Biophys Acta 1793(4): 664-673.
Fine, S. L., J. W. Berger, et al. (2000). "Age-related macular degeneration." N Engl J Med
342(7): 483-492.
Friedman, D. S., B. J. O'Colmain, et al. (2004). "Prevalence of age-related macular degeneration
in the United States." Arch Ophthalmol 122(4): 564-572.
Gilmore, A. P. (2005). "Anoikis." Cell Death Differ 12 Suppl 2: 1473-1477.
Gonzales, C. R. (2005). "Enhanced efficacy associated with early treatment of neovascular age-
related macular degeneration with pegaptanib sodium: an exploratory analysis." Retina
25(7): 815-827.
Gullapalli, V. K., I. K. Sugino, et al. (2005). "Impaired RPE survival on aged submacular human
Bruch's membrane." Exp Eye Res 80(2): 235-248.
27
Hadlock, T., S. Singh, et al. (1999). "Ocular cell monolayers cultured on biodegradable
substrates." Tissue Eng 5(3): 187-196.
Hageman, G. S., D. H. Anderson, et al. (2005). "A common haplotype in the complement
regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular
degeneration." Proc Natl Acad Sci U S A 102(20): 7227-7232.
Haines, J. L., M. A. Hauser, et al. (2005). "Complement factor H variant increases the risk of
age-related macular degeneration." Science 308(5720): 419-421.
Hartl, F. U. and M. Hayer-Hartl (2002). "Molecular chaperones in the cytosol: from nascent
chain to folded protein." Science 295(5561): 1852-1858.
Idelson, M., R. Alper, et al. (2009). "Directed differentiation of human embryonic stem cells into
functional retinal pigment epithelium cells." Cell Stem Cell 5(4): 396-408.
Jakobsdottir, J., Y. P. Conley, et al. (2005). "Susceptibility genes for age-related maculopathy on
chromosome 10q26." Am J Hum Genet 77(3): 389-407.
Katz, M. L., K. D. Wendt, et al. (2005). "RPE65 gene mutation prevents development of
autofluorescence in retinal pigment epithelial phagosomes." Mech Ageing Dev 126(4):
513-521.
Kijlstra, A., E. La Heij, et al. (2005). "Immunological factors in the pathogenesis and treatment
of age-related macular degeneration." Ocul Immunol Inflamm 13(1): 3-11.
Killingsworth, M. C., J. P. Sarks, et al. (1990). "Macrophages related to Bruch's membrane in
age-related macular degeneration." Eye (Lond) 4 ( Pt 4): 613-621.
Kinnunen, K., G. Petrovski, et al. (2012). "Molecular mechanisms of retinal pigment epithelium
damage and development of age-related macular degeneration." Acta Ophthalmol 90(4):
299-309.
28
Kiuchi, K., K. Yoshizawa, et al. (2002). "Morphologic characteristics of retinal degeneration
induced by sodium iodate in mice." Curr Eye Res 25(6): 373-379.
Klein, R. J., C. Zeiss, et al. (2005). "Complement factor H polymorphism in age-related macular
degeneration." Science 308(5720): 385-389.
Kopito, R. R. (2000). "Aggresomes, inclusion bodies and protein aggregation." Trends Cell Biol
10(12): 524-530.
Krill, A. E., P. A. Morse, et al. (1966). "Hereditary vitelliruptive macular degeneration." Am J
Ophthalmol 61(6): 1405-1415.
Li, L. and J. E. Turner (1991). "Optimal conditions for long-term photoreceptor cell rescue in
RCS rats: the necessity for healthy RPE transplants." Exp Eye Res 52(6): 669-679.
Li, W., D. C. Rodger, et al. (2011). "Parylene-based integrated wireless single-channel
neurostimulator." Sensors and Actuators A: Physical 166(2): 193-200.
Little, C. W., B. Castillo, et al. (1996). "Transplantation of human fetal retinal pigment
epithelium rescues photoreceptor cells from degeneration in the Royal College of
Surgeons rat retina." Invest Ophthalmol Vis Sci 37(1): 204-211.
Little, C. W., C. Cox, et al. (1998). "Correlates of photoreceptor rescue by transplantation of
human fetal RPE in the RCS rat." Exp Neurol 149(1): 151-160.
Lu, B., Z. Liu, et al. (2011). Ultrathin parylene-C semipermeable membranes for biomedical
applications. Micro Electro Mechanical Systems (MEMS), 2011 IEEE 24th International
Conference. Cancun, Mexico, IEEE Xplore: 505-508.
Lu, B., C. Malcuit, et al. (2009). "Long-term safety and function of RPE from human embryonic
stem cells in preclinical models of macular degeneration." Stem Cells 27(9): 2126-2135.
29
Lund, R. D., P. Adamson, et al. (2001). "Subretinal transplantation of genetically modified
human cell lines attenuates loss of visual function in dystrophic rats." Proc Natl Acad Sci
U S A 98(17): 9942-9947.
Lund, R. D., A. S. Kwan, et al. (2001). "Cell transplantation as a treatment for retinal disease."
Prog Retin Eye Res 20(4): 415-449.
Matthes, M. T. and M. M. LaVail (1989). "Inherited retinal dystrophy in the RCS rat:
composition of the outer segment debris zone." Prog Clin Biol Res 314: 315-330.
Moore, D. J. and G. M. Clover (2001). "The effect of age on the macromolecular permeability of
human Bruch's membrane." Invest Ophthalmol Vis Sci 42(12): 2970-2975.
Nandrot, E. F. and E. M. Dufour (2010). "Mertk in daily retinal phagocytosis: a history in the
making." Adv Exp Med Biol 664: 133-140.
Nowak, J. Z. (2006). "Age-related macular degeneration (AMD): pathogenesis and therapy."
Pharmacol Rep 58(3): 353-363.
Olson, J. H., J. C. Erie, et al. (2011). "Nutritional supplementation and age-related macular
degeneration." Semin Ophthalmol 26(3): 131-136.
Pang, J. J., B. Chang, et al. (2005). "Retinal degeneration 12 (rd12): a new, spontaneously arising
mouse model for human Leber congenital amaurosis (LCA)." Mol Vis 11: 152-162.
Pardue, M. T., M. J. Phillips, et al. (2005). "Neuroprotective effect of subretinal implants in the
RCS rat." Invest Ophthalmol Vis Sci 46(2): 674-682.
Petrilli, V., C. Dostert, et al. (2007). "The inflammasome: a danger sensing complex triggering
innate immunity." Curr Opin Immunol 19(6): 615-622.
30
Petrovski, G., E. Berenyi, et al. (2011). "Clearance of dying ARPE-19 cells by professional and
nonprofessional phagocytes in vitro- implications for age-related macular degeneration
(AMD)." Acta Ophthalmol 89(1): e30-34.
Phillips, S. J., S. R. Sadda, et al. (2003). "Autologous transplantation of retinal pigment
epithelium after mechanical debridement of Bruch's membrane." Curr Eye Res 26(2): 81-
88.
Pinilla, I., N. Cuenca, et al. (2009). "Intraretinal processing following photoreceptor rescue by
non-retinal cells." Vision Res 49(16): 2067-2077.
Rivera, A., S. A. Fisher, et al. (2005). "Hypothetical LOC387715 is a second major susceptibility
gene for age-related macular degeneration, contributing independently of complement
factor H to disease risk." Hum Mol Genet 14(21): 3227-3236.
Rodger, D. C., A. J. Fong, et al. (2008). "Flexible parylene-based multielectrode array
technology for high-density neural stimulation and recording." Sensors and Actuators B:
Chemical 132(2): 449-460.
Sauve, Y., H. Klassen, et al. (1998). "Visual field loss in RCS rats and the effect of RPE cell
transplantation." Exp Neurol 152(2): 243-250.
Sauve, Y., I. Pinilla, et al. (2006). "Partial preservation of rod and cone ERG function following
subretinal injection of ARPE-19 cells in RCS rats." Vision Res 46(8-9): 1459-1472.
Schaumberg, D. A., S. E. Hankinson, et al. (2007). "A prospective study of 2 major age-related
macular degeneration susceptibility alleles and interactions with modifiable risk factors."
Arch Ophthalmol 125(1): 55-62.
Schutt, F., S. Davies, et al. (2000). "Photodamage to human RPE cells by A2-E, a retinoid
component of lipofuscin." Invest Ophthalmol Vis Sci 41(8): 2303-2308.
31
Seddon, J. M., P. J. Francis, et al. (2007). "Association of CFH Y402H and LOC387715 A69S
with progression of age-related macular degeneration." JAMA 297(16): 1793-1800.
Seiler, M. J. and R. B. Aramant (1998). "Intact sheets of fetal retina transplanted to restore
damaged rat retinas." Invest Ophthalmol Vis Sci 39(11): 2121-2131.
Sheedlo, H. J., L. Li, et al. (1991). "Photoreceptor cell rescue at early and late RPE-cell
transplantation periods during retinal disease in RCS dystrophic rats." J Neural
Transplant Plast 2(1): 55-63.
Singh, S., S. Woerly, et al. (2001). "Natural and artificial substrates for retinal pigment epithelial
monolayer transplantation." Biomaterials 22(24): 3337-3343.
Stanzel, B. V., Z. Liu, et al. (2012). "Subretinal delivery of ultrathin rigid-elastic cell carriers
using a metallic shooter instrument and biodegradable hydrogel encapsulation." Invest
Ophthalmol Vis Sci 53(1): 490-500.
Steele, F. R., G. J. Chader, et al. (1993). "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 90(4): 1526-1530.
Strauss, O. (2005). "The retinal pigment epithelium in visual function." Physiol Rev 85(3): 845-
881.
Takeda, A. L., J. Colquitt, et al. (2007). "Pegaptanib and ranibizumab for neovascular age-related
macular degeneration: a systematic review." Br J Ophthalmol 91(9): 1177-1182.
Tezcaner, A., K. Bugra, et al. (2003). "Retinal pigment epithelium cell culture on surface
modified poly(hydroxybutyrate-co-hydroxyvalerate) thin films." Biomaterials 24(25):
4573-4583.
32
Thumann, G., A. Viethen, et al. (2009). "The in vitro and in vivo behaviour of retinal pigment
epithelial cells cultured on ultrathin collagen membranes." Biomaterials 30(3): 287-294.
Toth, C. A., D. J. Lapolice, et al. (2004). "Improvement in near visual function after macular
translocation surgery with 360-degree peripheral retinectomy." Graefes Arch Clin Exp
Ophthalmol 242(7): 541-548.
Tucker, B. A., S. M. Redenti, et al. (2010). "The use of progenitor cell/biodegradable MMP2-
PLGA polymer constructs to enhance cellular integration and retinal repopulation."
Biomaterials 31(1): 9-19.
Tuo, J., S. Grob, et al. (2012). "Genetics of immunological and inflammatory components in age-
related macular degeneration." Ocul Immunol Inflamm 20(1): 27-36.
Valter, K., J. Maslim, et al. (1998). "Photoreceptor dystrophy in the RCS rat: roles of oxygen,
debris, and bFGF." Invest Ophthalmol Vis Sci 39(12): 2427-2442.
Wang, S., B. Lu, et al. (2008). "Morphological and functional rescue in RCS rats after RPE cell
line transplantation at a later stage of degeneration." Invest Ophthalmol Vis Sci 49(1):
416-421.
Williams, R. L., Y. Krishna, et al. (2005). "Polyurethanes as potential substrates for sub-retinal
retinal pigment epithelial cell transplantation." J Mater Sci Mater Med 16(12): 1087-1092.
Winkler, B. S., M. E. Boulton, et al. (1999). "Oxidative damage and age-related macular
degeneration." Mol Vis 5: 32.
Wojcik, C. (2002). "Regulation of apoptosis by the ubiquitin and proteasome pathway." J Cell
Mol Med 6(1): 25-48.
Yang, Z., C. Stratton, et al. (2008). "Toll-like receptor 3 and geographic atrophy in age-related
macular degeneration." N Engl J Med 359(14): 1456-1463.
33
Zareparsi, S., K. E. Branham, et al. (2005). "Strong association of the Y402H variant in
complement factor H at 1q32 with susceptibility to age-related macular degeneration."
Am J Hum Genet 77(1): 149-153.
Zhu, D., X. Deng, et al. (2011). "Polarized secretion of PEDF from human embryonic stem cell-
derived RPE promotes retinal progenitor cell survival." Invest Ophthalmol Vis Sci 52(3):
1573-1585.
34
Chapter 2
In Vivo Functional Comparison of Polarized and Non-polarized hESC-RPE Cells
Transplantation in RCS Rats
ABSTRACT
Purpose: To evaluate the visual functional differences in dystrophic Royal College of Surgeons
(RCS) rats after transplantation of retinal pigment epithelial cells derived from H9 human
embryonic stem cells (hESC-RPE) delivered as a polarized monolayer cell sheet on a thin
parylene membrane and non- polarized cell suspension.
Methods: For transplantation of polarized monolayer cell sheet, hESC-RPE cells were cultured
for 3 weeks on parylene sheets; for transplantation of cell suspension, hESC-RPE cells were
cultured for 3 weeks and digested into cell a suspension before implantation. Subretinal
transplantation of hESC-RPE /parylene (approximately 2800 cells/implant, n=6) and subretinal
injection of hESC-RPE cells (5x10
4
/2µl, n=6) were performed in 27 to 29 day old RCS rats. Six
non-implanted RCS rats served as controls. Post-surgical evaluation was performed using
spectral-domain optical coherence tomography (SD-OCT). In order to assess visual function at 2
months post-surgery, we employed optokinetic head tracking (OHT) testing, luminance threshold
measurement from the superior colliculus (SC), and electroretinography (ERG).
Results: SC recording demonstrated significantly lower (p=0.008) luminance threshold in the
hESC-RPE/parylene implanted group (-5.15+0.27 log cd/m
2
) compared to the cell suspension
treated group (-3.31+0.49 log cd/m
2
). Based on SD-OCT, we observed preservation of the
retinal layers at the implanted area was observed in both groups. The SC luminance threshold
map showed good correlation with the location of the implant in the retina. In hESC-
35
RPE/parylene implanted animals, the head tracking duration was higher (4.06+0.76 sec/min)
compared to the suspension injection group (2.56+0.53 sec/min).
Conclusions: Whether the hESC-RPE was implanted as a monolayer or injected as cell
suspension, it maintained preservation of visual function in RCS rats up to 2 months after
surgery. Better visual preservation was observed in the hESC-RPE/parylene implanted animals
as evidenced by SC luminance threshold measurement and OHT testing. Our study suggests that
implantation of hESC-RPE/parylene may be a more desirable therapeutic approach then cell
suspension injection for disease conditions related to RPE dysfunction such as dry age-related
macular degeneration.
INTRODUCTION
In a normal retina, retinal pigment epithelium (RPE) forms a polarized monolayer between
neural retina and the choroid. It has several critical functions that support and help maintain the
various activities of the photoreceptors. The RPE phagocytizes photoreceptor outer segments
(POS), maintains the outer blood–retinal barrier, and controls the exchange of fluid and
molecules between the choroid and the retina (Strauss 2005).
Over the years, various preparations of RPE cells have been studied for developing cell-
based therapies for blindness caused by RPE dysfunction. Previous studies have shown that the
subretinal injection of different RPE preparations could delay the progression of the retinal
degeneration and improve vision rescue in the RCS (Pinilla, Cuenca et al. 2007; Vugler, Carr et
al. 2008; Lu, Malcuit et al. 2009; Schwartz, Hubschman et al. 2012). However, studies using
different RPE cell lines have demonstrated rosette-like structure cell clumps forming in the
subretinal space or macrophage attack of the transplanted cells, and transient visual function
rescue (Wongpichedchai, Weiter et al. 1992; Sauve, Pinilla et al. 2006; Vugler, Carr et al. 2008;
36
Carr, Vugler et al. 2009; Idelson, Alper et al. 2009). These varied findings of RPE
transplantation may occur due to the transplanted cells not being polarized when transplanted. In
vitro studies have shown that the polarized RPE behave more close to normal RPE compared to
the non-polarized RPE, based on the morphology of the polarized RPE, which shows the
presence of microvilli on the apical surface as well as formation of basal lamina on the basal
surface (Zhu, Deng et al. 2011). In the innate RPE such an organization is required for
interdigitation, with photoreceptors on one side and with the Bruch’s membrane on the other
side. Polarized RPE also demonstrates polarity in the expression of trophic factors, a
phenomenon observed in innate RPE (Zhu, Deng et al. 2011). In innate RPE, the apical surface
expressed more pigment epithelium-derived factor to support photoreceptors, and the basal side
exhibited more VEGF to support the metabolic exchange between RPE and choroidal
vasculature (Zhu, Deng et al. 2011). RPE is also known to be an anchorage dependent tissue
which needs to be attached to an extracellular matrix to survive and function normally (Gilmore
2005; Petrovski, Berenyi et al. 2011; Kinnunen, Petrovski et al. 2012).
While the survival of RPE cells delivered as a suspension is variable and depends on
unknown factors, other researchers have proposed that delivering RPE cells as a monolayer
might provide a better chance of long term survival and better preservation of the photoreceptors
because the cells do not need to reorganize while they are in a new environment (Hadlock, Singh
et al. 1999; Singh, Woerly et al. 2001; Tezcaner, Bugra et al. 2003; Williams, Krishna et al.
2005; Thumann, Viethen et al. 2009; Stanzel, Liu et al. 2012). Our approach is to culture and
deliver the RPE cells using a mesh-supported submicron parylene-C membrane, which is
comprised of a non-biodegradable biocompatible (USP class VI biocompatible polymer)
material. The parylene membrane is structurally designed with multiple ultra-thin areas (0.30μm
37
of thickness) which are permeable to molecules and oxygen transportation between the retina
and the choroidal vasculature (Lu, Zhu et al. 2012). Our previous study showed that it is capable
of providing support to the RPE cells during culture and surgery with minimal cell loss (Hu, Liu
et al. 2012).
Tissue replacement with human embryonic stem cells (hESC) is considered to be a
promising cell-base therapy in different fields of medicine for the past decade (Thomson 1998;
Huang, Enzmann et al. 2011). hESC have the potential to differentiate into various cell types and
provide unlimited source for cell replacement therapies. Our studies used hESC derived RPE
cells cultured on parylene membrane (hESC-RPE/parylene) for transplantation in Royal College
of Surgeon (RCS) rats to compare functional difference between polarized RPE monolayer
implantation vs. non-polarized hESC-RPE cell suspension injection. Post-surgical evaluation
was performed using spectral-domain optical coherence tomography (SD-OCT). Optokinetic
head tracking (OHT) testing, superior colliculus (SC), and electroretinography (ERG) were
employed for visual functional assessments at 2 months after surgery. We also compared the in
vivo functional testing to see how the results of non-invasive testing compare to the gold
standard of visual function testing, SC recording. The follow up period was selected as 2 months
after surgery, because this is the time point where the transient rescue effect from the sham
implant diminishes (Sauve, Pinilla et al. 2006).
MATERIALS AND METHODS
Animals
Pigmented dystrophic RCS rats were used for subretinal polarized hESC-RPE
transplantation (n=6) and cell suspension injection (non-polarized hESC-RPE) (n=6)
38
transplantation experiments. The RCS rat has a recessive mutation in the receptor tyrosine kinase
gene, MerTk. This mutation causes a defect in the RPE phagocytosis of photoreceptor outer
segments (POS) and subsequent photoreceptors degeneration. All animals were maintained on
oral prednisolone (0.002mg/liter) through drinking water starting 2 days prior the surgery and
until they were sacrificed. Rats were housed under a standard 12-hour light-dark cycle. Six
dystrophic RCS rats served as un-operated controls. All animals were treated in accordance with
the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and guidelines
from the Institutional Animal Care and Use Committee at the University of Southern California.
Cell Preparation
Human embryonic stem cells (H9) were used for derivation of RPE using the spontaneous
differentiation method as described previously (Zhu, Deng et al. 2011). RPE cells were
enzymatically enriched and cultured for 3 weeks. For the monolayer hESC-RPE implant, the
cells were grown on a custom made rectangular parylene membrane (0.4mm x 0.9mm x 6.5 µm)
described previously (Hu, Liu et al. 2012); for the hESC-RPE cell suspension, the cells were
grown on petri dish 3 weeks and cells were digested into cell suspension at the cell density of
2.5x10
7
/ml prior surgery.
Subretinal hESC-RPE Suspension Injection
Subretinal injections of hESC-RPE cells were performed in RCS rats 27 to 29 days post-
natal. Each rat was anaesthetized by intraperitoneal injection of ketamine/xylazine (3.75 mg/kg
ketamine and 5 mg/kg xylazine. Each cornea was treated with 0.5% tetracaine (Akorn Inc., Lake
Forest, Illinois) and the pupils dilated using 2.5% phenylephrine (Akorn Inc., Lake Forest,
Illinois) and 0.5% tropicamide (Akorn Inc., Lake Forest, Illinois). The eye was then rotated
39
inferiorly with a traction suture. Paracentesis was performed through a puncture at the limbus
with a 30 gauge needle to decrease pressure in the eye. The injection was performed using a 10
µl Hamilton syringe attached to a fine glass pipette (internal diameter 75–150 μm) through a
small scleral incision created by a 30 gauge needle. A subretinal injection of hESC-RPE cells
(5X104, 2µl) was performed. Immediately after injection, a fundus examination was performed
to confirm the presence of a retinal bleb.
Subretinal hESC-RPE Monolayer Transplantation
The procedures for hESC-RPE monolayer transplantation began similarly to the subretinal
hESC-RPE suspension transplantation and continued until the step where the subretinal injection
was performed. Instead of injecting the hESC-RPE suspension with a glass pipette needle,
balanced salt solution (BSS, Alcon Medical, Ft. Worth, TX) was injected into the subretinal
space using a 33 gauge needle attached to a 10 µl Hamilton syringe. Then a scleral incision was
enlarged from the needle punctured site where the BSS was injected. The hESC-RPE parylene
implant was inserted through the scleral incision using fine forceps. Following surgery, a fundus
examination was performed to evaluate the location of the implant inside the eye. An SD-OCT
examination was performed 1 week after implantation to further confirm the location of the
implant in the subretinal space. Rats without proper placement of the implant were eliminated.
Electroretinography
ERG responses were recorded 1 and 2 months after surgery. Each rat was anaesthetized by
intraperitoneal injection of ketamine/xylazine (3.75 mg/kg ketamine and 5 mg/kg xylazine). Both
corneas were treated with 0.5% tetracaine and the pupils dilated with topical application 2.5%
phenylephrine and 0.5% tropicamide on each cornea. ERG was recorded during full-field light
40
stimulation using gold-wired contact lens electrodes (Mayo Corporation, Aichi, Japan) placed on
the surface of the corneas. Subcutaneous 30-gauge platinum needle electrodes inserted into the
forehead and tail were used as the reference and ground electrodes respectively. Under dark-
adapted conditions, a 10-step intensity series from 0.001 to 10 cd*s/m
2
were presented to both
eyes. With increasing intensity, the interstimulus interval increase from 2 to 90 seconds.
Responses of 3 to 10 flashes were averaged to generate a waveform at each flash intensity.
Optokinetic Head Tracking Response (OHT)
OHT was performed every 2 weeks after surgery using a computer based testing apparatus
consisting of four 20-inch computer monitors arranged in square to form an optokinetic chamber.
During, testing the optokinetic stimuli (moving vertical stripes) was displayed on the computer
screen. The rats were placed in a transparent acrylic round cylinder chamber for 3 minutes or
until they had settled down. A Java-based computer program was used to generate the
optokinetic stimuli, consisting of alternating black and white stripes (3.6 cm wide). During each
test, the rats were exposed to optokinetic stimuli for 2 minutes (1 minute clockwise rotation and
1 minute counterclockwise rotation, randomly starting clockwise or counterclockwise). Three
trials were repeated with a 3 minute rest between each. The duration each rat spent tracking
within 1 minute was averaged for each eye.
Spectral-Domain Optical Coherence Tomography (SD-OCT)
Rats received OCT volume scan 2 months post-surgery. Rats were anesthetized by
intraperitoneal injection of ketamine/xylazine (3.75 mg/kg ketamine and 5 mg/kg xylazine). The
pupils were dilated using topical application 2.5% phenylephrine and 0.5% tropicamide on each
41
cornea. Implanted areas were scanned using spectralis HRA+OCT (Heidelberg Engineering,
Germany). An OCT volume scan consisting of 31 scan lines was performed.
All rats received OCT volume scan at the implanted quadrant and 3 rats in each group
received OCT volume scan in four quadrants of the retina 2 months after surgery in both eyes.
The scan line that showed preservation of outer plexiform layer (OPL) band was plotted into a
contour map using Origin Pro 8.6 (OriginLab, Northampton, MA) and overlapped with the
infrared fundus picture.
Superior Colliculus Responses
Electrophysiological responses were recorded from the superior colliculus (SC) for
assessing the visual function of the rat using methods previously described. Briefly, rats were
dark-adapted overnight and the preparation of the SC recording was performed under dim red
light. Rats received intraperitoneal injection of ketamine/xylazine (3.75 mg/kg ketamine and 5
mg/kg xylazine) for induction of anesthesia, and gas inhalant anesthetic (0.3 to 1% sevoflurane
in 100% O2, flow1.5 liter/minute) for maintenance of anesthesia. Their body temperatures were
maintained at 37°C with a self-regulating heating pad. The rats were held with ear bars and
mounted in a stereotactic apparatus (Kopf Instruments, Tujunga, CA). Then a black dome-
shaped shield was used to cover the left eye to prevent light stimulation during preparation. A
right craniotomy was performed and the cortex was aspirated to expose the SC. Multi-unit visual
responses were recorded extracellularly from the superficial laminae of the SC (stratum griseum
superficiale) using a nail polish coated tungsten microelectrode (custom made in our laboratory)
penetrating 100–200 μm into the SC. An alligator clip electrode attached to the scalp served as
the reference electrode, and a subcutaneous 30-gauge platinum needle electrode inserted into the
42
thigh served as the ground electrode. The recordings were performed under complete darkness
except when moving to different recording sites. Full-field flash stimuli were delivered with a
photic flash visual stimulator (Grass model PS 33 Photic stimulator, W. Warwick, RI) placed 10
cm in front of the rat's eye. At each recording site, we delivered flash stimuli ranging from dim
intensity of -5.4 to 0.6 log cd*s/ m
2
using neutral density filters to establish the luminance
threshold. All electrical activity was recorded using a digital data acquisition system (Powerlab;
ADI Instruments, Mountain View, CA) 100 msec before and 500 msec after the onset of the
stimulus, and we took the average of 6 responses at each site. The visual threshold was defined
as a trace that showed a train of response after light stimulation with amplitude at least twice of
the average background activity. Once the threshold level was found, 6 repeats of a light
stimulus were given (duration 10 µs, interstimulus interval of 10 sec). Recording sites were
moved with stereotactic apparatus (200–400 μm apart) covering the full extent of the SC except
in the area not accessible due overlying vascular structure. The stereotactic co-ordinates were
plotted into a contour map that presented the distribution of the threshold over the SC using
Origin Pro 8.6 (OriginLab, Northampton, MA).
Statistical Analysis
Parametric data are presented as mean ± SEM. Significance was calculated with two-way
ANOVA with Bonferroni post hoc test to compare the OHT tracking duration in the RCS rat at
different time points after treatment. Non-parametric data are presented as scatter plot.
Significance was calculated with Mann–Whitney U test. Pearson correlation was used to study
the correlation between the visual field area and maximum b-wave amplitude in ERG. P<0.05
was considered to be statistically significant.
43
RESULTS
Threshold of Visual Response in Superior Colliculus
The mean number of sites recorded from the SC of each animal was 37±2. The SC view
during recording is shown in figure 1 with the corresponding retinotopic visual field indicated.
The visual response threshold recorded in each animal was plotted on a contour map to
demonstrate a visual field-like map (Figure 2). The areas that showed sensitivity at the low-light
stimulation are mostly located in the anterolateral quadrant, which represents the projections
from the temporodorsal quadrant. This corresponds to the location where the implants were
placed in both the hESC-RPE/parylene and the hESC-RPE suspension treated groups.
Figure 1. The right superior colliculus (SC) exposed during superior colliculus
recording. SC maintains the retinotopic map of the visual field. The anterior part
of SC receives signals from the temporal retina; the lateral part receives signals
from the dorsal retina; the posterior part receives signals from the nasal retina; and
the medial part receives signals from the ventral retina.
Figure 2. The
threshold contour
map of the
superior
colliculus 2
months after
surgery from
each animal in all
three groups. The
warm colors
represent the
areas that
responded at low-
44
light intensity stimulation; the cold colors represent the areas that responded at high-light intensity. The black area
represents the area where no light response can be found at 0.6 log cd/m2, which was the highest level of stimulation
tested. For the hESC-RPE/parylene implanted group and the hESC-RPE suspension treated group, the lower
threshold responses are mostly observed in the anterolateral quadrant of the SC, which corresponds to the
dorsotemporal quadrant of the retina. The response corresponds to the area where the hESC-RPE/parylene and the
suspension were implanted. The non-implanted animal showed response at random areas only at high-intensity light
stimulation.
SC recording demonstrated significantly lower (p=0.008) luminance threshold in the hESC-
RPE/parylene implanted group (-5.15+0.27 log cd/m
2
) compared to the cell suspension treated
group (-3.31+0.49 log cd/m
2
). The area of visual field preservation was presented in figure 3.
The percentage of visual fields (sites recorded) that responded above a certain light intensity
were plotted. Five rats in the hESC-RPE/parylene implanted group maintained a small area that
responded to light stimulus of -4.7 log cd/m
2
or lower; for the hESC-RPE suspension treated
group, only one rat responded at -4.7 log cd/m2. The proportion of preservation visual field
preservation at different light intensities were not significantly different between the hESC-
RPE/parylene treated and hESC-RPE suspension treated group. At -5.4 log cd/m
2
, the Mann-
Whitney test cannot be carried out because all the rats in hESC-RPE suspension group were non-
responsive at this light level.
45
Figure 3. Visual threshold versus
visual field area based on SC
response in hESC-RPE/parylene,
hESC-RPE suspension, and control
group. While all rats in the hESC-
RPE/parylene transplanted group
maintained a small area that
responded to light stimulus of -3.9
log cd/m
2
or lower; for the hESC-
RPE suspension treated group, 3
rats responded at -3.9 log cd/m
2
.
Optokinetic Head Tracking (OHT)
At 8 weeks after surgery, the mean head tracking time is 4.06±0.76 sec in the hESC-
RPE/parylene implanted group and 2.56±0.48 sec in the hESC-RPE suspension treated group
(P>0.05) (Figure 4). Although the difference was not significant, there seems to be a trend of
improvement in OHT duration in the hESC-RPE/parylene treated group, whereas the OHT
duration decreased in the hESC-RPE suspension treated group compared to its previous OHT
duration at 6 weeks after surgery. Both the hESC-RPE/parylene implanted eyes (P<0.001) and
the hESC-RPE suspension treated eyes (P<0.05) showed significantly different OHT durations
compared with their untreated fellow eye.
46
Figure 4. Optokinetic head tracking
duration of hESC-RPE/parylene
implanted, hESC-RPE suspension
treated, and their fellow eyes at
different times after surgery. At 8
weeks after surgery, the mean head
tracking time is 4.06±0.76 sec in
hESC-RPE/parylene treated group
and 2.56±0.48 sec in the hESC-RPE
suspension treated group (P>0.05). The difference between the hESC-RPE/parylene implanted eyes and the fellow
eyes (mean 1.11±0.42 sec) was significant (P<0.001), and the difference between the hESC-RPE suspension treated
eyes and the fellow eyes (mean 0.05±0.29 sec) was significant (P<0.05).
Comparison of Optokinetic Head Tracking Duration and Superior Colliculus Response
Correlation between the OHT tracking duration and the visual field area of SC response was
performed in the hESC-RPE/parylene treated group, the hESC-RPE suspension treated group,
and the two groups combined. The OHT tracking duration is weakly correlated with the
proportion of the visual field area with a threshold of -5.0 log cd/m
2
when the two groups are
added together (P=0.04).
Electroretinography (ERG)
Dark-adapted full filed ERG at 1 month after surgery showed that the mean maximum b-
wave amplitude was 90.89±4.69 µv in the hESC-RPE/parylene implanted group and 305±85 µv
in the hESC-RPE suspension treated group (P=0.02). Two months after surgery, mean maximum
b-wave amplitude was 77.68 ± 15.66 in the hESC-RPE/parylene implanted group and 120.5 ±
47
20.76 µv in the hESC-RPE suspension treated group (P>0.05). The results showed that the
hESC-RPE suspension treatment have a greater transient rescue in retina response. However this
rescue effect rapidly vanishes 2 months after surgery.
Figure 5. The mean maximum b-wave amplitude of
dark-adapted full filed ERG at 1 month and 2 months
after surgery. The mean maximum b-wave amplitude
was 90.89±4.69 µv in the hESC-RPE/parylene
implanted group and 305±85 µv in the hESC-RPE
suspension treated group (P=0.02). The mean
maximum b-wave amplitude was 77.68 ± 15.66 in the
hESC-RPE/parylene implanted group and 120.5 ± 20.76 µv in the hESC-RPE suspension treated group (P>0.05).
ERG comparison with the SC response
The ERG response was compared with the SC response to see if the ERG responses have
predictive values for SC response. The visual field area recorded at different threshold levels
from the SC of each animal was compared with the maximum b-wave amplitude recorded from
the ERG. In the hESC-RPE suspension treated group, there was a significant correlation between
the visual field area and the maximum b-wave amplitude at threshold light intensities of -0.1 log
cd/m
2
(P=0.03), -0.9 log cd/m
2
(P=0.04), -1.6 cd/m
2
(P=0.03), -2.4 log cd/m
2
(P=0.01), -3.1 log
cd/m
2
(P=0.007), -3.9 log cd/m
2
(P=0.004) (Table 1). However, this correlation was not that
significant in the hESC-RPE/parylene implanted group. We only found the correlation between
the visual area and the maximum b-wave amplitude at the threshold level on -0.1 log cd/m
2
(P=0.04) (Table 1). The standard error of the mean of the visual field area preserved at each
threshold level demonstrated that the visual field area was more variable among the hESC-RPE
1 month 2 months
0
100
200
300
400
Suspension
Parylene w/ cells
Time after surgery
Mean B wave amplitude ( v)
48
suspension treated group compared to the hESC-RPE/parylene implanted group at each light
intensity.
Group Threshold (log
cd/m2)
0.6 -0.1 -0.9 -1.6 -2.4 -3.1 -3.9 -4.7 -5.4
hESC-RPE
/parylene
Pearson r -0.150 -0.900 -0.448 -0.148 -0.110 0.231 0.711 0.409 0.390
P value 0.810 0.037 0.449 0.812 0.860 0.709 0.178 0.494 0.516
P value summary ns * ns ns ns ns ns ns ns
R squared 0.022 0.811 0.201 0.022 0.012 0.053 0.506 0.168 0.152
SEM of VF 2.319 1.862 1.906 1.853 2.357 1.762 1.409 1.710 0.619
hESC-RPE
suspension
Pearson r 0.400 0.906 0.905 0.925 0.960 0.969 0.979
P value 0.504 0.034 0.035 0.025 0.010 0.007 0.004
P value summary ns * * * ** ** **
R squared 0.160 0.821 0.820 0.855 0.922 0.93 0.95
9 9
SEM of VF 4.595 7.917 8.977 9.212 8.531 6.146 3.962
Table 1. Correlation between the visual field area and the maximum b-wave amplitude at different threshold levels
in hESC-RPE/parylene and hESC-RPE suspension treated rats 2 months after surgery. The hESC-RPE suspension
treated rats showed a significant correlation between the visual field and the maximum b-wave amplitude at
threshold between -0.1 and -3.9 log cd/m
2
. Correlation could not be done at threshold intensity below -4.7 log cd/m2
because all the rats in the hESC-RPE suspension group had no visual response at these light levels. (SEM of VF,
standard error of the mean of the visual field area preserved at each threshold level.)
The SD-OCT in Transplanted RCS rat
The OCT performed in RCS rats 2 months after surgery of hESC-RPE/parylene implanted
rat showed that all rats had some level of preservation of the OPL band over the implanted and
adjacent area (Figure 6C), whereas 50% of the hESC-RPE suspension treated rats showed
preservation of the OPL band of the quadrant where cells were injected (Figure 6D). The SD-
OCT showed the area with preservation is more focused at the implanted area for hESC-
49
RPE/parylene implanted rats (Fig. 6A) while the hESC-RPE suspension treated rats showed
more scattered OPL preservation (Fig. 6B). The finding is compatible with the findings of SC
recording in that the area of response was located in the dorsotemporal quadrant of the retina and
the cell suspension treated rats had a relatively scattered distribution of response.
Figure 6. Infrared fundus photos (A, B) and OCT scan (C, D) of retina of RCS rat 2 months after surgery. The OCT
scan of hESC-RPE/parylene implanted retina (C) represented the scan of the implanted area (thick green line in A).
OCT showed a preservation of the OPL band in the implanted area indicated with whit arrow (C). The contour plot
of the OPL overlay with the infrared fundus photo showed that the preservation of OPL corresponds to the
implanted area (red). Infrared fundus photo of an RCS rat at the cell injected quadrant 2 months after subretinal
hESC-RPE suspension injection (C). The OCT scan (D) represented the scan of the implanted area (thick green line
in B). The OCT showed a focal preservation of the OPL band near the injection site (D).
50
DISCUSSION
The results of the SC recording showed that the hESC-RPE /parylene implanted rats
showed a significantly lower luminance threshold compared to the subretinal injection group and
the non-treated group. The lower luminance threshold response at scotopic conditions indicated a
higher level of functional preservation of the rod (Thomas, Aramant et al. 2005) which meant
that the hESC-RPE /parylene maintained a better rod driven visual circuitry. Girman et al studied
the course of deterioration of rod and cone function of RCS rats by measuring light- and dark-
adaptation curves from responses of the SC (Girman, Wang et al. 2005). They reported that the
rod function did not recover although the cone responses were preserved despite the fact that
both rod and cones were preserved in histologic evaluation. The results showed that, similar to
clinical observation of retinitis pigmentosa, the rod system is affected early on in the course of
the disease followed by the cone system (Berson 1981), which would make the rod system
harder to rescue. Thus, the preservation of low threshold of scotopic response in SC in hESC-
RPE /parylene would indicate a higher level of functional rescue.
The area that responded in SC at different luminance levels is very different between the
hESC-RPE /parylene and hESC-RPE suspension treated groups. The hESC-RPE/parylene
implanted rats showed a relatively consistent preservation of the visual field area among
different rats. However, the hESC-RPE suspension showed a diverse response. At higher
luminance, some rats in the hESC-RPE suspension had a larger visual field area while some had
a very small visual area (Figure 2, 3). This shows that the effect of hESC-RPE suspension is
variable (Carr, Vugler et al. 2009). There are several possible reasons. First, the survival rate of
the RPE cell suspension in the subretinal space is variable. While some groups showed that the
RPE cell suspensions was able to form monolayer on the surface of host Bruch’s membrane
51
(Schwartz, Hubschman et al. 2012), other studies showed the cell suspensions in the subretinal
space formed multilayer cell clumps or pigmented clumps that damaged the overlaying
photoreceptors which results in poor visual outcome (Wongpichedchai, Weiter et al. 1992). The
diverse visual function outcome in the hESC-RPE suspension treated group is likely related to
the variable survival rate of the implanted cells. Secondly, the amount of cells that maintained in
the subretinal space might be variable. Although the targeted dose for suspension injection was
5X10
4
cells/ 2 μl and a retinal bleb formation was confirmed at the end of each procedure, the
cells may reflux through the incision afterwards and cause the effect to be variable. On the other
hand, the parylene membrane delivers a less variable number of cells (2500-3000 cells
/membrane) (Hu, Liu et al. 2012). The subretinal injection may deliver twenty times more cells
compared to the parylene membrane. The retinal bleb observed during cell suspension injection
takes up around one quadrant of the retina where the parylene membrane takes up about 2% of
the total area of the retina. This is compatible with the finding that the visual field area in the SC
is larger in some of the hESC-RPE suspension treated rats compared to the hESC-RPE /parylene
implanted rats.
The OHT response showed that the difference between the implanted eye (both hESC-RPE
/parylene and suspension group) and the non-implanted eye became significant only at 8 weeks
post-implantation. This indicates that both treatments have some effect in maintaining the visual
function. Although the difference was not significant, the hESC-RPE/parylene implanted group
showed a trend of better OHT response compared to the suspension treated group at 8 weeks
after surgery.
ERG is widely used for evaluation of retinal function in subretinal cell suspension injection
studies (Girman, Wang et al. 2005; Pardue, Phillips et al. 2005; Pinilla, Lund et al. 2005). In the
52
hESC-RPE suspension treated rats, we found a significant correlation between the areas
responded in the SC at higher luminance levels and the b-wave amplitude. However, we did not
find this correlation in the hESC-RPE/parylene treated rats. Similar findings were reported by
Sauve et al where they compared the visual field area responded at different threshold level in
SC and the maximum b-wave amplitude of RCS rats. They found that at the b-wave amplitude
correlates with the area of visual field preservation (Sauve, Lu et al. 2004). The hESC-
RPE/parylene is implanted only in a consistent- sized area of the rat retina which is much smaller
than that of the suspension. Since Since b-wave amplitude is correlated with visual field size,
this smaller visual field area made it harder to detect the differences in b-wave amplitude in the
hESC-RPE/parylene implanted rats. However, the hESC-RPE suspension rats showed a highly
variable visual field area which allows the difference in the b-wave amplitude to be revealed. It
appears that the ERG is less valuable for studies that focus on the treatment of a smaller area.
SD-OCT provides a non-invasive anatomic evaluation of the whole retina. Seiler et al
evaluated the retinal sheet transplants in rats using a Fourier-domain optical coherence
tomography and reported that it is a great examination tool to image the retinal layers and the
laminar structure of the transplants (Seiler, Rao et al. 2010). We used SD-OCT to look at the
preservation of retinal layers. In the RCS rats, most changes in the retina occur in the outer retina
due to the RPE dysfunction. We used the presence of an OPL band instead of outer nuclear layer
(ONL) as an indicator of preservation of the outer retina because the ONL degenerates rapidly
and fuses with the debris layer that forms in the subretinal space, which makes it hard to
determine the actual thickness of the ONL in OCT in RCS rats. The OPL is a thin bright band
that is sandwiched by two dark bands, the inner nuclear layer and the ONL. It can be observed in
the RCS rat when all three layers maintain some level of thickness and it is usually not present
53
by P90 in the RCS rat, thus making it a useful indicator for outer retina preservation (our
unpublished data).
The contour map composited with SD-OCT showed the presence of OPL band over the
parylene implanted retina in the hESC-RPE/parylene treated rats or the cell transplanted quadrant
in the hESC-RPE suspension treated rats. This is compatible with the finding of the topographic
location of SC responses.
The major limitation of our study is that the amount of cells delivered to the subretinal
space in the hESC-RPE suspension rat was much higher than the hESC-RPE/parylene implanted
rat. The amount of cell suspension was planned based on previous studies to obtain a comparison
(Lu, Malcuit et al. 2009). And also our preliminary study showed that injection of 3000 cells
suspension is not practical and frequently end up with no cells in histology which makes it hard
to determine if it was due to cell death or failure of the injection procedure. Despite the low
amount of cells implanted, the hESC-RPE/parylene group showed a lower luminance threshold
compared to the suspension treated group which implies a better rod photoreceptor rescue. This
indicates that the RPE given as a monolayer may have better functional integration with the host
retina.
54
CHAPTER 2 REFERENCES
Abe, T., H. Tomita, et al. (2000). "Autologous iris pigment epithelial cell transplantation in
monkey subretinal region." Curr Eye Res 20(4): 268-275.
Berson, E. L. (1981). "Retinitis pigmentosa and allied diseases: applications of
electroretinographic testing." Int Ophthalmol 4(1-2): 7-22.
Carr, A. J., A. A. Vugler, et al. (2009). "Protective effects of human iPS-derived retinal pigment
epithelium cell transplantation in the retinal dystrophic rat." PLoS One 4(12): e8152.
Gilmore, A. P. (2005). "Anoikis." Cell Death Differ 12 Suppl 2: 1473-1477.
Girman, S. V., S. Wang, et al. (2005). "Time course of deterioration of rod and cone function in
RCS rat and the effects of subretinal cell grafting: a light- and dark-adaptation study."
Vision Res 45(3): 343-354.
Hadlock, T., S. Singh, et al. (1999). "Ocular cell monolayers cultured on biodegradable
substrates." Tissue Eng 5(3): 187-196.
Hu, Y., L. Liu, et al. (2012). "A novel approach for subretinal implantation of ultrathin substrates
containing stem cell-derived retinal pigment epithelium monolayer." Ophthalmic Res
48(4): 186-191.
Hu, Y., L. Liu, et al. (2012). "A Novel Approach for Subretinal Implantation of Ultrathin
Substrates Containing Stem Cell-Derived Retinal Pigment Epithelium Monolayer."
Ophthalmic research 48(4): 186-191.
Huang, Y., V. Enzmann, et al. (2011). "Stem cell-based therapeutic applications in retinal
degenerative diseases." Stem cell reviews 7(2): 434-445.
Idelson, M., R. Alper, et al. (2009). "Directed differentiation of human embryonic stem cells into
functional retinal pigment epithelium cells." Cell Stem Cell 5(4): 396-408.
55
Kinnunen, K., G. Petrovski, et al. (2012). "Molecular mechanisms of retinal pigment epithelium
damage and development of age-related macular degeneration." Acta Ophthalmol 90(4):
299-309.
Lu, B., C. Malcuit, et al. (2009). "Long-term safety and function of RPE from human embryonic
stem cells in preclinical models of macular degeneration." Stem Cells 27(9): 2126-2135.
Lu, B., D. Zhu, et al. (2012). "Mesh-supported submicron parylene-C membranes for culturing
retinal pigment epithelial cells." Biomedical microdevices 14(4): 659-667.
Pardue, M. T., M. J. Phillips, et al. (2005). "Neuroprotective effect of subretinal implants in the
RCS rat." Invest Ophthalmol Vis Sci 46(2): 674-682.
Petrovski, G., E. Berenyi, et al. (2011). "Clearance of dying ARPE-19 cells by professional and
nonprofessional phagocytes in vitro- implications for age-related macular degeneration
(AMD)." Acta Ophthalmol 89(1): e30-34.
Phillips, S. J., S. R. Sadda, et al. (2003). "Autologous transplantation of retinal pigment
epithelium after mechanical debridement of Bruch's membrane." Curr Eye Res 26(2): 81-
88.
Pinilla, I., N. Cuenca, et al. (2007). "Preservation of outer retina and its synaptic connectivity
following subretinal injections of human RPE cells in the Royal College of Surgeons rat."
Exp Eye Res 85(3): 381-392.
Pinilla, I., R. D. Lund, et al. (2005). "Measuring the cone contribution to the ERG b-wave to
assess function and predict anatomical rescue in RCS rats." Vision Res 45(5): 635-641.
Sauve, Y., B. Lu, et al. (2004). "The relationship between full field electroretinogram and
perimetry-like visual thresholds in RCS rats during photoreceptor degeneration and
rescue by cell transplants." Vision Res 44(1): 9-18.
56
Sauve, Y., I. Pinilla, et al. (2006). "Partial preservation of rod and cone ERG function following
subretinal injection of ARPE-19 cells in RCS rats." Vision Res 46(8-9): 1459-1472.
Schwartz, S. D., J. P. Hubschman, et al. (2012). "Embryonic stem cell trials for macular
degeneration: a preliminary report." Lancet 379(9817): 713-720.
Seiler, M. J., B. Rao, et al. (2010). "Three-dimensional optical coherence tomography imaging of
retinal sheet implants in live rats." J Neurosci Methods 188(2): 250-257.
Singh, S., S. Woerly, et al. (2001). "Natural and artificial substrates for retinal pigment epithelial
monolayer transplantation." Biomaterials 22(24): 3337-3343.
Stanzel, B. V., Z. Liu, et al. (2012). "Subretinal delivery of ultrathin rigid-elastic cell carriers
using a metallic shooter instrument and biodegradable hydrogel encapsulation." Invest
Ophthalmol Vis Sci 53(1): 490-500.
Strauss, O. (2005). "The retinal pigment epithelium in visual function." Physiol Rev 85(3): 845-
881.
Tezcaner, A., K. Bugra, et al. (2003). "Retinal pigment epithelium cell culture on surface
modified poly(hydroxybutyrate-co-hydroxyvalerate) thin films." Biomaterials 24(25):
4573-4583.
Thomas, B. B., R. B. Aramant, et al. (2005). "Light response differences in the superior
colliculus of albino and pigmented rats." Neurosci Lett 385(2): 143-147.
Thomson, J. A. (1998). "Embryonic Stem Cell Lines Derived from Human Blastocysts." Science
282(5391): 1145-1147.
Thumann, G., A. Viethen, et al. (2009). "The in vitro and in vivo behaviour of retinal pigment
epithelial cells cultured on ultrathin collagen membranes." Biomaterials 30(3): 287-294.
57
Vugler, A., A. J. Carr, et al. (2008). "Elucidating the phenomenon of HESC-derived RPE:
anatomy of cell genesis, expansion and retinal transplantation." Exp Neurol 214(2): 347-
361.
Williams, R. L., Y. Krishna, et al. (2005). "Polyurethanes as potential substrates for sub-retinal
retinal pigment epithelial cell transplantation." J Mater Sci Mater Med 16(12): 1087-1092.
Wongpichedchai, S., J. J. Weiter, et al. (1992). "Comparison of external and internal approaches
for transplantation of autologous retinal pigment epithelium." Invest Ophthalmol Vis Sci
33(12): 3341-3352.
Zhu, D., X. Deng, et al. (2011). "Polarized secretion of PEDF from human embryonic stem cell-
derived RPE promotes retinal progenitor cell survival." Investigative ophthalmology &
visual science 52(3): 1573-1585.
58
Chapter 3
Histologic Comparison of Polarized and Non-polarized hESC-RPE Cells
Transplantation in RCS Rats
ABSTRACT
Purpose: To evaluate the histologic differences in dystrophic Royal College of Surgeons (RCS)
rats after transplantation of retinal pigment epithelial cells derived from H9 human embryonic
stem cells (hESC-RPE) delivered as a polarized monolayer cell sheet on a thin parylene
membrane and non-polarized cell suspension.
Methods: For transplantation of the polarized monolayer cell sheet, hESC-RPE cells were
cultured for 3 weeks on parylene sheets; for transplantation of cell suspension, hESC-RPE cells
were cultured for 3 weeks and digested into cell suspension before implantation. Subretinal
implantation of hESC-RPE /parylene (approximately 2800 cells/implant, n=6) and subretinal
injection of hESC-RPE cells (5x10
4
/2µl, n=9) were performed in 27 to 29 day old RCS rats.
Animals were sacrificed and eyes enucleated at 2 months after surgery. Tissue sections
underwent histological evaluation using hematoxylin and eosin (H&E) staining and
immunofluorescent staining for RPE markers (RPE65), human marker (TRA-1-85), and immune
markers (CD68, GFAP). The function of the RPE was analyzed by the quantification of
phagosomes within transplanted cells by immunofluorescent staining for rhodopsin. An Aperio
ScanScope was used for the quantification of outer nuclear layer (ONL) cell counts.
Results: Survival of transplanted hESC-RPE was observed in all of the rats implanted with
hESC-RPE/parylene and 22% of the rats implanted with the hESC-RPE suspension based on
immunofluorescent staining (RPE65+TRA-1-85). The phagosomes count is higher in the
59
monolayer hESC-RPE (0.040 ± 0.005/µm
2
) compared to the non-polarized RPE (0.0029 ±
0.0019 /µm
2
, P < 0.0001). The ONL cell count showed more ONL preservation in the hESC-
RPE/parylene implanted rats compared to the hESC-RPE suspension implanted rats (P < 0.05).
The hESC-RPE suspension implanted rats have more CD68 expression around the implanted
cells. There is no difference in GFAP expression between the two groups.
Conclusions: The hESC-RPE implanted as a monolayer demonstrates a higher survival rate
compared to cells delivered as a cell suspension. The hESC-RPE monolayer demonstrates better
RPE function by phagocytosis assay. The low survival rate in hESC-RPE suspension may be
related to the CD68 positive macrophages around the implanted area. Our study suggests that
implantation of hESC-RPE/parylene may be a more desirable therapeutic approach for disease
conditions related to RPE dysfunction such as dry age-related macular degeneration.
INTRODUCTION
In a normal retina, retinal pigment epithelium (RPE) forms a polarized monolayer between
the neural retina and the choroid. It has several critical functions that support and help maintain
the various activities of the photoreceptors. The RPE phagocytize photoreceptor outer segments
(POS), maintain the outer blood–retinal barrier, and control the exchange of fluid and molecules
between the choroid and the retina (Strauss 2005).
Previous animal studies using models of retinal degenerative disease have used subretinal
injections of various cell preparations to demonstrate a slowing of the loss of photoreceptor
function (Little, Cox et al. 1998; Lund, Adamson et al. 2001; Lund, Kwan et al. 2001; Carr,
Vugler et al. 2009; Pinilla, Cuenca et al. 2009). However, many of these studies failed to
establish the degree of integration between the transplanted cells and the host retina needed for
long-term maintenance of phagocytic activity of the implanted RPE (Wongpichedchai, Weiter et
60
al. 1992; Vugler, Carr et al. 2008; Carr, Vugler et al. 2009; Idelson, Alper et al. 2009). It has
been speculated that RPE cells injected as a suspension do not survive in the subretinal space
long term because the cells were not polarized at the time of implantation, and thus do not have
the molecular and functional attributes of normal RPE (Binder, Stanzel et al. 2007; Zhu, Deng et
al. 2011). Transplantation of polarized RPE monolayer will enable better integration with the
host photoreceptors (Hadlock, Singh et al. 1999; Singh, Woerly et al. 2001; Tezcaner, Bugra et
al. 2003; Williams, Krishna et al. 2005; Thumann, Viethen et al. 2009; Stanzel, Liu et al. 2012).
These cells can remain functional and maintain the long-term preservation of photoreceptors.
Human embryonic stem cells have the potential to differentiate into various cell types and
can be an unlimited source for cell replacement therapies. Our transplantation studies used
human embryonic stem cells derived RPE (hESC-RPE) in Royal College of Surgeon (RCS) rats,
a rodent model of RPE dysfunction, to compare polarized RPE monolayer implantation with
non-polarized hESC-RPE cell suspension injection. We evaluated the morphology and function
of the transplanted cells, as well as the preservation of host photoreceptors.
MATERIALS AND METHODS
Animals
Pigmented dystrophic RCS rats were used for subretinal polarized hESC-RPE
transplantation (n=6) and cell suspension injection (non-polarized hESC-RPE) (n=9)
transplantation experiments. RCS rats have a recessive mutation in the receptor tyrosine kinase
gene, MerTk. This mutation causes a defect in the RPE phagocytosis of photoreceptor outer
segments (POS) and photoreceptor degeneration. All animals were maintained on oral
prednisolone (0.002mg/liter) through drinking water starting 2 days prior to surgery and
61
continuing until they were sacrificed. Rats were housed under the standard 12-hour light-dark
cycle. Five RCS rats were implanted with parylene membrane serving as a sham surgery group;
and 4 rats were injected with DMEM/F12 medium, serving as a sham injection group. Five
Copenhagen (normal) rats served as non-operated controls. All animals were treated in
accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision
Research and guidelines from the Institutional Animal Care and Use Committee at the University
of Southern California.
Cell Preparation
Human embryonic stem cells (H9) were used for the derivation of RPE using the
spontaneous differentiation method previously described (Zhu, Deng et al. 2011). RPE cells were
enzymatically enriched and cultured for 3 weeks. For the monolayer hESC-RPE implant, the
cells were grown on a custom made rectangular parylene membrane (0.4mm x 0.9mm x 6.5 µm)
described previously (Hu, Liu et al. 2012); for the hESC-RPE cell suspension, the cells were
grown on petri dishes for 3 weeks and digested into cell suspension at a cell density of
2.5x10
7
/ml prior to surgery.
Subretinal hESC-RPE Suspension Transplantation
Subretinal injections of hESC-RPE cells were performed in RCS rats 27 to 29 days post-
natal. Each rat was anaesthetized by intraperitoneal injection of ketamine/xylazine (3.75 mg/kg
ketamine and 5 mg/kg xylazine. Each cornea was treated with 0.5% tetracaine (Akorn Inc., Lake
Forest, Illinois) and the pupil dilated using 2.5% phenylephrine (Akorn Inc., Lake Forest, Illinois)
and 0.5% tropicamide (Akorn Inc., Lake Forest, Illinois). The eye was then rotated inferiorly
with a traction suture. Paracentesis was performed through a puncture at the limbus with a 30
62
gauge needle to decrease pressure in the eye. The injection was performed using a 10 µl
Hamilton syringe attached to a fine glass pipette (internal diameter 75–150 μm) through a small
scleral incision created by a 30 gauge needle. A subretinal injection of hESC-RPE cells (5X10
4
,
2µl) was performed. Immediately after injection, a fundus examination was performed to
confirm the presence of a retinal bleb.
Subretinal hESC-RPE Monolayer Transplantation
The procedures for hESC-RPE monolayer transplantation began similarly to the
subretinal hESC-RPE suspension transplantation up until the step where the subretinal injection
was performed. Instead of injecting hESC-RPE suspension with a glass pipette needle, a
balanced salt solution (BSS, Alcon Medical, Ft. Worth, TX) was injected into the subretinal
space using a 33 gauge needle attached to a 10 µl Hamilton syringe. Then a scleral incision was
enlarged from the needle-punctured site where the BSS was injected. The hESC-RPE parylene
implant was inserted through the scleral incision with fine forceps. Following surgery, a fundus
examination was performed in order to evaluate the location of the implant inside the eye. SD-
OCT examination was performed one week after implantation to further confirm the location of
the implant in the subretinal space. Rats without proper placement of the implant were
eliminated.
Histology
All animals were euthanized with 0.5 ml of pentobarbitol sodium 390 mg/ml and phenytoin
sodium 50 mg/ml (Euthasol, Virbac AH, Inc, Forth Worth, TX) at around 8AM in the morning 2
months after surgery. The eyes were fixated with Davidson’s solution and embedded with
paraffin. Serial sections (5 µm thick) were taken at the implanted quadrant. Sections
63
(approximately 250 µm apart) were stained using hematoxylin and eosin (H&E) to examine the
morphology of the implanted cells, starting from the first section where the parylene implant (for
the hESC-RPE/parylene implanted eyes) or the injected cells (for the hESC-RPE suspension
implanted eyes) were found. The outer nuclear layer (ONL) cells above the implant (400 µm
length) were counted with an automated cell counting algorithm using Aperio Scanscope (Aperio
Technologies, Vista, CA). The ONL cell count from three sections was averaged into a single
value for each animal.
Adjacent sections of those used for ONL cell count were used for immunofluorescent
imaging. The sections were deparaffinized and underwent heat-induced epitope retrieval with
antigen unmasking solution (H-3300, Vector). Characterization of human RPE was obtained
using immunofluorescent labeling of human-specific cell surface markers (anti-TRA-1-85,
1:100, R&D systems Inc., Minneapolis, MN) and an RPE marker (anti-RPE65, 1:500; Abcam
Inc., Cambridge, MA). The POS, including the phagosomes phagocytized by RPE, can thus be
stained by rhodopsin (anti-rhodopsin, 1:500; Abcam). Evaluation of local immune response was
done by double labeling of glial fibrillary acidic protein (GFAP), a marker for retinal glia cells
(1:2000, Abcam) and CD68, a marker for active microglia (1:50, Abcam). The images were
evaluated under a spinning disc confocal microscope (PerkinElmer Ultraviewer spinning disc
confocal, PerkinElmer Inc.; Waltham, MA). Quantification of the phagosomes within implanted
RPE and the activity of GFAP and CD68 were done using the image analysis software Volocity
(PerkinElmer Inc.).
64
Statistical Analysis
Parametric data are presented as mean ± SEM. Significance was calculated with two-way
ANOVA using the Bonferroni post hoc test. P<0.05 was considered to be statistic significant.
RESULTS
Survival of hESC-RPE after transplantation into the subretinal space of RCS rats
The survival of hESC-RPE is determined by the presence of pigmented cells in H&E
stained sections and the positive expression of the RPE marker, RPE65 (green), and the human
surface marker, TRA-1-85 (red), in immunofluorescent staining. H&E sections showed that all
rats receiving hESC/parylene had a pigmented monolayer of cells remaining on the surface of the
parylene membrane in the subretinal space 2 months after surgery (Figure 1A). The
immunofluorescence image showed that the cells express RPE65 and TRA-1-85 (Figure 1B).
Figure 1. Light microscopic and confocal images of a retinal section of a subretinal hESC-RPE/parylene
transplanted RCS rat 2 months after surgery. H&E staining showed a layer of pigmented hESC-RPE remaining on
the surface of the parylene membrane in the subretinal space (A). The immunofluorescence image of the adjacent
sections showed cells with positive expression of RPE65 (green) and TRA-1-85 (red) on the surface of the parylene
membrane (B). (Magnification x40)
65
Two (22%) of the hESC-RPE suspension treated rats showed cell survival in the subretinal
space 2 months after surgery. The others presented with only pigmentary aggregates in the
subretinal space (Figure 2A, arrows). One rat had a segment of monolayer of cells expressing
TRA-1-85 but not RPE65 (Figure 2C). The pigment clump did not express RPE65 and TRA-1-
85 and seemed to cause a focal photoreceptor loss (Figure 2A, arrows). An adjacent retinal
section from the same animal showed that the cells are no longer monolayer but organized into a
flower petal shape with a pinkish mucoid substance in the center (Figure 2D). These cells
express RPE65 and TRA-1-85. The other rat showed survival of the hESC-RPE as a large cell
aggregate of a mixed population of RPE65 and TRA-1-85 positive cells and large pigments
(Figure 2I).
66
Figure 2. Light microscopic transmission and confocal images of retina sections of hESC-RPE suspension
transplanted RCS rats that showed survival of hESC-RPE 2 months after surgery. Different patterns of cell survival
were found. (A) to (F) are retinal sections of the same eye at different locations. Pigment clumps located at the
subretinal space were found with focal loss of photoreceptors (arrows) (A). The immunofluorescence image of
adjacent sections showed a segment of monolayer of cells expressing TRA-1-85 (red) but not RPE65 (green).
Retinal sections of the same rat showed the hESC-RPE had organized into a flower-petal shape with a pinkish
mucoid substance in the center (D). Immunofluorescence image showed that these cells were positive for RPE65
(green) and TRA-1-85 (red) (F). Retinal sections of a different rat showing a large cell aggregate in the subretinal
space (G). The aggregate is comprised of a mix population of RPE65 and TRA-1-85 positive hESC-RPE and large
pigments (I).
Preservation of photoreceptors
The ONL cell count was quantified over a 400 μm length for the implanted animal at
locations immediately over the implant, adjacent to the implant, and across the optic nerve at the
opposite quadrant of the eye, The ONL cell count was significantly higher in hESC-
67
RPE/parylene implanted rats (356± 31/400μm) compared with the hESC-RPE suspension
implanted rats (214 ± 35/400μm) at the implanted area (P < 0.05). The ONL cell count of the
parylene only group (411± 95/400μm) was significantly higher compared with the hESC-RPE
suspension group (P < 0.005) and the medium injection group (P < 0.05). The ONL cell count
was identical among regions outside implanted area.
Figure 3. The outer nuclear layer (ONL) cell
count for controls and implanted animals at
different relative regions of the retina. The
ONL cell count was significantly higher in
the hESC-RPE/parylene implanted rats
compared to the hESC-RPE suspension
implanted rats at the implanted area (P <
0.05). The ONL cell count of the parylene-
only group was significantly higher
compared to the hESC-RPE suspension
group (P < 0.005) and the medium injection
group (P < 0.05). The ONL cell count was
identical among regions outside implanted
area.
Phagocytosis of photoreceptor outer segment by RPE
Immunostaining of the implanted rat retina showed positive phagosomes within
transplanted cells indicating that the RPE is functional (Figure4). The phagosomes contain
photoreceptor outer segments (POS) which consist of rhodopsin and hence can be identified by
immunostaining with rhodopsin antibody. Confocal images showed rhodopsin-positive
phagosomes within the transplanted RPE of RCS rats implanted with hESC-RPE/parylene. The
transplanted hESC-RPE (Figure 4B) showed more phagosomes and melanin pigments located
closer to the apical surface of RPE, similar to the distributions of these particles in normal
68
Copenhagen RPE (Figure 4A). There were only two rats in the hESC-RPE suspension treatment
groups that showed survival of the transplanted cells, thus only these two rats had positive
phagocytosis. The transplanted RPE in hESC-RPE suspension treated rats showed a mixed
population of cells with phagosomes and large pigment clumps which may be dead RPE (Figure
4c). The location of the phagosomes is, however, not close to photoreceptors but in the center of
a clump, which means these phagosomes are not obtained from the adjacent photoreceptors.
Figure 4. Confocal images of rhodopsin-positive phagosomes within the RPE of a normal Copenhagen rat (A), the
transplanted RPE of an RCS rat implanted with hESC-RPE/parylene (B), and the transplanted RPE of an RCS rat
implanted with hESC-RPE suspension (C). The RPE cells were identified by positive RPE65 (green) expression (A-
C). The right square brackets indicate the RPE. The transplanted hESC-RPE (B) showed more phagosomes (arrows)
and melanin pigments which block the fluorescence. These particles were located closer to the apical surface (the
upper portion where the RPE is in contact with photoreceptors). These features are similar to the normal
Copenhagen RPE (A). The transplanted RPE in the hESC-RPE suspension treated rat showed a mixed population of
cells with phagosomes and large pigment clumps. The location of the phagosomes is, however, not close to
photoreceptors but in the center of a clump. DAPI is used for nuclear staining (blue).
Quantification of the phagosomes showed that the phagosome count within the RPE of
Copenhagen rats (0.065±0.028 phagosomes μm
2
) was not significantly different from the
transplanted RPE of implanted hESC-RPE/parylene (0.040 ±0.013/ phagosomes μm
2
). The
phagosome count is significantly higher in the transplanted RPE of hESC-RPE/parylene
compared with the transplanted RPE of hESC-RPE suspension (0.003±0.006 phagosomes /μm
2
,
P<0.005). This indicates that the hESC-RPE implanted as a monolayer has a phagocytosis
behavior closer to normal RPE than the hESC-RPE suspension.
69
Figure 5. Phagosome count within the RPE of Copenhagen rats, transplanted RPE of rats implanted with hESC-
RPE/parylene, and transplanted RPE of rats implanted with an hESC-RPE suspension. The phagosome count is
significantly higher with hESC-RPE/parylene (0.040 ±0.013/ phagosomes μm
2
) compared to hESC-RPE suspension
(0.003±0.006 phagosomes /μm
2
, P<0.005).
Retinal Glia Cell Activation
In vitro studies suggested that polarized hESC-RPE suppress inflammatory reactions by
modulating macrophage activation and induce anti-inflammatory cytokine (Zamiri, Masli et al.
2006; Zhu, Deng et al. 2011). To understand the effect of the transplant on the host innate
immune response, we studied the activity of retinal glia cells. The activity of CD68 (a marker for
active microglia) and GFAP (a marker for astrocytes and active Muller cells) was examined.
Retinas implanted with hESC-RPE/parylene showed a mild expression of CD68 positive cells
near the implant (A). Retinas implanted with parylene showed prominent CD68 and GFAP
(arrow) expression only below the implant (B). Retinas implanted with hESC-RPE suspension
showed two different patterns. In the majority of cases, the cells became pigmented clumps, and
prominent CD68 and GFAP (arrow) expression over the clumps were found (C). In the rare cases
where there were cell survivals, the CD68 and GFAP expression was low around the surviving
cells. The CD68 cell count showed that the parylene-only implanted rats had more CD68 cells
near the implanted area compared with the hESC-RPE/parylene implanted rats (P<0.005) (A). As
70
for the area of CD68 expression, the suspension implanted group had significantly more area
expressing CD68 then the hESC-RPE/parylene rats (P<0.001) and the parylene-only group
(P<0.001) (B). This was due to the presence of large pigmented clumps with prominent CD68
expression. The GFAP expression was significantly higher in the parylene-only group compared
with the hESC-RPE/parylene group (P<0.05) (C) which indicates that the polarized RPE is
capable of suppressing Müller cell activation.
Figure 6. Confocal images of the CD68 and GFAP immunofluorescence of hESC-RPE/parylene implanted retina
(A), parylene-only retina (B), and hESC-RPE suspension implanted retina (C, D). The hESC-RPE/parylene retina
showed some CD68 (green) positive cells around the implant but no significant GFAP expression (A). The parylene
implant showed prominent CD68 and GFAP (arrow) expression below the implant (B). The hESC-RPE suspension
treated rats showed prominent expression of CD68 and GFAP over pigmented clumps (C). Remnants of the hESC-
RPE suspension that managed to survive, however, had relatively low CD68 and GFAP expression.
71
Figure 7. CD68 and GFAP expression adjacent to implant
area and away from it. The CD68 cell count showed that the
parylene only implanted rat have more CD68 cells near the
implanted area compared to the hESC-RPE/parylene
(P<0.005) (A). As for the area of CD68 expression, the
suspension implanted group have significant more area
expressing CD68 then the hESC-RPE/parylene (P<0.001)
and parylene only group (P<0.001) (B). The GFAP
expression was significantly higher in parylene only group
compared to the hESC-RPE/parylene group (P<0.05) (C).
DISCUSSION
The results of the study showed that the survival rate of hESC-RPE after transplantation
into the subretinal space of RCS rats is significantly higher when it is transplanted as a
monolayer rather than a cell suspension. The cells survived and maintained the RPE and human
characteristics in the subretinal space of RCS rats 2 months after surgery. The confocal images of
phagocytosis of the RPE demonstrated that the hESC-RPE transplanted as a monolayer had a
phagosomes distribution that is close to the normal physiological condition. The phagosomes
count was also similar to normal RPE which indicates that the monolayer hESC-RPE cells are
functional.
72
The effect of hESC-RPE suspension was variable. For the hESC-RPE suspension rats, most
cells became large pigmented clumps. In the rare cases where cell survival was found, two
different patterns of cell survival were noted. The first was as a segment of monolayer formation
on the surface of the host Bruch’s membrane. The second was as a clump of transplanted cells
with some RPE gaining connection with neighboring RPE cells, wrapped around a pigment or
some extracellular matrix (ECM) more or less in the form of a flower petal. This flower petal-
like organization may be a part of ongoing process of cell death, with the large pigment clumps
that we found later on being the remnant of the dead cells. This is supported by the finding of
active microglia and Müller cells that covered the pigment clumps. RPE is an anchorage
dependent cell that needs to attach to an ECM. A cell death phenomenon called anoikis occurs
when cells that have lost their support try to survive by attaching to a dead cell (Gilmore 2005;
Petrovski, Berenyi et al. 2011; Kinnunen, Petrovski et al. 2012). This seems to correspond to our
histologic findings. However, it is not clear whether the glia cell activation happened before or
after cell death. If it happened before cell death, it would mean that the non-polarized RPE cells
caused a local inflammatory response. If it were after cell death, it might mean that the reactions
were a response to scavenge the cell debris.
While some groups showed that the RPE cell suspensions were able to form a monolayer on
the surface of the host Bruch’s membrane (Schwartz, Hubschman et al. 2012), other studies
showed that the cell suspensions in the subretinal space formed multilayer cell clumps or large
pigmented clumps (Wongpichedchai, Weiter et al. 1992). Overall, these studies showed that the
survival of RPE delivered as a suspension is variable. The fate of the RPE suspension likely
depends on whether there is a bare healthy Bruch’s membrane to which the transplanted RPE
suspension can attach. This is supported by a study done in rabbits where an autologous RPE
73
suspension was able to survive when injected into an area where the Bruch’s membrane was
mechanically debrided (Phillips, Sadda et al. 2003). The rat in the cell suspension group that had
a monolayer of hESC-RPE development might have happened to have a bare Bruch’s membrane
due to the jet flow of the suspension in the subretinal space during injection.
The preservation of photoreceptors was higher in the hESC-RPE/parylene implanted
group compared to the suspension group. This is likely due to the fact that the transplanted cells
in the suspension rats failed to survive. As shown in our study, for the one animal that showed a
monolayer development of the transplanted cells, there was photoreceptor preservation. However,
focal photoreceptor loss was found in the same animal where pigment clumps were in contact
with the photoreceptors (Figure 2A). Other studies have also reported multilayer clumps causing
damage to the overlaying retina (Wongpichedchai, Weiter et al. 1992; Abe, Tomita et al. 2000).
RPE is known to be an immune-privileged tissue which can promote its own survival after
heterogeneous transplantation through the expression of a CD95 ligand (Wenkel and Streilein
2000). In vitro studies have suggested that polarized hESC-RPE secrets higher amounts of
pigment epithelium-derived factor (PEDF) compared to non-polarized RPE (Zhu, Deng et al.
2011) which is known for its neurotrophic and antiangiogenic activity. PEDF suppress
inflammatory reactions by modulating macrophage activation and induce anti-inflammatory
cytokine, interleukin-10 production from macrophages (Zamiri, Masli et al. 2006). The lower
glia cell activity observed in the hESC-RPE monolayer transplanted retinas compared to the
suspension treated retinas is likely due to the fact that transplanted RPE cells are polarized and
behave more similarly to normal RPE cells.
In summary, we found that the transplanted hESC-RPE cells need to survive in order to
preserve the photoreceptors for animals with retinal degeneration, as the death of the transplanted
74
cells may cause toxicity to the retina. In order for the RPE to survive long term, it needs to attach
to an extracellular matrix and obtain a monolayer structure. The hESC-RPE cultured on the
parylene membrane allows delivery of RPE cells as a monolayer, whereas, for the cell
suspension, the chance of forming a monolayer is small and depends on unknown factors. Our
study suggests that implantation of hESC-RPE/parylene may be a desirable therapeutic approach
for disease conditions related to RPE dysfunction such as dry age-related macular degeneration.
CHAPTER 3 REFERENCES
Abe, T., H. Tomita, et al. (2000). "Autologous iris pigment epithelial cell transplantation in
monkey subretinal region." Curr Eye Res 20(4): 268-275.
Binder, S., B. V. Stanzel, et al. (2007). "Transplantation of the RPE in AMD." Prog Retin Eye
Res 26(5): 516-554.
Carr, A. J., A. A. Vugler, et al. (2009). "Protective effects of human iPS-derived retinal pigment
epithelium cell transplantation in the retinal dystrophic rat." PLoS One 4(12): e8152.
Gilmore, A. P. (2005). "Anoikis." Cell Death Differ 12 Suppl 2: 1473-1477.
Hadlock, T., S. Singh, et al. (1999). "Ocular cell monolayers cultured on biodegradable
substrates." Tissue Eng 5(3): 187-196.
Hu, Y., L. Liu, et al. (2012). "A novel approach for subretinal implantation of ultrathin substrates
containing stem cell-derived retinal pigment epithelium monolayer." Ophthalmic Res
48(4): 186-191.
Idelson, M., R. Alper, et al. (2009). "Directed differentiation of human embryonic stem cells into
functional retinal pigment epithelium cells." Cell Stem Cell 5(4): 396-408.
75
Kinnunen, K., G. Petrovski, et al. (2012). "Molecular mechanisms of retinal pigment epithelium
damage and development of age-related macular degeneration." Acta Ophthalmol 90(4):
299-309.
Little, C. W., C. Cox, et al. (1998). "Correlates of photoreceptor rescue by transplantation of
human fetal RPE in the RCS rat." Exp Neurol 149(1): 151-160.
Lund, R. D., P. Adamson, et al. (2001). "Subretinal transplantation of genetically modified
human cell lines attenuates loss of visual function in dystrophic rats." Proc Natl Acad Sci
U S A 98(17): 9942-9947.
Lund, R. D., A. S. Kwan, et al. (2001). "Cell transplantation as a treatment for retinal disease."
Prog Retin Eye Res 20(4): 415-449.
Petrovski, G., E. Berenyi, et al. (2011). "Clearance of dying ARPE-19 cells by professional and
nonprofessional phagocytes in vitro- implications for age-related macular degeneration
(AMD)." Acta Ophthalmol 89(1): e30-34.
Phillips, S. J., S. R. Sadda, et al. (2003). "Autologous transplantation of retinal pigment
epithelium after mechanical debridement of Bruch's membrane." Curr Eye Res 26(2): 81-
88.
Pinilla, I., N. Cuenca, et al. (2009). "Intraretinal processing following photoreceptor rescue by
non-retinal cells." Vision Res 49(16): 2067-2077.
Schwartz, S. D., J. P. Hubschman, et al. (2012). "Embryonic stem cell trials for macular
degeneration: a preliminary report." Lancet 379(9817): 713-720.
Singh, S., S. Woerly, et al. (2001). "Natural and artificial substrates for retinal pigment epithelial
monolayer transplantation." Biomaterials 22(24): 3337-3343.
76
Stanzel, B. V., Z. Liu, et al. (2012). "Subretinal delivery of ultrathin rigid-elastic cell carriers
using a metallic shooter instrument and biodegradable hydrogel encapsulation." Invest
Ophthalmol Vis Sci 53(1): 490-500.
Strauss, O. (2005). "The retinal pigment epithelium in visual function." Physiol Rev 85(3): 845-
881.
Tezcaner, A., K. Bugra, et al. (2003). "Retinal pigment epithelium cell culture on surface
modified poly(hydroxybutyrate-co-hydroxyvalerate) thin films." Biomaterials 24(25):
4573-4583.
Thumann, G., A. Viethen, et al. (2009). "The in vitro and in vivo behaviour of retinal pigment
epithelial cells cultured on ultrathin collagen membranes." Biomaterials 30(3): 287-294.
Vugler, A., A. J. Carr, et al. (2008). "Elucidating the phenomenon of HESC-derived RPE:
anatomy of cell genesis, expansion and retinal transplantation." Exp Neurol 214(2): 347-
361.
Wenkel, H. and J. W. Streilein (2000). "Evidence that retinal pigment epithelium functions as an
immune-privileged tissue." Investigative Ophthalmology & Visual Science 41(11): 3467-
3473.
Williams, R. L., Y. Krishna, et al. (2005). "Polyurethanes as potential substrates for sub-retinal
retinal pigment epithelial cell transplantation." J Mater Sci Mater Med 16(12): 1087-1092.
Wongpichedchai, S., J. J. Weiter, et al. (1992). "Comparison of external and internal approaches
for transplantation of autologous retinal pigment epithelium." Invest Ophthalmol Vis Sci
33(12): 3341-3352.
77
Zamiri, P., S. Masli, et al. (2006). "Pigment epithelial growth factor suppresses inflammation by
modulating macrophage activation." Investigative Ophthalmology & Visual Science
47(9): 3912-3918.
Zhu, D., X. Deng, et al. (2011). "Polarized secretion of PEDF from human embryonic stem cell-
derived RPE promotes retinal progenitor cell survival." Invest Ophthalmol Vis Sci 52(3):
1573-1585.
78
Chapter 4
Long-term Efficacy and Safety of RPE Monolayer Derived From Human
Embryonic Stem Cell
ABSTRACT
Purpose: To compare the histological and visual functional differences in dystrophic Royal
College of Surgeons (RCS) rats after transplantation of retinal pigment epithelial cells derived
from H9 human embryonic stem cells (hESC-RPE) delivered as a polarized monolayer cell sheet
on a thin parylene membrane and non-polarized cell suspension.
Methods: For transplantation of polarized monolayer cell sheet, hESC-RPE cells were cultured
for 3 weeks on parylene sheets; for transplantation of cell suspension, hESC-RPE cells were
cultured for 3 weeks and digested into cell a suspension before implantation. Subretinal
transplantation of hESC-RPE /parylene (approximately 2800 cells/implant, n=20) and subretinal
injection of hESC-RPE cells (5x10
4
/2µl, n=18) was performed in 27 to 29 day old RCS rats.
Post-surgical visual function evaluation was performed using optokinetic head tracking (OHT)
testing. , Subretinal implantation of hESC-RPE /parylene (approximately 2800 cells/implant,
n=20) and subretinal injection of hESC-RPE cells (5x10
4
/2µl, n=18) were performed in RCS
rats. Animals were sacrificed at different time point for histologic evaluation. Sections of the
eyes were evaluated with hematoxylin and eosin (H&E) staining for photoreceptor survival and
immunofluorescent staining for transplanted cells.
Results: There was more preservation of photoreceptors in the hESC-RPE/parylene treated rat
than the suspension treated rats at 4 months after surgery. RPE marker (RPE65), human surface
marker (TRA-1-85) and rhodopsin were positive in the majority of the rats implanted with
79
hESC-RPE/parylene. In the suspension treated group, cell survival was found only up to 4
months after surgery. After that, only remnants of the cells can be found. Animals implanted
with hESC-RPE/parylene performed better than the hESC-RPE suspension and control groups
on a visual behavior tests from 2 months to up to 4 months after surgery (p<0.05).
Conclusions: The transplantation of hESC-RPE/parylene resulted in better preservation of
photoreceptors, visual function, and better long term survival compared to hESC-RPE delivered
as a suspension.
INTRODUCTION
Various preparations of RPE cells have been studied for developing cell-based therapies for
blindness caused by retinal pigment epithelial (RPE) dysfunction. Previous studies have shown
that the subretinal injection of different RPE could delay the progression of the retinal
degeneration and improve vision rescue in the RCS (Pinilla, Cuenca et al. 2007; Vugler, Carr et
al. 2008; Lu, Malcuit et al. 2009; Schwartz, Hubschman et al. 2012). However, studies using
different RPE cell lines have demonstrated transient visual function rescue and have shown
pigmented clumps forming in the subretinal space (Wongpichedchai, Weiter et al. 1992; Sauve,
Pinilla et al. 2006; Vugler, Carr et al. 2008; Carr, Vugler et al. 2009; Idelson, Alper et al. 2009).
While the survival of RPE cells delivered as a suspension is variable, other researchers have
proposed that delivering RPE cells as a polarized monolayer might provide a better chance of
long term survival and better preservation of the photoreceptors because the cells don’t need to
reorganize while they are in a new environment (Hadlock, Singh et al. 1999; Singh, Woerly et al.
2001; Tezcaner, Bugra et al. 2003; Williams, Krishna et al. 2005; Thumann, Viethen et al. 2009;
Stanzel, Liu et al. 2012).
80
In vitro studies showed that the polarized RPE behave more close to normal RPE
compared to non-polarized RPE (Zhu, Deng et al. 2011). However, the in vivo behavior
difference has not been compared between the polarized and non-polarized RPE in the past. The
purpose of our study is to compare the histological and functional difference of subretinal
implantation of polarized RPE monolayer and non-polarized cell suspension injection. The long
term survival of the hESC-RPE, preservation of photoreceptors, and visual behavior response
were assessed in this study.
Human embryonic stem cells (hESC) have the potential to differentiate into various cell
types and provide unlimited source for cell replacement therapies. Our studies used hESC
derived RPE cells cultured on parylene membrane (hESC-RPE/parylene) and hESC-RPE
suspension for transplantation in a rat model of RPE dysfunction, Royal College of Surgeon
(RCS) rats.
METHODS
Animals
All experiments were performed in compliance with the ARVO statement of the use of
Animals in Ophthalmic and Vision Research under a protocol approved by the Animal Care and
Use Committee at the University of Southern California. Subretinal implantation of hESC-RPE
seeded on a submicron parylene-C membranes (n=20) and subretinal injection of hESC-RPE
cells (n=18) were performed in dystrophic pigmented RCS rats. The RCS rat is widely used in
transplantation studies as a model of RPE dysfunction because it has a mutation in the MerTk
gene, which causes a dysfunction in the RPE's ability to perform phagocytosis of the shed outer
segment (POS) material leading to a progressive loss of photoreceptors (Yamamoto, Du et al.
81
1993; Little, Cox et al. 1998; Lund, Adamson et al. 2001; Lu, Malcuit et al. 2009; Engelhardt,
Tosha et al. 2012). Ten RCS rats received parylene membranes implantation without cells served
as a sham surgery group and 7 rats received subretinal injections of culture medium
(DMEM/F12) served as a sham treatment group. All the fellow eyes served as the control group.
Oral cyclosporine was given in a concentration of 210 mg/l from day 2 before the surgery until
the day of euthanasia. Dexamethasone (1.6mg/kg/Day) was injected 2 days prior to the surgery
and daily for 14 days. Animals were sacrificed at 4, 6, 8, and 12 months of age. Animals that did
not survive the study due to surgical complications nor had health problems such as infections or
cataracts were excluded from the study.
Cell Preparation
Human embryonic stem cells (H9) were used to derive RPE using the spontaneous
differentiation method as previously described (Zhu, Deng et al. 2011). hESC-RPE cells were
cultured and maintained in serum-free medium (X-VIVO 10, Lonza). For preparation of the
hESC-RPE/parylene implant, cells were seeded on vitronectin (AMS Biotechnology, Lake Forest,
CA) coated mesh-supported submicron parylene-C membrane (MSPM, 0.4mm x 0.9mm x 6.5
µm) (Lu, Zhu et al. 2012) at density of 10
5
/cm
2
and cultured in X-VIVO 10 medium for 3-4
weeks to allowed the cells to become confluent on the parylene membrane and reach a density of
approximately 2,700 cells/membrane. For preparation of hESC-RPE cell suspension, cells were
grown on a petri dish for 3-4 weeks and digested into cell suspension at the cell density of
2.5x10
7
/ml prior to surgery.
82
Subretinal hESC-RPE Monolayer Transplantation
RCS rats were transplanted at P27-29 days by a single surgeon as previously described
(Hu, Liu et al. 2012). Briefly, animals were anesthetized by intraperitoneal injection of
ketamine/xylazine (3.75 mg/kg ketamine and 5 mg/kg xylazine). The corneas were treated with
0.5% tetracaine (Akorn Inc., Lake Forest, Illinois) and the pupils dilated using 2.5%
phenylephrine (Akorn Inc.) and 0.5% tropicamide (Akorn Inc.). The eyes were rotated inferiorly
with traction sutures. A sclerotomy approximately 1.0 mm in width was performed 1 mm
posterior to the limbus with a 30-gauge needle. A paracentesis was performed through
puncturing the cornea with a 30 gauge needle to decrease the pressure in the eye. A retinal bleb
was created by the subretinal injection of 5μl of balance salt solution (Alcon Laboratories, Inc.
Fort Worth, TX), using a 32-gauge blunt needle attached to a 10-µl Hamilton syringe through the
sclerotomy. The parylene membranes (with or without cells) were placed into the subretinal
space through the sclerotomy using fine forceps. The fundus was checked to confirm the correct
placement. Immediately after the procedure, Optical Coherence Tomography (Spectralis,
Heidelberg, Germany) B-Scans were performed to confirm the proper subretinal placement of
the implant (Fig 1). After imaging, the eyes were protected with neomycin and polymyxin B
sulfates and dexamethasone ophthalmic ointment (Alcon Laboratories, Inc. Fort Worth, TX).
83
Figure 1: Combined Infrared (left) and OCT image (right) demonstrating the subretinal implant of the parylene
membrane. The OCT image corresponds to the location of the scan line (green arrow) on the infrared image.
Subretinal hESC-RPE Suspension Transplantation
RCS rats were transplanted at P27-29 days. Rats were prepared similarly as subretinal
hESC-RPE monolayer transplantation. A paracentesis was performed through a puncture at the
cornea with a 30 gauge needle to decrease pressure in the eye. The injection was performed
using a 10 µl Hamilton syringe attached to a fine glass pipette (internal diameter 75–150 μm)
through a small scleral incision created by a 30 gauge needle. Subretinal injections of hESC-RPE
cells (5X10
4
, 2µl) or DMEM/F12 solution were performed. Immediately after injection, the
fundus was examined in order to confirm the presence of a retinal bleb.
Optokinetic Head Tracking Test
OHT was tested every month until 6 months after surgery or until the animals were
sacrificed. This testing was performed as previously describe(Prusky, West et al. 2000; McGill,
Douglas et al. 2004; Thomas, Seiler et al. 2004; Douglas, Alam et al. 2005; Thomas, Samant et
al. 2007). Briefly, the animal was placed on a transparent acrylic chamber surrounded by four
flat screen computer monitors that projected white and black stripes; each stripe had a width of
84
3.60 cm and they moved at a speed of 6.9cm/s resulting in a constant spatial frequency of 0.04
cycles per degree. Each trial consisted of 60 seconds clockwise/ counterclockwise (right eye test)
movement and 60 seconds counterclockwise/clockwise (left eye test). Each test consisted of
three trials with a 3 minute rest in order to avoid fatigue. The OHT testing was recorded and the
duration of time that the rat spent tracking the stripes was quantified by two experienced
technicians.
Histology
All animals were euthanized with 0.5 ml of pentobarbitol sodium 390 mg/ml and
phenytoin sodium 50 mg/ml (Euthasol, Virbac AH, Inc, Forth Worth, TX). The eyes were
fixated in Davidson’s solution and embedded with paraffin. Serial sections (5 µm thick) were
taken at the implanted quadrant. For hESC-RPE/parylene implanted eyes, the sections
(approximately 250 µm apart) were stained using hematoxylin and eosin (H&E) to exam the
morphology of the implanted cells starting from the first section where the parylene implant were
found. For hESC-RPE suspension treated eyes, the sections were examined approximately every
125 μm and stained every 250 μm at the implanted quadrant using H&E staining to identify
implanted cells. An Aperio ScanScope (Aperio Technologies, Vista, CA) was used to quantify
the outer nuclear layer cells (ONL). For the hESC-RPE/parylene implanted eyes, ONL was
calculated over the implanted cells of a 400 μm length. For the hESC-RPE suspension group,
two areas of 400 μm length in implanted quadrant were quantified. For sections that showed no
obvious implanted cells, two randomly selected areas of 400 μm were quantified. Each rat had
six measurements from three non-adjacent sections and averaged into one number.
Characterization of human RPE was obtained by using human-specific surface markers
(anti-TRA-1-85, 1:100, R&D systems Inc., Minneapolis, MN), RPE markers (anti-RPE65,
85
1:500; Abcam Inc., Cambridge, MA) and rhodopsin (anti-rhodopsin, 1:500, Abcam). Rhodopsin
is contained in the POS and phagocytosis of POS is one of RPE’s critical functions (Kevany and
Palczewski 2010).. Thus, in a functioning RPE cell containing phagosomes, the phagosomes can
be visualized by immunofluorescent staining with anti-rhodopsin antibodies.
Data Analysis
Statistical analysis was done using two-way ANOVA with Bonferroni post hoc test for
OHT results. P<0.05 was considered to be statistic significant.
RESULTS
Survival of hESC-RPE after Transplantation into the Subretinal Space of RCS Rats
Survival of hESC-RPE is determined by the presence of pigmented cells in H&E stained
sections and the positive expression of the RPE marker, RPE65 and human surface marker,
TRA-1-85 in immunofluorescent staining. In the hESC-RPE/parylene implanted groups, H&E
sections showed a pigmented monolayer of cells remaining on the surface of the parylene
membrane in the subretinal space 4 months after transplantation (Figure 2A).
The immunofluorescence image showed that the transplanted cells in the hESC-
RPE/parylene implanted group that express RPE65 (green) and TRA-1-85 (red) (Figure 2B) also
showed phagosomes of rhodopsin positive POS in the cytoplasm indicating that the cells
maintain the physiologic functioning of RPE (Figure 2D).
86
Figure 2. Light microscopic
images (A) and confocal images
(B,D) of hESC-RPE/parylene
transplanted retinas of RCS rat
and non-operated dystrophic
control retinas (C) at 4 months
after surgery. The ONL was still
evident in the implanted
animals (A) at this point,
whereas in the non-operated
control retina (C), only a few
photoreceptors remained. The
parylene implant was located in the subretinal space with hESC-RPE remaining on the surface of the membrane
(Magnification x20) (A). Immunofluorescent staining showed cells with positive expressions of RPE65 (green) and
TRA-1-85 (red) on the surface of the parylene membrane (Magnification x40) (B). The transplanted cells that
showed positive expressions of RPE65 and TRA-185 also showed rhodopsin (red) (arrow) positive phagosomes
within the cells with positive RPE65 expression (green) (D) (Magnification x100).
Two hESC-RPE suspension-treated animals showed cell survival in the subretinal space
4 months after surgery. The cells appear to be a monolayer of cells expressing RPE65 and TRA-
1-85 (Figure 3A, B). The expression of RPE and the human marker was segmental, although the
pigmented layer of cells appears continuous on the H&E stain. RCS rats treated with hESC-RPE
8 months and over did not show viable cells when examined with immunofluorescence (data not
shown) although the H&E stain showed pigmented clumps (Figure 3C) and a pinkish mucoid
substance (Figure 3D-F) that indicated the previous presence of transplanted cells. Phagosomes
were not found in the hESC-RPE suspension treated group (data not shown).
87
Figure 3. Light microscopic images (A,C-F) and
confocal image (B) of retina section of hESC-RPE
suspension transplanted RCS rats. H&E staining
showed a pigmented layer of cell on the surface of
Bruch’s membrane in the subretinal space of an
RCS rat’s retina 4 months after implantation (A).
Immunofluorescence of adjacent sections showed a
part of the pigmented layer was positive for RPE65
(green) and TRA-1-85 (red) (white arrow) (B). The
pigmented clump was present in the suberetinal
space (black arrow, C) in an RCS rat 8 months after
implantation. Sections from different rats 8 months
after surgery showed pinkish mucoid substance
(arrowhead) in the subretinal space with localized
loss of the nuclear layers of the retina (D-F).
Survival rate of the implanted hESC-RPE is summarized in table 1. The hESC-
RPE/parylene implanted group showed over a 50% survival rate at different ages; the hESC-RPE
suspension implanted group showed a 50% survival rate 4 months after surgery; however, no
survival was observed at 8 months or over. No cell migration or tumor formation was observed
in either group. However, in the hESC-RPE suspension treated group, we found pigmented
clumps in the nuclear layers of the retina which we could not determine either to be previously
viable cells that migrated or cells that we injected into the retina.
88
Time after surgery 4 months 6 months 8 months 12 months
hESC-RPE/parylene 50%(3/6) 83%(5/6) 50%(3/6) 100%(2/2)
hESC-RPE suspension 50%(2/4) - 0%(0/8) 0%(0/6)
Table 1. Survival of hESC-RPE after transplantation into subretinal space of RCS rat. For the hESC-RPE/parylene
group, the survival rate ranged from 50% to 100%. For the hESC-RPE suspension treated group, cell survival is
29% at 4 months post-surgery. No survival is observed at 8 months and over after surgery.
Preservation of Photoreceptors
Retinas from all three groups were examined. hESC-RPE/parylene implanted animals
had a higher ONL cell count (153± 69/ 400μm) at 4 months after surgery compared to the
suspension treated group (19± 7/ 400μm, P<0.05). The parylene implanted group and the non-
treated control group had almost no ONL preserved since 4 months after surgery. ONL cell count
over 8 months did not show a difference among the groups, with only a few scattered
photoreceptors present in all animals.
Visual Function Assessment
Visual function assessment of animals that received cell transplantation was done by
optokinetic head-tracking response. The animals that received hESC-RPE/parylene implanted
(4.3 ± 0.5 sec/min) tracked significantly better than hESC-RPE suspension (2.5 ± 0.5 sec,
P<0.05), medium (4.3 ± 1.3 sec, P<0.001), and non-operated dystrophic control animals (0.8 ±
0.2 sec, P<0.001) at 2 months after surgery, the difference maintain significant up until 4 months
after surgery (Figure 4). The difference between the hESC-RPE/parylene treated animals and the
parylene treated animals was significantly at 4 months after surgery. The hESC-RPE suspension
treated animals tracked better than non-operated animals at 2 months after surgery, however the
89
differences were not significant after this time point. There was no significant difference between
the hESC-RPE suspension treated animals and the parylene treated animals. While the hESC-
RPE/parylene treated animals maintained a certain level of OHT response at 5 months after
surgery, the parylene treated animals and the non-operated animals were not able to track beyond
this age.
Figure 4. Optokinetic head tracking durations at
different times after surgery. At 2 months after
surgery, the animals that received the hESC-
RPE/parylene implant (4.3 ± 0.5 sec/min) tracked
significantly better than the animals treated with
hESC-RPE suspension (2.5 ± 0.5 sec, P<0.05),
medium (4.3 ± 1.3 sec, P<0.001), and the non-operated
dystrophic control animals (0.8 ± 0.2 sec, P<0.001).
This difference was still significant up until 4 months
after surgery. The hESC-RPE suspension treated
animals tracked better than the non-operated animals at 2 months after surgery; however the differences between
them were not significant after this point. There was no significant difference between the hESC-RPE suspension
treated animals and the parylene treated animals. While the hESC-RPE/parylene treated animals maintain a certain
level of OHT response at 5 months after surgery, the parylene treated animals, medium injected, and the non-
operated animals were not able to track beyond this age.
DISCUSSION
The results of our study showed that subretinal implantation of a polarized, monolayer
hESC-RPE cell line seeded on a submicron parylene-C membrane can survive and maintain as a
monolayer up to one year after surgery. However, the survival rate was variable among different
age groups and was not correlated with the duration after surgery. We speculate that this is
90
related to the surgical manipulation during transplantation because the subretinal implant surgery
requires a learning curve for beginning surgeons and the surgeries on some of the eyes that
showed no cell survival were performed at an earlier point.
The survival of hESC-RPE suspension was only found up to 4 months after surgery. In the
later time points, only a pinkish mucoid-like substance was found in the subretinal space. When
RPE cells lose their anchorage, a cell death phenomenon called anoikis occurs, where the cells
try to survive by attaching to a dead cell or any available extracellular matrix (Gilmore 2005;
Petrovski, Berenyi et al. 2011; Kinnunen, Petrovski et al. 2012).. We speculate that in the later
age groups there were remnants of the injected hESC-RPE cells, because in our short term study,
we found RPE cell clumps forming around materials that were similar to the pinkish substance
which formed a flower petal-like configuration (our unpublished data). The flower petal of RPE
cells may be an expression of the ongoing phenomenon of anoikis and the pinkish mucoid
substance may indicate the final stage of cell death. This also explains our finding where any
RPE that survived long term was in the form of a monolayer. This indicates that the injected
hESC-RPE cells that didn’t manage to attach to Bruch’s membrane and regain monolayer
structure will eventually undergo anoikis.
The ONL cell count at 4 months after surgery showed that the hESC-RPE monolayer
transplanted eyes had more photoreceptor preservation compared to the hESC-RPE suspension
treated eyes. However, no rescue of photoreceptors was found in either group at a later time
point. This can be related to the fact that we are performing transplantation surgeries at a later
time point compared to most of the other study group. At the age of P28 (Dowling and Sidman
1962; LaVail and Battelle 1975), the RCS rat has already demonstrated significant loss of
photoreceptors and electroretinographic activity. The remaining photoreceptors may already be
91
moving towards the irreversible process of apoptosis, which would result in the less prominent
rescue of the photoreceptors that is also reported by other groups (Sheedlo, Li et al. 1991; Wang,
Lu et al. 2008).
The presence of phagosomes within the transplanted RPE cells in the hESC-RPE monolayer
transplanted group indicates that transplanted cells maintained the characteristic functions of
normal RPE. The presence of phagosomes was not found in the transplanted cells in the hESC-
RPE suspension treated group, possibly because the retina was in an advanced stage of
degeneration and no POS were present to be phagocytized.
The visual behavior testing showed that the hESC-RPE monolayer implanted RCS rats
showed better head tracking response from 2 months after surgery until 4 months after surgery.
This is compatible with the findings of ONL cell count where differences between the two
groups were observed at 4 months after surgery but not afterward. The hESC-RPE monolayer
implanted group continues to show tracking at least up to 6 months after surgery, while the sham
implanted group and the non-operated group had no tracking at all at this point. Although the
difference was not significant, we think it shows that the few photoreceptors remaining allowed
the RCS rat to maintain some level of visual function.
One limitation of the study is that we performed the subretinal implant surgery at a later
time point compared to other studies which makes the results less comparable to other groups.
This was due to the fact that the subretinal parylene implant surgery is more difficult to perform
in RCS rats under P21 where the eye has not fully developed. However, performing a surgery at
a time were the manifestations of disease have already started better resembles clinical
conditions, where patients seeking intervention already have a certain amount of visual loss.
92
Thus, this experiment gives us a more realistic expectation of the new therapy than previous
studies.
In summary, hESC-RPE implanted as a monolayer can survive long term, maintaining the
RPE characteristics in the subretinal space and reducing the speed of progression of retinal
degeneration in the RCS rats. We propose that this is a more feasible option for future clinical
therapy for diseases such as age related or macular degeneration. For future studies we will
execute more frequent and regular in vivo examinations, using fundus infrared imaging and OCT
to follow up the implantation to understand if there is any progressive change in the transplanted
cells that results in the non-survival of the cells if and when it occurs.
93
CHAPTER 4 REFERENCES
Carr, A. J., A. A. Vugler, et al. (2009). "Protective effects of human iPS-derived retinal pigment
epithelium cell transplantation in the retinal dystrophic rat." PLoS One 4(12): e8152.
Douglas, R. M., N. M. Alam, et al. (2005). "Independent visual threshold measurements in the
two eyes of freely moving rats and mice using a virtual-reality optokinetic system."
Visual neuroscience 22: 677-684.
Dowling, J. E. and R. L. Sidman (1962). "Inherited retinal dystrophy in the rat." J Cell Biol 14:
73-109.
Engelhardt, M., C. Tosha, et al. (2012). "Functional and morphological analysis of the subretinal
injection of retinal pigment epithelium cells." Visual neuroscience 29(2): 83-93.
Gilmore, A. P. (2005). "Anoikis." Cell Death Differ 12 Suppl 2: 1473-1477.
Hadlock, T., S. Singh, et al. (1999). "Ocular cell monolayers cultured on biodegradable
substrates." Tissue Eng 5(3): 187-196.
Hu, Y., L. Liu, et al. (2012). "A Novel Approach for Subretinal Implantation of Ultrathin
Substrates Containing Stem Cell-Derived Retinal Pigment Epithelium Monolayer."
Ophthalmic research 48(4): 186-191.
Idelson, M., R. Alper, et al. (2009). "Directed differentiation of human embryonic stem cells into
functional retinal pigment epithelium cells." Cell Stem Cell 5(4): 396-408.
Kevany, B. M. and K. Palczewski (2010). "Phagocytosis of retinal rod and cone photoreceptors."
Physiology (Bethesda) 25(1): 8-15.
94
Kinnunen, K., G. Petrovski, et al. (2012). "Molecular mechanisms of retinal pigment epithelium
damage and development of age-related macular degeneration." Acta Ophthalmol 90(4):
299-309.
LaVail, M. M. and B. A. Battelle (1975). "Influence of eye pigmentation and light deprivation on
inherited retinal dystrophy in the rat." Exp Eye Res 21(2): 167-192.
Little, C. W., C. Cox, et al. (1998). "Correlates of photoreceptor rescue by transplantation of
human fetal RPE in the RCS rat." Experimental neurology 149(1): 151-160.
Lu, B., C. Malcuit, et al. (2009). "Long-term safety and function of RPE from human embryonic
stem cells in preclinical models of macular degeneration." Stem Cells 27(9): 2126-2135.
Lu, B., D. Zhu, et al. (2012). "Mesh-supported submicron parylene-C membranes for culturing
retinal pigment epithelial cells." Biomedical microdevices 14(4): 659-667.
Lund, R. D., P. Adamson, et al. (2001). "Subretinal transplantation of genetically modified
human cell lines attenuates loss of visual function in dystrophic rats." Proceedings of the
National Academy of Sciences of the United States of America 98(17): 9942-9947.
McGill, T. J., R. M. Douglas, et al. (2004). "Quantification of spatial vision in the Royal College
of Surgeons rat." Investigative ophthalmology & visual science 45(3): 932-936.
Petrovski, G., E. Berenyi, et al. (2011). "Clearance of dying ARPE-19 cells by professional and
nonprofessional phagocytes in vitro- implications for age-related macular degeneration
(AMD)." Acta Ophthalmol 89(1): e30-34.
95
Pinilla, I., N. Cuenca, et al. (2007). "Preservation of outer retina and its synaptic connectivity
following subretinal injections of human RPE cells in the Royal College of Surgeons rat."
Exp Eye Res 85(3): 381-392.
Prusky, G. T., P. W. West, et al. (2000). "Behavioral assessment of visual acuity in mice and
rats." Vision research 40(16): 2201-2209.
Sauve, Y., I. Pinilla, et al. (2006). "Partial preservation of rod and cone ERG function following
subretinal injection of ARPE-19 cells in RCS rats." Vision Res 46(8-9): 1459-1472.
Schwartz, S. D., J. P. Hubschman, et al. (2012). "Embryonic stem cell trials for macular
degeneration: a preliminary report." Lancet 379(9817): 713-720.
Sheedlo, H. J., L. Li, et al. (1991). "Photoreceptor cell rescue at early and late RPE-cell
transplantation periods during retinal disease in RCS dystrophic rats." J Neural
Transplant Plast 2(1): 55-63.
Singh, S., S. Woerly, et al. (2001). "Natural and artificial substrates for retinal pigment epithelial
monolayer transplantation." Biomaterials 22(24): 3337-3343.
Stanzel, B. V., Z. Liu, et al. (2012). "Subretinal delivery of ultrathin rigid-elastic cell carriers
using a metallic shooter instrument and biodegradable hydrogel encapsulation." Invest
Ophthalmol Vis Sci 53(1): 490-500.
Tezcaner, A., K. Bugra, et al. (2003). "Retinal pigment epithelium cell culture on surface
modified poly(hydroxybutyrate-co-hydroxyvalerate) thin films." Biomaterials 24(25):
4573-4583.
96
Thomas, B. B., D. M. Samant, et al. (2007). "Behavioral evaluation of visual function of rats
using a visual discrimination apparatus." Journal of neuroscience methods 162(1-2): 84-
90.
Thomas, B. B., M. J. Seiler, et al. (2004). "Optokinetic test to evaluate visual acuity of each eye
independently." Journal of neuroscience methods 138(1-2): 7-13.
Thumann, G., A. Viethen, et al. (2009). "The in vitro and in vivo behaviour of retinal pigment
epithelial cells cultured on ultrathin collagen membranes." Biomaterials 30(3): 287-294.
Vugler, A., A. J. Carr, et al. (2008). "Elucidating the phenomenon of HESC-derived RPE:
anatomy of cell genesis, expansion and retinal transplantation." Exp Neurol 214(2): 347-
361.
Wang, S., B. Lu, et al. (2008). "Morphological and functional rescue in RCS rats after RPE cell
line transplantation at a later stage of degeneration." Invest Ophthalmol Vis Sci 49(1):
416-421.
Williams, R. L., Y. Krishna, et al. (2005). "Polyurethanes as potential substrates for sub-retinal
retinal pigment epithelial cell transplantation." J Mater Sci Mater Med 16(12): 1087-
1092.
Wongpichedchai, S., J. J. Weiter, et al. (1992). "Comparison of external and internal approaches
for transplantation of autologous retinal pigment epithelium." Invest Ophthalmol Vis Sci
33(12): 3341-3352.
Yamamoto, S., J. Du, et al. (1993). "Retinal pigment epithelial transplants and retinal function in
RCS rats." Investigative ophthalmology & visual science 34(11): 3068-3075.
97
Zhu, D., X. Deng, et al. (2011). "Polarized secretion of PEDF from human embryonic stem cell-
derived RPE promotes retinal progenitor cell survival." Investigative ophthalmology &
visual science 52(3): 1573-1585.
98
Chapter 5
Using optical coherence tomography as a proxy for evaluation of RPE implants in
the RCS rat.
ABSTRACT
Purpose: To evaluate progressive changes in the optical coherence tomography (OCT) and to
compare the OCT with histologic findings in dystrophic Royal College of Surgeons (RCS) rats.
The study also examines the feasibility of using this information to evaluate the RPE-stem cell
implanted animals.
Methods: Spectral-domain optical coherence tomography (SD-OCT) was performed at postnatal
age 18, 25, 32, 39, 46, 60, 90 day (n=3 for each age group) old RCS rats. Three normal
Copenhagen rats served as controls. The rats were then sacrificed and eyes enucleated for
histologic examination using hematoxylin and eosin (H&E) staining. The changes in SD-OCT
were evaluated and compared with the histology. The RCS rats underwent implantation of retinal
pigment epithelial cells derived from human embryonic stem cells (hESC-RPE) and delivered as
a polarized cell sheet on a thin parylene membrane (n=3). The animals were imaged using SD-
OCT volume scan at 4 quadrants of the retina 2 months post-surgery. The retinal area with
preservation of the outer retinal structure (outer plexiform layer, OPL) was mapped using
information from the volume scan.
Results: The H&E staining of the retina of RCS rats showed that the most significant changes of
aging occurred in the outer retina. The outer nuclear layer (ONL) thickness decreased rapidly
between the age of P18 to P90, and the changes were detectable on OCT. The outer segment
degeneration made the photoreceptor inner segment/ outer segment (IS/OS) junction invisible by
99
P25 on OCT. OCT showed that the outer plexiform layer (OPL) band becomes invisible by the
age of P60. The more significant change in histology between age P46 and P60 was that both the
thicknesses of ONL and OPL fall below 5µm. For the RCS rat that received subretinal hESC-
RPE/parylene transplantation, mapping of the retinal area that demonstrated the presence of
outer retinal structures showed that the preservation was at the treated quadrant. The parylene
membrane used for delivering the hESC-RPE has the benefit of being visible on the infrared
fundus picture and thus the preservation can be clearly identified as being on and adjacent to the
implanted area.
Conclusions: The OPL on SD-OCT became invisible at the age of P60 in the RCS rat due to the
degeneration of the outer retina. This can be used as a quick in vivo evaluation for studies
looking at therapeutic effects such as stem cell implantation/transplantation in animal models of
outer retinal dystrophies.
INTRODUCTION
The RCS rat has a naturally occurring mutation of the receptor tyrosine kinase gene (MerTk)
(D'Cruz, Yasumura et al. 2000; Nandrot, Dufour et al. 2000). This mutation causes a defect in
RPE cell phagocytosis which is critical for maintenance of normal photoreceptor physiology
(Dowling and Sidman 1962). This results in undigested photoreceptor outer segments to form a
debris zone between the photoreceptor and retinal pigment epithelial cells and subsequent
photoreceptor cell death (LaVail and Battelle 1975). Optical coherence tomography (OCT) is
widely used clinically for the diagnosis of retinal diseases and for the evaluation of treatment
modalities (Schuman 2013). Seiler et al for example, reported using OCT for evaluation of fetal
retinal sheet transplants in RCS rats (Seiler, Rao et al. 2010). The location of the implant and
gross changes of retina structure could be determined non-invasively in this way. The purposes
100
of the current study are 1) to establish and evaluate the progressive structural changes in the RCS
rat retina by OCT; 2) evaluate the possibility of using the specific OCT findings as a quick
evaluation tool in a large retinal area, thus, obviating the need of processing large numbers of
histologic samples, and 3) preliminarily, determine if OCT would be helpful in assessment of
treatments such as stem cell therapy.
MATERIALS AND METHODS
Baseline Animals
For all experimental procedures, animals were treated in accordance with the NIH
guidelines for the care and use of laboratory animals and the ARVO Statement for the Use of
Animals in Ophthalmic and Vision Research, under a protocol approved by the Institutional
Animal Care and Use Committee of the Doheny Eye Institute, University of Southern California.
All efforts were made to minimize animal suffering and to use only the minimum number of
animals necessary. Pigmented dystrophic RCS rats at postnatal day (P) 18(n=3), 25 (n=3), 32
(n=3), 39 (n=3), 46 (n=3), 60 (n=3), 90 (n=3) and P120 (n=3) were used for evaluation of the
progressive changes in retina. Three 3 month-old Copenhagen rats were used as a control group.
Spectral-domain optical coherence tomography (SD-OCT) Four Quadrant Volume Scan
Rats were anesthetized by intraperitoneal injection of ketamine/xylazine (3.75 mg/kg
ketamine and 5 mg/kg xylazine). The pupils were dilated using topical application 2.5%
phenylephrine and 0.5% tropicamide on each cornea. The OCT of four quadrants of the retina
(center, superior, inferior, nasal, and temporal) of each eye was scanned using a spectralis
HRA+OCT (Heidelberg Engineering, Germany). An OCT volume scan consisting of 31 scan
lines was performed.
101
Histology
Rats were euthanized with an overdose of sodium pentobarbital (Sigma-Aldrich, St.
Louis, MO) after OCT examination. The nasal side of each cornea was marked with a burn
created by a fine tip ophthalmic cautery. Then the eyes were enucleated and fixed with
Davidson’s solution for at least 18 hours and embedded in paraffin. The eyes were processed for
horizontal sections to compare with the OCT data at corresponding locations. Sections (5 µm
thick) crossing the center of optic nerve were stained with H&E stain and scanned with an
Aperio ScanScope (Aperio Technologies, Vista, CA). The thickness of the inner nuclear layer
(INL), inner plexiform layer (IPL), the outer plexiform layer (OPL), and ONL were measured in
the mid-peripheral area (midpoint between optic nerve and pars plana) of the nasal and temporal
retina using the Aperio ScanScope.
Implantation and OCT scanning in RCS rat
Implantation of hESC-RPE cells cultured on a thin parylene membrane was performed in
RCS rats between postnatal days 27 to 29 (P27-29) using methods previously described (need
reference here). Briefly, the hESC-RPE cells were differentiated from H9 embryonic stem cells.
Rectangular pieces (0.4mm x 0.9mm) of specially machined parylene membrane were seeded
with hESC-RPE cells. hESC-RPE formed a confluent monolayer on the parylene membranes
after about 3 weeks of culture. Implantation surgeries were performed under ketamine and
xyalazine anesthesia. A transcleral approach was used for delivery of the subretinal implant after
induction of a focal retinal detachment with injection of balanced salt solution into the subretinal
space. The three implanted rats received an OCT volume scan in 4 quadrants of the retina 2
months post-surgery in both eyes. The scan line that showed preservation of the OPL band was
102
plotted onto a contour map using Origin Pro 8.6 (OriginLab, Northampton, MA) and overlapped
with the infrared fundus picture.
Statistical Analysis
Data are presented as mean ± SEM. Significance was calculated with one-way ANOVA
with Bonferroni post hoc test to compare the retinal layer thicknesses of the RCS rats at different
ages.
RESULTS
SD-OCT and histology in normal Copenhagen rat retinas
The OCT in the P90 Copenhagen rat showed that the nerve fiber layer (NFL), INL (dark
band), OPL (bright band), ONL (thick dark band), and IS/OS junction (bright band) can be
clearly identified (Figure 1). However, it is difficult to clearly identify the NFL, ganglion cell
layer (GCL) and inner plexiform layer (IPL). The NFL, GCL, and IPL are the 3 layers that are
referred to as the “ganglion cell complex” clinically because of the relevance to glaucoma
progression. These three layers are marked as “ganglion cell complex” in our present study
because it is difficult to differentiate the three layers. The histology of the corresponding area
showed the relevance between OCT and histology.
103
Figure 1. A. OCT scan of a normal adult Copenhagen rat. B. Histology of the corresponding area showing the
relevance between the OCT scans and the histology. (GCC, ganglion cell complex; INL, inner nuclear layer; OPL,
outer plexiform layer; ONL, outer nuclear layer; IS/OS, photoreceptor inner segment/outer segment junction)
SD-OCT in dystrophic RCS rat retinas
The OCT scans in RCS rats from P18 to 60 showed that the retinal layers seemed to be
comparable to the retinal layers of the normal Copenhagen rats at P18 with a visible ganglion
cell complex, INL, OPL, ONL, and IS/OS junction (compare Figures 1 and 2). The thickness of
the ONL visibly declined progressively after the age of P25. The OPL band disappeared focally
at P46 and became totally invisible by P60 (Figure 2E, F). The OCT maintains a similar pattern
from P60 up to P120 (data not shown).
Figure 2. The horizontal OCT scans of
nasal retina at the level of optic nerve of
the RCS rat from P18 to P60. At P18, the
OCTs scans were similar to a normal
Copenhagen rat (A). At P25, the ONL
thickness started to decline rapidly and
the IS/OS junction band disappear (B).
At P25 (B) and P32 (C) broad
hyperreflective band started to form in
the subretinal space as a result of debris
accumulation (arrowhead). At P39, the
OPL band was less defined (arrow) (D).
104
At P46, the OPL band was only present focally (arrow) (E). At P60, The OPL band is completely diminished (F).
(OPL, outer plexiform layer; ONL, outer nuclear layer; IS/OS junction, inner segment/outer segment junction)
Histology of Dystrophic RCS rat Retinas
The histology of the RCS rats showed that the IS/OS junction is present until P25 (Figure
3A, B). At P25, the OS is visibly elongated (Figure 3B). At P32, the IS/OS junction became
blurry when the IS and OS gradually turn into a debris layer (Figure 3C). The ONL layer
thickness declined rapidly from P18 to P60 (Figure 3). After P90, the ONL declined to an
incontinous single layer (data not shown). The measured thickness of IPL, INL, OPL, and ONL
from histology sections showed that the most significant change in the RCS rat’s retina between
the ages of P18 to P90 is in the ONL thickness; it is decreased by 93% (Figure 4). The OPL is
decreased by 79% and the INL by 20% between P18 to P90. The change in the IPL thickness
was less consistent.
105
Figure 3. Histologic section of RCS rats from the age of P18 to P60. The ONL underwent rapid degeneration from
P18 to P90 (A-F). At P18 and P25, the ONL was thicker than the INL and the photoreceptor OS and IS boundary
was visible (A, B). At P32, the IS/OS boundary is less clear (C). At P39, the IS/OS structure became invisible (D).
At P46, the ONL layer was significantly thinner than the INL (E). At P60, the ONL was only around 3 layers thick
(F). (ONL, outer nuclear layer; OS, outer segment; IS, inner segment; INL, inner nuclear layer)
Because of the significant changes in the OPL between the age of P39 and P46, we compared the thickness changes
in the OPL and the adjacent layers- INL and ONL in histology. The INL thickness is 41.3±1.7 μm at P39 and
40.1±2.6 μm at P46 (P=0.8). The OPL thickness is 6.7±1.0μm at P39 and 5.6±0.8 μm at P46 (P=0.16). The ONL
thickness is 31.5±1.5 μm at P39 and 24.1±1.1 μm at P46 (P<0.001).
106
Figure 4. The thickness measure of the retinal layers in the mid-peripheral retina of different ages RCS rats The
OPL and ONL thickness decreased rapidly from P18 to P90. (IPL, inner plexiform layer; INL, inner nuclear layer;
OPL, outer plexiform layer; ONL, outer nuclear layer.)
OCT evaluation of hESC-RPE/parylene membranes implanted into the subretinal space of
RCS rats
OCT analyses performed in RCS rats 2 months after the subretinal implantation of the
hESC-RPE/parylene membrane showed that, in all 3 rats examined, the presence of the OPL
band could be discerned in the implanted area (Figure 5 top inset). The contour map shows the
area with preservation of the OPL band (orange color). Importantly, the results clearly
demonstrated that the preservation of the OPL band was only immediately over the implant or
adjacent area (Figure 5).
61.1
48.1
37.8
31.5
24.1
10.2
4.1
12.0
10.6
7.9
6.7
5.6
5.5
2.6
46.6
42.8
41.5
41.3
40.1
33.9
33.2
47.0
46.6
40.0
41.1
46.9
45.0
37.8
0
20
40
60
80
100
120
140
160
180
18 25 32 39 46 60 90
Thickness (µm)
Age (post-natal days)
IPL thickness( μm)
INL thickness( μm)
OPL thickness( μm)
ONL thickness( μm)
107
Figure 5. Infrared fundus photo of an RCS rat 2 months after subretinal transplantation of hESC-RPE/parylene
membrane. The OCT scan (upper inset) represented the scan of the implanted area (thick green line in infrared photo)
(upper inset). The OCT showed a preservation of the OPL band in the implanted area (orange). The contour plot of
the OPL showed that the preservation of OPL corresponds to the implanted area. The other quadrants were scanned
and presented by the OCT scan lines. However, these areas did not show any preservation of the OPL band on OCT.
The Histologic section of the implanted area showed preservation of the ONL, and the parylene implant is indicated
by black arrow (lower inset).
108
DISCUSSION
The OPL band appears as a hyperreflective band on OCT and the adjacent layers, INL
and ONL, are hyporeflective (Schuman 2013). The OPL band became less clear on OCT at the
age of P39 and became discontinuous at P46 in dystrophic RCS rats. The cause of discontinuity
of the IPL can be due to the changes in the thickness of the OPL layer or the adjacent
hyporeflective layers which help contrast the hyperreflective OPL band between the age of P39
and P46. It is more likely that the disappearance of the OPL was due to the decrease of the ONL
layer thickness, because of the three layers that help to form a visible OPL band on OCT, the
ONL thickness was the only one that has changed significantly. It is likely that the ONL
thickness falls below the vertical resolution of the OCT, and the OPL band blends with the outer
retinal debris which also creates a hyperreflectance. The other possibility is that the OPL
thickness falls below the OCT vertical resolution at P46. However, the mean difference of OPL
between P39 and P46 was only 1.1μm. It is less likely that the dramatic difference was cause by
a minor change in OPL thickness. Either the OPL or ONL thickness change is responsible for the
disappearance of the OPL band on OCT; this indicates that the degeneration in the outer retina is
extensive enough to change the OCT pattern and can be used as an indicator of advanced retinal
degeneration.
In our study, we tried to determine if the finding of the preservation of the OPL band on
OCT could be valuable marker for developing a rapid in vivo evaluation for any retinal
regeneration study using animal models of retinal degeneration. We found that the preservation
of the OPL band in OCT can be clearly observed in RCS rats 2 months after implantation
surgery even though it usually disappears completely by P60 in the untreated retina. This
indicates that the rescue effect is sufficient to retain at least the OPL and ONL 2 months after the
109
surgery. This is an especially useful finding for the evaluation of large numbers of implanted
animals without the need of processing tissue sections. For other studies focusing on the area of
rescue, the contour map offers a better way to demonstrate the effect because the OCT contour
provides a three dimensional evaluation in contrast to histologic sections which are two
dimensional. It will also be useful for regeneration studies that do not have a clearly visible
implant such as in gene therapy or cell injection therapy. The 4 quadrant OCT scan can give
researchers an idea where to find the preservation of the retinal layers and focus the processing
of the histology sections over that particular area.
Note that when interpreting OCT, OCT scans are not a direct measurement of the
histology. As our OCT and histology comparison demonstrated, the OPL and ONL layer
disappear on OCT before the total loss of these layers on histology. This is likely due to the
limitation of the vertical resolution of SD-OCT. However, the loss of the OPL band will still be a
good marker to indicate the advanced loss of outer retinal tissue.
A limitation of our study is that the contour map was plotted based on the presence or
absence of a visible OPL. The OCT would have been more valuable if each layer of the retina
had been evaluated separately. However, the outer retinal structure is in an advanced stage of
degeneration by the age of P46 in the RCS rat, and thus, it become difficult to discriminate
between the different layers in OCT. An automated segmentation software will be required to
perform a more comprehensive analysis of the OCT.
In summary, we have demonstrated that there is good correlation between the SD-OCT
scans that show the retinal layers and the underlying histology. The outer retina layer is lost in
OCT scans, however, before the total loss is observed in histology as would be expected due to
the resolution limit of the OCT. We believe though that SD-OCT appears to be a powerful tool
110
for the evaluation of the effect of subretinal implants in RCS rats as the volume scan can non-
invasively evaluate a broad area in a short time.
111
CHAPTER 5 REFERENCES
D'Cruz, P. M., D. Yasumura, et al. (2000). "Mutation of the receptor tyrosine kinase gene Mertk
in the retinal dystrophic RCS rat." Hum Mol Genet 9(4): 645-651.
Dowling, J. E. and R. L. Sidman (1962). "Inherited retinal dystrophy in the rat." J Cell Biol 14:
73-109.
LaVail, M. M. and B. A. Battelle (1975). "Influence of eye pigmentation and light deprivation on
inherited retinal dystrophy in the rat." Exp Eye Res 21(2): 167-192.
Nandrot, E., E. M. Dufour, et al. (2000). "Homozygous deletion in the coding sequence of the c-
mer gene in RCS rats unravels general mechanisms of physiological cell adhesion and
apoptosis." Neurobiol Dis 7(6 Pt B): 586-599.
Schuman, J. S. (2013). Optical coherence tomography of ocular diseases. Thorofare, NJ, SLACK
Inc.
Seiler, M. J., B. Rao, et al. (2010). "Three-dimensional optical coherence tomography imaging of
retinal sheet implants in live rats." J Neurosci Methods 188(2): 250-257.
112
Chapter 6
Histological correlates of Changes in the Fundus Autofluorescence Pattern in a Rat
Retinal Degeneration Model
ABSTRACT
Purpose: To investigate the histological correlates of progressive changes in the fundus
autofluorescence (FAF) pattern in Royal College of Surgeon (RCS) rats observed using
Spectralis HRA+OCT imaging. This study is aimed at implementing a reliable non-invasive
technique to estimate the degenerative status of the retina and hence for suitable planning of
therapeutic strategies based on gene therapy and cell replacement techniques.
Methods: FAF imaging in RCS rats was performed at postnatal age 18, 25, 32, 39, 46, 60, 90,
120, 150, 300, 415. To examine the possible correlation between FAF pattern and the
degenerative status of the retina, histologic examination of the retina was performed during the
above time points. With the help of an Aperio ScanScope, H&E stained retinal sections were
imaged. The thickness and nuclear layer cell count was measured using the Aperio ScanScope
software. To determine the mechanism underlying heterogeneity observed in the FAF pattern in
RCS rats during the course of degeneration, immunofluorescent staining for CD68 (macrophage
marker) and GFAP (glia cell marker) was performed and imaged using confocal microscopy.
Spectral analysis was used to estimate the composition of the hyperautofluorescent spots
observed in the debris layer of the degenerating retina at various time points.
Results: In young RCS rats (until postnatal day (P) 32), the FA pattern remained more or less
comparable to the normal (control) Copenhagen rats. After P32, signs of hyperautofluorescence
began to appear in the entire fundus. The above pattern persisted until P90. After P90,
113
considerable changes in the FAF were apparent with the development of a prominent
hypoautofluorescent area surrounding the optic nerve (ON) head. With increase in age, this
hypoautofluorescent area progressively extended to the peripheral regions of the fundus. During
this period, spicule like hyperautofluorescent spots began to appear in the fundus. Histological
evaluation of the retina at ages corresponding to FAF imaging revealed that loss of debris is
initiated at the area surrounding the ON head where hypoautofluorescence was first observed.
Aperio ScanScope measurements suggested that changes in the fundus autofluorescence pattern
are concomitant with the loss of outer nuclear layer (ONL) thickness and cell count. The area
that developed hypoautofluorescence showed significant loss of ONL thickness and cell count.
During the earlier stages of degeneration (P39-P46), significant differences were observed for
ONL thickness and count when superior quadrants and nasal-inferior quadrants of the retina
were compared (p<0.05). The above difference in the ONL thickness and count was not strictly
corresponding to the progressive changes observed in the FAF pattern. Morphological evaluation
of the RCS retina demonstrated that the bright hyperautofluorescent spots in the FAF originates
from the apical surface of the RPE and not from inside the RPE cells. Spectral analysis of such
increased autofluorescent spots revealed a peak emission at 538 nm.
Conclusions: In RCS rats, changes in the FAF are predominantly related to the status of the
debris zone in the subretinal space. Based on the FAF imaging data, it is possible to predict the
progressive state of the disease. The study ruled out the role of macrophages and RPE
phagosomes as a possible source for the isolated hyperautofluorescent spots observed during
advanced stages of degeneration. The information obtained from this study may help to
understand the histological correlates of progressive changes in FAF in human retinal disease
conditions and its correlation with the degenerative status of the retina. The above technique can
114
be used as a reliable tool for planning suitable treatment strategies for progressive retinal
degeneration diseases.
INTRODUCTION
Progressive changes in the fundus autofluorescence (FAF) pattern in human disease conditions
(Nandakumar, Buzney et al. 2012) and in animal models of retinal degeneration (RD) has been
well documented (Hirata, Yasukawa et al. 2009; Luhmann, Robbie et al. 2009; Wang, Fine et al.
2009; Boretsky, Motamedi et al. 2011; Charbel Issa, Singh et al. 2012; Secondi, Kong et al.
2012). In patients with age-related macular degeneration (AMD), such changes appear as
hypoautofluorescence areas surrounding the fovea and gradually spread into the nearby areas.
The source of FAF in human is mostly contributed by lipofuscin accumulated in the retina
pigment epithelium (RPE) (Delori, Dorey et al. 1995). The major lipofuscin fluorophore is
bisretinoid A2E. A2E is demonstrated to affect the acidity in the intracellular organelles, causes
loss of membrane integrity and the blue region of the Light spectrum has a marked ability to
induce apoptosis of A2E-laden RPE cells. Investigations of the photochemical events that trigger
the cell death have revealed that upon blue light irradiation, A2E self-generates singlet oxygen
and later react with A2E at carbon–carbon double bonds along the retinoid-derived side-arms of
the molecule to form epoxides (Ben-Shabat, Itagaki et al. 2002; Sparrow, Zhou et al. 2002).
These highly reactive epoxides (A2E-epoxides), rather than singlet oxygen, may be the
intermediates that ravage the cells (Sparrow, Fishkin et al. 2003). This has been reported also in
abcr1/2 Mice, an animal model with a phenotype similar to recessive Stargardt’s disease (STGD)
and AMD in Humans (Mata, Tzekov et al. 2001). Based on the AMD studies, it has been
concluded that any change in the pattern of autofluorescence (AF) such as increase or decrease in
autofluorescence is indicative of dysfunctional or dead RPE cells. Based on the information
115
available from animal models of RD, change in the FAF pattern is related to the accumulation of
degenerative substances in the retina. Ccl2 knockout mice that possess a defective monocyte
recruitment pattern develop drusen and other features of AMD, such as the accumulation of
lipofuscin in RPE cells, progressive outer retinal degeneration, and geographic/RPE atrophy
(Luhmann, Robbie et al. 2009). A significantly higher number of hyperautofluorescent spots in
16- to 25-month-old Ccl2-/- mice have been reported. The regular pattern, size, and location of
the cells indicate that the hyperautofluorescent spots in the AF-SLO fundus images in senescent
Ccl2-/- mice were subretinal macrophages containing autofluorescent material (Luhmann,
Robbie et al. 2009). From the above investigations, it may suggest that the origin of changes in
the FAF pattern is different across diseases and may vary between disease models. Hence, it is
difficult to make a conclusion of the disease status based on the FAF pattern without making in
depth analysis of the retinal histology. Lipofuscin accumulation (A2E) inside the RPE and its
structural changes is considered to be the basis of changes in the FAF pattern in human AMD
patients (Shaban and Richter 2002; Wu, Yanase et al. 2010). In Royal College of Surgeons (RCS)
rats, based on the histological examination the source of autofluorescence is considered to be the
extensive debris accumulation observed in the subretinal space (Katz, Eldred et al. 1987; Spaide
2008). Since no major investigation has been conducted to evaluate the FAF pattern in the RCS
eyes it will be interesting to study the possible changes in the FAF pattern consequent to
progressive changes in the retinal degenerative condition. In RCS rats, the degeneration of the
photoreceptors is initiated by the RPE dysfunction diseases caused by the Mer tyrosine kinase
(MerTK) protein (D'Cruz, Yasumura et al. 2000; Nandrot, Dufour et al. 2000) This genetic
defect makes the RPE incapable of phagocytosis and hence the debris accumulates in the retina
that ultimately leads to degeneration of the photoreceptors. (Dowling and Sidman 1962; Holz,
116
Bellman et al. 2001; Hwang, Chan et al. 2006) Based on the histological examinations performed
in RCS rats, the major source of retinal autofluorescence is the debris layer.(Katz, Eldred et al.
1987) This may be because of the fact that the debris layer mainly composes of the A2-PE,
precursor of A2E, that is generated in the photoreceptor outer segment membrane.(Liu, Itagaki et
al. 2000). This unlike in the majority of the human RP or AMD patients where the source of
autofluorescence is the lipid derivatives of the phototransduction cascade in the RPE(Yin 1996;
Sparrow and Boulton 2005) However, in human RPE dysfunction diseases that are comparable
to the RPE dysfunction observed in RCS rats, the source of FAF could be different. This has
been recently studied in a subset of patients with retinitis pigmentosa and having mutation in
MerTK as observed in RCS rats. Charbel et al report one family with mutation in MerTK and
noted one younger patient had multifocal increase FAF signal which is similar to what we found
in RCS rat (Charbel Issa, Bolz et al. 2009). In this study, progressive changes in the FAF pattern
is observed, however no major histological correlates has been made mostly due to the
unavailability of tissue sample representing different stages of the degeneration . This suggests
that, although the increased accumulation of autofluorescent pigments inside the RPE may be
one of the causes for photoreceptor death in human AMD patients (Sparrow and Boulton 2005),
the FAF generating debris in the RCS retina is not directly involved in the photoreceptor
degeneration in these animals. Hence, it is important that progressive changes in the pattern of
FAF and its correlation with histology changes is studied using animal models. For example,
LaVail et al observed a significantly slower loss of outer nuclear layer in the superior quadrant of
RCS rat from P45 to P85 (LaVail and Battelle 1975). The difference was not significant at P95.
Because RCS rat is often used in translational medicine studies as a model of RPE dysfunction
for development of new treatment strategy such as transplantation of RPE cells, and outer
117
nuclear layer count is often used as measurement to determine the efficacy of the treatment, it is
critical to develop a non-invasive technique to easily understand if there is difference in the
degree of degeneration at different locations in the retina and possible correlation with FAF
pattern. This can also help to decide the area of administration of the therapeutic intervention in
the eye.
MATERIALS AND METHODS
Animals
For all experimental procedures, animals were treated in accordance with the NIH
guidelines for the care and use of laboratory animals and the ARVO Statement for the Use of
Animals in Ophthalmic and Vision Research, under a protocol approved by the Institutional
Animal Care and Use Committee of the Doheny Eye Institute, University of Southern California.
All efforts were made to minimize animal suffering and to use only the minimum number of
animals necessary to provide an adequate sample size for statistically meaningful scientific
conclusions. Pigmented dystrophic RCS rats at P18 (n=3), 25 (n=3), 32 (n=3), 39 (n=3), 46 (n=3),
60 (n=3), 90 (n=3), 120 (n=3), 150(n=2), 300(n=2) 415(n=3) were used for evaluation of the
natural progression of autofluorescence. Normal Copenhagen rats were used as a control group
for studying the autofluorescence pattern in a normal rat fundus.
Autofluorescence Scanning Laser Ophthalmoscopy (AF-SLO) and Spectral-domain
Optical Coherence Tomography (SD-OCT)
Rats were anesthetized by intraperitoneal injection of ketamine/xylazine (3.75 mg/kg
ketamine and 5 mg/kg xylazine). The pupils were dilated using topical application 2.5%
phenylephrine and 0.5% tropicamide on each cornea.
118
FAF imaging was performed using Spectralis HRA+OCT (Heidelberg Engineering, Germany) at
excitation wavelength of 488 nm. A 55° angle lens was used for autofluorescence imaging with
sensitivity of 67%. Five images at different quadrants of the retina (center, superior, inferior,
nasal, and temporal) were taken for each eye.
Histology
Rats were euthanized with an overdose of sodium pentobarbital (Sigma-Aldrich, St.
Louis, MO) at various time points. The nasal side of each cornea was marked with a burn created
by a fine tip ophthalmic cautery and used as a marker to locate various retinal areas during
histological evaluation. Both right and left eyes were enucleated and fixed using Davidson’s
solution for at least 18 hours before embedded in paraffin. The left eyes were processed for
obtaining sagittal sections and the right eyes were used for obtaining transverse sections. Each
microtome section (5 micrometer thick) passing through the center of optic nerve was stained
using hematoxylin and eosin (H&E). Vertical sections were used for measuring the retinal
thickness and outer nuclear layer cell count from the superior and inferior quadrant of the retina.
Horizontal sections were used for measurements representing the nasal and temporal retina areas.
From each section, measurements were made from the mid peripheral area (midpoint between
optic nerve and pars plana). Retinal imaging and quantitative analysis were performed using an
Aperio ScanScope CS instrument and analyzed in ImageScope
TM
using the modified IHC
Nuclear Analysis v9.1 tool.
Immunofluorescence Imaging
Sections adjacent to those used for morphologic evaluation based on H&E staining were
selected for immunofluorescent imaging. Sections were deparaffinized with xylene, dehydrated
119
in graded alcohols, washed in PBS. Antigen was retrieved with antigen unmasking solution (H-
3300, Vector, Burlingame, CA) using a pressure cooker for 3 minutes. Blocking was performed
using 5% bovine serum albumin with 0.1% Triton-X100 (Sigma) for 20 minutes at room
temperature (RT). The sections were subsequently incubated with primary antibodies diluted
with PBS in a humidified chamber at 37 degree water bath for 1 hour, and secondary antibodies
diluted in PBS for 45 minutes at RT. The primary antibody used was against glial fibrillary
acidic protein (GFAP, dilution 1:1000, Abcam, Cambridge, MA) and macrophage marker CD68
(dilution 1:50, Abcam, Cambridge, MA). The secondary antibodies used were fluorescein
isothiocyanate (FITC)-conjugated goat anti-mouse IgG (dilution 1:200) and rhodamine-
conjugated goat anti-rabbit IgG (dilution 1:50) (both from Jackson ImmunoResearch
Laboratories, Inc, West Grove, PA). After washing three times with PBS, the sections were
mounted with mounting medium containing the nuclear dye DAPI (VECTASHIELD,
Burlingame, CA) and images taken under a spinning disc confocal microscope at X40
(PerkinElmer Ultraviewer spinning disc confocal, PerkinElmer Inc., Waltham,MA).
Evaluation of autofluorescence: To evaluate the presence of autofluorescence in the tissue
samples, histological sections adjacent to those used for immunofluorescent imaging were
deparaffinized and mounted with mounting medium containing the nuclear dye DAPI.
Representative images (n=6) that nearly cover the entire section was taken using a laser scanning
microscope (LSM 510 Meta; Carl Zeiss Microimaging, Jena, Germany) at X40. Spectral images
(excitation 488 nm) were obtained from 528 nm to 688 nm in 10 nm intervals. For each image,
10 increased autofluorescent spots located in the subretinal space were selected and the emission
spectra of the spots obtained using computer software, LSM 510Meta. The peak emission
120
spectrum was analyzed for 3 age group, P60, P90, and P120 where changes in FAF pattern were
found.
Statistical Analysis
Retinal layer thickness and outer nuclear layer cell count are presented as mean ± SEM.
Statistical comparisons were made using two-way ANOVA followed by the Bonferroni post hoc
test for multiple comparisons. Pearson correlation analysis was performed between the length of
hypoautofluorescence on FAF and debris loss on histologic section (Graph Pad Prism 5.0, La
Jolla, CA).
RESULTS
Autofluorescence Scanning Laser Ophthalmoscopy
The FAF image of RCS rats showed a uniform appearance of the fundus at P18 (Figure
1). At P32, the FAF slightly increased making the retinal vessel in the background more visible
which is comparable to that of a normal Long-Evans rat (Figure 1). At P90, a
hypoautofluorescent area started to develop surrounding the optic nerve head and slowly
enlarged with age. By P180, the hypoautofluorescent area occupied about 80% of the fundus
(Figure 1). The fundus was almost entirely covered by this dark area when examined at P415
(Figure 1). Presence of bright hyperautofluorescent spots inside the dark area and more
frequently at the margin between the hypo and hyperautofluorescent regions in the fundus began
to appear at P90.Abundance of such spots are observed in the fundus in the subsequent time
points (up to P415, data not shown).
121
Figure 1. Fundus autofluorescence image taken at different age in RCS rats. A hypoautofluorescent area centered on
optic nerve is found at P90.The size of hypoautofluorescent patch continues to increase with age. Punctate
hyperautofluorescent spots developed within the patch of hypoautofluorescence around P90. Hyperautofluorescent
spots extended outside the hypoautofluorescent area around P150. The FAF pattern remains similar after P150.
Retina Layer Thickness and Outer Nuclear Layer Cell Count
In order to examine possible correlation between changes in the FAF pattern and retinal
morphology, quantitative estimation of the retinal layers was undertaken from H&E stained
images using Aperio ScanScope technique. For this, the thickness of inner plexiform layer, inner
nuclear layer, outer plexiform layer, outer nuclear layer (ONL) and ONL cell count were
122
measured in RCS rats from four different quadrants of the retina, at different time points from
P18 to P415. Although progressive decrease in thickness was observed in various layers of the
RCS retina, loss was more dramatic at the level of outer nuclear layer where the thickness
decreased from 60 μm (P18) to near total loss at P120(Figure 2). Approximately 20% loss of
ONL count was observed at each week from P18 to P32 (Figure 2E). The ONL cell count and
the thickness measurement maintained a negative linear relationship with age until advanced
stages of degeneration (P60) when it is difficult to perform microscopic measurement of ONL
thickness (Figure 2D, E).
Figure 2. Progressive loss of the different layers of retina at different quadrant in RCS rats. The different quadrant
showed significant difference of outer nuclear layer count between inferior and nasal quadrant and inferior and
temporal quadrant at P46. The difference was not significant after P46.
123
Progressive Loss of Debris in Retina of RCS rat
Based on histological evaluation, the progressive loss of debris layer in the RCS retina
starts from the optic nerve head area and gradually extends to the peripheral retina. Based on the
measurements made at each quadrant, the progressive loss of the debris layer was found to be
uniform across the four quadrants until P300. At this age significantly higher loss of debris layer
was observed in the inferior quadrant of the retina (Figure 3A).
Correlation between Autofluorescence and Debris Loss
Histological examination of the RCS retina confirmed that the major source of autofluorescence
is the debris layer. Hypoautofluorescence in the area surrounding the ON was observed at P90, the
earliest age at which apparent loss of debris layer from the subretinal area is first noticed.
Quantitative estimation of the progressive changes in the size of the hypoautofluorescent region
(indicative of loss of FAF) (Figure 3B) suggested statistically significant linear association
between the loss of debris zone and the loss of autofluorescence on FAF (p<0.05). This shows
that the loss of autofluorescence in FAF image is due to the loss of debris zone. When a retinal
detachment was made using saline to cause focal disruption of the debris, a hypoautofluorescent
area appeared in the fundus corresponding to the location of the bleb formation (Figure 4). This
further demonstrated that the absence of debris causes decrease in the FAF.
124
HE debris length
0 100 200 300 400 500
0
500
1000
1500
2000
2500
Sup
Nasal
Inf
Temp
Post-natal age (days)
Length ( m)
Debris length correlation
0 1000 2000 3000
0
500
1000
1500
2000
2500
Debris length measureed from HE sections
Debris length measured from FAF images
A B
Figure 3. The length of loss of debris zone measured from the optic nerve (A). The length measured from optic
nerve at each quadrant showed that the debris loss is more extensive in the inferior quadrant at P300. However, it is
only in this age group. The subretinal space without debris measured from center of optic nerve of HE sections and
length of hypoautofluorescence measured from FAF image showed a significant linear correlation (B).
Figure 4. A FAF imaging of an RCS rat that received subretinal injection of saline 2
weeks before. The FAF showed a hypoautofluorescent patch.
Evaluation of the Role of Microglia in the Preservation of FAF in RCS rats
The CD68 expression in the RCS retina showed presence of active microglia in the
outer plexiform and outer nuclear layer as early as P25 (Figure 5). At a later stage (P46) when
the inner and outer segment boundaries and the outer segment stack-disks configuration are
totally lost, active microglia became predominantly expressed in the area of the outer segment
125
(Figure 5). This suggests that the microglia is also participating in the clearance of debris
resulted from the non-phagocytized photoreceptor outer segment. On the other hand, the GFAP
expression was observed only in few animals at P60 and P90 and the level of expression was
very weak suggesting a minor role for astrocyte and Muller cells in the clearance of outer retina
debris layer in RCS rats during the earlier stages of the disease (Figure 5). No significant
correlation between microglia expression and the hyperautofluorescent spots present in fundus
during advanced stages of degeneration (after P90) could be observed. Such
hyperautofluorescent spots were found to be originating from the apical surface of the RPE but
not within the RPE cells.
Figure 5. Immunohistochemistry of RCS rat retina at different age using anti-CD68 antibody (active microglia,
green) and GFAP (astrocyte and active Muller cell, red) showing the CD68 is mostly expressed in the outer
plexiform layer and outer nuclear layer around P25 and becomes actively expressed in the photoreceptor outer
segment at P46 where the outer segment losses the typical stacked-disks configuration significantly and the inner
segment is no longer present. GFAP expression was observed in P60 RCS rats. However, the expression was not as
prominent as CD68.
126
Spectral Analysis
The spectral image of the RCS retina showed that the debris layer is accentuated
with less background autofluorescence at emission wavelength of 570 nm (Figure 6B). The
spectral analysis of hyperautofluorescent spots in debris layer at different age showed that 40%
of the spot have peak emission at 538 nm at P60 and P90. However, at P120 there is a right shift
of peak emission towards a higher wavelength. The peak is also more scattered towards different
wavelength. This indicates that the hyperautofluorescent debris layer in RCS retina maybe
undergoing a change composition around P120 causing the emission spectrum to be more
diverse among different spots analyzed.
Figure 6. Spectral image of autofluorescence of a P60 RCS rat retina at emission wavelength of 538 nm (A) and 570
nm (B) showing that at 570 nm there is less background autofluorescence. Peak emission spectrum of
hyperautofluorescent spots were analyzed in P60, P90, and P120 RCS rats (C). The results showed 40% of the spots
analyzed in P60 and P90 RCS rats have peak emission at 538 nm. However, in the P120 rats, the peak emission
becomes more variable with more hyperautofluorescent spots having a peak emission at longer wavelength.
DISCUSSION
Changes in the FAF is observed in many human RD disease conditions (Holz, Bellman et
al. 2001; Charbel Issa, Bolz et al. 2009; Schmitz-Valckenberg, Fleckenstein et al. 2009; Durrani
and Foster 2012; Riaz, Jampol et al. 2012; Almeida, Kaliki et al. 2013; Cuba and Gomez-Ulla
127
2013; Pepple, Cusick et al. 2013) and animal models of RD diseases (Hirata, Yasukawa et al.
2009; Luhmann, Robbie et al. 2009; Wang, Fine et al. 2009; Boretsky, Motamedi et al. 2011;
Charbel Issa, Singh et al. 2012; Secondi, Kong et al. 2012). In human AMD patients, attempts to
correlate progressive changes in the FAF pattern with degenerative status of the retina is limited
due to the non-available of the histology data at various time points. In RCS rat which is widely
used as a model to study the effect of various therapeutic interventions to cure blindness, no
major investigations have been conducted to assess the possible correlation between the FAF and
histology of the retina. Correlation between histological changes and FAF pattern can be
implemented as a suitable non-invasive technique to assess regional difference in the inner
retinal layers at various stages of the disease progression. Our study demonstrates that, with the
help of new imaging techniques, it is possible to monitor the changes in the FAF pattern in a
diseased retina.
We used RCS rat, an animal model for RD disease to better obtain an insight into the
changes in the FAF pattern and its correlation with the histological status of the retina. RCS rat is
widely used for studying the therapeutic benefits of various treatment strategies. In RCS rats, the
RPE fail to phagocytize the shed photoreceptor outer segments (Dowling and Sidman 1962), a
condition observed in some of the human AMD patients (Charbel Issa, Bolz et al. 2009).
Although the presence of the debris in RCS retina may not be crucial in triggering cell damage,
its presence might be a crucial factor that determines the success of treatment strategies for
treating RD diseases. The debris accumulates in the subretinal space and blocks the diffusion of
materials from the choroid to the photoreceptors and hence induces photoreceptor death (Valter,
Maslim et al. 1998; Smith, Chan et al. 2006)( We observed that progressive changes take place
in the FAF pattern in RCS rats starting from P32 until the advanced stages of degeneration
128
(P415). In this study, the source of FAF in RCS retina and its correlation with debris
accumulation and the progressive changes in the retinal degeneration is demonstrated. The FAF
showed patches of decrease AF that enlarge progressively with age. Such hypoautofluorescence
initially appears in the central area (close to ON head) which is consistent with the debris loss.
Human patients at late stage of non-neovascular age-related macular degeneration (NNVAMD)
with geographic atrophy (GA), hypoautofluorescence in the FAF is considered area of RPE death
(Bearelly, Cousins, 2010). On the other hand, in several other RD models, the accumulation of
RPE lipofuscin is most marked in central retina, the area having the greatest concentration of
visual chromophore and thus the most pronounced capacity for photon catch. (Faulkner and
Kemp 1984; Liem, Keunen et al. 1996; Tornow and Stilling 1998; Sparrow, Fishkin et al. 2003).
In RCS rats, the loss of FAF appears as a patch that extends more peripherally in the inferior
quadrant at P300, however, the difference was only significant at this age group.
The decrease in AF generally progress eccentrically from optic nerve although the
extension of the area was not symmetric for each quadrant. The inferior quadrant of the fundus
had a more extensive loss of AF. The length of the hypofluorescent area measured from the optic
nerve (FAF image) showed positive linear relationship with the area without debris (estimated
based on H&E stained images, figure 3). This demonstrates that the decrease in the FAF
observed in RCS rats is due to the progressive loss of the debris layer. In order further establish
this, we made a retinal detachment using saline that inevitably causes focal disruption of the
debris. During FAF imaging, a hypoautofluorescent area appeared in the fundus corresponding to
the location of the bleb formation suggesting that the absence of debris can cause decreased AF
in the fundus images. Previously (Katz, Drea et al. 1986; Katz, Eldred et al. 1987)( it was
demonstrated that a lipofuscin-like autofluorescence was found to develop in the debris zone in
129
RCS rats, providing the most direct evidence available that components of the outer segments
can be directly converted into RPE lipofuscin fluorophores. The debris layer in RCS rats first
began to disappear from the central and inferior quadrants of the retina and gradually extends to
the periphery. The finding is corresponding to the previous finding by LaVail et al that the
superior quadrant that the debris accumulated in the superior quadrant is markedly greater than
other meridians. (LaVail and Battelle 1975) However, we didn’t find significant difference in the
thickness of outer nuclear layer as described by the same group. However, we did not find any
correlation between changes in the AF pattern and the loss of outer nuclear layer cell count. This
could be due to the accumulation of the subretinal debris in the RCS retina at a very early age
(by P32) before the complete loss of photoreceptors. Further, the loss of photoreceptors in the
RCS retina is not uniform across the various regions of the retina.
Increase in the loss of debris from the central area of the retina suggests structural
changes may be taking place in the debris and this may be initiated by light exposure (photon
catch). It may be speculated that, following structural changes, the RCS RPE may express their
ability to phagocytize these debris. This argument is supported by the previous in vitro studies
suggesting that the dystrophic RPE is capable of phagocytosis comparable to normal RPE when
suitable materials are used as beads (Seyfried-Williams, McLaughlin et al. 1984; Braun, Kage et
al. 1999). Such structural changes in the debris can play a crucial role in the degradation and
gradual disappearance of the subretinal debris in RCS rats at later stages of degeneration (when
macrophages are not active). The above possibility is supported by our spectral analysis data
suggesting progressive changes taking place in the chemical composition of the
hyperautofluorescent spots appearing in the FAF as isolated patches.
130
In RCS rats, presence of hyperautofluorescent spots inside the dark area and more frequently at
the margin of this area started to develop at the age of P90. Comparable phenomena has been
reported in human patients where the FAF image also shows an area of focal increase in
autofluorescence around part of the perimeter of GA or rim area focal hyperautofluorescence
suggested to be indicative of increased lipofuscin load in these RPE cells. (Bearelly and Cousins
2010).
In this study we used Aperio ScanScope for quantitative assessment of the retinal layer
thickness and cell counting. The retina layer measurements showed the loss of retina thickness is
mostly at the outer nuclear layer with 20% loss of thickness and cell count between from P18 to
P32. This is in agreement with the previous investigations (LaVail and Battelle 1975) in which
retinal thickness was measured manually. Our study demonstrates that the quantitative
assessments made using Aperio ScanScope is a reliable and more easy way of understanding the
progressive changes in the retina during diseased conditions. The other retina layers undergo a
more gradual decrease in thickness. The results are in agreement with the pathogenesis of the
disease in which a dysfunctional RPE is the cause of degeneration thus most changes are
observed in the outer retina.
The immunofluorescence imaging of the CD68 and GFAP showed that there is active
involvement of microglia and little involvement of GFAP in the POS. The CD68 is mostly
observed in the outer nuclear layer and outer plexiform layer at P18 and P25. However, active
microglia starts to express prominently at P46 when the POS starts to lose its typical stacked-
disk configuration. And the expression decreased when the debris is gone. We speculate that the
microglia plays a major role in the removal of toxic debris formed by non-digested POS. Ccl2-
knockout (Ccl2-/-) mice have been reported to develop drusen and phenotypic features similar to
131
AMD, The drusen-like lesions of Ccl2-/- mice comprised accelerated accumulation of swollen
CD68, F4/80 macrophages in the subretinal space that were apparent as autofluorescent foci on
AF-SLO. These macrophages contained pigment granules and phagosomes with outer segment
and lipofuscin inclusions that may account for their autofluorescence (Luhmann, Robbie et al.
2009).
We try to correlate the increased AF spots with histology to find the source of it. Other
studies using mouse model of enhanced S-one syndromes (NR2E3 mutation) (Wang, Lu et al.
2008) and mouse model of AMD (Ccl2-knowout) (Luhmann, Robbie et al. 2009) showed similar
increased AF spots. The source of AF was found to be within macrophage. However, in our
study, the increase AF spots were not located within the glial cells. The increased
hyperautofluorescent spots were located on the surface of the RPE which was also observed by
Katz et al (Katz, Drea et al. 1986) and this zone is rich in expanded RPE cell processes (Matthes
and LaVail 1989). This is reasonable because the genetic defect in RCS rat causes inability for
RPE to phagocytize, however, the binding of POS is regulated by αvβ5 integrin which is not
defective in RCS rat, thus it is likely that the undigested POS debris remain attached to surface of
RPE (Finnemann, Bonilha et al. 1997). Kim et al demonstrated that photooxidation of
bisretinoids will enhances their fluorescence intensity (Kim, Jang et al. 2010), although the main
component of the POS debris, A2-PE was not analysed in the study, it is possible that it
undergoes photooxidation with time and results in the increased fluorescence intensity as the
RCS rats age.
The increased AF spots can be observed up to P415 (Figure 1) in the retina even when
the background AF of the debris are gone. The microglia (Figure 4) expression also decreased at
this stage. This indicates that these spots may not be phagocytized by microglia. The possible
132
explanation maybe the composition of these particles changed over time and became inert to
immune system. The other explanation is that these particles are wrapped by RPE processes thus
lost the chemotactic activity (Matthes and LaVail 1989). According to LaVail et al 1972, in the
RCS retina, many disorganized outer segment saccules are observed in continuity with longer
membranous lamellae and large lamellar whorls of the RPE. These extra lamellar materials are
considered to be derived from the rod photoreceptor debris and pigment epithelial cells (LaVail,
Sidman et al. 1972).
Spectral analysis of the increased AF spots showed that the peak emission spectrum of
more spots is 538 nm (Figure 5). However, the peak emission shifts to a higher wavelength for
P120 RCS rat. Sparrow et al analyzed fluorescence emission of A2PE and found that it has a
lower peak emission wavelength compared to A2E (Sparrow, Yoon et al. 2010). This suggests
that, in the RCS retina, there is a change in the composition of the POS debris over time.
Our study evaluated the RCS rat which is a model of RPE dysfunction frequently used in
transplantation studies. Using the Aperio Scanscope, we were able to count the numbers of
nuclei in outer nuclear layer and provide a more objective evaluation of any treatment. The FAF
imaging correlated with the progress of retina degeneration; however, the focal AF did not
correlate with a focal change in degeneration. The observation that the timing of loss the visible
boundary between photoreceptor inner segment and outer segment and the normal stacking
configuration of the POS coincide with the increased number of reactive microglia in the debris
zone indicates that microglia might take a major role in removal of the debris when the RPE is
dysfunctional.
133
The fact that increased autofluorescent spots are located in the subretinal space in the
RPE dysfunction RCS rat imply that the increased autofluorescence seen in patients with AMD
may not originated from RPE.
134
CHAPTER 6 REFERENCES
Almeida, A., S. Kaliki, et al. (2013). "Autofluorescence of intraocular tumours." Curr Opin
Ophthalmol.
Bearelly, S., S.W. Cousins, R.E. (2010). Fundus autofluorescence imaging in age-related
macular degeneration and geographic atrophy. In Anderson, R. E., M. M. LaVail, et al
(Eds.), retinal degenerative diseases : laboratory and therapeutic investigations( pp.395-
409). New York, Springer.
Ben-Shabat, S., Y. Itagaki, et al. (2002). "Formation of a nonaoxirane from A2E, a lipofuscin
fluorophore related to macular degeneration, and evidence of singlet oxygen
involvement." Angew Chem Int Ed Engl 41(5): 814-817.
Boretsky, A., M. Motamedi, et al. (2011). "Quantitative evaluation of retinal response to laser
photocoagulation using dual-wavelength fundus autofluorescence imaging in a small
animal model." Invest Ophthalmol Vis Sci 52(9): 6300-6307.
Braun, M., A. Kage, et al. (1999). "Retinal pigment epithelial cells from Royal College of
Surgeons dystrophic rats can take up melanin granules." Graefes Arch Clin Exp
Ophthalmol 237(1): 67-71.
Charbel Issa, P., H. J. Bolz, et al. (2009). "Characterisation of severe rod-cone dystrophy in a
consanguineous family with a splice site mutation in the MERTK gene." Br J Ophthalmol
93(7): 920-925.
Charbel Issa, P., M. S. Singh, et al. (2012). "Optimization of in vivo confocal autofluorescence
imaging of the ocular fundus in mice and its application to models of human retinal
degeneration." Invest Ophthalmol Vis Sci 53(2): 1066-1075.
135
Cuba, J. and F. Gomez-Ulla (2013). "Fundus autofluorescence: applications and perspectives."
Arch Soc Esp Oftalmol 88(2): 50-55.
D'Cruz, P. M., D. Yasumura, et al. (2000). "Mutation of the receptor tyrosine kinase gene Mertk
in the retinal dystrophic RCS rat." Hum Mol Genet 9(4): 645-651.
Delori, F. C., C. K. Dorey, et al. (1995). "In vivo fluorescence of the ocular fundus exhibits
retinal pigment epithelium lipofuscin characteristics." Invest Ophthalmol Vis Sci 36(3):
718-729.
Dowling, J. E. and R. L. Sidman (1962). "Inherited retinal dystrophy in the rat." J Cell Biol 14:
73-109.
Durrani, K. and C. S. Foster (2012). "Fundus autofluorescence imaging in posterior uveitis."
Semin Ophthalmol 27(5-6): 228-235.
Faulkner, D. J. and C. M. Kemp (1984). "Human rhodopsin measurement using a T.V.-based
imaging fundus reflectometer." Vision Res 24(3): 221-231.
Finnemann, S. C., V. L. Bonilha, et al. (1997). "Phagocytosis of rod outer segments by retinal
pigment epithelial cells requires αvβ5 integrin for binding but not for internalization."
Proc Natl Acad Sci U S A 94(24): 12932-12937.
Hirata, M., T. Yasukawa, et al. (2009). "Fundus autofluorescence and fate of glycoxidized
particles injected into subretinal space in rabbit age-related macular degeneration model."
Graefes Arch Clin Exp Ophthalmol 247(7): 929-937.
Holz, F. G., C. Bellman, et al. (2001). "Fundus autofluorescence and development of geographic
atrophy in age-related macular degeneration." Invest Ophthalmol Vis Sci 42(5): 1051-
1056.
136
Hwang, J. C., J. W. Chan, et al. (2006). "Predictive value of fundus autofluorescence for
development of geographic atrophy in age-related macular degeneration." Invest
Ophthalmol Vis Sci 47(6): 2655-2661.
Katz, M. L., C. M. Drea, et al. (1986). "Influence of early photoreceptor degeneration on
lipofuscin in the retinal pigment epithelium." Exp Eye Res 43(4): 561-573.
Katz, M. L., G. E. Eldred, et al. (1987). "Lipofuscin autofluorescence: evidence for vitamin A
involvement in the retina." Mech Ageing Dev 39(1): 81-90.
Kim, S. R., Y. P. Jang, et al. (2010). "Photooxidation of RPE lipofuscin bisretinoids enhances
fluorescence intensity." Vision Res 50(7): 729-736.
LaVail, M. M. and B. A. Battelle (1975). "Influence of eye pigmentation and light deprivation on
inherited retinal dystrophy in the rat." Exp Eye Res 21(2): 167-192.
LaVail, M. M., R. L. Sidman, et al. (1972). "Photoreceptor-pigment epithelial cell relationships
in rats with inherited retinal degeneration. Radioautographic and electron microscope
evidence for a dual source of extra lamellar material." J Cell Biol 53(1): 185-209.
Liem, A. T., J. E. Keunen, et al. (1996). "Clinical applications of fundus reflection
densitometry." Surv Ophthalmol 41(1): 37-50.
Liu, J., Y. Itagaki, et al. (2000). "The biosynthesis of A2E, a fluorophore of aging retina,
involves the formation of the precursor, A2-PE, in the photoreceptor outer segment
membrane." J Biol Chem 275(38): 29354-29360.
Luhmann, U. F., S. Robbie, et al. (2009). "The drusenlike phenotype in aging Ccl2-knockout
mice is caused by an accelerated accumulation of swollen autofluorescent subretinal
macrophages." Invest Ophthalmol Vis Sci 50(12): 5934-5943.
137
Mata, N. L., R. T. Tzekov, et al. (2001). "Delayed dark-adaptation and lipofuscin accumulation
in abcr+/- mice: implications for involvement of ABCR in age-related macular
degeneration." Invest Ophthalmol Vis Sci 42(8): 1685-1690.
Matthes, M. T. and M. M. LaVail (1989). "Inherited retinal dystrophy in the RCS rat:
composition of the outer segment debris zone." Prog Clin Biol Res 314: 315-330.
Nandakumar, N., S. Buzney, et al. (2012). "Lipofuscin and the principles of fundus
autofluorescence: a review." Semin Ophthalmol 27(5-6): 197-201.
Nandrot, E., E. M. Dufour, et al. (2000). "Homozygous deletion in the coding sequence of the c-
mer gene in RCS rats unravels general mechanisms of physiological cell adhesion and
apoptosis." Neurobiol Dis 7(6 Pt B): 586-599.
Pepple, K. L., M. Cusick, et al. (2013). "SD-OCT and autofluorescence characteristics of
autoimmune retinopathy." Br J Ophthalmol 97(2): 139-144.
Riaz, K. M., L. M. Jampol, et al. (2012). "Fundus autofluorescence imaging in punctate inner
choroidopathy with blind spot enlargement." Ocul Immunol Inflamm 20(6): 460-463.
Schmitz-Valckenberg, S., M. Fleckenstein, et al. (2009). "Fundus autofluorescence and
progression of age-related macular degeneration." Surv Ophthalmol 54(1): 96-117.
Secondi, R., J. Kong, et al. (2012). "Fundus autofluorescence findings in a mouse model of
retinal detachment." Invest Ophthalmol Vis Sci 53(9): 5190-5197.
Seyfried-Williams, R., B. J. McLaughlin, et al. (1984). "Phagocytosis of lectin-coated beads by
dystrophic and normal retinal pigment epithelium." Exp Cell Res 154(2): 500-509.
Shaban, H. and C. Richter (2002). "A2E and blue light in the retina: the paradigm of age-related
macular degeneration." Biol Chem 383(3-4): 537-545.
138
Smith, R. T., J. K. Chan, et al. (2006). "Autofluorescence characteristics of early, atrophic, and
high-risk fellow eyes in age-related macular degeneration." Invest Ophthalmol Vis Sci
47(12): 5495-5504.
Spaide, R. (2008). "Autofluorescence from the outer retina and subretinal space: hypothesis and
review." Retina 28(1): 5-35.
Sparrow, J. R. and M. Boulton (2005). "RPE lipofuscin and its role in retinal pathobiology." Exp
Eye Res 80(5): 595-606.
Sparrow, J. R., N. Fishkin, et al. (2003). "A2E, a byproduct of the visual cycle." Vision Res
43(28): 2983-2990.
Sparrow, J. R., K. D. Yoon, et al. (2010). "Interpretations of fundus autofluorescence from
studies of the bisretinoids of the retina." Invest Ophthalmol Vis Sci 51(9): 4351-4357.
Sparrow, J. R., J. Zhou, et al. (2002). "Involvement of oxidative mechanisms in blue-light-
induced damage to A2E-laden RPE." Invest Ophthalmol Vis Sci 43(4): 1222-1227.
Tornow, R. P. and R. Stilling (1998). "Variation in sensitivity, absorption and density of the
central rod distribution with eccentricity." Acta Anat (Basel) 162(2-3): 163-168.
Valter, K., J. Maslim, et al. (1998). "Photoreceptor dystrophy in the RCS rat: roles of oxygen,
debris, and bFGF." Invest Ophthalmol Vis Sci 39(12): 2427-2442.
Wang, N. K., H. F. Fine, et al. (2009). "Cellular origin of fundus autofluorescence in patients and
mice with a defective NR2E3 gene." Br J Ophthalmol 93(9): 1234-1240.
Wang, S., B. Lu, et al. (2008). "Morphological and functional rescue in RCS rats after RPE cell
line transplantation at a later stage of degeneration." Invest Ophthalmol Vis Sci 49(1):
416-421.
139
Wu, Y., E. Yanase, et al. (2010). "Structural characterization of bisretinoid A2E photocleavage
products and implications for age-related macular degeneration." Proc Natl Acad Sci U S
A 107(16): 7275-7280.
Yin, D. (1996). "Biochemical basis of lipofuscin, ceroid, and age pigment-like fluorophores."
Free radical biology & medicine 21(6): 871-888.
140
Chapter 7
CONCLUSION
The studies in this dissertation aim to understand the feasibility of using polarized retinal
pigment epithelium (RPE) monolayer for transplantation experiments in order to develop an
effective treatment strategy for curing blinding eye diseases such as age-related macular
degeneration (AMD). Although the pathogenesis of AMD is not totally understood, studies
demonstrated that degeneration of the RPE plays a major role (Ambati, Ambati et al. 2003;
Nowak 2006; Kinnunen, Petrovski et al. 2012) in the disease. Therefore, RPE transplantation is
considered as a potential treatment. Furthermore, embryonic stem cells (ESC) have the potential
to develop into any cell type and provide unlimited resource of cells. Thus ESC derived cells are
considered to be a feasible cell source for transplantation studies. While Idelson et al (Idelson,
Alper et al. 2009) reported that subretinal injection of hESC-RPE suspension did not result in
integration between the host and transplanted cells; Schwart et al (Schwartz, Hubschman et al.
2012) demonstrated that hESC-RPE suspension can develop into a monolayer in the subretinal
space. Despite of the differences of results between these groups, both of the studies agreed that
a monolayer formation in the subretinal space is critical for the success of the therapy.
Since formation of the monolayer is critical for the survival and function of RPE, we
hypothesize that delivering the cells as a monolayer will have better outcomes than injecting the
cells as a suspension. Our approach was to deliver a polarized monolayer of hESC-RPE with
support from a parylene membrane. The results of our study demonstrated that the RPE
monolayer implantation group had a significant higher visual sensitivity than the subretinal
injection group and the non-treated group. Histological evaluation also revealed that the
transplanted RPE monolayer could remain as a monolayer in the subretinal space and maintain
141
the RPE (RPE65) and human (TRA-1-85) characteristics while the cell survival was rare in the
RPE suspension treated group. The monolayer RPE cells also showed the presence of rhodopsin
positive phagosomes, indicating that they are capable of phagocytosis, which is one of the most
critical functions of RPE. The monolayer of RPE cells has melanin pigment distributed more at
the apical surface of the RPE which is also a feature of physiological RPE.
Our results support the theory that the RPE implanted as a monolayer has a higher
survival rate and better functional rescue than the non-polarized RPE suspension. For future
studies, there are particularly two important aspects need to be addressed, as the project moves
towards the next phase of human clinical trial. The first aspect involves in the host versus graft
interaction, including the integration of host photoreceptors into the grafted cells and the
immunologic responses of the host. The second aspect involves in execution of the
transplantation procedure that includes the implant design and surgical procedures.
With regard to the immunologic response of the host to date, there has been no consensus
about immunological response that occurs in the subretinal space towards transplanted RPE.
Although the RPE is known to be immune privileged tissue (Wenkel and Streilein 2000; Zamiri,
Masli et al. 2006) and the results of our studies showed that the polarized monolayers
demonstrate less innate immune response than non-polarized cells, it has been shown in rabbit
models that the RPE suspension fails to survive long term even with triple immunosuppressants
(Crafoord, Algvere et al. 2000; Del Priore, Ishida et al. 2003). For the RCS rat, most of the RPE
transplantation studies chose to adapt the immunosuppressant regimen from RPE cell suspension
transplantation study reported by Coffey et al (Coffey, Girman et al. 2002). However, other
studies had shown survival of transplanted RPE cells in RCS rats without any
immunosuppressant (Zhu, Carido et al. 2013). And, the RPE is known to be an immune-
142
privileged tissue that may be partly contributed by the secretion of anti-inflammatory factors
such as PEDF (Zamiri, Masli et al. 2006). Moreover, the polarized RPE behave more similar to
innate RPE and secret more PEDF than the non-polarized RPE (Zhu, Deng et al. 2011).
Therefore, we hypothesize that other RPE properties that contribute to the immune-privilege of
RPE may also express more when the cells are polarized. The results of our study successfully
demonstrated that the polarized RPE cells are less likely to be targeted by host microglia,
implicating that polarized RPE cells possess more immune-privileged properties. In summary,
studying the host immunological reactions towards the transplanted RPE will allow better
understanding of the immune response triggered by such transplants. This will also allow
administration of a more precise immunosuppressant regimen into patients as the study moves
towards the next phase human clinical trial while avoiding unnecessary toxicity from
immunosuppressive agents.
The other important research area for transplantation studies is the combination therapy
with growth factors or intracellular matrix proteins that may be capable of facilitating the host
versus graft integration. Seiler et al reported using brain-derived neurotrophic factor or glial-
derived neurotrophic factor as a coating for fatal retinal sheet transplantation in retinal
degenerative rats (Yang, Seiler et al. 2010). They demonstrated better preservation of visual
sensitivity in the superior colliculus when compared to non-coated retinal transplants. It would
not be surprising if these growth factors can also facilitate RPE and host photoreceptors
integration; because the recipient of transplant would be in a stage of where retinal degeneration
has already taken place. It is reasonable to infer that the sooner the transplanted RPE cells can
integrate with the host retina, the better support for photoreceptor by the RPE cell would be
143
observed. Also there will be less damage caused by surgery or the post-surgical recovery phase if
the retina can restore its physiological contact with RPE early after the transplantation.
There are other molecules that have the potential to promote the integration between
transplanted RPE and the host retina is the components of interphotoreceptor matrix (IPM). It
has been reported that IPM is a viscous material, which lies between the photoreceptor and the
RPE, composed largely of proteins, glycoproteins, and proteoglycans (Adler and Klucznik 1982).
It was also suggested that these material create bondings between the RPE and photoreceptors
via specific receptors that bind IPM components to the cell membranes (Marmor, Yao et al. 1994;
Hageman, Marmor et al. 1995). Although the function and mechanism of this bonding
contributed by IPM is not completely understood, it has been shown that enzymes that degrade
IPM components, such as chondroitinase can reduce retinal adhesiveness (Lazarus and Hageman
1992). This also indicates that the presence of IPM is critical for normal photoreceptors and RPE
attachment. For an RPE transplant, the presence of IPM components may facilitate retina
adhesion to the transplanted RPE cells and photoreceptors, thus facilitating the integration of
graft and the host.
The studies in this dissertation were performed on rodents which are good for
quantitative analysis for studies of efficacy. However, it is not an ideal model to mimic human
visual system. The major difference is that the size of the eye of rodent is much smaller than that
of human. For transplantation surgery in larger animals, an internal approach clinically known as
vitrectomy is used unlike rodent surgeries where an external approach was performed. Therefore,
a custom-made surgical tool that can deliver the implant without damaging the cells while it
allows smaller incision of the eye by folding the implant is required. It is necessary to develop a
specific surgical model of larger animals for pre-clinical testing, as well as reliable functional
144
and imaging evaluations which will serve as a guideline for follow up in the clinical settings. In
addition to the development of surgical instrument, it is also necessary to modify the curvature or
the rigidity of the implant so that the implant can be more flexible yet manageable during surgery
allowing which to be lay flat in the subretinal space. This is also a critical element as the project
moves toward human clinical trials.
Future researches will answer the questions that incur when the therapy goes to clinical
application. The research will be continued by the surgical and research team at Doheny Eye
Institute.
145
CHAPTER 7 REFERENCES
Adler, A. J. and K. M. Klucznik (1982). "Proteins and glycoproteins of the bovine
interphotoreceptor matrix: composition and fractionation." Exp Eye Res 34(3): 423-434.
Ambati, J., B. K. Ambati, et al. (2003). "Age-related macular degeneration: Etiology,
pathogenesis, and therapeutic strategies." Survey of Ophthalmology 48(3): 257-293.
Coffey, P. J., S. Girman, et al. (2002). "Long-term preservation of cortically dependent visual
function in RCS rats by transplantation." Nat Neurosci 5(1): 53-56.
Crafoord, S., P. V. Algvere, et al. (2000). "Cyclosporine treatment of RPE allografts in the rabbit
subretinal space." Acta Ophthalmol Scand 78(2): 122-129.
Del Priore, L. V., O. Ishida, et al. (2003). "Triple immune suppression increases short-term
survival of porcine fetal retinal pigment epithelium xenografts." Invest Ophthalmol Vis Sci 44(9):
4044-4053.
Hageman, G. S., M. F. Marmor, et al. (1995). "The interphotoreceptor matrix mediates primate
retinal adhesion." Arch Ophthalmol 113(5): 655-660.
Idelson, M., R. Alper, et al. (2009). "Directed differentiation of human embryonic stem cells into
functional retinal pigment epithelium cells." Cell Stem Cell 5(4): 396-408.
Kinnunen, K., G. Petrovski, et al. (2012). "Molecular mechanisms of retinal pigment epithelium
damage and development of age-related macular degeneration." Acta Ophthalmol 90(4): 299-309.
Lazarus, H. S. and G. S. Hageman (1992). "Xyloside-induced disruption of interphotoreceptor
matrix proteoglycans results in retinal detachment." Invest Ophthalmol Vis Sci 33(2): 364-376.
Marmor, M. F., X. Y. Yao, et al. (1994). "Retinal adhesiveness in surgically enucleated human
eyes." Retina 14(2): 181-186.
146
Nowak, J. Z. (2006). "Age-related macular degeneration (AMD): pathogenesis and therapy."
Pharmacol Rep 58(3): 353-363.
Petrovski, G., E. Berenyi, et al. (2011). "Clearance of dying ARPE-19 cells by professional and
nonprofessional phagocytes in vitro- implications for age-related macular degeneration (AMD)."
Acta Ophthalmol 89(1): e30-34.
Schwartz, S. D., J. P. Hubschman, et al. (2012). "Embryonic stem cell trials for macular
degeneration: a preliminary report." Lancet 379(9817): 713-720.
Yang, P. B., M. J. Seiler, et al. (2010). "Trophic factors GDNF and BDNF improve function of
retinal sheet transplants." Exp Eye Res 91(5): 727-738.
Wenkel, H. and J. W. Streilein (2000). "Evidence that retinal pigment epithelium functions as an
immune-privileged tissue." Investigative Ophthalmology & Visual Science 41(11): 3467-3473.
Zamiri, P., S. Masli, et al. (2006). "Pigment epithelial growth factor suppresses inflammation by
modulating macrophage activation." Investigative Ophthalmology & Visual Science 47(9):
3912-3918.
Zhu, D., X. Deng, et al. (2011). "Polarized secretion of PEDF from human embryonic stem cell-
derivRPE promotes retinal progenitor cell survival." Invest Ophthalmol Vis Sci 52(3): 1573-158
Zhu, Y., M. Carido, et al. (2013). "Three-dimensional neuroepithelial culture from human
embryonic stem cells and its use for quantitative conversion to retinal pigment epithelium."
PLoS One 8(1): e54552.
Abstract (if available)
Abstract
Age-related Macular Degeneration (AMD) is the leading cause of blindness among the elderly in United States. It is categorized into two types: neovascular (wet) AMD, caused by abnormal blood vessel growth in the choriocapillaris, and atrophic (dry) AMD, which is characterized by the extensive loss of retinal pigment epithelium, followed by the loss of photoreceptors. While several treatments are available for wet AMD, there is no effective way to address dry AMD. Although tablet formulation of select vitamins and minerals has been shown to slow the progression of dry AMD in some patients, there are currently no medical or surgical treatments. An effective therapy for dry AMD is needed. ❧ The purpose of this study is to evaluate the feasibility of using a cell based therapy to replace the dysfunctional retinal pigment epithelium (RPE), which is one of the contributing factors of AMD. The scientific evidence supporting such treatment comes from clinical studies using a surgical approach called macular translocation, in which the area with the diseased RPE is moved to an area with healthy RPE support. However, the surgery is complicated and carries a large risk of ocular complications, so it is rarely performed nowadays. The experiments outlined in this dissertation examined the feasibility of using an alternative approach by in vitro expansion of an RPE cell line derived from human embryonic stem cells. ❧ The specific aim of the study was to examine the functional outcome of transplantation of RPE cells in the dystrophic RCS rat by performing both in vivo functional examination and histological evaluation. The study also compared the outcome difference of cells delivered in different formats, either as a monolayer or a cell suspension. We evaluated the correlation of several non-invasive examinations and functional outcomes to see if these non-invasive techniques have the potential for predicting the treatment effects once the study moves to the human clinical trial. ❧ Our results showed that RPE transplantation as a monolayer had better visual function preservation compared to cell suspension injection and sham therapy. Histologic evaluation showed better preservation of retinal structure in animals transplanted with RPE monolayer. In vivo imaging showed correlation with the implant that may be translatable for future clinical evaluation of human clinical trials. Future studies will continue to develop the surgical technique in larger mammals for future application in human patients.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The role of bone morphogenetic protien 4 in the pathogenesis of choroidal neovascularization in age-related macular degeneration
PDF
Characterization of human embryonic stem cell derived retinal pigment epithelial cells for age-related macular degeneration
PDF
RPE secretome for the treatment of retinal degeneration in the RCS rat
PDF
Cortical and subcortical responses to electrical stimulation of rat retina
PDF
Towards a high resolution retinal implant
PDF
Manipulation of RGCs response using different stimulation strategies for retinal prosthesis
PDF
Cryopreserved umbilical cord mesenchymal stem cells therapy for the treatment of knee osteoarthritis: in-vitro evaluation and phase I clinical trial protocol
PDF
Cell and gene therapy in the murine model of adenosine deaminase deficiency
PDF
Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells
PDF
The calcium-sensing receptor in the specification of normal and malignant hematopoietic cell localization in the bone marrow microenvironment
PDF
Transcriptomic maturation of developing human cone precursors in fetal and 3D hESC-derived tissues
PDF
Effects of a GSK-3 inhibitor on retroviral-mediated gene transfer to human CD34+ hematopoietic progenitor cells
PDF
Targeting glioma cancer stem cells for the treatment of glioblastoma multiforme
PDF
Mechanisms that dictate beta cells’ response to stress in the context of genetic mutation, pregnancy, and infection
PDF
An iPSC-based biomarker strategy to identify neuroregenerative responders to allopregnanolone
PDF
The role of vitamin D in pediatric hematopoietic stem cell transplantation
PDF
Characterization of new stem/progenitor cells in skin appendages
PDF
Clinical outcomes of allogeneic hematopoietic stem cell transplant in acute lymphoblastic leukemia patients: a quality improvement project and systematic review meta-analysis
PDF
IPS and CNS cell models of peroxisomal disorders
PDF
The role of endoplasmic reticulum chaperones in regulating hematopoietic stem cells and hematological malignancies
Asset Metadata
Creator
Liu, Laura
(author)
Core Title
Morphological and functional evaluation of hESC-RPE cell transplantation in RCS rats
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Cell and Neurobiology
Publication Date
07/15/2015
Defense Date
04/02/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
human embryonic stem cells,OAI-PMH Harvest,Retinal degeneration,retinal pigment epithelium,Royal College of Surgeon rats
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Humayun, Mark S. (
committee chair
), McNeill, Thomas H. (
committee chair
), Adams, Gregor B. (
committee member
), Hinton, David R. (
committee member
), Schauwecker, P. Elyse (
committee member
), Thomas, Biju B. (
committee member
)
Creator Email
laurajl@gmail.com,lauraliu@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-290592
Unique identifier
UC11287923
Identifier
etd-LiuLaura-1778.pdf (filename),usctheses-c3-290592 (legacy record id)
Legacy Identifier
etd-LiuLaura-1778.pdf
Dmrecord
290592
Document Type
Dissertation
Rights
Liu, Laura
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
human embryonic stem cells
retinal pigment epithelium
Royal College of Surgeon rats