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
/
Cellular mechanisms controlling the mosaic of surviving cones in retinitis pigmentosa retinas
(USC Thesis Other)
Cellular mechanisms controlling the mosaic of surviving cones in retinitis pigmentosa retinas
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
ii
CELLULAR MECHANISMS CONTROLLING
THE MOSAIC OF SURVIVNG CONES IN
RETINITIS PIGMENTOSA RETINAS
By
Yerina Ji
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
NEUROSCIENCE
Copyright 2013 Yerina Ji
TABLE OF CONTENTS
_______________________________________________________________________________________
ABSTRACT……………………………………………..………………………….…...i
ACKNOWLEDGEMENTS…………………………………………....………...…...ii
LIST OF FIGURES …………………………………………………….……….……iii
CHAPTER ONE: Rearrangement of the Cone Mosaic in the Retina of the Rat
Model of Retinitis Pigmentosa
1. ABSTRACT ……………………………………………………………….….1
2. INTRODUCTION………………………………………………………….….2
3. MATERIALS AND METHODS ………………………………………….….3
3.1 Animals ………………………………………………………………...3
3.2 Tissue Preparation ……………………………………………………..4
3.3 Immunohistochemistry ………………………………………………...4
3.4 Hematoxylin Staining ………………………………………………….6
3.5 TUNEL Staining ……………………………………………………….7
3.6 Quantification and Statistics …………………………………………...7
4. RESULTS ………………………………………………………………….…9
i
4
4.1 Remodeling of M-opsin immunoreactive cones in developing RP
retinas …………………………………………………………………..9
4.2 Reorganization of M-opsin and S-opsin-immunoreactive cones
in orderly array of rings ……………………………………………….11
4.3 Cone rings are not associated with retinal foldings …………….……..13
4.4 Clusters of cell death are observed inside the rings …………………...15
4.5 Rings first develop at around P15 and start to lose their form
from around P180 ……………………………………………………...15
4.6 Rings in RP retinas grow significantly with age in both sizes
and numbers …………………………………………………………...18
4.7 M-ospin and S-opsin-immunoreactive cones in RP retinas
do not die until P180 …………………………………………………..19
5. DISCUSSION ………………………………………………………………...23
5.1 Remodeling of cone morphology in RP retinas ……………………….23
5.2 Remodeling of the spatial distribution of cones into orderly
array of rings in RP retinas ……………………………………………24
5.3 Ring formation is triggered by rod deaths not by the mechanical
disruption of the ONL …………………………………………………25
5.4 The spatial and temporal distribution of rings in RP retina …………...26
5.5 No cone degeneration until rings start to lose their form ……………..27
ii
5
5.6 Possible mechanisms underlying reorganization of cones in
rings in RP retinas ……………………………………………………..28
6. REFERENCES ……………………………………………………………….31
7. FIGURE LEGENDS …………………………………………………………48
7.1 Figure 1. ……………………………………………………………….48
7.2 Figure 2………………………………………………………………...48
7.3 Figure 3 ………………………………………………………………..48
7.4 Figure 4 ………………………………………………………………..49
7.5 Figure 5 ………………………………………………………………..49
7.6 Figure 6 ………………………………………………………………..50
7.7 Figure 7 ………………………………………………………………..50
CHAPTER TWO: Spatiotemporal Pattern of Rod Degeneration in the Retina of
the Rat Model of Retinitis Pigmentosa
1. ABSTRACT ………………………………………………………………….52
2. INTRODUCTION ……………………………………………………………53
3. MATERIALS AND METHODS …………………………………………….54
3.1 Animals ………………………………………………………………...54
3.2 Tissue Preparation ……………………………………………………...55
3.3 Immunohistochemistry ………………………………………………...55
iii
6
3.4 Hematoxylin Staining ………………………………………………….57
3.5 TUNEL Staining ……………………………………………………….57
3.6 Quantification and Statistical Analysis of Rod Holes ………………....57
4. RESULTS ……………………………………………………………………58
4.1 Change in vertical sections of the outer layer (ONL) of
S334ter-line-3 rat retina ………………………………………………..58
4.2 Rhodopsin-immunoreactive rods in developing RP retinas …………...59
4.3 Rhodopsin-immunoreactive rods in developing RP retinas …………...59
4.4 Spatiotemporal pattern of rods in S334ter-line-3 rat retinas …………..60
4.5 Holes in RP retinas increase significantly in diameter over time ……...62
4.6 Spatial correlation between rods and dying rods ………………………63
4.7 Microglial cells fill the holes in the early stage of RP ………………...65
4.8 Remodeled Muller-cell processes fill the center of rod holes after
the disappearance of active microglial cells …………………………...68
4.9 Rod deaths trigger modification in the spatial-distribution
patterns of cones ……………………………………………………….70
5. DISCUSSION ………………………………………………………………..71
5.1 Spatiotemporal pattern of rod death …………………………………...71
5.2 A possible mechanism responsible for this rod death pattern ………....73
5.3 Activation of retinal glial cells in RP retina …………………………...74
iv
7
6. REFERENCES ………………………………………………………………77
7. FIGURE LEGENDS …………………………………………………………84
7.1 Figure 1. ………………………………………………………………..84
7.2 Figure 2. ………………………………………………………………..84
7.3 Figure 3 ………………………………………………………………...84
7.4 Figure 4 ………………………………………………………………...85
7.5 Figure 5 ………………………………………………………………...85
7.6 Figure 6 ………………………………………………………………...85
7.7 Figure 7 ………………………………………………………………...86
7.8 Figure 8 ………………………………………………………………...86
CHAPTER THREE: Role of Mü ller Cells in Cone-mosaic Rearrangement in the
Retina of the Rat Model of Retinitis Pigmentosa
1. ABSTRACT ………………………………………………………………….88
2. INTRODUCTION ……………………………………………………………89
3. MATERIALS AND METHODS …………………………………………….89
3.1 Animals ………………………………………………………………..89
3.2 Administration of Alpha-Aminoadipidic Acid (AAA) ………………..91
3.2 Tissue Preparation …………………………………………………….91
3.3 Immunohistochemistry ………………………………………………...92
v
8
3.4 Statistical analysis …………………………………………………….93
4. RESULTS …………………………………………………………………...94
4.1 Contribution of Mü ller cells to cone migration ……………………....94
4.2 Remodeling of Mü ller cell processes into holes of rods and cones ......94
4.3 Elimination of rings of cones by disrupting metabolism of
Mü ller cells …………………………………………………………...100
5. DISCUSSION ……………………………………………………………….101
6. REFERENCES ……………………………………………………………...104
7. FIGURE LEGENDS ………………………………………………………...115
7.1 Figure 1. ………………………………………………………………115
7.2 Figure 2. ……………………………………………………………....115
7.3 Figure 3 ……………………………………………………………….116
7.4 Figure 4 ……………………………………………………………….116
7.5 Figure 5 ……………………………………………………………….116
CHAPTER FOUR: Manipulation of cone mosaics in the Retina of the Rat Model
of Retinitis Pigmentosa
1. ABSTRACT ………………………………………………………………....117
2 INTRODUCTION …………………………………………………………...118
3 MATERIALS AND METHODS …………………………………………...120
vi
9
3.1 Animals ……………………………………………………………….120
3.2 Administration of Tissue Inhibitor of Metalloproteinase ……………120
3.3 Tissue Preparation …………………………………………………....121
3.4 Immunohistochemistry ……………………………………………….122
3.5 Statistical analysis …………………………………………………….123
4 RESULTS …………………………………………………………………..125
4.1 Disturbance of the M-opsin-immunoreactive cone mosaic
in normal rat retinas with TIMP-1 ……………………………………125
4.2 M-opsin-immunoreactive cone spacings in normal retinas with TIMP-1
loses regularity and becomes close to random ………………………..127
4.3 No glial activation in TIMp-1 treated retina …………………………..131
4.4 Mosaic of M-opsin-immunoreactive cone can be manipulated
with TIMP-1 ..…………………………………………………………132
4.5 Mosaic of Mü ller cell processes can be manipulated with TIMP-1 …..135
4.6 M-opsin-immunoreactive cone spacings in RP retina with TIMP-1
gains homogeneity, loses regularity, and become close to random …..137
5 DISCUSSION ……………………………………………………………….142
5.1 Saline does not change M-opsin-immunoreactive cone mosaic ……...142
5.2 M-opsin-immunoreactive cone mosaic can be manipulated
by TIMP-1…………………………………………………………….144
vii
10
5.3 Possible mechanisms underlying modulation of M-opsin-immunoreactive
cone mosaic with TIMP-1 …………………………………………….146
5.4 Possible mechanisms controlling the regularity of cone mosaics
in the retina ……………………………………………………………147
6 REFERENCES ………………………………………………………………148
7 FIGURE LEGENDS ………………………………………………………....156
7.1 Figure 1. ……………………………………………………………….156
7.2 Figure 2. ……………………………………………………………….156
7.3 Figure 3 ………………………………………………………………..157
7.4 Figure 4 ………………………………………………………………..157
7.5 Figure 5 ………………………………………………………………..157
7.6 Figure 6 ………………………………………………………………..158
7.7 Figure 7 ………………………………………………………………..159
7.8 Figure 8 ………………………………………………………………..159
SUMMARY ………………………………………………………………………161
SUPPLEMENTARY RESULTS……………………………………………….163
viii
11
LIST OF FIGURES
_____________________________________________________________________________________
CHAPTER ONE: Rearrangement of the Cone Mosaic in the Retina of the Rat
Model of Retinitis Pigmentosa
Figure 1: M-opsin-immunoreactive cones remodel in RP retinas ………...………....10
Figure 2: M-opsin and S-opsin-immunoreactivity cones remodel into
orderly array of rings………………………………………………………12
Figure 3: Cone rings are not associated with retinal foldings …………………...…..14
Figure 4: Clusters of cell death are observed inside a ring……………………….......15
Figure 5: The development and change of rings………………………………….......17
Figure 6: Growth of RP rings in size and quantity ..…………………………….......19
Figure 7: Prolonged survival of M-opsin and S-opsin-immunoreactive
cones in RP retinas ……………….………………………………………………...…22
CHAPTER TWO: Spatiotemporal Pattern of Rod Degeneration in the Retina of
the Rat Model of Retinitis Pigmentosa
ix
12
Figure 1: Changes in the outer nuclear layer (ONL) of S334ter-line-3 rat retina .....59
Figure 2: Rhodopsin-immunoreactive rods in developing RP retinas………………60
Figure 3: Spatiotemporal pattern of rods in developing S334ter-line-3 rat retinas…62
Figure 4: Holes in RP retinas increase significantly in diameter over time ….…......63
Figure 5: Spatial correlation between rods and dying rods ………………...…….....65
Figure 6: Microglial cells fill the holes of rods in the early stage of RP……...……..67
Figure 7: Active microglial cells and remodeled Mü ller-cell processes …………....69
Figure 8: Rod deaths trigger modification in the spatial-distribution
patterns of cones…………………………………………………………....70
CHAPTER THREE: Role of Mü ller Cells in Cone-mosaic Rearrangement in the
Retina of the Rat Model of Retinitis Pigmentosa
Figure 1: Contribution of Mü ller cells to cone migration ……………………..….…94
Figure 2: Remodeling of Mü ller-cell processes inside cone rings …………………..96
Figure 3: Positive GS and GFAP immunoreactivity inside each ring…………….….97
Figure 4: ZO-1 immunoreactivity around rings at the junction between
cones and Mü ller cells ……………..............................................................................99
x
13
Figure 5: Elimination of rings of cones by disrupting metabolism
of Mü ller cells…………………………………………………………….………...101
CHAPTER FOUR: Manipulation of cone mosaics in the Retina of the Rat Model
of Retinitis Pigmentosa
Figure 1: Disturbance of the M-opsin-immunoreactive cones mosaic
in normal rat retinas with TIMP-1.…............................................………126
Figure 2: ND analysis distributions of control and TIMP-1 normal retinas …….....129
Figure 3: Voronoi domain analysis distribution for control and TIMP-1
normal retinas …………………………………………………………...131
Figure 4: No glial activation in TIMP-1 treated retina ………………………….....132
Figure 5: Mosaic of M-opsin-immunoreactive cone can be manipulated
with TIMP-1……………………………………………………..……....134
Figure 6: Close association between cones and Mü ller cell processes
during mosaic change ……………………………………………………136
Figure 7: ND analysis distributions of control and TIMP-1 RP retinas….................139
Figure 8: Voronoi domain analysis distribution for control and TIMP-1
RP retinas …………………………………………………………..….....141
xi
14
SUPPLEMENTARY FIGURE ………………………………………………………163
xii
15
ACKNOWLEDGEMENTS
______________________________________________________________________________________
First, I wish to express my deepest regards to my family members for their endless
support, patience and understanding.
Equally, I would like to thank my advisors; Dr. Norberto Grzywacz and Dr. Eun-
Jin Lee. Throughout my years in USC, I have always felt very fortunate to have
been under their invaluable discussions, guidance and support. All my work was
possible due to their incomparable mentorship and care, which I cannot explain in
words alone. In particular, Dr. Grzywacz was always encouraging and cheerful,
and such professional character was the pillar of endurance during tough times. Dr.
Lee‘s genuine care was always felt and was a character to learn from.
I would also like to thank committee members for their knowledge and support.
My sincere thanks go to Dr. Cheryl Craft for her kind gift of valuable resources for
research. Her professional guidance yet mother-like warmth was deeply
appreciated. I equally thank Dr. David Hinton for his professional ideas and
discussion, which led to great motivation and guidance. My thanks extends to Dr.
Jeannie Chen for her guidance and feedback; to Dr. Biju Thomas for his kind gift
of resources for research; and Dr. Matthew M. LaVail for kind supply of animals
for research.
xiv
16
I would like to thank my lab colleagues. I specially thank WanQing Yu for her
help with analysis using statistical programs developed by Dr. Grzywacz and
herself. I thank Arvind Iyer for his ever friendly support and cheerful philosophy. I
thank JunKwan Lee for his generous help and supportive words. I thank Nadav
Izvan for being a provider of helpful ideas support. I thank Colleen Zhu for not
only being a helpful character to work with but also a friend to share with. My
thank extends to Farouk Bruce, Xiwu Cao, Aditi Ray, Joaquin Rapela, Soyun Kim,
YoungKyung Kim, Annette Eom, Divya Nair, Lynette Song, Albert Chang, Walid
Kayali and Steven Walston for being a great colleagues and friends. I finally thank
Denise Steiner for her administrative support; Consuelo Correa for providing
warmth and kindness to the lab; and Darrel Adams for help with the animal
management; and all my other life-long friends.
xv
1
Rearrangement of the Cone Mosaic in the Retina of the Rat Model
of Retinitis Pigmentosa
ABSTRACT
In retinitis pigmentosa (RP), the death of cones normally follows some time after the
degeneration of rods. Recently, surviving cones in RP have been studied and reported in detail.
These cones undergo extensive remodeling in their morphology. In this chapter, we report an
extension of the remodeling study to consider possible modifications of spatial-distribution
patterns. For this purpose, we used S334ter-line-3 transgenic rats, a transgenic model developed
to express a rhodopsin mutation causing RP. In this study, retinas were collected at post-natal (P)
days P5 – 30, 90, 180, and P600. We then immunostained the retinas to examine the morphology
and distribution of cones, and to quantify the total cone numbers. Our results indicate that cones
undergo extensive changes in their spatial distribution to give rise to a mosaic comprising an
orderly array of rings. These rings first begin to appear at P15 at random regions of the retina and
become ubiquitous throughout the entire tissue by P90. Such distribution pattern loses its clarity
by P180 and mostly disappears at P600, at which time the cones are almost all dead. In contrast,
the numbers of cones in RP and normal conditions do not show significant differences at stages
as late as P180. Therefore, rings do not form by cell death at their centers, but by cone migration.
We discuss its possible mechanisms, and suggest a role for hot spots of rod death and the
remodeling of Mü ller-cell process into zones of low density of photoreceptors.
Key words: Retinitis Pigmentosa; cone mosaics; reorganization; retina
2
INTRODUCTION
A wide variety of mutations that affect rods in Retinitis pigmentosa (RP) first lead to their
degeneration (Blanks et al., 1974; Farber and Lolley, 1974; Bowes et al., 1990; Rosenfeld et al.,
1992; Marc et al., 2003). Then, rod degeneration frequently results in the death of cones,
although the extent of their apoptosis varies across patients (Ripps, 2002; Hartong et al., 2006)
and animal models (Carter-Dawson et al., 1978; Garcia-Fernandez et al., 1995; Jimenez et al.,
1996; LaVail et al., 1998). Eventually, these cones undergo almost complete degeneration in
human and retinal degenerative animal models (Blanks et al., 1974; Berson, 1993; Chang et al.,
1993; Farber et al., 1994; Li et al., 1994; Milam et al., 1996; Milam et al., 1998). Thus,
understanding the cellular level of cones in degenerative animal models may influence
therapeutic efforts of cone repopulation, transplantation, and retinal prosthesis.
Recently, some studies have shown in detail remodeling of cones in retinal degenerate
animal models (Barhoum et al. 2008; Lin et al., 2009, Hombrebueno et al., 2010). In these
studies, shortening or loss of cone outer segments (COS), loss of their normal vertical alignment,
disorganization of axon terminals and outgrowth of new processes from the cell body and the
axon process were reported. These cones maintained much of their abnormal phenotype until
postnatal (P) P180 (Hombrebueno et al., 2010) of retinal degeneration. Furthermore, past studies
reported on the discrepancies in the quantity or the densities of surviving M- and S-cones in
different regions of the retinas of different retinal degenerate animal models; ventral versus
dorsal (LaVail and Battelle, 1975; Carter-Dawson et al., 1978; Garcia-Fernandez et al., 1995;
Jimenez et al., 1996; LaVail et al., 1997) and central versus peripheral regions (LaVail et al.,
1982; Lin et al., 2009, Hombrebueno et al., 2010). However, the pattern of mosaic in which S-
3
and M-cones are distributed in retinal degenerative models is not well studied. Hence, the present
work focuses on examining the distribution patterns of cones in developing retinas of the
S334ter-line-3 model. (This model shall be referred to as the RP model in the rest of this paper.)
Because the maintenance of prolonged survival of cones is important for the treatment of retinal
degeneration, a thorough understanding of mosaic formation by cones in the progression of the
disease will influence therapeutic efforts of cone repopulation, transplantation, and retinal
prosthesis.
MATERIALS AND METHODS
Animals
The third line of albino Sprague-Dawley rats homozygous for the truncated murine opsin gene
(stop codon at residue 334; S334ter-line-3) was obtained from M.M. LaVail (University of
California, San Francisco, CA). Homozygous S334ter-line-3 breeding pairs were mated with
normal Copenhagen rats to produce offspring heterozygous for the S334ter transgene that was
subsequently used in this study. Heterozygous animals were used instead of homozygous in
order to avoid any changes in the retina due to albinism (O'Steen and Anderson 1972; O'Steen,
Anderson et al. 1974; Baker, Dovey et al. 2005). A line of homozygous RP rats was also kept in
breed for comparison study. For control, age-matched Sprague Dawley rats (Harlan, Indianapolis,
IN) were used. All rats were housed under cyclic 12/12-hour light/dark conditions with free
access to food and water. Both sexes of normal (control) and RP rats were used. All procedures
4
were in conformance with the Guide for Care and Use of Laboratory Animals (National
Institutes of Health). The University of Southern California Institutional Animal Care and Use
Committee reviewed and approved all procedures.
Tissue Preparation
The animals at P5 – 30, 90, 180, and P600 were used (n = 15 for each stage). All animals were
dark-adapted for at least 1 hour prior to sacrifice in the dark. Animals were deeply anesthetized
by intra-peritoneal injection of pentobarbital (40 mg/kg body weight) and the eyes were
enucleated. Animals were then killed with an overdose of pentobarbital. The anterior segment
and crystalline lens were removed and the eyecups were fixed in 4% paraformaldehyde in 0.1 M
phosphate buffer (PB), pH 7.4, for 30 minutes to 1 hr at 4 C. Following fixation, the retinas
were carefully isolated from the eyecups and were transferred to 30% sucrose in PB for 24 h at 4
C. For storage, all retinas (for cryostat sections and whole mounts) were then frozen in liquid
nitrogen, and stored at -70º C, thawed, and rinsed in 0.01 M phosphate buffered saline (PBS; pH
7.4). For cryostat sections, eyecups were embedded in OCT embedding medium (Tissue-Tek,
Elkhart, IN), then quickly frozen in liquid nitrogen and subsequently sectioned along the vertical
meridian on a cryostat at a thickness of 20 m.
Immunohistochemistry
For fluorescence immunohistochemistry, 20-µ m-thick cryostat sections were incubated in 10%
normal goat serum (NGS, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA,
5
dilution 1:1,000) or normal donkey serum (NDS, Jackson ImmunoResearch Laboratories, Inc.,
West Grove, PA, dilution 1:1,000) for 1 hr at room temperature. Sections were then incubated
overnight with either marker for middle-wavelength-sensitive opsin (M-opsin, kindly received
from Dr. C. Craft, Doheny Eye Institute, University of Southern California, dilution 1:1500) or short-
wavelength-sensitive opsin marker (S-opsin, Santa Cruz Biotechnology, Santa Cruz, CA, dilultion
1:1500) or rhodopsin marker (Rho 1D4, Gift of Dr. B. Thomas, Doheny Eye Institute, University of
Southern California, dilution 1:100) or Proliferating cell nuclear antigen (PCNA, Dako Corp.,
Carpinteria, CA, Clone PC10, dilution 1:100). Each antiserum was diluted in PBS containing 0.5%
Triton X-100 at 4º C. Retinas were washed in PBS for 45 min (3 x 15 min) and afterwards
incubated for 2 h at room temperature in either carboxymethylindocyanine-3 (Cy3)-conjugated
affinity-purified donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, Inc., West
Grove, PA, dilution 1:500) or carboxymethylindocyanine-5 (Cy5)-conjugated affinity-purified
donkey anti-mouse IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, dilution
1:300) or Alexa 488 anti-goat IgG (Molecular Probes, Eugene, OR, dilution 1:300). The sections
were washed for 30 min with 0.1M PB and coverslipped with Vectashield mounting medium
(Vector Labs, Burlingame, CA). For whole mount immunostaining, the same
immunocytochemical procedures described above were used. However, we used longer
incubation times with primary antibodies (three nights with anti-S-opsin, two nights with anti-M-
opsin, rho 1D4, and PCNA) and secondary antibodies (4 hours either with Alexa 488 donkey
anti-goat IgG or with Cy3-conjugated donkey anti-rabbit IgG or Cy5-conjugated donkey anti-
mouse IgG).
For double-label studies, whole mounts were incubated for three nights in a mixture of
S-opsin and anti-M-opsin markers. Incubation with these antibodies used 0.5% Triton X-100 in
6
0.1 M PBS at 4 ° C. After this incubation, whole mounts were rinsed for 30 min with 0.1 M PBS.
Afterwards, we incubated them with Alexa 488 donkey anti-goat and Cy3-conjugated donkey
anti-rabbit IgG for two nights at 4 ° C. For triple-label studies, whole mounts were first incubated
for two nights in a mixture of anti-M-opsin and rho 1D4 antibody. Again, incubation with these
antibodies used 0.5% Triton X-100 in 0.1 M PBS at 4 ° C. After this, the whole mounts were
rinsed for 30 min with 0.1M PBS before incubating them with Cy3-conjugated donkey anti-
rabbit IgG and Cy5-conjugated goat anti-mouse IgG for two nights at 4 ° C. Finally, the whole
mounts were stained with Terminal deoxynucleotidyl transferase dUTP nick end labeling
(TUNEL).
For nuclear layer staining, we used TOPRO-3 (Invitrogen Corporation, Carlsbad, CA
(T3605), dilution 1:1,000). TOPRO-3 was incubated for 10 min then washed for 30 min with 0.1
M PB and cover-slipped with Vectashield mounting medium. In controls, the primary antibody
was omitted from the incubation solution. Whole mounts were then washed again for 30 min
with 0.1 M PB and cover-slipped with Vectashield mounting medium. Sections and whole
mounts were then analyzed using a Zeiss LSM 510, (Zeiss, NY) confocal microscope.
Immunofluorescence images were processed in Zeiss LSM-PC software. The brightness and
contrast of the images were adjusted using Adobe Photoshop 7.0 (Adobe Systems, Mountain
View, CA). For presentation, all Photoshop adjustments (brightness and contrast only) were
carried out equally across sections.
7
Hematoxylin Staining
The anterior segments of the eyeballs were removed and the eyecups fixed in 4%
paraformaldehyde in 0.1 M PB for 2 h at 4 C. Following fixation, eyecups were then washed by
several changes of PB and transferred to 30% sucrose in PB for 5 h at 4 C. We then embedded
the eyecups in OCT embedding medium (Tissue-Tek, Elkhart, IN, USA). They were next fast-
frozen in liquid nitrogen and sectioned along the vertical meridian on a cryostat at a thickness of
10 m. We then collected sections on gelatin-coated slides for hematoxylin staining and dipped
them in hematoxylin for 5 min. They were then washed in tap water, dehydrated in alcohol,
cleared in xylene, and mounted in xylene-based medium (Richard-Allan Scientific, Kalamazoo,
MI, USA).
TUNEL Staining
Cell death was visualized by a modified TUNEL technique, according to the manufacturer‘s
instructions (In Situ Cell Detection kit, Boehringer Mannheim, Mannheim, Germany). The P15
RP whole mount retinas harvested were incubated with proteinase K (10 g/ml in 10 mM
Tris/HCl, pH 7.4–8.0) for 10 min at 37 C. After rinsing in PBS, the sections were incubated
with TUNEL reaction mixture (terminal deoxynucleotidyl transferase plus nucleotide mixture in
reaction buffer) for 60 min at 37 C. Sections were then washed again for 30 min with 0.1 M PB
and coverslipped with Vectashield mounting medium.
8
Quantification and Statistics
The size (n = 2 for all stages) and the total number (n = 3 for all stages) of rings formed by M-
opsin- and S-opsin-immunoreactive cones in RP retinas were measured at P30 and P90. For P30
RP retinas (n = 3), the numbers of rings were also measured separately for the dorsal and the
ventral hemispheres (divided by an imaginary line running through the optic disc horizontally).
The size of the ring was defined as the mean distance of two cell bodies separated in the opposite
side of the arrangement, completely across the ring. For rings were not always completely
circular in shape, the Zeiss LSM Image Browser Software was used to estimate the largest and
the smallest diameter for each ring and the values were averaged to get the mean size of the ring.
The total of 123 rings from P30 RP retinas and 148 rings from P90 RP retinas were arbitrarily
selected for the measurement of their diameters. The retinal-area for these retinas were also
measured, P30 (n = 3), 180 (n = 2) and P600 (n = 3) by ImageJ (National Institutes of Health,
Bethesda, MD, USA). In addition, the total number of M-opsin- and S-opsin-immunoreactive
cones in both normal and RP retinas were manually counted at different stages – P30 (n = 2), 180
(n = 3) and P600 (n = 3). For comparison study, the homozygous RP rat retinas (n = 2) were also
examined for the total number of M-opsin-immunoreactive cones at P180. Finally, both
heterozygous and homozygous RP retinas (n = 4 each) were examined to compare their densities
of M-opsin-immunoreactive cone cell body at P180 in whole mounts. An area of 0.16mm
2
in the
middle region of dorsal retina was selected from each retina for measure. We made sure that no
more than three of what seemed to have been rings were within the tested areas. All the
measurements were expressed as mean ± standard errors. Student‘s t-tests were used to examine
the difference between two different means. The tests were performed by MATLAB version
7.4.0 (The MathWorks Inc., Natick, MA, USA) and all graphs were generated by Microsoft
9
Excel spreadsheet, 2010 (Microsoft Corporation, Redmond, WA, USA). A difference between
the means of separate experimental conditions was considered statistically significant at P < 0.05.
RESULTS
Remodeling of M-opsin-immunoreactive cones in developing RP retinas
Recently, S-opsin-immunoreactive cones were reported to undergo extensive morphological
modifications in RP retinas (Hombrebueno et al., 2010). We examined M-opsin-immunoreactive
cones in vertical sections of normal (N) and RP retinas at P15, 30, and P90. In normal retinas at
P15 (data not shown), P30, and P90 (data not shown), we observed M-opsin immunoreactivity in
the segments, cell bodies, axon processes, and pedicles of cones (Fig. 1A). All M-opsin-
immunoreactive cones were upright and vertically aligned. These results are consistent with
previous data (Rohrer et al., 2005; Fujieda et al., 2009; Hombrebueno et al., 2010). The entire M-
opsin-immunoreactive cones are labeled in RP retinas (Figs. 1B-D). In P15 RP retina, M-opsin-
immunoreactive cones were upright (Fig. 1B), similar to that seen in normal retinas. This image
was taken from the central part of the retinal section. The cone outer segments (COS) are
shortened and distorted in orientation (arrow) compared to that in normal condition. By P30, M-
opsin immunoreactive cones have shortened remarkably in length; from the COS to the pedicle
(Fig. 1C). The COS were shortened and distorted. The overall orientation of some cones were
not vertical but were slightly aquiline/curved. By P90, all M-opsin-immunoreactive cones have
lost their upright orientation completely (Fig. 1D). All cones were positioned ‗flat‘ against the
10
outer part of the inner nuclear layer (INL). We also observed separate regions full of clusters of
cell bodies followed by regions devoid of cell bodies but rich in processes. These two regions of
different cellular structures alternated along the length of the vertical retinal sections.
Fig. 1 M-opsin-immunoreactive cones remodel in RP retinas
11
Reorganization of M-opsin- and S-opsin-immunoreactive cones in orderly array of rings
To investigate the distribution pattern of cones in RP retina, we used M-opsin and S-opsin
antibodies to identify cones in whole mount retinas. Figure 2 shows example of whole mounts
processed for M-opsin (Figs. 2A, D, G) and S-opsin (Figs. 2B, E, H) immunoreactivities. The
images were taken from the mid-peripheral part of the inferior (3-mm away from the optic disk)
of P90 normal (Figs. 2A-C) and P90 RP (Figs. 2D-I) whole mount retinas. The results showed
that there were more M-opsin-immunoreactive cones compared to S-opsin-immunoreactive
cones in all retinas. In P90 normal retinas, M-opsin- (Fig. 2A) and S-opsin- (Fig. 2B)
immunoreactive cones were distributed homogeneously throughout the retinas. Double-labeling
experiments showed no co-localization of M-opsin and S-opsin immunoreactivity (Fig. 2C). In
P90 RP retinas, we observed strikingly different mosaic of opsin-immunoreactive cones (Figs.
2D-F). M-opsin- (Fig. 2D) and S-opsin- (Fig. 2E) immunoreactive cones were distributed in
arrangements that resembled rings. When both M-opsin and S-opsin immunoreactivity are shown
together (Fig. 2F), one can see that M-opsin- and S-opsin-immunoreactive cones formed rings at
the same locations of the retina. A high-magnification view of part of a ring marked by the inset
rectangle revealed that M-opsin- (Fig. 2G) and S-opsin- (Fig. 2H) immunoreactive cones share a
specific orientation. Almost all the COS and the cell bodies were near the rims of the rings
whereas the processes were extended towards the center of the rings. Some M-opsin-
immunoreactive cones and few S-opsin-immunoreactive cones showed abnormal processes
sprouting from either their cell bodies or their axon processes. Double exposure showed how cell
bodies were aligned very close to each other (Fig. 2I). Also, there was no co-localization of M-
opsin and S-opsin-immunoreactive cones.
12
Fig. 2 M-opsin and S-opsin-immunoreactive cones remodel into orderly array of rings
13
Cone rings are not associated with retinal foldings
We aimed to investigate whether cones rearranging themselves in rings have any relationship
with rosettes first described by Flexner (1891) and Wintersteiner (1897). Mostly observed in
retinoblastoma, rosettes are spherical folding of the ONL, mostly composed of photoreceptors
(Ts‘o et al., 1970; Tansley, 1933; Gallie et al., 1999). In order to study whether the arrangements
of cones in rings in RP retinas reflect rosettes or not, we examined for any presence of physical
folding at the level of the ONL. Rings arranged of M-opsin-immunoreactive cones were seen in
P30 whole mount RP retina (Fig. 3A). Light micrographs taken under differential interference
contrast (DIC) mode at the same focal plane as M-opsin-immunoreactive cones showed no
obvious retinal folds (Fig. 3B). The merged image of the two micrographs confirmed that rings
in P30 RP retina were not associated with physical foldings of the ONL of the retina (Fig. 3C).
Similarly, rings of M-opsin-immunoreactive cones in P90 RP retinas (Fig. 3D) were also not
associated with physical retinal foldings (Fig. 3E). The merged image of the two modes
confirmed that rings in P90 RP retina were not rosettes (Fig. 3F). In order to examine in vertical
sections, we processed P15 and P30 retinal sections with Hematoxylin stain. The result showed
all the retinal layers (Figs. 3G-J). Hematoxylin staining of P15 RP retinal sections showed
multiple rows of nuclei in the ONL (Fig. 3G). A high-magnification view of the ONL indicated
within the inset rectangle showed that its thickness along the length of the section was uniform
(Fig. 3H). However, multiple spaces empty of cell bodies were spotted within the ONL (arrow).
P30 RP retinal sections, on the other hand, indicated that the thickness of the ONL was not
uniform (Fig. 3I). Grooves were often observed in the ONL as marked within the inset rectangle.
At the trough of the groove (arrow), cell bodies were lacking (Fig. 3J). To study the arrangement
of cell bodies in more detail, we processed P30 RP whole mount retinas with antibodies against
14
M-opsin plus TOPRO-3 nuclear stain. When the image was taken at the level of the M-opsin-
immunoreactive cell bodies at the ONL, nuclei at the center of the ring were out of focus (Fig.
3K). High-magnification view of the center of the ring confirmed this (Fig. 3L). These nuclei
were in the INL just below the level of M-opsin-immunoreactive cones. Such results suggest that
the center of the ring is the trough of the grooves seen in the vertical sections. Taken together,
our rings are not the same as rosettes in their physical architecture.
Fig. 3 Cone rings are not associated with retinal foldings
15
Clusters of cell death are observed inside the rings
We processed P15 whole mount RP retinas for M-opsin (Fig. 4A) with rhodopsin (Fig. 4B) and
TUNEL (Fig. 4C) labeling to observe the spatial correlation of cones, rods, and dying rods. We
observed co-localization of local zones with low densities of cones and rods (Figs. 4A, B, D).
Triple labeling of M-opsin, rhodopsin, and TUNEL showed clusters of dying cells inside zones
with no cones and rods (Fig. 4D). Such clusters were consistent with our previous study showing
massive rod cell death around P15 (Ray et al., 2010; Lee et al., 2011; Li et al., 2011). Hence, the
holes emerged at least in part because of clusters of rod death.
Fig. 4 Clusters of cell death are observed inside a ring
Rings first develop at around P15 and start to lose their form from around P180
In order to examine when the rings first develop and how they change over time, we
immunostained whole mount RP retinas at P5 – 30 (n ≥ 5), 180 and P600 (n = 3). The
distribution of M-opsin- and S-opsin-immunoreactive cones in P5-14 RP retinas (data not
shown) resembled that of in normal retinas (Figs. 2A-C). At P15, a small region of abnormal
distribution of cones was observed for the first time (Fig. 5A). The relative position of the initial
16
ring-formation in the retinas was random (data not shown). The immature ring in P15 was much
smaller in size in comparison to ones found in later postnatal stages. A high-magnification view
showed that the orientation of all cones was the same as previously observed; COS forming the
rims of the rings and the other parts of the cones being near the center of the rings (Fig. 5B). By
P30, rings have grown larger in both their number and their size (Fig. 5C). M-opsin- and S-
opsin-immunoreactive cones formed rings at the same regions in the RP retina. All cones had the
same orientation and the center of the ring was filled with processes (Fig. 5D). Compared to
cones in P15 RP retinas, it was much easier to view the entire cones (from the COS to the
pedicle) in one focal plane suggesting that cones in P30 RP retinas were closer to losing their
vertical alignment within the ONL. By P180, rings have somewhat lost their shape (Fig. 5E). A
higher-magnification view of a part of what seemingly used to be a ring revealed that cones were
comparatively more disorderly (Fig. 5F). The orientation shared by all cones until P90 (Fig. 2I)
was not as evident anymore. Most processes were no longer extended straight towards the center
of the ring thereby leaving a large area in the central zone of the ring devoid of cell processes.
Most cones were clustered together and were far from arranged in orderly array. In addition, new
processes emerged from either the cell bodies or from the remaining axon processes. Also, a lot
of cones had lost their COS. In P600 whole mount RP retinas, there were larger areas of space
devoid of M-opsin- and S-opsin-immunoreactive cones compared to P180 RP retinas (Fig. 5G).
A higher-magnification view on some cones indicated that they have generated extensive
branches in their processes that were not typical in the morphology of normal cones (Fig. 5H).
Their morphology seemed to resemble bipolar or amacrine cells. Also, most have lost their COS.
17
Fig. 5 The development and change of rings
18
Rings in RP retinas grow significantly with age in both sizes and numbers
The sizes and quantities of rings at P30 and P90 were examined. We did not use P180 retinas for
these measurements since the distribution pattern has mostly lost its form of rings and we did not
want to introduce personal sampling errors to data. Composite images of P30 (Fig. 6A) and P90
(Fig. 6B) whole mount RP retinas showing M-opsin immunoreactivity were constructed. In all
P30 RP retinas, more rings were observed in the peripheral region of the retina compared to the
areas closer to the optic disk. Also, when the mean total counts of rings in the dorsal region (138
± 5 — mean ± standard error) were compared with those in the ventral region, (54 ± 4; Fig. 6C),
the dorsal region had significantly greater number of rings (p < 0.002, one-tailed Student‘s t-test).
The dorsal and the ventral regions are shown as areas divided by the dotted lines in Fig. 6A. The
mean total number of rings in P90 RP retinas (350 ± 10) retinas was significantly larger
compared to that in P30 RP retinas (191 ± 6, p < 0.003, one-tailed Student‘s t-test; Fig. 6D). By
P90, rings were seen throughout the retina. The mean diameter of rings formed by M-opsin- and
S-opsin-immunoreactive cones were significantly larger in P90 (275 ± 3 µ m) compared to P30
(168 ± 1 µ m, p < 7e
-023
, one-tailed Student‘s t-test; Fig. 6E). These results showed that rings
increase significantly in both their mean size and their quantity between P30 and P90 with
progression of the disease.
19
Fig. 6 Growth of RP rings in size and quantity
M-opsin and S-opsin-immunoreactive cones in RP retinas do not die until P180
We aimed to examine when cones degenerate in RP retinas. M-opsin- and S-opsin-
immunoreactive cones in P30, P180 and P600 normal and RP whole mount retinas were counted
(Fig. 7A). One-tailed Student‘s was used to test the significance between the different means. In
P30 normal retinas, the mean total number of M-opsin- and S-opsin-immunoreactive cones were
88,700 ± 700 and 35,000 ± 2,000 (mean ± standard error) respectively. In P30 RP retinas, the
20
mean total number of M-opsin and S-opsin-immunoreactive cones was 87,000 ± 4,000 and
36,000 ± 4,000 respectively. These mean counts for the normal and the RP retinas were not
statistically significantly different. In P180 normal retinas, the mean total number of M-opsin-
and S-opsin-immunoreactive cones was 88,600 ± 400 and 32,600 ± 100 respectively. In P180 RP
retinas, the mean total number of M-opsin- and S-opsin-immunoreactive cones was 87,500 ± 300
and 31,200 ± 700 respectively. Again, the test showed no significant differences between the
counts in the normal and RP retinas. In P600 normal retinas, the mean total number of M-opsin-
and S-opsin-immunoreactive cones was 70,000 ± 10,000 and 26,000 ± 7,000 respectively. The
decrease in the counts seen in P600 normal retinas compared to P30 and P180 normal retinas
were not statistically significant. In P600 RP retinas, the mean total number of M-opsin- and S-
opsin-immunoreactive cones was 12,000 ± 2,000 and 9,100 ± 900 respectively. The mean count
for both M-opsin- and S-opsin-immunoreactive cones in the RP retinas were significantly lower
compared to the normal retinas (p < 0.004 and 0.04, one-tailed Student‘s t-test). The counts for
P180 RP retinas showed no significant difference from the counts for P30 RP retinas. Taken
together, these results suggest that cones of the S334ter-line-3 transgenic rat do not degenerate
until after P180. However, significant degree of degeneration is detected by P600. Furthermore,
we have examined proliferating cells using Proliferating cell nuclear antigen (PCNA) labeling in
RP whole mount retinas (P15, 30 and 180). We have not observed proliferating cells in RP
retinas, certainly not within the arrangement of rings (data not shown). Hence, the steady total
cell counts mean that there has been no cone death in RP retinas until P180.
For control, we also measured the areas of the retinas used for counting cones to ensure
that no sampling errors have occurred between the normal and the RP retinas (Fig. 7B). The
mean retinal areas for P30 normal and RP retinas were 24.5 ± 0.5 mm
2
and 23.9 ± 0.6 mm
2
,
21
respectively. For P180 normal and RP retinas, they were 52 ± 1 mm
2
and 51.7 ± 0.9 mm
2
,
respectively. And for P600 normal and RP retinas, they were 62 ± 3 mm
2
and 63 ± 2 mm
2
respectively. Therefore, the retinas were shown to grow with age. However, the Student‘s t-test
(two-tailed) indicated no significant differences in the areas between the normal and the RP
retinas at all the postnatal stages measured. Hence, the drops in the P600 RP cone-counts were
neither due to selection of unusually small RP retinas nor due to shrinkage in their areas due to
the disease.
To compare the morphology, distribution and orientation of M-opsin-immunoreactive
cones in homozygous RP rat retinas with those in heterozygous RP rats, we processed their
retinas using M-opsin antibody. The whole mount retinas from both heterozygous and
homozygous P20 RP rats (Fig. 7C, left panel) showed what we have previously observed (Figs.
2D-I, 5). The COS of the M-opsin-immunoreactive cones were in the rim of the ring and the
processes were pointed towards the center of the ring (Fig. 7C). However, there were more rings
in homozygous retinas than in heterozygous ones (data not shown).
To compare the survival of M-opsin-immunoreactive cones in heterozygous and
homozygous RP retinas, we have further counted the total number of the cells in the P180
homozygous RP whole mounts. The mean total number of M-opsin-immunoreactive cones was
85,000 ± 2,000 (mean ± standard error). The two-tailed Student‘s t -test revealed that this number
was not statistically significantly different from that from P180 heterozygous retinas.
Furthermore, we have examined the regional densities of the M-opsin-immunoreactive cone cell
bodies in heterozygous and homozygous RP rat retinas at P180. The mean density of M-opsin-
immunoreactive cone cell bodies in the dorsal wing in heterozygous RP retinas was 310 ± 20.
22
The mean density in the homozygous RP retinas was 290 ± 30. The two-tailed Student‘s t-test
indicated that the two density measures were not statistically significantly different from each
other.
Fig. 7 Prolonged survival of M-opsin and S-opsin-immunoreactive cones in RP retinas
23
DISCUSSION
Remodeling of cone morphology in RP retinas
In the present study, we observed entire parts of cones stained by S-opsin and M-opsin
antibodies in normal (Fig.1A) and RP retinas (Figs. 1B-D). This expression pattern of cone opsin
proteins has been found in normal mice and rats (Rohrer et al., 2005, Fujieda et al., 2009;
Hombrebueno et al., 2010), retinal pigment epithelium-specific 65 kDa protein (RPE65)-
deficient mice (Rohrer et al., 2005; Karan et al., 2008; Zhang et al., 2008), cyclic nucleotide
gated channel alpha 3 (CNGA3)-deficient mice (Michalaski et al., 2005) and guanylate cyclase
(GC)-knockout mice (Karan et al., 2008). Such expression pattern aided in examining how cones
remodel in RP retinas as the disease progressed. We found that the remodeling of M-opsin-
immunoreactive cones was similar to that of S-opsin-immunoreactive cones previously described
in the same RP animal model (Hombrebueno et al., 2010). Many of the main morphological
changes we observed were also shared with different animal models of RP. For example, the
shortening and distortion of the COS observed from P15 in our RP retinas (Fig. 1B), were
reported also in rd1 mice from as early as P8 (Fei, 2002, Lin et al., 2009), and in rd10 mice from
around P30 (Barhoum et al., 2008). The loss of the COS we observed at P90, 180 and P600 (Figs.
2I, 5F, H) was reported in rd1 mice at P12 (Fei, 2002, Lin et al., 2009). The abnormal sprouting
of processes from either the cell bodies or the axon processes of M-opsin-immunoreactive cells
(Figs. 2I, 5F, H) was also observed from S-opsin-immunoreactive cones in the same animal
model (Hombrebunos et al., 2010). Such abnormal sprouting from cones was also reported in rd1
mice (Fei, 2002, Lin et al., 2009). Furthermore, the loss of vertical alignment and positioning of
cones at the outer part of the INL (Fig. 1D) was described in P90 RP rats (S-opsin-
24
immunoreactive cones; Hombrebueno et al., 2010), and also in rd1 mice at P75 (Lin et al., 2009).
Thus, we can conclude that the way that cones remodel in RP rats is similar to that seen in some
other animal models of RP. It is likely that similar signal from the affected cascade from
mutation in RP initiate such morphological changes of cones.
Remodeling of the spatial distribution of cones into orderly array of rings in RP retinas
Contrary to the pattern of morphological remodeling that is shared across some animal models of
RP, the distribution pattern of cones in rings is not ubiquitous. We found orderly array of rings
formed by both M-opsin- and S-opsin-immunoreactive cones in S334ter-line-3 RP retinas (Figs.
2, 5). Such distinct distribution pattern of cones was also observed in human patients with eye
disease due to retinal dystrophy, inherited retinal degeneration, central ring scotomas and genetic
perturbations in the photopigment in M-cones, (Carroll et al., 2004; Choi et al., 2006; Duncan et
al., 2007; Joeres et al., 2008; Rossi et al., 2011). In these patients, circular areas of dark spaces or
patchy regions on the ONL of their retinas were observed. In addition, cyclin D1-deficient mice
(Ma et al., 1998) showed ‗holes of photoreceptors‘ that resemble our rings (Ma et al., 1998). W e
do not yet know why such ring-like cone distribution is present in these various cases. Previously,
Lee et al. (2011) have hypothesized that remodeled Mü ller cell processes, which contribute to the
formation of the rings in S334ter-line3 RP retinas are significant for the survival of the cones.
Thus, studying the cellular basis of ring formation in our RP model may help understand the
neural-glial interactions, survival and reorganization of photoreceptors in some human patients.
Possible reasons why rings were not observed in some other animal models of RP could
be due either to the way that different mutations affect photoreceptors or to the experimental
25
protocol used to reach the conclusion. If the degeneration rate is fast, and most rods and cones
die a short time after birth, such as in rd1 mice (Carter-Dawson et al., 1978; Jimé nez et al., 1996),
cones may not have enough time to form rings. It could also be possible that the period during
which cones show rings before they quickly degenerate is short. Thus, if an experiment sampled
retinas at relative coarse time bins, then it might not catch the moment when rings are present.
We have shown that our moderately slow-degenerative RP rats represent an apt model to study
the progression of cone spatial rearrangement into a distribution pattern that is also reported in
some important cases including human patients.
Ring formation is triggered by rod deaths not by the mechanical disruption of the ONL
Cones in RP retinas begin forming rings at P15 (Fig. 5A). Because almost all pups used for this
experiment have opened their eyes at P16, patterned visual input does not appear to be necessary
for ring formation. One must thus find non-visual, epigenetic mechanisms for the emergence of
rings. We observed numerous spaces empty of photoreceptor nuclei in the ONL of P15 RP
vertical retinal sections (Fig. 3H; arrow). And there were apoptotic cells in the ONL of P15 RP
whole mount retinas (Fig. 4C). These apoptotic cells are dying rod cells as rod deaths but no
cone deaths have been reported at around P15 in RP retinas (Li et al., 2010; Ray et al., 2010; Lee
et al., 2011). Previously, Ray et al. (2010) reported the largest number of apoptotic cells in the
ONL of RP retinas at P15. Our results indicated a close temporal and spatial correlation of dying
rods with cone rings formation. Where rings were not present, rod death was scattered randomly
in the retina (Fig. 4D). However, where a ring was present, the focal death of rods was observed
in a cluster (Fig. 4D). Taken together our results suggest that the clusters of rod deaths may be
26
causing ring formation. Clusters of photoreceptor deaths were also reported to lead to the
formation of ‗holes of photoreceptors‘ in cyclin D1-deficient mice (Ma et al., 1998). Rod
degeneration in hot spots could be due to local reduction of survival factors intrinsically secreted
by rods after their death (LaVail et al., 1998). Previous studies have also reported how
degenerating rods can often lead to apoptosis in their neighbors (Huang et al., 1993; Kedzierski
et al., 1998).
Furthermore, our results also indicated that our rings were not formed by mechanical
collapse nor foldings of the ONL (Figs. 3A - J). Our rings were very regular and consistent in
terms of their spatial arrangement and orientation of cones (Fig. 2D-I). We have also observed
alternating regions full of cell bodies and regions full of processes along the length of the RP
vertical sections (Fig. 1D). Such preciseness suggests that rings are not formed by mere
mechanical disruption following rod loss. Continual expansion of rings in their quantity and size
until P90 (Fig. 6C-E), long after all rods are degenerated in RP retinas (Li et al., 2010; Ray et al.,
2010), also support this. In all, we can conclude that ring formation is triggered by rod deaths not
by the mechanical disruption of the ONL.
The spatial and temporal distribution of rings in RP retina
Our results indicated that despite the initial formation of a ring being random in its relative
position in the RP retina at P15, there was a greater occurrence of rings in the peripheral and
dorsal regions at P30 (Fig. 6A, C) before spreading throughout the retina by P90 (Fig. 6B). This
may be due to the intrinsic disparity in these regions as previously described in rats with retinal
dystrophy (LaVail and Battelle, 1975). Previous studies have shown that the degeneration of
27
cones in the central retina precedes the peripheral region in rd1 mice (Carter-Dawson et al.,
1978; Lin et al., 2009). Also, hemispheric asymmetry in the number of surviving cones was
previously reported in rd1 mice (Garcia-Fernandez et al., 1995; Jimenez et al., 1996; LaVail et
al., 1997). These studies reported slower rate of cone degeneration in the dorsal retina compared
to the ventral retina. Therefore, it is possible that ring formation is also influenced by the same
mechanisms that induce such regional differences in survival of cones. Further studies are
required to examine whether cones survive better in these regions in late stage of RP.
No cone degeneration until rings start to lose their form
Interestingly, our total cone cell body counts indicated no cell death until P180 (Fig. 7A) – that is,
until when rings start to lose their clear form and cones lose their specific orientation (Figs. 5E-
H). The growth in the retinal area we observed from P30 to P600 in both normal and RP rats (Fig.
7B) is consistent with previous studies (McCall et al., 1987; Harman et al., 2003). Thus, no
sampling errors were introduced to our cone counts. We also did not observe proliferating cells
in the ONL of RP retinas at P15, 30, and 180 (data not shown). Proliferation of photoreceptors
from P15 was also not reported in any RP models. Although, Mü ller cells are known to exhibit
some neuronal stem cell properties (Fischer and Reh, 2001, 2003; Ooto et al., 2004), Mü ller cells
transdifferentiating into photoreceptors was not reported to date in mammalian retinas. Hence,
these findings collectively indicate that there was no cone degeneration until P180. This was
reported otherwise in Li et al. (2010). Li et al. (2010) immunostained cones in homozygous
S334ter-line-3 rat retinas using Peanut Agglutinin (PNA, marks COS) and cone arrestin (CAR,
marks entire cones). Li et al. (2010) supported cone death using their evidence of falling COS
28
densities and weakening of CAR immunoreactivity with the progression of the disease. However,
for examining the degeneration of cones, it is more accurate to measure by cone cell bodies
rather than by the COS as many cones were found to lose their COS in later stages (Fig. 2I, 5F,
H). Also, one needs to take extra caution when measuring the density instead of the total cell
counts in the RP retinas because the distribution of cones is not homogenous. Depending on
which regions you select for measurement, the density may vary.
We have neither observed significant differences in the total number of M-opsin-
immunoreactive cones nor observed differences in their cell body densities in heterozygous
versus homozygous P180 RP retinas (Fig. 7D). Hence, some discrepancies between Li et al.
(2010) and our study could have been due to different staining protocols. If the staining was
relatively weaker, the cellular structures inside the rings could have been difficult to discern.
Taken together, our data suggest that the formation of rings is not a product of secondary
degeneration of cones.
Possible mechanisms underlying reorganization of cones in rings in RP retinas
The absence of cone degeneration and proliferation until P180 (Fig. 7A) and their expansion in
rings (Figs. 6D, E), suggest that cones are migrating. As to how the cones may migrate, one
mechanism was proposed by Lee et al (2011). In their study, they showed that remodeled
processes of Mü ller cells filled the inside of the rings. Also, the cone processes in RP retinas
were intertwined with the processes of Mü ller cells (Lee et al., 2011). Moreover, when glial-cell
toxin DL-alpha-aminoadipic acid (AAA) was injected intravitreally, the rings disappeared (Lee
29
et al., 2011). These results suggest that Mü ller cells interact with cones, compelling them to
migrate into the array of rings in RP retinas. Another significant finding regarding Mü ller cells
was reported by Xia et al. (2011). They applied intravitreal injection of Oncostatin M (OSM), a
member of the IL-6 family of cytokines (Rose and Bruce, 1991), to P20 and 35 homozygous RP
rats and observed the abolishment of rings. Xia et al. (2011) interpreted the resulting
homogenous immunoreactivity of cones as OSM protecting and regeneration the COS of cones
inside the rings. They also reported that the effects of OSM on cones were mediated by Mü ller
cells. Because we do not normally observe cone cell death in RP retinas until P180, our data
suggest that OSM is not inducing further survival of photoreceptors in the Xia et al.‘s study
(2011). An alternate interpretation of their and our data is that OSM causes the redistribution of
cones perhaps by regulating tight junctions between Mü ller cells and cones (Lee et al., 2011).
The disappearance of rings upon application of OSM could have been due to loosening of tight
junctions between Mü ller cells and cones. Several past studies have reported about the role of
cytokines on tight junction alteration in retina (Zech et al., 1998; Abe et al., 2003; Villarroel et
al., 2009; Aveleira et al., 2010).
At P600, we observed significant cone death in RP retinas (Fig. 7A). The decline of the
total cone number in the normal retinas at P600 compared to those at their earlier stages (Fig.
7A) was statistically not significant and was probably due to the effect of aging (McCall et al.,
1987; Dorey et al., 1989; Harman et al., 2003). Cones in rings maintained their abnormal
phenotypes for prolonged periods until their degeneration (Figs. 5F, H). Such extended cone
survival in RP retinas long after massive rod death was previously reported. In rd1 mice, cones
were reported to survive up to at least 18 months (Carter-Dawson, 1978). There must be some
mechanisms that help cones survive for long periods in harsh conditions. An intriguing
30
possibility is that close-gathering of cones may aid in their survival. Within rings, cones are
aligned very close to another (Fig. 2I). This may allow them better share of self-secreted trophic
factors. Future studies on the effects of different distribution pattern of cones on their survival in
RP should enlighten what possible effects rings may have on cones and what mediates their
rearrangement. Understanding whether and how rings in RP retinas improve the survival of
cones would provide the scientific and clinical communities with better knowledge of how to
prolong cone survival in RP.
31
REFERENCES
Abe T, Sugano E, Saigo Y , Tamai M. 2003. Interleukin-1beta and barrier function of retinal
pigment epithelial cells (ARPE-19): aberrant expression of junctional complex molecules. Invest
Ophthalmol Vis Sci. 44(9): 4097-104.
Aveleira CA, Lin CM, Abcouwer SF, Ambró sio AF, Antonetti DA. 2010. TNF-α signals through
PKCζ/NF-κB to alter the tight junction complex and increase retinal endothelial cell permeability.
Diabetes. 59(11): 2872-82.
Baker GE, Dovey M, Davda P, Guibal C, Jeffery G. 2005. Protein kinase C immunoreactivity in
the pigmented and albino rat retina. Eur J Neurosci 22(10): 2481-2488.
Barhoum R, Martí nez-Navarrete G, Corrochano S, Germain F, Fernandez-Sanchez L, de la Rosa
EJ, de la Villa P, Cuenca N. 2008. Functional and structural modifications during retinal
degeneration in the rd10 mouse. Neuroscience. 155(3):698-713.
Berson EL. 1993. Retinitis pigmentosa. The Friedenwald Lecture. Invest Ophthalmol Vis Sci.
34(5):1659-76.
32
Blanks JC, Adinolfi AM, Lolley RN. 1974. Photoreceptor degeneration and synaptogenesis in
retinal-degenerative (rd) mice. J Comp Neurol. 156(1):95-106.
Bonilha VL, Fishman GA, Rayborn ME, Hollyfield JG. 2009. Retinal pathology of a patient with
Goldmann-Favre syndrome. Ophthalmic Genet. 30(4):172-80.
Bowes C, Li T, Danciger M, Baxter LC, Applebury ML, Farber DB. 1990. Retinal degeneration
in the rd mouse is caused by a defect in the beta subunit of rod cGMP-phosphodiesterase. Nature.
347(6294):677-80.
Bravo R, Frank R, Blundell PA, Macdonald-Bravo H. 1987. Cyclin/PCNA is the auxiliary
protein of DNA polymerase-delta. Nature. 326(6112):515-7.
Carroll J, Neitz M, Hofer H, Neitz J, Williams DR. 2004. Functional photoreceptor loss revealed
with adaptive optics: An alternate cause for color blindness. Proc Natl Acad Sci U S
A. 101(22):8461-6.
33
Carter-Dawson LD, LaVail MM, Sidman RL. 1978. Differential effect of the rd mutation on rods
and cones in the mouse retina. Invest Ophthalmol Vis Sci. 17(6):489-98.
Chalmel F, Lé veillard T, Jaillard C, Lardenois A, Berdugo N, Morel E, Koehl P, Lambrou G,
Holmgren A, Sahel JA, Poch O. 2007. Rod-derived Cone Viability Factor-2 is a novel
bifunctional-thioredoxin-like protein with therapeutic potential. 8:74.
Chang B, Heckenlively JR, Hawes NL, Roderick TH. 1993. New mouse primary retinal
degeneration (rd-3). Genomics. 16(1):45-9.
Chen J, Nathans J. 2007. Genetic ablation of cone photoreceptors eliminates retinal folds in the
retinal degeneration 7 (rd7) mouse. Invest Ophthalmol Vis Sci. 48(6):2799-805.
Choi SS, Doble N, Hardy JL, Jones SM, Keltner JL, Olivier SS, Werner JS. 2006. In vivo
imaging of the photoreceptor mosaic in retinal dystrophies and correlations with visual function.
Invest Ophthalmol Vis Sci. 47(5):2080-92.
34
Cideciyan AV , Hood DC, Huang Y , Banin E, Li ZY , Stone EM, Milam AH, Jacobson SG. 1998.
Disease sequence from mutant rhodopsin allele to rod and cone photoreceptor degeneration in
man. Proc Natl Acad Sci U S A. 95(12):7103-8.
Curcio CA, Sloan KR, Kalina RE, Hendrickson AE. 1990. Human photoreceptor topography. J
Comp Neurol. 292(4): 497-523.
Delyfer MN, Lé veillard T, Mohand-Saï d S, Hicks D, Picaud S, Sahel JA. 2004. Inherited retinal
degenerations: therapeutic prospects. Biol Cell. 96(4):261-9.
DeVries SH, Qi X, Smith R, Makous W, Sterling P. 2002. Electrical coupling between
mammalian cones. Curr Biol. 12(22):1900-7.
Dorey CK, Wu G, Ebenstein D, Garsd A, Weiter JJ. 1989. Cell loss in the aging retina.
Relationship to lipofuscin accumulation and macular degeneration. Invest Ophthalmol Vis Sci.
30(8):1691-9.
35
Duncan JL, Zhang Y, Gandhi J, Nakanishi C, Othman M, Branham KE, Swaroop A, Roorda A.
2007. High-resolution imaging with adaptive optics in patients with inherited retinal
degeneration. Invest Ophthalmol Vis Sci. 2007 Jul;48(7):3283-91.
Farber DB, Lolley RN. 1974. Cyclic guanosine monophosphate: elevation in degenerating
photoreceptor cells of the C3H mouse retina. Science. 186(4162):449-51.
Farber DB, Flannery JG, Bowes-Rickmann C. 1994. The rd mouse story: seventy years of
research on an animal model of inherited retinal degeneration. Prog Ret Eye Res. 13(1): 31–64.
Fei Y . 2002. Cone neurite sprouting: an early onset abnormality of the cone photoreceptors in the
retinal degeneration mouse. Mol Vis. 27(8):306-14.
Fischer AJ, Reh TA.. 2001. Mü ller glia are a potential source of neural regeneration in the
postnatal chicken retina. Nat Neurosci. 4(3): 247-52.
Fischer AJ, Reh TA. Fischer AJ, Reh TA. 2003. Potential of Mü ller glia to become neurogenic
retinal progenitor cells. Glia. 43(1): 70-6. Review.
36
Flexner S. 1891. A peculiar glioma (neuroepithelioma) of the retina. Johns Hopkins Hospital
Bull. 15:115.
Fujieda H, Bremner R, Mears AJ, Sasaki H. 2009. Retinoic acid receptor-related orphan receptor
alpha regulates a subset of cone genes during mouse retinal development. J Neurochem.
108(1):91-101.
Gallie BL, Campbell C, Devlin H, Duckett A, Squire JA. 1999. Developmental basis of retinal-
specific induction of cancer by RB mutation. Cancer Res. 59(7 Suppl):1731s-1735s.
Garcí a-Ferná ndez JM, Jimenez AJ, Foster RG. 1995. The persistence of cone photoreceptors
within the dorsal retina of aged retinally degenerate mice (rd/rd): implications for circadian
organization. Neurosci Lett. 187(1):33-6.
Glubrecht DD, Kim JH, Russell L, Bamforth JS, Godbout R. 2009. Differential CRX and OTX2
expression in human retina and retinoblastoma. J Neurochem. 111(1):250-63.
Gordon WC, Casey DM, Lukiw WJ, Bazan NG. 2002. DNA damage and repair in light-
induced photoreceptor degeneration. Invest Ophthalmol Vis Sci. 43(11):3511-21.
37
Gupta N, Brown KE, Milam AH. 2003. Activated microglia in human retinitis pigmentosa, late-
onset retinal degeneration, and age-related macular degeneration. Exp Eye Res. 76(4):463-71.
Harman AM, MacDonald A, Meyer P, Ahmat A. 2003. Numbers of neurons in the retinal
ganglion cell layer of the rat do not change throughout life. Gerontology. 49(6):350-5.
Hartong DT, Berson EL, Dryja TP. 2006. Retinitis Pigmentosa. Lancet. 368(9549):1795-809.
Haverkamp S, Wä ssle H, Duebel J, Kuner T, Augustine GJ, Feng Ge, Euler T. 2005. The
primordial, blue-cone color system of the mouse retina. J Neurosci. 25(22):5438-45.
Hombrebueno JR, Tsai MM, Kim HL, De Juan J, Grzywacz NM, Lee EJ. 2010. Morphological
changes of short-wavelength cones in the developing S334ter-3 transgenic rat. Brain Res.
1321:60-6.
Humphries MM, Rancourt D, Farrar GJ, Kenna P, Hazel M, Bush RA, Sieving PA, Sheils DM,
McNally N, Creighton P, Erven A, Boros A, Gulya K, Capecchi MR, Humphries P. 1997.
Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat Genet. 15(2):216-
9.
38
Huang PC, Gaitan AE, Hao Y , Petters RM, Wong F. 1993. Cellular interactions implicated in the
mechanism of photoreceptor degeneration in transgenic mice expressing a mutant rhodopsin
gene. Proc Natl Acad Sci U S A. . 90(18): 8484-8.
Jaskulski D, deRiel JK, Mercer WE, Calabretta B, Baserga R. 1988. Inhibition of cellular
proliferation by antisense oligodeoxynucleotides to PCNA cyclin. Science. 240(4858):1544-6.
Jimé nez AJ, Garcí a-Ferná ndez JM, Gonzá lez B, Foster RG. 1996. The spatio-temporal pattern of
photoreceptor degeneration in the aged rd/rd mouse retina. Cell Tissue Res. 284(2):193-202.
Karan S, Zhang H, Li S, Frederick JM, Baehr W. 2008. A model for transport of membrane-
associated phototransduction polypeptides in rod and cone photoreceptor inner segments. Vision
Res. 48(3):442-52.
Kedzierski W, Bok D, Travis GH. 2010. Non-cell-autonomous photoreceptor degeneration in rds
mutant mice mosaic for expression of a rescue transgene. J Neurosci. 18(11):4076-82.
39
LaVail MM, Battelle BA. 1975. Influence of eye pigmentation and light deprivation on inherited
retinal dystrophy in the rat. Exp Eye Res. 21(2):167-92.
LaVail MM, Blanks JC, Mullen RJ. 1982. Retinal degeneration in the pcd cerebellar mutant
mouse. I. Light microscopic and autoradiographic analysis. J Comp Neurol. 212(3):217-30.
LaVail MM, Matthes MT, Yasumura D, Steinberg RH. 1997. Variability in rate of cone
degeneration in the retinal degeneration (rd/rd) mouse. Exp Eye Res. 65(1):45-50.
LaVail MM, Yasumura D, Matthes MT, Lau-Villacorta C, Unoki K, Sung CH, Steinberg RH.
1998. Protection of mouse photoreceptors by survival factors in retinal degenerations. Invest
Ophthalmol Vis Sci. 39(3):592-602.
Lee EJ, Ji Y, Zhu CL, Grzywacz NM. 2011. Role of Mü ller Cells in Cone-mosaic
Rearrangement in a Rat Model of Retinitis Pigmentosa (in press). Glia.
40
Li ZY, Jacobson SG, Milam AH. 1994. Autosomal dominant retinitis pigmentosa caused by the
threonine-17-methionine rhodopsin mutation: retinal histopathology and immunocytochemistry.
Exp Eye Res. 58(4):397-408.
Li F, Cao W, Anderson RE. 2003. Alleviation of constant-light-induced photoreceptor
degeneration by adaptation of adult albino rat to bright cyclic light. Invest Ophthalmol Vis Sci.
44(11):4968-75.
Li Y , Tao W, Luo L, Huang D, Kauper K, Stabila P, Lavail MM, Laties AM, Wen R. 2010. CNTF
induces regeneration of cone outer segments in a rat model of retinal degeneration. PLoS One.
5(3):e9495.
Lin B, Masland RH, Strettoi E. 2009. Remodeling of cone photoreceptor cells after rod
degeneration in rd mice. Exp Eye Res. 88(3):589-99.
Ma C. Papermaster D. Cepko C. 1998. A unique pattern of photoreceptor degeneration in cyclin
D1 mutant mice. Proc. Natl. Acad. Sci. USA. 95(17): 9938-43.
Marc RE, Jones BW. 2003. Retinal remodeling in inherited photoreceptor degenerations. Mol
Neurobiol. 28(2):139-47.
41
Matsumoto K, Moriuchi T, Koji T, Nakane PK. 1987. Molecular cloning of cDNA coding for rat
proliferating cell nuclear antigen (PCNA)/cyclin. EMBO J. 6(3):637-42.
McCall MJ, Robinson SR, Dreher B. 1987. Differential retinal growth appears to be the primary
factor producing the ganglion cell density gradient in the rat. Neurosci Lett. 79(1-2):78:84.
Messersmith EK, Redburn DA. 1992. gamma-Aminobutyric acid immunoreactivity in multiple
cell types of the developing rabbit retina. Vis Neurosci. 8(3):201-11.
Michalakis S, Geiger H, Haverkamp S, Hofmann F, Gerstner A, Biel M. 2005. Impaired opsin
targeting and cone photoreceptor migration in the retina of mice lacking the cyclic nucleotide-
gated channel CNGA3 (in press). Invest Ophthalmol Vis Sci. 46(4):1516-24.
Milam AH, Li ZY, Cideciyan AV, Jacobson SG. 1996. Clinicopathologic effects of the Q64ter
rhodopsin mutation in retinitis pigmentosa. Invest Ophthalmol vis Sci. 37(5):753-65.
Milam AH, Li ZY, Fariss RN. 1998. Histopathology of the human retina in retinitis pigmentosa.
Prog Ret Eye Res. 17(2):175-205.
42
Mohand-Said S, Hicks D, Dreyfus H, Sahel JA. 2000. Selective transplantation of rods delays
cone loss in a retinitis pigmentosa model. Arch Ophthalmol. 118(6):807-11.
Molday RS, MacKenzie D. 1983. Monoclonal antibodies to rhodopsin: characterization, cross-
reactivity, and application as structural probes. Biochemistry. 22(3):653-60.
Ooto S, Akagi T, Kageyama R, Akita J, Mandai M, Honda Y, Takahashi M. 2004.
Potential for neural regeneration after neurotoxic injury in the adult mammalian retina. Proc Natl
Acad Sci U S A. 101(37): 13654-9.
O'Steen WK, Shear CR, Anderson KV. 1972. Retinal damage after prolonged exposure to visible
light. A light and electron microscopic study. Am J Anat. 134(1):5-21.
O'Steen WK, Anderson KV, Shear CR. 1974. Photoreceptor degeneration in albino rats:
dependency on age. Invest Ophthalmol. 13(5):334-9.
Puller C, Haverkamp S. 2011. Cell-type-specific localization of protocadherin β16 at AMPA
and AMPA/Kainatereceptor-containing synapses in the primate retina. J Comp Neurol.
43
519(3):467-79.
Punzo C, Kornacker K, Cepko CL. 2009. Stimulation of the insulin/mTOR pathway delays cone
death in a mouse model of retinitis pigmentosa. Nat Neurosci. 12(1):44-52.
Rakoczy PE, Brankov M, Fonceca A, Zaknich T, Rae BC, Lai CM. 2003. Enhanced recombinant
adeno-associated virus mediated vascular endothelial growth factor expression in the adult
mouse retina: a potential model for diabetic retinopathy. Diabetes. 52(3):857-63.
Ray A, Sun GJ, Chan L, Grzywacz NM, Weiland J, Lee EJ. 2010. Morphological alterations in
retinal neurons in the S334ter-line3 transgenic rat. Cell Tissue Res. 339(3):481-91.
Ripps H. 2002. Cell death in retinitis pigmentosa: gap junctions and the 'bystander' effect. Exp
Eye Res. 74(3):327-36.
Roberts MR, Hendrickson A, McGuire CR, Reh TA. 2005. Retinoid X receptor (gamma) is
necessary to establish the S-opsin gradient in cone photoreceptors of the developing mouse retina.
Invest Opthalmol Vis Sci. 46(8):2897-904.
Rohrer B, Lohr HR, Humphries P, Redmond TM, Seeliger MW, Crouch RK. 2005. Cone opsin
44
mislocalization in Rpe65-/- mice: a defect that can be corrected by 11-cis retinal. Invest
Ophthalmol Vis Sci. 46(10):3876-82.
Rose TM, Bruce AG. 1991. Oncostatin M is a member of a cytokine family that includes
leukemia-inhibitory factor, granulocyte colony-stimulating factor, and interleukin 6. Proc Natl
Acad Sci USA. 88(19):8641-5.
Rosenfeld PJ, Cowley GS, McGee TL, Sandberg MA, Berson EL, Dryja TP. 1992. A null
mutation in the rhodopsin gene causes rod photoreceptor dysfunction and autosomal recessive
retinitis pigmentosa. Nat Genet. 1(3):209-13.
Rossi EA, Chung M, Dubra A, Hunter JJ, Merigan WH, Williams DR. 2011.
Imaging retinal mosaics in the living eye. Eye (Lond). 25(3):301-8.
Sato K, Nakazawa M, Takeuchi K, Mizukoshi S, Ishiguro S. 2010. S-opsin protein is
incompletely modified during N-glycan processing in Rpe65(-/-) mice. Exp Eye Res. 91(1):54-
62.
Schiviz AN, Ruf T, Kuebber-Heiss A, Schubert C, Ahnelt PK. 2008. Retinal cone topography of
artiodactyl mammals: influence of body height and habitat. J Comp Neurol. 507(3):1336-50.
Sigulinsky CL, Green ES, Clark AM, Levine EM. 2008. Vsx2/Chx10 ensures the correct timing
45
and magnitude of Hedgehog signaling in the mouse retina. Dev Biol. 317(2):560-75.
Tan CK, Castillo C, So AG, Downey KM. 1966. An auxiliary protein for DNA polymerase-delta
from fetal calf thymus. J Biol Chem. 261(26):12310-6.
Tanito M, Kaidzu S, Anderson RE. 2007. Delayed loss of cone and remaining rod photoreceptor
cells due to impairment of choroidal circulation after acute light exposure in rats. Invest
Ophthalmol Vis Sci. 48(4):1864-72.
Tansley K. 1933. The formation of rosettes in the rat retina. Br J Ophthlmol. 17(6):321-36.
Ts'o MO, Zimmerman LE, Fine BS. 1970. The nature of retinoblastoma. I. Photoreceptor
differentiation: a clinical and histopathologic study. Am J Ophthalmol. 69(3):339-49.
van de Pavert SA, Kantardzhieva A, Malysheva A, Meuleman J, Versteeg I, Levelt C, Klooster J,
Geiger S, Seeliger MW, Rashbass P, Le Bivic A, Wijnholds J. 2004. Crumbs homologue 1 is
required for maintenance of photoreceptor cell polarization and adhesion during light exposure. J
Cell Sci. 117(Pt 18):4169-77.
Villarroel M, Garcí a-Ramí rez M, Corraliza L, Herná ndez C, Simó R. 2009.
46
Effects of high glucose concentration on the barrier function and the expression of tight junction
proteins in human retinal pigment epithelial cells. Exp Eye Res. 89(6):913-20.
Waseem NH, Lane DP. 1990. Monoclonal antibody analysis of the proliferating cell nuclear
antigen (PCNA). Structural conservation and the detection of a nucleolar form. J Cell Sci.
96(Pt1): 121-9.
Wä ssle H, Regus-Leidig H, Haverkamp S. 2006. Expression of the vesicular glutamate
transporter vGluT2 in a subset of cones of the mouse retina. J Comp Neurol. 496(4):544-55.
Wintersteiner, H. (1897). Das Neuroẽpithelioma retinae: Eine anatomische und klinische Studie.
Leipzig: Franz Deuticke.
Xu J, Morris L, Fliesler SJ, Sherry DM, Ding XQ. 2011. Early Onset, Slow Progression of Cone
Photoreceptor Dysfunction and Degeneration in CNG Channel Subunit CNGB3 Deficiency.
Invest Ophthalmol Vis Sci. 52(6): 3557-66.
Yamamoto C, Ogata N, Yi X, Takahashi K, Miyashiro M, Yamada H, Uyama M, Matsuzaki K.
1996. Immunolocalization of basic fibroblast growth factor during wound repair in rat retina
after laser photocoagulation. Graefes Arch Clin Exp Ophthalmol. 234(11):695-702.
47
Zabouri N, Bouchard JF, Casanova C. 2011. Cannabinoid receptor type 1 expression during
postnatal development of the rat retina. J Comp Neurol. 519(7):1258-80.
Zech JC, Pouvreau I, Cotinet A, Goureau O, Le Varlet B, de Kozak Y . 1998. Effect of cytokines
and nitric oxide on tight junctions in cultured rat retinal pigment epithelium. Invest Ophthalmol
Vis Sci. 39(9): 1600-8.
Zeiss CJ, Johnson EA. 2004. Proliferation of microglia, but not photoreceptors, in the outer
nuclear layer of the rd-1 mouse. Invest Ophthalmol Vis Sci. 45(3):971-6.
Zhang H, Fan J, Li S, Karan S, Rohrer B, Palczewski K, Frederick JM, Crouch RK, Baehr W.
2008. Trafficking of membrane-associated proteins to cone photoreceptor outer segments
requires the chromophore 11-cis-retinal. J Neurosci. 28(15):4008-14.
Zhu X, Brown B, Li A, Mears A, Swaroop A, Craft CM. 2003. GRK1-dependent
photophorylation of S and M opsins and their binding to cone arrestin during cone
phototransduction in the mouse retina. J Comp Neurol. 23(14):6152-6160.
48
FIGURE LEGENDS
Figure 1. Confocal micrographs taken from vertical sections of retinas processed for M-opsin
immunoreactivity. The micrographs are for P30 N (A), P15RP (B), P30RP (C), and P90 RP (D).
In P30 N retinas, entire M-opsin-immunoreactive cones are labeled. In P15 RP retinas, the OS
are distorted in orientation (arrow). In P30 RP retinas, M-opsin immunoreactive cones are
shortened in length and show disorganized axon terminals (C). In P90 RP retinas, M-opsin-
immunoreactive cones are positioned ‗flat‘ against the outer part of the INL. ONL, outer nuclear
layer; OPL, outer plexiform layer; INL, Inner nuclear layer; OS, outer segment; N, normal; RP,
retinitis pigmentosa. All scale bars = 20 µ m.
Figure 2. Confocal micrographs taken from whole mounts processed for M-opsin and S-opsin
immunoreactivities. Low-power micrographs illustrate the homogeneous distributions of M-
opsin (A) and S-opsin (B) cones in P90 normal retina. Double exposure (C) demonstrates no co-
localization of M-opsin and S-opsin immunoreactivity. Low-power micrographs show that M-
opsin (D) and S-opsin (E) cones in P90 RP retinas exhibit spatial organizations in matrices of
rings. Double exposures (F) demonstrates that both types of cones form rings at the same
locations in the RP retinas. High-power micrographs of part of a ring marked with inset
rectangles in D, E, F, are shown in G, H, I, resepctively. The orientation of M-opsin (G) and S-
opsin (H) immunoreactive cones in rings are shown. Double exposures (I) demonstrates the same
orientation of M-opsin and S-opsin cones. All scale bars = 100 µ m.
Figure 3. Confocal micrograph taken from P30 (A-C) and P90 (D-F) whole mount RP retinas
49
processed for M-opsin immunoreactivities showing rings in their distribution (A, D). Light
micrograph taken at the same retinal location under DIC mode shows no retinal folds (B, E).
Double exposure (C, F) confirms no retinal folds are associated with rings. Light micrographs
taken from RP vertical retinas processed with Hematoxylin staining (G- J). At P15, the thickness
of the ONL is uniform (G). H: Higher-power micrograph of G is shown. At P30, the ONL show
‗grooves‘. J: Higher -power micrograph of groove is shown. No nuclei are visible at the trough
of the groove (arrow). Confocal micrograph taken from P30 whole mount RP retina processed
with M-opsin antibody (red) and TOPRO-3 (blue). Nuclei at the center of the ring are not in the
same focal plane as M-opsin-immunoreactive cones (K). L: Higher-power micrograph of K is
shown. DIC, differential interference contrast; ONL, outer nuclear layer; INL, inner nuclear
layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar = 100 µ m (A-J); 50 µ m (K,
L).
Figure 4. Confocal micrograph taken from P15 whole mount RP retinas processed for M-opsin
(A), rhodopsin (B) immunoreactivities and for apoptotic cells (C). Triple exposure (A, B, C)
indicates a cluster of apoptotic cells inside the ring. Where rings are not observed, apoptotic cells
are scattered randomly. Scale bar = 100 µ m.
Figure 5. Confocal micrograph taken from whole mount RP retina processed for M-opsin (red)
and S-opsin (green) immunoreactivities at P15 (A, B), 30 (C, D), 180 (E, F) and P600 (G, H).
Double exposure shows a ring of M-opsin and S-opsin immunoreactive cone at P15 (A). A
higher-power micrograph of a ring is shown (B). It illustrates the change of orientation of cones;
50
starting to lie flat with their processes pointing towards the center of the ring. Many rings are
visible by P30 (C). D: Higher-power micrograph of a ring is shown (C). All the COS and the cell
bodies are near the rims of the rings whereas the processes are pointing towards the center of the
ring. At P180, rings start to lose their form (E). Higher-power micrograph of a part of what
probably used to be a ring reveal M-opsin- and S-opsin-immunoreactive cones are no longer
organized in the previously observed orientation (F). A lot of cones show growth of abnormal
processes and loss of their OS. At P600, rings are no longer clear (G). Higher-power micrograph
illustrates the cones‘ extensive growth of processes and their loss of OS (H). All scale bars = 100
µ m.
Figure 6. Composite image of confocal micrographs taken from the whole mount RP retinas
processed for M-opsin immunoreactivities at P30 (A) and at P90 (B). At P30, there are
comparatively more rings in the dorso-peripheral region of the retina. At P90, rings are seen
throughout the entire retina. A graph of the mean total number of rings versus retinal regions of
the P30 RP retinas (n = 3) suggest significantly greater number of rings in the dorsal region of
the retinas compared to the ventral region (C). A graph of mean total number of rings versus
postnatal age (n = 3; D) and a graph of mean diameter of rings (µ m) versus postnatal age (n = 2;
E) indicate rings grow both in their number and size from P30 to P90. Data presented as mean ±
standard error. The symbol * represents p < 0.005 or better. All scale bars = 1 mm.
Figure 7. A graph of mean total number of immunoreactive cells versus postnatal age (A)
shows no significant differences between the normal and the RP retinas at P30 (n = 2) and P180
51
(n = 3). Significant reduction is seen in P600 RP retinas (n = 3) for both M-opsin- and S-opsin-
immunoreactive cone counts. A graph of RP retinal area (mm
2
) versus postnatal age indicates
growth of retina in size with age (B – P30, n = 3; P180, n = 2; P600, n = 3). There are no
significant differences in the retinal area between the normal and the RP retinas. Confocal
micrographs taken from whole mount heterozygous (left) and homozygous (right) RP retinas
processed for M-opsin immunoreactivities (C). Both show the same morphology, arrangement
and orientation of M-opsin-immunoreactive cones – the COS forming the rim of the ring and
their processes in the inside of the ring. The densities of M-opsin-immunoreactive cones cell
bodies in the dorsal wing of P180 heterozygous and homozygous RP retinas (n = 4 each)
indicated no significant difference (D). Data presented as mean ± standard error. The symbol *
represents p < 0.005 or better. Scale bar = 100 µ m.
52
Spatiotemporal Pattern of Rod Degeneration in the Retina of the
Rat Model of Retinitis Pigmentosa
ABSTRACT
In the first chapter, remodeling of surviving cones in retinitis pigmentosa (RP) retinas into an
orderly array of rings was explained. Here, we report the spatiotemporal pattern of healthy rods,
their relationship with dying rods and the way that rod death stimulates the modification of cone
spatial-distribution patterns and Mü ller-glia processes in the S334ter-line-3 rat, a transgenic
model expressing a rhodopsin mutation that causes RP. The spatial patterns of rods, cones,
microglial and Mü ller cells were labeled by immunocytochemistry with cell-type-specific
markers at various stages of development in rat whole-mount retinas. Spatial patterns of dying
cells were examined by TUNEL staining. The S334ter rod mosaic began to develop small holes
around postnatal day 10. These hot-spots of cell death progressively increased in size, leaving
larger rod-less holes behind. The holes were temporarily occupied by active microglial cells,
before being replaced by remodeled Mü ller-cell processes. Our data suggest that the hot spots of
rod death create holes in the rod mosaic early in retinal degeneration and that the resulting
pattern triggers the modification of the spatial-distribution patterns of cones and glia cells.
Keywords: Photoreceptor mosaics Cell death Glia cells Retina
53
INTRODUCTION
Retinitis Pigmentosa (RP) is characterized by rod degeneration in the early stages, then cone
degeneration followed by complete blindness (Berger et al. 2010). In many RP models, rod loss
occurs evenly throughout the retina, such that eventually all rods degenerate (Bowes et al. 1990;
Humphries et al. 1997). However, focal degeneration of rods can be seen in some animal models
of RP, e.g., retinal degenerative mice (Huang et al. 1993) and cyclin D1-deficient mice (Ma et
al. 1998). Similarly, most cases of human RP patients begin with a focal loss of visual acuity (a
scotoma). The scotoma usually enlarges, especially in the mid-peripheral retina and may lead to
the destruction of the whole retina (Berson 1993). However, none of these studies have described,
in detail, the spatiotemporal progression of deaths and the rearrangement of the glial cells during
the focal degeneration of rods.
In the retina, Mü ller and microglial cells maintain homeostasis (Bringmann et al., 2006),
serving the functions assumed by oligodendrocytes and ependymal cells in other parts of the
central nervous system (CNS; Newman and Reichenbach, 1996). Furthermore, Mü ller cells have
been suggested to provide structural support in the retina (Rich et al. 1995). In retinal
degeneration, Mü ller cells begin to hypertrophy; their processes fill the space left by dying
photoreceptors and grow along the exposed outer surface of the retina (Lewis et al. 1995; Ma et
al. 1998; Lewis and Fisher, 2000; Marc and Jones, 2003; Marc et al., 2003; Gargini et al., 2007;
Ray et al., 2010). Such remodeling is not surprising, as Mü ller cells contain trophic receptors for
most of the molecules involved in photoreceptor rescue (Harada et al. 1995; Wen et al. 1995). In
turn, microglia cells are the immunocompetent cells throughout the CNS, maintaining
homeostasis by scavenging viruses, bacteria, dead cells and cell debris (Glybina et al. 2010).
54
Accumulating evidence suggests that chronic-microglia activation is associated with various
neurodegenerative diseases, including the retinal dystrophies (Schuetz and Thanos 2004a, 2004b).
Several studies have now demonstrated early microglia activation in animal models of inherited
photoreceptor degeneration (Zeiss and Johnson 2004; Zeng et al. 2005; Gehrig et al. 2007). Zeiss
and Johnson (2004) have shown microglia migration and proliferation in the outer nuclear layer
(ONL) of retinal-degeneration mice, suggesting that interactions between photoreceptors and
glial cells do indeed occur during degeneration.
In this study, we have investigated the spatiotemporal patterns of rod, Mü ller and
microglia cells in developing S334ter-line-3 rat retinas.
MATERIALS AND METHODS
Animals
The third line of albino Sprague Dawley rats homozygous for the truncated murine opsin gene
(stop codon at residue 334; S334ter-3) was obtained from Dr. Matthew M. LaVail (University of
California, San Francisco, Calif., USA). The mutation in the S334ter-line-3 retina triggers the
cell death of rods, similar to a mutation that causes human RP. These rats were mated with
normal Copenhagen rats to produce the S334ter transgene heterozygous rats that were used in
this study. For simplicity, we will refer to these heterozygous offspring rats as ―RP rats‖ . The RP
rats were killed at post-natal (P) days P8-P28 (n=5 for each stage). Controls were age-matched
normal Sprague Dawley rats (n=5 for each stage; Harlan, Indianapolis, Ind., USA). All rats were
55
housed under cyclic 12/12-h light/dark conditions with free access to food and water. Both sexes
of normal (control) and RP rats were used. All procedures conformed with the Guide for Care
and Use of Laboratory Animals (National Institutes of Health). The University of Southern
California Institutional Animal Care and Use Committee reviewed and approved all procedures.
Tissue Preparation
Animals were deeply anesthetized by intra-peritoneal injection of pentobarbital (40 mg/kg body
weight) and the eyes were enucleated. After enucleation, animals were killed with an overdose of
sodium pentobarbital. The anterior segments of the eyes were removed and the eyecups were
fixed by immersion in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4, for 1.5–2 h.
Following fixation, the retinas were dissected and transferred to 30% sucrose in PB for 24 h at
4° C. All retinas (for cryostat sections and whole-mounts) were then frozen in liquid nitrogen and
stored at −70°C. Before being processed, retinas were thawed and rinsed in 0.01 M phosphate-
buffered saline (PBS; pH 7.4).
Immunohistochemistry
For fluorescence immunocytochemistry, whole-mount retinas were incubated in 10% normal
goat serum (NGS) or normal donkey serum (NDS) and 1% Triton X-100 in PBS for 1 h at room
temperature. Whole-mounts were then incubated for two nights with a rabbit polyclonal antibody
directed against glial fibrillary acidic protein (GFAP; Sigma, dilution 1:500), ionized calcium-
binding adaptor molecule (Iba1; Wako, dilution 1:1000), red/green opsin (M-opsin; kindly
56
provided by Dr. Cheryl Craft from the Doheny Eye Institute, University of Southern California,
dilution 1:1000), mouse monoclonal antibody directed against glutamine synthetase (GS;
Chemicon, dilution 1:100) and rhodopsin (rho 1D4; kindly provided by Dr. Biju Thomas from
the Doheny Eye Institute, dilution 1:1000). Each antiserum was diluted with PBS containing
0.5% Triton X-100 at 4° C. The retinas were then washed in PBS for 45 min (3× 15 min).
Afterwards, we incubated the retinas for 1 day in carboxymethylindocyanine (Cy3)-conjugated
affinity-purified donkey anti-rabbit IgG (Jackson Immuno Labs, West Grove, Pa., USA, dilution
1:500); Alexa 488 anti-mouse (Molecular Probes, Eugene, Ore., USA, dilution 1:300) at 4° C.
Finally, whole-mounts were washed for 30 min with 0.1 M PB and cover -slipped with
Vectashield mounting medium (Vector Labs, Burlingame, Calif., USA).
For double-labeling, whole-mounts were incubated in a mixture of the following antibodies:
rhodopsin with GFAP or Iba-1; Iba-1 and GS; GS and GFAP. Whole-mounts were then rinsed in
PBS and incubated with appropriate secondary antibodies. Following this incubation, some of
the whole-mounts were stained with TUNEL, washed for 30 min with 0.1 M PB and cover -
slipped with Vectashield mounting medium. Whole-mounts were then analyzed by using a Zeiss
LSM 510 (Zeiss, N.Y., USA) confocal microscope. Immunofluorescence images were processed
with Zeiss LSM-PC software. The brightness and contrast of the images were adjusted by using
Adobe Photoshop 7.0 (Adobe Systems, Mountain View, Calif., USA) for clarity. All Photoshop
manipulations (brightness and contrast only) were carried out equally across sections.
57
Hematoxylin staining
Eyecups were embedded in OCT embedding medium (Tissue-Tek, Elkhart, IN, USA) and fast-
frozen in liquid nitrogen. Embedded eyecups were sectioned along the vertical meridian at a
thickness of 10 m on a cryostat. Sections were collected on gelatin-coated slides for
hematoxylin staining: slides were dipped in hematoxylin stain for 5 min, washed in tap water,
dehydrated in alcohol, cleared in xylene, and mounted in xylene-based medium (Richard-Allan
Scientific, Kalamazoo, MI, USA).
TUNEL Staining
Cell death was visualized by a modified terminal deoxynucleotidyl transferase-mediated dUTP
nick-end labeling (TUNEL) technique, according to the manufacturer‘s instructions (In Situ Cell
Detection kit, Boehringer Mannheim, Germany). Whole-mounts were incubated with proteinase
K (10 μg/ml in 10 mM TRIS/HCl, pH 7.4 –8.0) for 10 min at 37° C. After being rinsed in PBS,
the sections were incubated with TUNEL reaction mixture (terminal deoxynucleotidyl
transferase plus nucleotide mixture in reaction buffer) for 90 min at 37° C. Whole-mounts were
then washed again for 30 min with 0.1 M PB and cover -slipped with Vectashield mounting
medium.
Quantification and Statistical Analyses of Rod Holes
The size (n=2 each for all stages) of holes formed by rhodopsin-immunoreactive rods in RP
retinas was measured at P10, P14, P17 and P21. For each P10 RP retina, all holes within a
58
100× 100 μm2area near the optic disc were measured. For each P14, 17 and P21 retina, the hole
sizes were measured in two areas: near the optic disc area (within 2-mm steps from the optic
disc) and peripheral area (2-mm steps away from optic disc). The size of a hole was estimated
from the lengths of chords measured from the center of one hole-boundary cell body to the center
of another, as far as possible, across the hole. Holes were not perfectly circular in shape, so the
―diameter‖ of each hole was estimated by averaging the largest and the smallest chord
measurements. All hole measurements were taken by using Zeiss LSM Image Browser Software.
The estimated diameters for each stage (P10, P14, P17 and P21) were then expressed as means ±
standard deviations. Analysis of variance (ANOVA) and post-hoc Student-Newman-Keuls
(SNK) tests were used to evaluate the significance of hole sizes across postnatal stages.
Statistical computations were performed in MATLAB version 7.4.0 (The MathWorks, Natick,
Mass., USA) and all graphs were generated by Microsoft Excel spreadsheet, 2010 (Microsoft
Corporation, Redmond, Wash., USA). A difference between the means of separate experimental
conditions was considered statistically significant at P < 0.05.
RESULTS
Changes in vertical sections of the outer nuclear layer (ONL) of S334ter-line-3 rat retina
We began by examining the changes in the ONL attributable to photoreceptor degeneration in
vertical sections of S334ter-line-3 retinas. To study these changes, hematoxylin staining was
performed in P15 normal (Fig. 1a), P15 RP (Fig. 1b) and P21 RP (Fig. 1c) retinas. The thickness
of the RP ONL was less than half that of age-matched normal ONLs, similar to the findings of
59
another S334ter-line-3 heterozygous study (Liu et al. 1999). As opposed to the tightly packed,
organized alignment of photoreceptor nuclei in P15 normal ONLs (Fig. 1a), P15 RP retinas
showed many loosely filled spaces (Fig. 1b, arrows) and a disorganized alignment of nuclei that
continued into P21 (Fig. 1b, c). At the P21 stage, the RP retinas had an uneven thickness and
many zones with few observed cell bodies (Fig. 1c, arrow).
Fig. 1 Changes in the outer nuclear layer (ONL) of S334ter-line-3 rat retina
Rhodopsin-immunoreactive rods in developing RP retinas
We examined rhodopsin-immunoreactive rods in vertical sections of normal and RP retinas at
P10, P15 and P21, taking images from the central area of the sections. In normal retinas at P10,
P15 (data not shown) and P21 (data not shown), we observed rhodopsin immunoreactivity in the
segments and cell bodies of rods (Fig. 2a). Rhodopsin immunoreactivity was similar to that seen
in the normal retinas (Fig. 2a) in P10 RP (Fig. 2b) but by P15 RP (Fig. 2c), rhodopsin
immunoreactivity in the ONL was reduced compared with that at P10 RP. P21 RP retinas
showed some rods positioned ―flat‖ against the outerplexiform layer (OPL). We also observed
two distinct regions that alternated repeatedly: regions that were full of clusters of cell bodies
and regions that were devoid of cell bodies but that had some processes (Fig. 2d, e).
60
Fig. 2 Rhodopsin-immunoreactive rods in developing RP retinas
Spatiotemporal pattern of rods in S334ter-line-3 rat retinas
To characterize further the abnormal features observed in the ONL of the vertical sections, we
investigated the distribution of rods in whole-mount retinas. Whole-mount RP retinas were
stained with rhodopsin antibodies to visualize the rod spatial pattern in early postnatal stages
(Fig. 3). As previously established by using vertical sections of retinas, rods in our RP model
began to degenerate at P8-P11 and had completely died off by P25-P30 (Liu et al. 1999;
Sagdullaev et al. 2003; Li et al.2010; Martinez-Navarrete et al. 2011; Ray et al. 2010).
Accordingly, we gathered our experimental retinal samples between P8 and P24. Figure 3 shows
61
P15 normal (Fig. 3a), P10 RP (Fig. 3b, e), P14 RP (Fig. 3f), P17 RP (Fig. 3c, g) and P21 RP
(Fig. 3d, h) retinas. Findings are summarized in a series of images of the superior strip of
representative retinas, shown with the optic disc to the left and periphery to the right. The P15
normal retina showed a homogeneous distribution of rods across the retina (Fig. 3a). We
observed a similar pattern in P8-P24 of normal retinas (data not shown). In RP retinas, the
earliest apparent changes in the rod mosaic were the spatially random formation of small ―holes‖
found at P10 (Fig. 3b, e). At P14, larger and roughly circular holes began to form in the rod
mosaic (Fig. 3f). As RP progressed (P17), these holes became larger, and the rhodopsin-
immunoreactive cell bodies could be faintly seen within rings, just inside the perimeter of each
hole (Fig. 3c, g). By P21, many of the photoreceptors near the optic disc had degenerated.
Labeling showed a greater density of rods in the peripheral region of the retina (Fig. 3h) but
regions nearest to the optic disc were too sparse to form distinguishable cell distribution patterns.
Many rods did not have intact outer segments (Fig. 3d, h). This finding is consistent with the
S334ter line being an optic-disc-to-periphery degeneration model (Liu et al. 1999; Ray et
al. 2010). However, the rod-holes were still observed in peripheral regions of the retina,
demonstrating that the various distribution patterns of rods overlapped spatially and temporally
across a single retina. For example, the superior-peripheral region of a P14 RP retina could
present large rod-holes, whereas the retina could still exhibit small holes in other regions
(Fig. 3f). Likewise, in later stages of the disease, sparse and scattered cells were observed as
large rod-holes within the same retina (Fig. 3h).
62
Fig. 3 Spatiotemporal pattern of rods in developing S334ter-line-3 rat retinas
Holes in RP retinas increase significantly in diameter over time
The development of large rod-holes during early postnatal stages, i.e., at P14, P17 and P21
(Fig. 4a–c) was examined by measuring the diameter of the holes (Fig. 4d). The mean diameter
of rod-holes at P14 was 70± 10 μm, whereas those at P17 and P21 were 110±30 μm and
63
130± 30 μm, respectively (Fig. 4e). ANOVA statistical analysis (P<0.05) revealed a significant
increase in hole size as the disease progressed from P14 to P21. No statistical comparisons were
made between P10 and the three other stages, because no definitive large rod-holes occurred in
P10 RP retinas. The mean diameter of the small holes observed in P10 was found to be
6.1± 0.6 μm.
Fig. 4 Holes in RP retinas increase significantly in diameter over time
Spatial correlation between rods and dying rods
We examined the spatial relationships between surviving rods and dying rods. Figure 5 shows
examples of this relationship in P14 RP (Fig. 5a–c), P17 RP (Fig. 5d–f) and P21 RP (Fig. 5g–i)
whole-mounts. We processed these whole-mounts for rhodopsin (RHO in Fig. 5a, d, g) and
TUNEL (Fig. 5b, e, h) labeling. At P14, hot spots of cell death, represented by clusters of cell
64
death in rod-holes (Fig. 5a), could be distinguished from the otherwise ―random‖ distribution of
cell death (Fig. 5b, c). Each TUNEL-stained cluster was concentric with a rod-hole (Fig. 5c). As
holes became more numerous by P17 (Fig. 5d), more hot spots of clustered cell death filled each
hole (Fig. 5e) and the spatial correlation between cell-death clusters and rod-holes was well
established (Fig. 5f). At P21, rod-holes were enlarged (Fig. 5g) and the clusters of cell death
propagated out radially and became rings (Fig. 5h), seen just inside the perimeter of each rod-
hole (Fig. 5i). Overall, TUNEL-staining analysis suggested that small and large rod-holes were
both formed from voids that dying rods left behind. Additionally, cell death began as hot spots of
clusters and then formed rings as the disease progressed.
65
Fig. 5 Spatial correlation between rods and dying rods
Microglial cells fill the holes of rods in the early stage of RP
In the retina, microglial cells are normally located in the inner retina (Bringmann et al. 2006).
However, in inherited retinal degeneration, microglial cells migrate to the outer retina, where
66
photoreceptor degeneration occurs (Thanos 1992; Roque et al. 1996). These findings are not
surprising, given that microglia are common in areas with extensive neuronal cell death (Hume
et al. 1983; Wong and Hughes 1987). Deaths of neurons have been proposed as a factor that
attracts microglial precursors into the CNS (Perry and Gordon 1991; Moore and Thanos 1996).
In our RP model, we observed clusters of cell death (Fig. 5) forming holes in the rod mosaic.
Accordingly, we tested the hypothesis that microglial cells could be aggregating inside these
holes, attracted by neuron death. For this test, we performed double-labeling experiments with
antibodies against rhodopsin (Fig. 6a, d, g) and Iba1 (Fig. 6b, e, h), a marker expressed in resting
and in activated phagocytic microglial cells (Graeber et al. 1998; Ito et al. 1998, 2001; Nakajima
et al. 1998). In RP retina, we observed rod-holes (Fig. 6a, d, g) and the clustering of the
microglial cells (Fig. 6b, e). Double labeling of rhodopsin and Iba1 showed that microglial cells
were occupying the holes and some could even be seen phagocytozing cell bodies (Fig. 5c, f). By
P21 in the RP retina, active microglial cells had become absent in the centers of the holes
(Fig. 6i).
67
Fig. 6 Microglial cells fill the holes of rods in the early stage of RP
68
Remodeled Mü ller-cell processes fill the center of rod holes after the disappearance of
active microglial cells
Like microglial cells, Mü ller cells have been found to activate and remodel their processes in
several photoreceptor degeneration models (Lewis et al. 1995; Ma et al. 1998; Lewis and
Fisher 2000; Marc and Jones 2003; Marc et al. 2003; Gargini et al. 2007; Ray et al. 2010; Lee et
al. 2011; Ji et al.2012). A typical consequence of Mü ller-cell remodeling during retinal
degeneration is a glial reaction involving the up-regulation of GFAP (Roque and Caldwell 1990;
DiLoreto et al. 1995; Tanihara et al.1997). To explore the behavior of Mü ller cells in our RP
model, whole-mounts were labeled with antibodies to GS, a marker of Mü ller-cells and GFAP.
The results demonstrated the up-regulation of GFAP by Mü ller cells similar to that found
previously (Ray et al. 2010; Lee et al. 2011). Figure 7shows an example of P21 normal (Fig. 7a)
and P21 RP (Fig. 7b–l) retinal whole-mounts. Several observations were made from an
examination of the focal plane at the outer limiting membrane (OLM) in such figures. In the
normal retina, GS immunoreactivity displayed the normal, spatially homogeneous mesh network
of Mü ller-cell processes (Fig. 7a). In contrast, RP retinas showed clusters/bunches of processes
of Mü ller cells that resembled broccoli florets (Fig. 7b). GFAP immunoreactivity was also
observed in the processes of Mü ller cells (Fig. 7c). Double labeling of GS (Fig. 7d) and GFAP
(Fig. 7e) revealed that GFAP immunoreactivity was present in the region in which the Mü ller-
cell processes exhibited clustered outgrowth (Fig. 7f). The insets in Fig. 7d–f show a higher
magnification view of the same focal plane of the retina as is shown in Fig. 7b, c. Double
labeling of GS (Fig. 7g) and Iba1 (Fig. 7h) indicated that the remodeled Mü ller-cell processes
filled the center of holes after the activated microglial cells had disappeared from the area
(Fig. 7i). We also performed double labeling for rhodopsin (Fig. 7j) and GFAP (Fig. 7k) to
69
confirm that these Mü ller-cell processes were indeed filling the rod-holes. Thus, our results
established that Mü ller-cell processes filled the holes left in the rod mosaic (Fig. 7l). Furthermore,
the data in Fig. 7 suggested that the remodeling of Mü ller-cell processes filling the regions of
low rod density in the ONL occurred only after the activated microglia had disappeared (Fig. 7g–
l).
Fig. 7 Active microglial cells and remodeled Mü ller-cell processes
70
Rod deaths trigger modification in the spatial-distribution patterns of cones
Whole-mount retinas were immunostained for rhodopsin (Fig. 8a, d) and M-opsin (Fig. 8b, e). In
P17 RP retinas (i.e., early on in the progression of RP), local zones of low density in rods and
cones were seen to co-localize (Fig. 8c, f). Such spatial correlation was consistent with our
previous study showing clusters of rod deaths within zones of cones at low density (Lee et
al. 2011; Ji et al. 2012).
Fig. 8 Rod deaths trigger modification in the spatial-distribution patterns of cones
71
DISCUSSION
Spatiotemporal pattern of rod death
In the present study, we have observed the expression of rhodopsin throughout the RP rod cell
(Figs. 2, 3, 4, 5, 6, 7, 8), an expression pattern that has also been found in rhodopsin transgenic
pigs (Li et al. 1998) and human retinas with RP (Fariss et al. 2000).
We observed dying rods between the first 2–4 weeks of development in RP animals
(Figs. 1, 2, 5) as described in previous studies in vertical retinal sections of the eyes of S334ter
rats (Liu et al. 1999; Sagdullaev et al. 2003; Li et al. 2010; Martinez-Navarrete et al. 2011; Ray et
al. 2010). Rod deaths first occurred in hot spots as opposed to in a uniform and scattered pattern
across the retina (Bowes et al. 1990; Humphries et al. 1997). These clusters of cell death were
embedded amongst a background of random cell death throughout the retina (Fig. 5b, e). Such
degeneration in hot spots suggests an inductive mechanism of cell death, consistent with human
and animal model studies demonstrating that a degenerating rod often leads to deaths in its
immediate neighbors (Huang et al.1993; Kedzierski et al. 1998). The rod deaths then propagated
and expanded radially away from the clusters (Fig. 5h) thereby creating holes in the mosaic of
surviving rods (Fig. 5a, d, g).
Our statistical analysis of large rod-hole diameters (Fig. 4e) further supported the
proposed pattern of expanding holes. Because the mean hole diameters increased significantly
from stages P14 to P21, the holes probably were initially small and expanded as degeneration
progressed, as opposed to them randomly appearing at a fixed-size. We thus hypothesize the
following temporal progression for rod-hole-pattern formation in a single RP rat retina: new rod-
72
holes appear at random times and positions throughout the earlier postnatal stages. As the animal
ages, each hole becomes increasingly larger following its own independent time course, resulting
in a mosaic of smaller holes interspersed with larger, previously formed holes. Independent hole
expansion creating holes with lifecycles ―out of phase‖ with each other would account for the
large variances in hole sizes seen within one retina.
In addition to the variability in the sizes of the holes in a single retina, we have found that
the time course of development is not exact and varies slightly amongst different animals, even
those from the same litter. We have reported that the large rod-holes appear at about P14 but the
appearance time has variabilities of ± 1–2 days within the same litter (data not shown). The
presence of the rod-hole pattern stages that have been described, like small holes and ―scattered‖
cells, can overlap spatially and temporally within the same retina (Fig. 3f–h). However, we can
nevertheless report that, generally, the progression of the spatiotemporal patterns described
above is consistent across various retinas: small holes (Fig. 3b, e) appear before large ones
(Fig. 3c, f, g) and both emerge before cells become scattered (Fig. 3d, h). However, the different
regions of a single retina might begin their progression through these patterns at different times.
The large rod-holes that we have observed are not a completely new finding. Our results
are congruous with data found in a previous study documenting a unique focal degeneration
pattern in cd1-deficient mice. Radially spreading cell death produces holes in the photoreceptor
layer (Ma et al.1998). The similarity in death patterns suggests that the same mechanism is
controlling the photoreceptor mosaics in both cd1-deficient mice and S334ter-line-3 rat
degeneration models. However, the previous study did not report the specific differentiation
between rods and cones or any of the earlier appearing, smaller rod-holes that we have observed.
73
Furthermore, this pattern of rod death has not been described in depth in any other type of animal
model of RP. This probably has at least two different reasons. First, the techniques used on the
retinal samples for analysis (i.e., the use of vertical sections versus whole-mounts) affect the
ability to observe the spatial distribution of cells across an entire retina. We have, for the first
time, examined, in depth, the rod distributions in an RP model by using whole-mounts instead of
vertical sections. Second, differences are known in the rate of degeneration among the different
types of RP models. For example, rd1 mice show a much faster degeneration rate compared with
our RP model (Carter-Dawson et al. 1978; Jimé nez et al. 1996). Thus, the rod-holes, if present
in rd1 mice, might appear for only a brief period of time, making it difficult temporally to pin-
point the relevant changes.
A possible mechanism responsible for this rod death pattern
The radial outward propagation of the large rod-holes leads to a few hypotheses for the
mechanisms that cause rod-mosaic change. For example, a single rod death attributable to the RP
mutation would cause a local deficit of survival factors (LaVail et al. 1998) and increase the
probability of its neighbors degenerating (Huang et al. 1993). The neighboring rod degeneration
would then cause the trophic factor deficiency to spread to a wider area. Alternately, dying rods
might release toxic factors, causing rods nearby to die (Nir et al. 1990; Kedzierski et al. 1998).
However, typical toxic factors such as fas ligand and tumor-necrosis factor have not been found
to be not over-expressed in or around the holes observed in cd1-deficient mice (Ma et al. 1998).
Another possibility involves the spread of intracellular second messengers and the gap junctions
that allow such signals to diffuse through neighboring rod cells. Indeed, evidence has been
74
presented showing that gap junctions mediate bystander cell death in the normal developing
retina (Cusato et al. 2003, 2006).
Activation of retinal glial cells in RP retina
Activated microglial cells clustered inside the rod-holes (Fig. 6c, f) and appeared, over time, to
take the rough pattern of expanding rings (Fig. 6i) that was also seen by TUNEL staining (Fig. 5c,
f, l). Microglia thus seem to ―follow‖ the propagation of rod death. Such precise temporal and
spatial positioning of activated microglial cells in relation to TUNEL-positive rods could be
attributable to interactions between microglia and dying neurons. Several other studies with
animal models of RP during early photoreceptor degeneration have observed activated microglia
migrating to and proliferating in the ONL, where they phagocytoze photoreceptor cell debris
(Gupta et al. 2003). These migrating activated microglia have also been found to secrete
cytokines, chemokines, neurotoxins and other molecules that are toxic signals to healthy
photoreceptors, leading to the ―b ystander effect‖ discussed above (Gupta et al. 2003). Overall, the
presence of activated microglia in our data could be a significant factor in the morphology and
expansion of RP rod-holes.
At P21, we observed that activated microglia began to disappear from inside the rod-
holes (Fig. 6g–i), while branching Mü ller-cell processes began to emerge into the holes (Fig. 7b–
e). Mü ller cells are known to react to most pathological changes in the retina, such as those
related to RP, with gliosis; gliosis can involve the normal function of Muller cells of supporting
75
photoreceptor survival through the release and mediation of neurotrophic factors (Bringmann et
al. 2006) but can also contribute to cell degeneration (Jablonski and Iannaccone 2000).
The consecutive hole occupations of microglia and Mü ller cells are perhaps the most interesting
finding of this study. The temporal correlation of the activation of these cells is an indication that
the functional roles of Mü ller cells and microglia are interdependent in RP. Indeed, previous
studies have suggested that activated microglial cells secrete factors that increase the production
of trophic factors by Mü ller cells, leading to increased photoreceptor survival (Harada et al. 2002;
Taylor et al. 2003; Bringmann et al. 2006).
Mü ller-cell processes might have another interesting effect in RP. The remodeled
processes of Mü ller cells might help to link rod-holes to the eventual remodeling of the cone
mosaic into rings of cones, which we saw co-localizing with the holes (Fig. 8). In a previous
study, we observed the way that Mü ller-cell processes surrounded cone cell bodies positioned
along the perimeters of the rings (Lee et al. 2011) and postulated that Mü ller cells control the
reorganization and maintenance of the cone mosaic in the later stages of RP (Lee et al. 2011).
Because we observed the cone rings in the same location as rod-holes with both formations
showing similar patterns of Mü ller cell gliosis, Mü ller cells might also contribute to the spatial
patterning of rod-holes, perhaps in an effort to act as a scaffold for the retina during neural
degeneration.
We show that the rod deaths caused by RP trigger multi-cellular interactions.
Furthermore, the commonality of the degeneration patterns (large rod-holes) in two different
degeneration models, namely cd1-deficient mice (Ma et al. 1998) and S334ter-line-3 rats,
76
suggests that the mechanisms controlling cell-death patterns in RP are relevant to retinal-
degenerative diseases in other species, perhaps even in humans. Ultimately, we hope that
research into these patterns of cell death and their spatial distribution across degenerative retinas
will lead to insights for developing RP therapies involving glial cell regulation.
77
REFERENCES
Berger W, Kloeckener-Gruissem B, Neidhardt J (2010) The molecular basis of human retinal
and vitreoretinal diseases. Prog Retin Eye 9:335–375
Berson EL (1993) Retinitis pigmentosa. The Friedenwald Lecture.Invest Ophthalmol Vis Sci
34:1659–1676
Bowes C, Li T, Danciger M, Baxter LC, Applebury ML, Farber DB (1990) Retinal degeneration
in the rd mouse is caused by a defect in the beta subunit of rod cGMP-phosphodiesterase. Nature
347:677–680
Bringmann A, Pannicke T, Grosche J, Francke M, Wiedemann P, Skatchkov SN, Osborne NN,
Reichenbach A (2006) Mü ller cells in the healthy and diseased retina. Prog Retin Eye Res
25:397–424
Carter-Dawson LD, LaVail MM, Sidman RL (1978) Differential effect of the rd mutation on
rods and cones in the mouse retina. Invest Ophthalmol Vis Sci 17:489–498
Cusato K, Bosco A, Rozental R, Guimarã es CA, Reese BE, Linden R, Spray DC (2003) Gap
junctions mediate bystander cell death in developing retina. J Neurosci 23:6413–6422
Cusato K, Ripps H, Zakevicius J, Spray DC (2006) Gap junctions remain open during
cytochrome c-induced cell death: relationship of conductance to ―bystander‖ cell killing. Cell
Death Differ 13:1707–1714
DiLoreto DA Jr, Martzen MR, Cerro C del, Coleman PD, Cerro M del (1995) Muller cell
78
changes precede photoreceptor cell degeneration in the age-related retinal degeneration of the
Fischer 344 rat. Brain Res 698:1–14
Fariss RN, Li ZY, Milam AH (2000) Abnormalities in rod photoreceptors, amacrine cells, and
horizontal cells in human retinas with retinitis pigmentosa.Am J Ophthalmol 129:215–223
Gargini C, Terzibasi E, Mazzoni F, Strettoi E (2007) Retinal organization in the retinal
degeneration 10 (rd10) mutant mouse: a morphological and ERG study. J Comp Neurol
500:222–238
Gehrig A, Langmann T, Horling F, Janssen A, Bonin M, Walter M, Poths S, Weber BHF (2007)
Genome-wide expression profiling of the retinoschisin-deficient retina in early postnatal mouse
development. Invest Ophthalmol Vis Sci 48:891–900
Glybina IV, Kennedy A, Ashton P, Abrams GW, Iezzi R (2010) Intravitreous delivery of the
corticosteroid fluocinoloneacetonide attenuates retinal degeneration in S334ter-4 rats.
Ophthalmol Vis Sci 51:4243–4252
Graeber MB, Ló pez-Redondo F, Ikoma E, Ishikawa M, Imai Y, Nakajima K, Kreutzberg GW,
Kohsaka S (1998) The microglia/macrophage response in the neonatal rat facial nucleus
following axotomy. Brain Res 813:241–253
Gupta N, Brown KE, Milam AH (2003) Activated microglia in human retinitis pigmentosa, late-
onset retinal degeneration, and age-related macular degeneration. Exp Eye Res 76:463–471
Harada T, Imaki J, Hagiwara M, Ohki K, Takamura M, Ohashi T, Matsuda H, Yoshida K (1995)
Phosphorylation of CREB in rat retinal cells after focal retinal injury. Exp Eye Res 61:769–771
79
Harada T, Harada C, Kohsaka S, Wada E, Yoshida K, Ohno S, Mamada H, Tanaka K, Parada LF,
Wada K (2002) Microglia-Mü ller glia cell interactions control neurotrophic factor production
during light-induced retinal degeneration.J Neurosci 22:9228–9236
Huang PC, Gaitan AE, Hao Y, Petters RM, Wong F (1993) Cellular interactions implicated in
the mechanism of photoreceptor degeneration in transgenic mice expressing a mutant rhodopsin
gene. Proc Natl Acad Sci USA 90:8484–8488
Hume DA, Perry VH, Gordon S (1983) Immunohistochemical localization of a macrophage-
specific antigen in developing mouse retina: phagocytosis of dying neurons and differentiation of
microglial cells to form a regular array in the plexiform layers. J Cell Biol 97:253–257
Humphries MM, Rancourt D, Farrar GJ, Kenna P, Hazel M, Bush RA, Sieving PA, Sheils DM,
McNally N, Creighton P, Erven A, Boros A, Gulya K, Capecchi MR, Humphries P (1997)
Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat Genet 15:216–
219
Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S (1998) Microglia-specific
localisation of a novel calcium binding protein, Iba1. Brain Res Mol Brain Res 57:1–9
Ito D, Tanaka K, Suzuki S, Dembo T, Fukuuchi Y (2001) Enhanced expression of Iba1, ionized
calcium-binding adapter molecule 1, after transient focal cerebral ischemia in rat brain.Stroke
32:1208–1215
Jablonski MM, Iannaccone A (2000) Targeted disruption of Muller cell metabolism induces
photoreceptor dysmorphogenesis. Glia 32:192–204
80
Ji Y, Zhu CL, Grzywacz NM, Lee EJ (2012) Rearrangment of the cone mosaic in the retina of
the rat model of retinitis pigmentosa. J Comp Neurol 520:874–888
Jimé nez AJ, Garcí a-Ferná ndez JM, Gonzá lez B, Foster RG (1996) The spatio-temporal pattern
of photoreceptor degeneration in the aged rd/rd mouse retina. Cell Tissue Res 284:193–202
Kedzierski W, Bok D, Travis GH (1998) Non-cell-autonomous photoreceptor degeneration in
rds mutant mice mosaic for expression of a rescue transgene. J Neurosci 18:4076–4082
LaVail MM, Yasumura D, Matthes MT, Lau-Villacorta C, Unoki K, Sung CH, Steinberg RH
(1998) Protection of mouse photoreceptors by survival factors in retinal degenerations. Invest
Ophthalmol Vis Sci 39:592–602
Lee EJ, Ji Y, Zhu CL, Grzywacz NM (2011) Role of Muller cells in cone mosaic rearrangement
in a rat model of retinitis pigmentosa. Glia 59:1107–1117
Lewis GP, Fisher SK (2000) Muller cell outgrowth after retinal detachment: association with
cone photoreceptors. Invest Ophthalmol Vis Sci 41:1542–1545
Lewis GP, Matsumoto B, Fisher SK (1995) Changes in the organization and expression of
cytoskeletal proteins during retinal degeneration induced by retinal detachment. Invest
Ophthalmol Vis Sci 36:2404–2416
Li YW, Tao W, Luo LY, Huang DQ, Kauper K, Stabila P, LaVail MM, Laties AM, Wen R
(2010) CNTF induces regeneration of cone outer segments in a rat model of retinal degeneration.
PLoS One 5:9495
Li ZY, Wong F, Chang JH, Possin DE, Hao Y, Petters RM, Milam AH (1998) Rhodopsin
81
transgenic pigs as a model for human retinitis pigmentosa.Invest Ophthalmol Vis Sci 39:808–
819
Liu CD, Li YW, Peng M, Laties AM, Wen R (1999) Activation of caspase-3 in the retina of
transgenic rats with the rhodopsin mutation S334ter during photoreceptor degeneration. J
Neurosci 19:4778–4785
Ma C, Papermaster D, Cepko CL (1998) A unique pattern of photoreceptor degeneration in
cyclin D1 mutant mice. Proc Natl Acad Sci USA 95:9938–9943
Marc RE, Jones BW (2003) Retinal remodeling in inherited photoreceptor degenerations. Mol
Neurobiol 28:139–147
Marc RE, Jones BW, Watt CB, Strettoi E (2003) Neural remodeling in retinal degeneration. Prog
Retin Eye Res 22:607–655
Martinez-Navarrete G, Seiler MJ, Aramant RB, Fernandez-Sanchez L, Pinilla I, Cuenca N
(2011) Retinal degeneration in two lines of transgenic S334ter rats. Exp Eye Res 92:227–237
Moore S, Thanos S (1996) The concept of microglia in relation to central nervous system disease
and regeneration. Prog Neurobiol 48:441–460
Nakajima K, Kikuchi Y, Ikoma E, Honda S, Ishikawa M, Liu Y, Kohsaka S (1998)
Neurotrophins regulate the function of cultured microglia. Glia 24:272–289
Newman E, Reichenbach A (1996) The Mü ller cell: a functional element of the retina. Trends
Neurosci 19:307–312
82
Nir I, Agarwal N, Papermaster DS (1990) Opsin gene expression during early and late phases of
retinal degeneration in rds mice. Exp Eye Res 51:257–267
Perry VH, Gordon S (1991) Macrophages and the nervous system.Int Rev Cytol 125:203–244
Ray A, Sun GJ, Chan L, Grzywacz NM, Weiland J, Lee EJ (2010) Morphological alterations in
retinal neurons in the S334ter-line3 transgenic rat. Cell Tissue Res 339:481–491
Rich KA, Figueroa SL, Zhan YT, Blanks JC (1995) Effects of Muller cell disruption on mouse
photoreceptor cell-development. Exp Eye Res 61:235–248
Roque RS, Caldwell RB (1990) Muller cell changes precede vascularization of the pigment
epithelium in the dystrophic rat retina. Glia 3:464–475
Roque RS, Imperial CJ, Caldwell RB (1996) Microglial cells invade the outer retina as
photoreceptors degenerate in Royal College of Surgeons rats. Invest Ophthalmol Vis Sci 37:196–
203
Sagdullaev BT, Aramant RB, Seiler MJ, Woch G, McCall MA (2003) Retinal transplantation-
induced recovery of retinotectal visual function in a rodent model of retinitis pigmentosa. Invest
Ophthalmol Vis Sci 44:1686–1695
Schuetz E, Thanos S (2004a) Neuro-glial interactions in the adult rat retina after reaxotomy of
ganglion cells: examination of neuron survival and phagocytic microglia using fluorescent
tracers. Brain Res Bull 62:391–396
Schuetz E, Thanos S (2004b) Microglia-targeted pharmacotherapy in retinal neuro-degenerative
diseases. Curr Drug Targets 5:619–627
83
Tanihara H, Hangai M, Sawaguchi S, Abe H, Kageyama M, Nakazawa F, Shirasawa E, Honda Y
(1997) Up-regulation of glial fibrillary acidic protein in the retina of primate eyes with
experimental glaucoma. Arch Ophthalmol 115:752–756
Taylor S, Srinivasan B, Wordinger RJ, Roque RS (2003) Glutamate stimulates neurotrophin
expression in cultured Mü ller cells. Mol Brain Res 111:189–197
Thanos S (1992) Sick photoreceptors attract activated microglia from the ganglion cell layer: a
model to study the inflammatory cascades in rats with inherited retinal dystrophy. Brain Res
588:21–28
Wen R, Song Y, Cheng T, Matthes MT, Yasumura D, Lavail MM, Steinberg RH (1995) Injury-
induced up-regulation of BFGF and CNTF messenger-RNAs in the rat retina. J Neurosci
15:7377–7385
Wong ROL, Hughes A (1987) Role of cell death in the topogenesis of neuronal distributions in
the developing cat retinal ganglion cell layer. J Comp Neurol 262:496–511
Zeiss CJ, Johnson EA (2004) Proliferation of microglia, but not photoreceptors, in the outer
nuclear layer of the rd-1 mouse. Invest Ophthalmol Vis Sci 45:971–976
Zeng H, Zhu X, Zhang C, Yang L, Wu L, Tso MOM (2005) Identification of sequential events
and factors associated with microglial activation, migration, and cytotoxicity in retinal
degeneration in rd mice. Invest Ophthalmol Vis Sci 46:2992–2999
84
FIGURE LEGENDS
Figure 1. Effect of retinitis pigmentosa (RP) on the outer parts of the retina. a–c Light
micrographs taken from cryostat vertical sections stained with hematoxylin. a Normal retina at
postnatal day 15 (P15N) showing healthy, organized and tightly packed nuclei within the outer
nuclear layer (onl). opl Outerplexiform layer. b The thickness of the ONL (onl) of the P15 RP
retina (P15 RP) is less than half the ONL of the age-matched normal retinas (arrows examples of
spaces in ONL loosely filled by nuclei). c P21 RP retina revealing a greater reduction in the
thickness in the ONL. The thickness of the ONL is uneven and regions in which fewer cell
bodies are present are often observed (arrow). Scale bars = 200 μm
Figure 2. Rhodopsin-immunoreactive rods in developing RP retinas. Confocal micrographs
taken from vertical sections of retinas processed for rhodopsin immunoreactivity. a, bRhodopsin
immunoreactivity is observed in the segments and cell bodies of rods in normal and RP retinas
(onl outer nuclear layer, opl outerplexiform layer). c The thickness of the rhodopsin-
immunoreactive ONL in RP retinas at P15 is reduced compared with that of the P10 RP
retinas. d P21 RP retinas show rods positioned ―flat‖ against the outerplexiform layer. d, e Two
distinct regions are observed to alternate repeatedly: regions full of clusters of cell bodies and
regions lacking cell bodies but with some processes. Scale bars = 50 μm
Figure 3. Rod-holes in the RP retina. Series of confocal micrographs taken from rhodopsin-
labeled retinal whole-mounts labeled from near the optic disc (left) toward the peripheral (right)
85
part of P15 N (a), P10 RP (b, e), P14 RP (f), P17 RP (c, g) and P21 RP (d, h) retinas. Rhodopsin-
immunoreactive cells have a homogeneous distribution in normal retinas but holes occur in RP
retinas. Three main rod degeneration patterns are demonstrated at high-power (not taken from f–
h): small holes (b), large holes (c) and scattered cells with few segments (d). Low power
micrographs show that RP progressively develops from small rod-holes to large rod-holes to
scattered rods. Scale bars = 0.5 mm (a), 1 mm (e–h), 50 μm ( b–d)
Figure 4. Increase of rod-hole size with progression of RP. Retinal whole-mounts labeled with
rhodopsin at P14 (a), P17 (b) and P21 (c). Examples of typical large rod-holes from retinas are
shown. The size of the holes were measured with respect to their diameter (d). The analysis
summarized in a bar graph indicates a significant increase in the size of the large rod-holes
between P14, P17 and P21 (e). *P<0.05 or better. Scale bars = 20 μm
Figure 5. Spatial relationship between holes (in surviving rod mosaic) and dying rods in RP
retinas. Confocal micrographs of retinal whole-mounts labeled with rhodopsin (RHO,red) and
TUNEL (green) in P14 (a–c), P17 (d–f) and P21 RP (g–i) retinas. The surviving rod mosaic
show holes (a, d, g), which increase in size and quantity with disease progression. The TUNEL-
positive dying cells (b, e, h) are observed to fill the center of each rod-hole (c, f) before
becoming ―rings‖ and leaving the central region of the holes rod -less (i). Scale bar = 50 μm
Figure 6. Relationship between holes and activated microglial cells in RP retinas. Confocal
micrographs of retinal whole-mounts labeled for rhodopsin (green) and IBA-1 (red) in P14 (a–c),
86
P17 RP (d–f) and P21 RP (g–i) retinas. The surviving rods show holes in their mosaic pattern
(a, d, g). Single exposure of IBA-1 demonstrates the clustering of activated microglia at P14 and
P17 (b low power, e high power). Double exposure of rhodopsin and IBA-1 shows that activated
microglia cells cluster inside the rod-holes (c, f). The high-power image shows activated
microglia engulfing rod cell bodies (f). At P21, activated microglia are observed to expand out,
away from the center (h, i). Scale bars = 50 μm
Figure 7. Relationship between holes and remodeled Mü ller-cell processes in RP retinas.
Confocal micrographs of retinal whole-mounts labeled with GS (green), GFAP (red), Iba-1 (red)
and rhodopsin (green). In P21 normal retinas, the mesh network of apical processes of Mü ller
cells are observed at the outer retina (a). In P21 RP retinas, Mü ller cells remodel by extending
their processes into a clustered bunch (b). GFAP immunoreactivity is also present in the
clustered processes (c). Double labeling of GS (d) and GFAP (e) in RP retinas demonstrates that
the clustered remodeled processes of Mü ller cells also express GFAP (f). Double labeling of GS
(g) and Iba-1 (h) reveals that the Mü ller cells processes are remodeled when the activated
microglial cells are no longer present inside the holes (i). Double labeling of rhodopsin (j) and
GFAP (k) indicates that the remodeled Mü ller-cell processes are filling in the holes left by rod
deaths in their mosaic (l) and also shows (inset) that the GFAP-immunoreactive Mü ller cell
processes wrap around the cell bodies of rods, as seen along the perimeter of the hole. Scale bars
= 100 μm ( a–l), 50 μm (inset in l)
Figure 8. Spatial correlation between rods and cones in early RP. Confocal micrographs of RP
retinal whole-mounts labeled with rhodopsin (green) and M-opsin (red). Both rods (a, d) and M-
87
cones (b, e) show holes in their mosaic at P17. These zones of low density in rods and cones co-
localize(c, f). Scale bars = 50 μm
88
Role of Mü ller Cells in Cone-mosaic Rearrangement in the Retina of
the Rat Model of Retinitis Pigmentosa
ABSTRACT
In the previous two chapters, the remodeling of cone mosaics and its trigger by rod degeneration
was explained in detail. In this chapter, the mechanism underlying the maintenance of cone rings
will be explained. The Mü ller-glia processes were found to envelop cones and fill the center of
each ring. Zonula occludens-1 located between the photoreceptor inner segments and the apical
processes of Mü ller cells were around the rings. These rings of Zonula occludens-1 were formed
before the onset of cone cell deaths and were maintained until late stages of RP. From these
observations, we hypothesize that cone-Mü ller-cell interactions mediate and maintain the rings.
To test of this hypothesis, DL-alpha-aminoadipic acid (AAA), a gliotoxin, was intravitreally
injected. A single intravitreal injection of AAA at P50 disrupted the rings of cones 3 days after
the injection. These findings indicate that the rearrangement of cones in rings is modulated by
Mü ller cells in RP. Thus, if the relationship between photoreceptors and Mü ller-glia is better
understood, the latter could potentially be manipulated for effective neuroprotection or the
restoration of normal cone arrays.
Key words: photoreceptor mosaics; reorganization; cell death; retina
89
INTRODUCTION
Retinitis pigmentosa (RP) is currently an untreatable disease (Bird, 1995). Studies in animal
models of RP have shown that rod degeneration leads to inner-retinal (Marc and Jones, 2003;
Strettoi et al., 2004; Jones and Marc, 2005; Gargini et al., 2007; Martinez-Navarrete et al., 2010;
Ray et al., 2010) remodeling, resulting in abnormal circuitry. In the retina, rods and cones are
packed together with processes of Mü ller glia for structural and metabolic support in the outer
nuclear layer (ONL — Rich et al., 1995). In addition, Mü ller cells are one of the main cell types
involved
in the formation of subretinal cellular membranes (Fisher et al., 1994). In detached and
retinal-degeneration retinas, Mü ller cells begin to
hypertrophy, filling the space left by dying
photoreceptors,
and then grow along the exposed outer surface of the
retina (Lewis et al., 1995;
Lewis and Fisher, 2000; Marc and Jones, 2003; Gargini et al., 2007; Ray et al., 2010). Such
remodeling is not surprising, as Mü ller cells contain trophic receptors for most of the molecules
involved in photoreceptor rescue (Harada et al., 1995; Wen et al., 1995; Bringmann et al., 2006).
Hence, Mü ller cells become stimulated after retinal insults such as mechanical injury (Harada et
al., 1995; Wen et al., 1995), and light-induced degeneration (Wen et al., 1998). Thus, interaction
between photoreceptors and Mü ller cells occur during retinal degeneration.
In this part of my research project, we investigated the mechanisms modulating the
spatial distribution of cones in the S334ter-line-3 rat. In particular, we investigated the role of
remodeled processes of the Mü ller cell in this reorganization. For instance, we used intravitreal
injections of Alpha-aminoadipic acid (AAA), which disrupt the Mü ller-cell metabolism within
the mammalian retina (Karlsen et al., 1982; Rich et al., 1995; Jablonski and Iannaccone, 2000;
West et al., 2008).
90
MATERIALS AND METHODS
Animals
The third line of albino Sprague-Dawley rats homozygous for the truncated murine opsin gene
(stop codon at residue 334; S334ter-3) was obtained from Matthew M. LaVail (University of
California, San Francisco, CA). The mutation in the S334ter-line-3 retina triggers the cell death
of rods, similar to those causing human RP. For simplicity, we will refer to this rat‘s disease as
RP in the rest of this paper. These rats were mated with normal Copenhagen rats to produce
offspring heterozygous for the S334ter RP transgene that were subsequently used in this study.
RP rats were sacrificed at post-natal (P) days P16, 30, 50, 60 and 90 (N = 12 for each stage).
Controls were age-matched Sprague Dawley rats (N = 12 for each stage; Harlan, Indianapolis,
IN). All rats were housed under cyclic 12/12-hour light/dark conditions with free access to food
and water. Surgeries on rats were performed under anesthesia induced by intra-peritoneal
injection of ketamine (100 mg/kg; KETASET, Fort Dodge, IA) and xylazine (20 mg/kg; X-Ject
SA, Butler, Dublin, OH). Surgical procedures were performed according to the guidelines of the
Association for Research in Vision and Ophthalmology (ARVO). All procedures were in
conformance with the Guide for Care and Use of Laboratory Animals (National Institutes of
Health). The University of Southern California Institutional Animal Care and Use Committee
reviewed and approved all procedures.
91
Administration of Alpha-Aminoadipic Acid (AAA)
DL-a-Aminoadipic acid (AAA; Sigma) was prepared in phosphate-buffered saline (PBS),
adjusted to pH 7.5 and sterile-filtered before administration. AAA was administered by
intravitreal injection with a fine-glass microelectrode through the sclera at the level of the
temporal peripheral retina. The high concentration of AAA is known to affect the photoreceptors
(Jablonski et al., 2000; West et al., 2008). In order to minimize the effects upon photoreceptors
and disrupt the Mü ller cell metabolism, we used three concentrations of AAA (10 g/ l, 50 g/ l,
100 g/ l) to find the optimal concentration (50 g/ l). Various concentrations of AAA in 5 l of
sterile saline were used. Sham injections, for control, consisted of 5 l sterile saline. For each
animal, one eye was used to inject AAA and the other eye was used to inject saline for
comparison. The entire injection procedure required only a few minutes and we finished before
the animals recovered from anesthesia.
Tissue Preparation
Animals were deeply anesthetized by intra-peritoneal injection of pentobarbital (40 mg/kg body
weight) and the eyes were enucleated. Animals were then killed with an overdose of
pentobarbital. Their eyes‘ anterio r segments were then removed and the eyecups were fixed by
immersion in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4, for 1.5-2 h.
Following fixation, the retinas were carefully dissected and transferred to 30% sucrose in PB for
24 h at 4 C. For cryostat sections, eyecups were embedded in OCT embedding medium
(Tissue-Tek, Elkhart, IN), then quickly frozen in liquid nitrogen and subsequently sectioned
92
along the vertical meridian on a cryostat at a thickness of 20 m. The sections were collected on
gelatin-coated slides for immuno and TUNEL staining.
Immunohistochemistry
For fluorescence immunocytochemistry, 20- m-thick cryostat sections were incubated in 10%
normal goat serum (NGS) or 10% normal donkey serum (NDS) and 1% Triton X-100 in PBS for
1 h at room temperature. Sections were then incubated overnight with a rabbit polyclonal
antibody directed against: red/green opsin (M-opsin — kindly provided by Dr. Cheryl Craft from
the Doheny Eye Institute, University of Southern California — dilution 1:1000) and glial
fibrillary acidic protein (Sigma, dilution 1:500); goat polyclonal antibody directed against the N-
terminus of human sensitive blue opsin (S-opsin — Santa Cruz, dilution 1:500); mouse
monoclonal antibody directed against glutamine synthetase (GS — Chemicon; dilution 1:100),
and zonula occludens (ZO-1 – Invitrogen; 1:500). Each antiserum was diluted with PBS
containing 0.5% Triton X-100 at 4 C. Retinas were washed in PBS for 45 min (3 15 min).
Afterwards, we incubated the retinas for 2 h in carboxymethylindocyanine (Cy3)-conjugated
affinity-purified, donkey anti-rabbit IgG (Jackson Immuno Labs, West Grove, PA, USA; dilution
1:500); Alexa 488 anti-goat IgG, or Alexa 488 anti-mouse (Molecular Probes, Eugene, dilution
1:300); Cy5-conjugated, donkey anti-goat IgG (Jackson Immuno Labs; dilution 1:300) at room
temperature. The sections were washed for 30 min with 0.1M PB and coverslipped with
Vectashield mounting medium (Vector Labs, Burlingame, CA). For whole-mount
immunostaining, the same procedure was used. For rhodopsin, S-opsin, M-opsin, GFAP, and
93
GS, the primary antibody incubation was for 3 days and the secondary antibody incubation was
for 2 days.
For double and triple labeling, sections and whole mounts were incubated in a mixture of
following antibodies: S-opsin and M-opsin; rhodopsin with S-opsin and M-opsin; GS and M-
opsin; GS and GFAP; ZO-1, GS, and M-opsin. Some of the whole mounts were stained with
TUNEL, and then washed for 30 min with 0.1M PB and coverslipped with Vectashield mounting
medium. Sections and whole-mounts were then analyzed using a Zeiss LSM 510, (Zeiss, NY)
confocal microscope. Immunofluorescence images were processed with Zeiss LSM-PC
software. The brightness and contrast of the images were adjusted using Adobe Photoshop 7.0
(Adobe Systems, Mountain View, CA). For presentation, all Photoshop manipulations
(brightness and contrast only) were carried out equally across sections.
Statistical analysis
Data were expressed as mean ± standard deviation and Student‘s t -tests were used for
comparisons with normal retinas. The t-tests were used to examine the difference between two
means in the the total number of M-opsin- and S-opsin-immunoreactive cones. The total number
of M-opsin- and S-opsin-immunoreactive cones in both normal (n = 3) and RP (n = 3) retinas
were manually counted. All statistical tests were performed using Stat View (Abacus Concepts,
Berkeley, CA, USA). A difference between the means of two conditions was considered
statistically significant at P < 0.05.
94
RESULTS
Contribution of Mü ller cells to cone migration
Mü ller cells contribute to migration of neurons in the retina (Sullivan et al., 2003). In the retina,
rods and cones are normally in contact with processes of Mü ller cells for structural and
metabolic support in the ONL (Rich et al., 1995). We thus decided to test whether these
processes are also in contact with cones in RP. For this purpose, double-labeling experiments
were performed in P16 and P60 RP retinas, using antisera against both M-opsin and GS, the
latter being a Mü ller-cell marker (Wä ssle and Haverkamp, 2000; Lee et al., 2002). We found
that processes of Mü ller cells were surrounding cell bodies of cones (Figs. 1A, B). In addition,
cone processes were following those of the Mü ller cells (Fig. 1C). These results demonstrate
that cones migrate through processes of M ller cells to form rings in the RP retina
Fig. 1 Contribution of Mü ller cells to cone migration
95
Remodeling of Mü ller-cell processes into holes of rods and cones
We have seen great extent of remodeling of cones in the previous chapters. In order to explore
the behavior of Mü ller cells in RP, whole mounts were labeled with antibodies to M-opsin, S-
opsin (data not shown), and GS, a marker of Mü ller-cell. Figure 5 shows an example of P50
normal and P50 RP retinal whole mounts processed for these antibodies. In this figure, the focal
plane of the outer limiting membrane (OLM) was examined. The normal retina showed labeled
M-opsin cones segments throughout the photoreceptor array (Fig. 2A). GS immunoreactivity
displayed the normal spatially homogeneous mesh network of Mü ller-cell processes (Fig. 2B).
Double labeling of M-opsin and GS confirmed that photoreceptor inner segments and the apical
processes of Mü ller cells had a close association (Fig. 2C). In contrast, in RP retinas, we again
observed an array of rings of cones (Fig. 2D). Furthermore, we observed a clustering of the
processes of Mü ller cells in broccoli-like shapes (Fig. 2E). Double labeling of M-opsin and GS
showed that the rings of cones surrounded the remodeled Mü ller-cell processes (Fig. 2F). Hence,
these processes filled the centers of the rings. Figures 2G-I show high-magnification
micrographs from the same focal plane of the retina. The M-opsin immunoreactivity in the cell
bodies and the axon-like processes of cones was associated with aggregated processes of Mü ller
cells. However, we observed no extensions of the Mü ller-cell processes onto the inner and outer
segments of cones (Fig. 2I). Overall, the data in Fig. 2 suggest that a remodeling of Mü ller-cell
processes fills the regions of low densities of cones and rods.
96
Fig. 2 Remodeling of Mü ller-cell processes inside cone rings
A typical consequence of the remodeling of Mü ller cells in retinal degeneration is a glial
reaction involving the upregulation of GFAP (DiLoreto et al., 1995; Tanihara et al., 1997). Our
RP model also showed such upregulation of GFAP by Mü ller cells (Ray et al., 2010). To
97
examine whether GFAP followed the remodeled processes of Mü ller cells, we performed double
labeling of GS (Fig. 3A) and GFAP (Fig. 3B) on P50 RP retinas. In RP retinas, GFAP
immunoreactivity was present in the regions of the broccoli-shaped clusters of processes of
Mü ller-cells (Fig. 3C).
Fig. 3 Positive GS and GFAP immunoreactivity inside each ring
Because both the mosaic of cones and the Mü ller-cell processes remodel in RP, one
would expect that the OLM does so, too. For this purpose, we first performed
immunocytochemisty on vertical sections of P30 normal (Figs. 4A-D) and P30 RP (Figs. 4E-J)
retinas. The OLM contains the specialized adherens junction associated protein, ZO-1, which is
normally present between the photoreceptor inner segments and the apical processes of Mü ller
cells (Tserentsoodol et al., 1998; Paffenholz et al., 1999; Campbell et al., 2006; Pearson et al.,
2010). We thus wished to establish whether ZO-1 might be changed in the RP retina. Vertical
sections immunolabeled for GS (Figs. 4A, B) and ZO-1 (Figs. 4B, F) are shown. In normal
98
retinas, Mü ller-cell processes were present at the outer part of the retina (Fig. 4A) and ZO-l
staining appeared at the OLM (Fig. 4B). In the OLM region, ZO-1 was expressed and
colocalized with GS (Figs. 4C, D, arrow). ZO-1 formed a continuous line at the OLM. In
contrast, the RP retina showed discontinuous and fragmented staining for ZO-1 at the OLM (Fig.
4F). Double labeling of GS (Fig. 4E) and ZO-1 (Fig. 4F) showed that the ZO-1 expression was
weak and fragmented but still colocalized with GS at the OLM (Fig. 4G, H, arrow). Next, we
observed the expression of ZO-1 in whole mounts of developmental RP retinas. In P16 RP
retina, we observed ZO-1 in a few rings (Fig. 4I). By P30 RP, the retina showed labeled ZO-1 in
a network of rings (Fig. 4J). Double labeling of ZO-1 (Fig. 4K) and M-opsin (Fig. 4L) in P50 RP
retinas showed that the ZO-1 rings coincided in location with those of the cones (Fig. 4M). We
found axon-like processes of cones projected onto the center of the rings and segments of the
cones present at the boundaries of the rings. ZO-1 expression was present at the boundaries of
the rings. Furthermore, we observed a clustering of the processes of Mü ller cells in broccoli-like
shapes (Figs. 2, 4O). Double labeling of ZO-1 (Fig. 4N) and GS (Fig. 4O) showed that the rings
of ZO-1 outlined the remodeled Mü ller-cell processes (Fig. 4P). Triple labeling of ZO-1 (Fig.
4Q), M-opsin (Fig. 4R), and GS (Fig.4S) showed that the OLM also remodeled to allow the
processes of Mü ller cells and cones to establish contacts (Fig. 4T, arrow).
99
Fig. 4 ZO-1 immunoreactivity around rings at the junction between cones and Mü ller cells
100
Elimination of rings of cones by disrupting metabolism of Mü ller cells
The results so far demonstrate that the migration of cones forming the rings happens through
remodeled processes of Mü ller cells. We wanted to test whether Mü ller-cell processes are simply
tracks for migration or are necessary for the rings to exist. Previous studies showed that
disrupting Mü ller cells during mouse retinal development with the specific glial-cell toxin DL-
alpha-aminoadipic acid (AAA) caused clumping of photoreceptor nuclei (Rich et al., 1995).
Therefore, to test the importance Mü ller-cell processes for the rings, we injected AAA. The
effect of AAA was compared to saline and monitored at 1, 3, and 5 days after the injections in
P50 RP eyes. The results were same in all three days after the injections (data not shown).
Consequently, we here illustrate only the result 3 days after injection of AAA. The data related to
saline injection appear in Figs. 5A and C, while those for AAA appear in Figs. 5B and D. RP
eyes injected with saline showed the aforementioned arrays of M-opsin cone rings throughout
the retina (Figs. 5A, C). In contrast, RP eyes injected with AAA no longer showed rings (Figs.
5B, D). These effects of AAA indicate that interactions between cones and Mü ller-cell processes
are necessary for the maintenance of rings in the RP retina.
101
Fig. 5 Elimination of rings of cones by disrupting metabolism of Mü ller cells
DISCUSSION
We have previously seen that rod degeneration trigger the reorganization of the cone
mosaic into an array of rings. What maintain such rings in shape was unclear. In this part of the
research project, we have observed that the Mü ller-cell processes remodel significantly to fill
center of the rings and colocalize with cones (Fig. 2). This result is consistent with a previous
report demonstrating that remodeled processes of Mü ller cells are closely associated with cones
(Lewis and Fisher, 2000). Such results suggest that cones migrate through processes of Mü ller
cells (Fig. 1).
102
Further evidence that Mü ller-cell processes are important for the maintenance of cone
mosaics in rings is shown by the elimination of rings after injection of AAA (Fig. 5). This drug,
an analogue of L-glutamate, accumulates at toxic levels within Mü ller cells (Rich et al., 1995),
causing cytotoxicity both in vivo (Rich et al., 1995) and in vitro (Garthwaite and Regan,1980;
Huck et al., 1984). We found that a single intravitrial injection of AAA at P50 disrupted the rings
of cones in RP retinas 3 days after the injection (Fig. 5). This disruption indicates that the rings
of cones in the RP depend on interactions with Mü ller cells. We do not currently know how the
processes of Mü ller cells contribute to the holes and rings of cones. One hypothesis is that cones
may be releasing a factor to interact with Mü ller cells in the diseased retina (Lewis et al., 1992).
Adherens junction associated protein, ZO-1, (Tserentsoodol et al., 1998; Paffenholz et al.,
1999; Campbell et al., 2006), is present between the inner segments of rods and cones, and the
apical processes of Mü ller cells (Pearson et al., 2010). In the RP retina, we observed ZO-1 in the
network of rings of cones (Fig. 4). This protein appeared between the photoreceptor segments
and the processes of Mü ller cells, and between the processes of Mü ller cells (Fig. 4). One
possibility is that the processes of Mü ller cells in the center of the rings are not equivalent to the
normal apical ones. Rather, the processes of Mü ller cells filling the center may represent the glial
sealing that we see in the vertical section of the RP retina (Li et al., 1995; Jones and Marc, 2005;
Ray et al., 2010). In addition, we observe discontinuous and fragmented staining for ZO-1 at the
OLM at P30 vertically sectioned RP retinas (Fig. 4F). This result is similar to previous reports
showing discontinuous and fragmented expression of β-catenin at the OLM of retinas of rd8 and
Crb(−/−) mice (Mehalow et al., 2003; van de Pavert et al., 2004). We do not know whether
fragmented or discontinuous expression of other adherens junction associated proteins will show
in rings. However, the aberrant integrity of the OLM indicated by ZO-1 expression may arise
103
from rod cell death in RP (Fig. 4I - Campbell et al., 2007). Moreover, our results suggest that
other adherens junction associated proteins (e.g., β -catenin and N-cadherin) in the OLM may
also modify. In future experiments, we will study localization of such proteins in our RP model.
Our study is a step towards understanding the mechanisms controlling normal and
pathological mosaics of cones. We raise hypotheses on how these mosaics form through
interactions with rods, Mü ller glia, and other processes. First, rods die in hot spots, which
progressively increased as circular waves, leaving rod-less zones behind. Second, the pattern of
rod death triggers reorganization of the cone mosaic by migration through Mü ller cells into an
orderly array of rings. Third, the OLM remodels such as to allow the processes of Mü ller cells
and cones to establish contacts. In particular, we investigated the role of remodeled processes of
the Mü ller cell in this reorganization. This Mü ller-glia-based experiment may be important for
the treatment of RP and other retinal degenerative diseases. Mü ller cells contribute to the
survival of photoreceptors through release and mediation of neurotrophic factors (Bringmann et
al., 2006). Moreover, Mü ller cells react with gliosis to most pathological changes in the retina.
The gliosis can cause the aforementioned secretion of neurotrophic factors or contribute to cell
degeneration. Thus, one may be able to manipulate Mü ller cells might for effective
neuroprotection in RP and other retinal degenerative diseases, if one understands better the
relationship between Mü ller cells and photoreceptors. Moreover, those actions of Mü ller glia are
specially important given that cones have a long survival period in RP as suggested by our data.
Therefore, our study may teach how to repopulate cones while they survive, possibly leading to
therapeutic efforts. Such ideas of cone repopulation may also be beneficial for retinal
transplantation, a technique in which one initially inserts this cell haphazardly into a host retina
(Humayun et al., 2000; Seiler et al., 2010).
104
REFERENCES
Anderson RE, Maude MB, McClellan M, Matthes MT, Yasumura D, LaVail MM. 2002. Low
docosahexaenoic acid levels in rod outer segments of rats with P23H and S334ter rhodopsin
mutations. Molecular Vision 8: 351-358.
Bird AC. 1995. Retinal photoreceptor dystrophies LI. Edward Jackson Memorial Lecture. Am J
Ophthalmol 119: 543-562.
Bringmann A, Pannicke T, Grosche J, Francke M, Wiedemann P, Skatchkov SN, Osborne NN.
2006. Mü ller cells in the healthy and diseased retina. Prog Retin Eye Res. 25(4):397-424.
Campbell M, Humphries M, Kennan A, Kenna P, Humphries P, Brankin B. 2006. Aberrant
retinal tight junction and adherens junction protein expression in an animal model of autosomal
dominant Retinitis pigmentosa: the Rho(-/-) mouse. Exp Eye Res 83:484-92.
Campbell M, Humphries M, Kenna P, Humphries P, Brankin B. 2007. Altered expression and
interaction of adherens junction proteins in the developing OLM of the Rho(-/-) mouse. Exp Eye
Res 85:714-20.
105
Carterdawson LD, Lavail MM, Sidman RL. 1978. Differential Effect of Rd Mutation on Rods
and Cones in Mouse Retina. Investigative Ophthalmology & Visual Science 17: 489-498.
Chalmel F, Leveillard T, Jaillard C, Lardenois A, Berdugo N, Morel E, Koehl P, Lambrou G,
Holmgren A, Sahel JA, Poch O. 2007. Rod-derived Cone Viability Factor-2 is a novel
bifunctional-thioredoxin-like protein with therapeutic potential. Bmc Molecular Biology 8: 74.
DiLoreto DA, Jr., Martzen MR, del Cerro C, Coleman PD, del Cerro M. 1995. Mü ller cell
changes precede photoreceptor cell degeneration in the age-related retinal degeneration of the
Fischer 344 rat. Brain Research 698: 1-14.
Gargini C, Terzibasi E, Mazzoni F, Strettoi E. 2007. Retinal organization in the retinal
degeneration 10 (rd10) mutant mouse: a morphological and ERG study. J Comp Neurol 500:
222-238.
Garthwaite J, Regan CM. 1980. Toxic effects of alpha-aminoadipate on cultured cerebellar cells.
Brain Research 194: 603-607.
106
Gupta N, Brown KE, Milam AH. 2003. Activated microglia in human retinitis pigmentosa, late-
onset retinal degeneration, and age-related macular degeneration. Experimental Eye Research 76:
463-471.
Harada T, Imaki J, Hagiwara M, Ohki K, Takamura M, Ohashi T, Matsuda H, Yoshida K. 1995.
Phosphorylation of CREB in rat retinal cells after focal retinal injury. Experimental Eye
Research 61: 769-771.
Hartong DT, Berson EL, Dryja TP. 2006. Retinitis pigmentosa. Lancet 368: 1795-1809.
Haverkamp S, Wassle H. 2000. Immunocytochemical analysis of the mouse retina. J Comp
Neurol 424: 1-23.
Hombrebueno JR, Tsai MM, Kim HL, De Juan J, Grzywacz NM, Lee EJ. 2010. Morphological
changes of short-wavelength cones in the developing S334ter-3 transgenic rat. Brain Research
1321: 60-66.
107
Huang P, Gaitan A, Hao Y, Petters RM, Wong F. 1993. Cellular Interactions Implicated in the
Mechanism of Widespread Photoreceptor Degeneration in Transgenic Mice Expressing a Mutant
Rhodopsin Gene. Investigative Ophthalmology & Visual Science 34: 768-768.
Huck S, Grass F, Hortnagl H. 1984. The glutamate analogue alpha-aminoadipic acid is taken up
by astrocytes before exerting its gliotoxic effect in vitro. J Neurosci 4: 2650-2657.
Humayun MS, de Juan E, Jr., del Cerro M, Dagnelie G, Radner W, Sadda SR, del Cerro C. 2000.
Human neural retinal transplantation. Invest Ophthalmol Vis Sci 41:3100-3106.
Jablonski MM, Iannaccone A. 2000. Targeted disruption of Mü ller cell metabolism induces
photoreceptor dysmorphogenesis. Glia 32: 192-204.
Jimenez AJ, Garcia-Fernandez JM, Gonzalez B, Foster RG. 1996. The spatio-temporal pattern of
photoreceptor degeneration in the aged rd/rd mouse retina. Cell Tissue Res 284: 193-202.
Jones BW, Marc RE. 2005. Retinal remodeling during retinal degeneration. Experimental Eye
Research 81: 123-137.
108
Karlsen RL, Pedersen OO, Schousboe A, Langeland A. 1982. Toxic Effects of Dl-Alpha-
Aminoadipic Acid on Mü ller Cells from Rats Invivo and Cultured Cerebral Astrocytes.
Experimental Eye Research 35: 305-311.
Kedzierski W, Bok D, Travis GH. 1998. Non-cell-autonomous photoreceptor degeneration in rds
mutant mice mosaic for expression of a rescue transgene. J Neurosci 18: 4076-4082.
Komeima K, Rogers BS, Campochiaro PA. 2007. Antioxidants slow photoreceptor cell death in
mouse models of retinitis pigmentosa. Journal of Cellular Physiology 213: 809-815.
LaVail MM, Yasumura D, Matthes MT, Lau-Villacorta C, Unoki K, Sung CH, Steinberg RH.
1998. Protection of mouse photoreceptors by survival factors in retinal degenerations.
Investigative Ophthalmology & Visual Science 39: 592-602.
Lee EJ, Min DS, Lee MY, Chung JW, Chun MH, Oh SJ. 2002. Differential expression of
phospholipase D1 in the developing retina. Eur J Neurosci 15: 1006-1012.
Lee ES, Flannery JG. 2007. Transport of truncated rhodopsin and its effects on rod function and
degeneration. Investigative Ophthalmology & Visual Science 48: 2868-2876.
109
Lewis GP, Erickson PA, Guerin CJ, Anderson DH, Fisher SK. 1992. Basic fibroblast growth
factor: a potential regulator of proliferation and intermediate filament expression in the retina. J
Neurosci 12: 3968-3978.
Lewis GP, Fisher SK. 2000. Mü ller cell outgrowth after retinal detachment: Association with
cone photoreceptors. Investigative Ophthalmology & Visual Science 41: 1542-1545.
Lewis GP, Guerin CJ, Anderson DH, Matsumoto B, Fisher SK. 1994. Rapid changes in the
expression of glial cell proteins caused by experimental retinal detachment. Am J Ophthalmol
118: 368-376.
Li YW, Tao W, Luo LY, Huang DQ, Kauper K, Stabila P, LaVail MM, Laties AM, Wen R.
2010. CNTF Induces Regeneration of Cone Outer Segments in a Rat Model of Retinal
Degeneration. Plos One 5: 9495.
Lin B, Masland RH, Strettoi E. 2009. Remodeling of cone photoreceptor cells after rod
degeneration in rd mice. Experimental Eye Research 88: 589-599.
110
Liu CD, Li YW, Peng M, Laties AM, Wen R. 1999. Activation of caspase-3 in the retina of
transgenic rats with the rhodopsin mutation S334ter during photoreceptor degeneration. Journal
of Neuroscience 19: 4778-4785.
Marc RE, Jones BW. 2003. Retinal remodeling in inherited photoreceptor degenerations. Mol
Neurobiol 28: 139-147.
Martinez-Navarrete G, Seiler MJ, Aramant RB, Fernandez-Sanchez L, Pinilla I, Cuenca N. 2010.
Retinal degeneration in two lines of transgenic S334ter rats. Exp Eye Res .
Mehalow AK, Kameya S, Smith RS, Hawes NL, Denegre JM, Young JA, Bechtold L, Haider
NB, Tepass U, Heckenlively JR, Chang B, Naggert JK, Nishina PM. 2003. CRB1 is essential for
external limiting membrane integrity and photoreceptor morphogenesis in the mammalian retina.
Hum Mol Genet 12:2179-89.
Mohand-Said S, Hicks D, Dreyfus H, Sahel JA. 2000. Selective transplantation of rods delays
cone loss in a retinitis pigmentosa model. Archives of Ophthalmology 118: 807-811.
111
Paffenholz R, Kuhn C, Grund C, Stehr S, Franke WW. 1999. The arm-repeat protein NPRAP
(neurojungin) is a constituent of the plaques of the outer limiting zone in the retina, defining a
novel type of adhering junction. Experimental Cell Research 250: 452-464.
Pearson RA, Barber AC, West EL, MacLaren RE, Duran Y, Bainbridge JW, Sowden JC, Ali RR.
2010. Targeted disruption of outer limiting membrane junctional proteins (Crb1 and ZO-1)
increases integration of transplanted photoreceptor precursors into the adult wild-type and
degenerating retina. Cell Transplant 19: 487-503.
Petters RM, Alexander CA, Wells KD, Collins EB, Sommer JR, Blanton MR, Rojas G, Hao Y,
Flowers WL, Banin E, Cideciyan AV, Jacobson SG, Wong F. 1997. Genetically engineered large
animal model for studying cone photoreceptor survival and degeneration in retinitis pigmentosa.
Nat Biotechnol 15: 965-970.
Punzo C, Kornacker K, Cepko CL. 2009. Stimulation of the insulin/mTOR pathway delays cone
death in a mouse model of retinitis pigmentosa. Nature Neuroscience 12: 44-52.
Ray A, Sun GJ, Chan L, Grzywacz NM, Weiland J, Lee EJ. 2010. Morphological alterations in
retinal neurons in the S334ter-line3 transgenic rat. Cell Tissue Res 339: 481-491.
112
Rich KA, Figueroa SL, Zhan YT, Blanks JC. 1995. Effects of Mü ller Cell Disruption on Mouse
Photoreceptor Cell-Development. Experimental Eye Research 61: 235-248.
Ripps H. 2002. Cell death in retinitis pigmentosa: Gap junctions and the "bystander' effect.
Experimental Eye Research 74: 327-336.
Rohrer B, Lohr HR, Humphries P, Redmond TM, Seeliger MW, Crouch RK. 2005. Cone opsin
mislocalization in Rpe65-/- mice: a defect that can be corrected by 11-cis retinal. Invest
Ophthalmol Vis Sci 46: 3876-3882.
Sagdullaev BT, Aramant RB, Seiler MJ, Woch G, McCall MA. 2003. Retinal transplantation-
induced recovery of retinotectal visual function in a rodent model of retinitis pigmentosa.
Investigative Ophthalmology & Visual Science 44: 1686-1695.
Seiler MJ, Aramant RB, Thomas BB, Peng Q, Sadda SR, Keirstead HS. 2010. Visual restoration
and transplant connectivity in degenerate rats implanted with retinal progenitor sheets. Eur J
Neurosci 31:508-520.
113
Semo M, Vugler AA, Jeffery G. 2007. Paradoxical opsin expressing cells in the inner retina that
are augmented following retinal degeneration. Eur J Neurosci 25: 2296-2306.
Strettoi E, Mears AJ, Swaroop A. 2004. Recruitment of the rod pathway by cones in the absence
of rods. J Neurosci 24: 7576-7582.
Sullivan R, Penfold P, Pow DV. 2003. Neuronal migration and glial remodeling in degenerating
retinas of aged rats and in nonneovascular AMD. Invest Ophthalmol Vis Sci 44: 856-865.
Tanihara H, Hangai M, Sawaguchi S, Abe H, Kageyama M, Nakazawa F, Shirasawa E, Honda
Y. 1997. Up-regulation of glial fibrillary acidic protein in the retina of primate eyes with
experimental glaucoma. Arch Ophthalmol 115: 752-756.
Tserentsoodol N, Shin BC, Suzuki T, Takata K. 1998. Colocalization of tight junction proteins,
occludin and ZO-1, and glucose transporter GLUT1 in cells of the blood-ocular barrier in the
mouse eye. Histochem Cell Biol 110:543-51.
114
Tso MO, Li WW, Zhang C, Lam TT, Hao Y, Petters RM, Wong F. 1997. A pathologic study of
degeneration of the rod and cone populations of the rhodopsin Pro347Leu transgenic pigs. Trans
Am Ophthalmol Soc 95: 467-479; discussion 479-483.
van de Pavert SA, Kantardzhieva A, Malysheva A, Meuleman J, Versteeg I, Levelt C, Klooster J,
Geiger S, Seeliger MW, Rashbass P, Le Bivic A, Wijnholds J. 2004. Crumbs homologue 1 is
required for maintenance of photoreceptor cell polarization and adhesion during light exposure. J
Cell Sci 117:4169-77.
Wen R, Cheng T, Song Y, Matthes MT, Yasumura D, LaVail MM, Steinberg RH. 1998.
Continuous exposure to bright light upregulates bFGF and CNTF expression in the rat retina.
Current Eye Research 17: 494-500.
Wen R, Song Y, Cheng T, Matthes MT, Yasumura D, Lavail MM, Steinberg RH. 1995. Injury-
Induced up-Regulation of Bfgf and Cntf Messenger-Rnas in the Rat Retina. Journal of
Neuroscience 15: 7377-7385.
West EL, Pearson RA, Tschernutter M, Sowden JC, MacLaren RE, Ali RR. 2008.
Pharmacological disruption of the outer limiting membrane leads to increased retinal integration
of transplanted photoreceptor precursors. Experimental Eye Research 86: 601-611.
115
Zhang H, Fan J, Li S, Karan S, Rohrer B, Palczewski K, Frederick JM, Crouch RK, Baehr W.
2008. Trafficking of membrane-associated proteins to cone photoreceptor outer segments
requires the chromophore 11-cis-retinal. J Neurosci 28: 4008-4014.
FIGURE LEGENDS
Figure 1. Confocal micrographs of vertical sections labeled with M-opsin (red) and GS (green)
in P16 (A) and P60 (B, C) RP retinas. Double exposure of M-opsin and GS shows that they have
a close association with each other. M-opsin immunoreactive cones are followed (arrow) and
surrounded (arrowhead) by processes of Mü ller cells. ONL, outer nuclear layer; OPL, outer
plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer.
Scale bar = 10 m.
Figure 2. Confocal micrographs of whole mounts labeled with M-opsin (red) and GS (green) in
P50 normal (A-C) and P50 RP (D-I) retinas. Double labeling of M-opsin (A) and GS (B) shows
that cone inner segments and the apical processes of Mü ller cells have a close association in P50
normal retinas (C). Double labeling of M-opsin (D) and GS (E) in RP retinas shows remodelled
Mü ller-cell processes surrounding cones and filling the center of their rings (F). G-I, High-
power micrographs of D-F. Scale bar = 100 m.
116
Figure 3. Confocal micrographs of whole mounts processed for GS (A - blue) and GFAP (B -
red) staining in P50 RP retina. GFAP immunoreactivity is present in the regions of clustered
processes of Mü ller cells (C). Scale bar = 50 m.
Figure 4. Confocal micrographs of a vertical P30 normal (A-D) and P30 RP (E-H) retinal
sections processed for GS (A, E) and ZO-1 (B, F) immunoreactivities. Double exposure shows
that the ZO-1 is expressed and colocalized with GS at the OLM (C, D - arrow). Double labeling
of GS (E) and ZO-1 (F) in P30 RP retina shows that the ZO-1 expression is weak and fragmented
but still colocalized with GS at the OLM (G, H - arrow). In P16 RP retina, ZO-1
immunoreactivity is in a few rings (I). In P30 RP, the retina showed labeled ZO-1 in a network
of rings (J). Double labeling of ZO-1(K) and M-opsin (L) in P50 RP retinas showed that the ZO-
1 rings coincided in location with those of the cones (M). Double labeling of ZO-1 (N) and GS
(O) showed that the rings of ZO-1 outlined the remodeled Mü ller-cell processes (P). The inset
shows colocalization of ZO-1 and GS (P, arrow). Triple labeling of ZO-1 (Q), M-opsin (R), and
GS(S) show that the ZO-1 is closely associated with segments of photoreceptors and processes of
Mü ller cells (T, arrow). A-C, E-G, Scale bar = 50 m; D, H, Scale bar = 20 m; I-T, Scale bar =
100 m.
Figure 5. Confocal micrographs of whole mounts processed for M-opsin staining in saline-
injected (A, C) and AAA-injected (B, D) P50 RP eyes. C, D: High-power micrographs of
saline-injected (C) and AAA-injected (D) P50 RP eyes. AAA-treated retinas show disruption of
M-opsin rings. Scale bar = 100 m.
117
Manipulation of cone mosaics in the Retina of the Rat Model of
Retinitis Pigmentosa
ABSTRACT
In the previous chapters, we have observed the degeneration of rods triggering cone mosaic
remodeling in regular array of rings and maintenance of these rings by the remodeled processes
of Mü ller cell processes. After all the complex interplay of various retinal cells, the cone mosaic
still reflects regularity in terms of spacing. The fundamental question of what controls the
crystalline-regular mosaic of photoreceptors in normal retinas is still is an open problem for
scientists. In the last part of the research project, a broader question of what controls the precise
spacing of cones in both normal and in S334ter-line-3 rat retinas was addressed. This study
implemented the retinal treatment with Tissue inhibitor of Metalloproteinase, which is known to
be involved in the mediation of degradation and turnover of the extracellular matrix. The M-
opsin-immunoreactive cone mosaics were followed before and after the treatment and were
statistically quantified of their regularity and homogeneity. The results suggest that Tissue
inhibitor of Metalloproteinase disrupts the regularity of the cone mosaic in normal and Retinitis
Pigmentosa retinas. Also, the evidence collectively propose the importance of proper cone-ECM
interaction for maintenance of regularity of cone mosaic.
Key words: Cone mosaics; regularity; Tissue Inhibitor of Metalloproteinase
118
INTRODUCTION
It is well known that in normal rat retinas, the mosaic of photoreceptors exhibits spatial
homogeneity with local quasi-crystalline regularity (Bremiller and Larison, 1990; Wikler and
Rakic, 1991; Wikler and Rakic, 1994; Raymond et al., 1995; Bruhn and Cepko, 1996; Lin et al.,
2004; Kram et al., 2010). The regularity and the homogeneity of the mosaic of photoreceptors
are important and significant in that it is necessary to uniformly sample the visual space (French
et al., 1997; Manning and Brainard, 2009). For instance, the spacing of cones determines the
visual resolution and the density of rods constraints visual sensitivity (Williams and Coletta,
1987; Wä ssle and Boycott, 1991). Recently, cones in S334ter-line-3 rat model of Retinitis
Pigmentosa were shown to survive for a long time and remodel in their distribution pattern into
orderly array of rings (Lee et al., 2011; Ji et al., 2012; Zhu et al., 2013) - a pattern that resemble
closely to that found in some human patients with various eye diseases (Carroll et al., 2004; Choi
et al., 2006; Duncan et al., 2007; Joeres et al., 2008; Rossi et al., 2011). Even after such change
of cone mosaic in S334ter-line-3 rats, the regularity was maintained in that the cone rings were
repeatedly arranged across the retinas in uniform pattern.
However, the morphogenesis of the spatially organized photoreceptor mosaic still
remains as an open problem of tissue patterning in developmental retina. Multiple mechanisms
proposed to control for cell spacing within retinal mosaics include cell migration, cell death, cell
fate and cell-cell interaction (McCabe et al., 1999; Reese et al., 1995, 1999; Cook and Chalupa,
2000; Rockhill et al., 2000; Reese and Galli-Resta, 2002; Raven et al., 2003). Tissue inhibitor of
metalloproteinase (TIMP; Murphy et al., 1991) is an inhibitor of Metalloproteinase (MMP),
which is the major enzyme that degrades the ECM (Matrisian, 1992). Past studies (Akahane et al.,
119
2004; German et al., 2008; Hansson et al., 2011) have reported that the balance in the level of
MMP and TIMP is important for proper cell-ECM interaction and modulation of the migration of
neuronal cells including photoreceptors. The present study aimed to investigate whether or not
TIMP-1 can affect the spatial organization of cones in the normal and inS334ter-line-3 rat retinas.
(This model shall be referred to as the RP model in the rest of the article.)
Statistical measures are often utilized to quantitatively analyze mosaic. One of the most
commonly used measures is the regularity index (RI). RI for any mosaic is the mean of
distribution of the nearest-neighbor distance (ND) for all neurons divided by its standard
deviation (Wä ssle and Riemann, 1978). Another method is Voronoi domain analysis through
which the space regarding the mosaic is divided for each neuron. Each neuron in the mosaic is
enclosed by each Voronoi polygon that include region of space closest to itself than any other
neuron (Voronoi, 1907). No past studies to date have quantitatively analyzed the cone mosaics in
ring. Thus, the present work has utilized the ND and Voronoi domain analyses to statistically
quantify and better understand different cone mosaics. Being able to manipulate cone mosaics
and statistically analyze them will add to therapeutic approach for treating conditions, where
mosaic change is associated, such as RP and the above-mentioned eye diseases.
120
MATERIALS AND METHODS
Animals
The third line of albino Sprague-Dawley rats homozygous for the truncated murine opsin gene
(stop codon at residue 334; S334ter-line-3) was obtained from M.M. LaVail (University of
California, San Francisco, CA). These rats were kept in breed. For control, age-matched Sprague
Dawley rats (Harlan, Indianapolis, IN) were used. All rats were housed under cyclic 12/12-hour
light/dark conditions with free access to food and water. Both sexes of normal (control) and RP
rats were used. All procedures were in conformance with the Guide for Care and Use of
Laboratory Animals (National Institutes of Health). The University of Southern California
Institutional Animal Care and Use Committee reviewed and approved all procedures.
Administration of Tissue Inhibitor of Metalloproteinase 1
Tissue inhibitor of metalloproteinase 1 (TIMP-1; Sigma, T8947-5UG) was prepared in
phosphate-buffered saline (PBS; Sigma, P3813), adjusted to pH 7.4 and sterile-filtered before
administration. TIMP-1 was administered by intravitreal injection with a fine glass
microelectrode through the sclera at the level of the temporal peripheral retina. For preliminary
testing, 4 µ L of several different final concentrations of the TIMP-1 (10, 25 and 50 µ g/mL) were
applied on normal and RP rats at P20, 30, 45 and P60. Survival periods of 1-3 hours, 3, 5 days
and 1-6 weeks were tested. Both 25 and 50 µ g/mL gave similar end results in terms of degree of
change in M-opsin-immunoreactive cone mosaic, thus 4 µ L of 25 µ g/mL was utilized for the rest
of the experiment. It was also decided that the optimal stage for the injection was P45 – when
121
cones in RP retinas arranged in rings are lying flat against the surface of the retinal layer instead
of being upright, and made the observation of changes in mosaics easier and clearer. As for
survival periods, 1 hour, 2 weeks and 6 weeks were used as they best described the progress of
cone mosaic changes with TIMP-1 application. Sham injections, for control, consisted of 4 µ L of
the same sterile-filtered PBS. For each animal, one eye was used to inject TIMP-1, and the other
eye was used to inject saline for comparison. The entire injection procedure required only a few
minutes, and we finished before the animals recovered from anesthesia.
Tissue Preparation
The animals at P45 (1 hour survival period), P59 (2 weeks survival period) and P87 (6 weeks
survival period) were used (n = 15 for each stage). All animals were dark-adapted for at least 1
hour prior to sacrifice in the dark. Animals were deeply anesthetized by intra-peritoneal injection
of pentobarbital (40 mg/kg body weight) and the eyes were enucleated. Animals were then killed
with an overdose of pentobarbital. The anterior segment and crystalline lens were removed and
the eyecups were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4, for 30
minutes to 1 hr at 4 C. Following fixation, the retinas were carefully isolated from the eyecups
and were transferred to 30% sucrose in PB for 24 h at 4 C. For storage, all retinas (for cryostat
sections and whole mounts) were then frozen in liquid nitrogen, and stored at -70º C, thawed, and
rinsed in 0.01 M phosphate buffered saline (PBS; pH 7.4). For cryostat sections, eyecups were
embedded in OCT embedding medium (Tissue-Tek, Elkhart, IN), then quickly frozen in liquid
nitrogen and subsequently sectioned along the vertical meridian on a cryostat at a thickness of 20
m.
122
Immunohistochemistry
For fluorescence immunohistochemistry, 20-µm-thick cryostat sections were incubated in 10%
normal goat serum (NGS, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA,
dilution 1:1,000) or normal donkey serum (NDS, Jackson ImmunoResearch Laboratories, Inc.,
West Grove, PA, dilution 1:1,000) for 1 hr at room temperature. Sections were then incubated
overnight with either marker for middle-wavelength-sensitive opsin (M-opsin; kindly provided
by Dr. Cheryl Craft, Doheny Eye Institute, University of Southern California, Los Angeles,
dilution 1:1,500) or glutamine synthetase (GS; Chemicon; 1:300). Each antiserum was diluted in
PBS containing 0.5% Triton X-100 at 4º C. Retinas were washed in PBS for 45 min (3 x 15 min)
and afterwards incubated for 2 h at room temperature in either carboxymethylindocyanine-3
(Cy3)-conjugated affinity-purified donkey anti-rabbit IgG (Jackson ImmunoResearch
Laboratories, Inc., West Grove, PA, dilution 1:500) or Alexa 488 anti-goat IgG (Molecular
Probes, Eugene, OR, dilution 1:300). The sections were washed for 30 min with 0.1M PB and
coverslipped with Vectashield mounting medium (Vector Labs, Burlingame, CA). For whole
mount immunostaining, the same immunocytochemical procedures described above were used.
However, we used longer incubation times with primary antibodies (two nights with anti-M-
opsin and with anti-GS) and secondary antibodies (5 hours either with Cy3-conjugated donkey
anti-rabbit IgG or with Alexa 488 donkey anti-goat IgG).
For double-label studies, whole mounts were incubated for two nights in a mixture of
anti-M-opsin and anti-GS markers. Incubation with these antibodies used 0.5% Triton X-100 in
0.1 M PBS at 4 ° C. After this incubation, whole mounts were rinsed for 30 min with 0.1 M PBS.
Afterwards, we incubated them with Cy3-conjugated donkey anti-rabbit IgG and Alexa 488
donkey anti-goat for two nights at 4 ° C. In controls, the primary antibody was omitted from the
123
incubation solution. Whole mounts were then washed again for 30 min with 0.1 M PB and cover-
slipped with Vectashield mounting medium. Sections and whole mounts were then analyzed
using a Zeiss LSM 510, (Zeiss, NY) confocal microscope. Immunofluorescence images were
processed in Zeiss LSM-PC software. The brightness and contrast of the images were adjusted
using Adobe Photoshop 7.0 (Adobe Systems, Mountain View, CA). For presentation, all
Photoshop adjustments (brightness and contrast only) were carried out equally across sections.
Statistical analysis
Confocal micrographs (objective of 20x) of the retinas (n = 3-5 for each group) were taken at
focal level of M-opsin-immunoreactive cones cell bodies to cover a 1mm2 area at the mid-
peripheral region of the superior wing of the retina. The micrographs taken were used to
compose a collage for each of the samples. Each cell body were then marked by a dot of
appropriate size. These position map images were used for the ND and Voronoi domain analyses.
For the ND analysis, the distance to the nearest neighboring cell was measured for each dot
(Wä ssle and Riemann, 1978). The distributions were plotted in a histogram. The RI was
calculated for each of the tested mosaic by dividing the mean of the distribution by its standard
deviation (Wassle and Riemann, 1978). For the Voronoi domain analysis, the voronoi polygon
for each cell was generated and the areas of each polygon were calculated and plotted in a
histogram. In order to remove the artifacts induced by the edge, we did not include polygons
around the boundaries. These histograms were then compared to simulation distribution
generated from our random distribution model. This model was programmed to yield an
expected distribution for mosaic, which is completely random in the spacing of cells. This model
was programmed to take into account of the constraint in spacing induced by the cone nuclei size
124
(~5 µ m). The curve generated by the model was overlaid on the ND and Voronoi domain
analysis histograms for comparison. The sknewness of the Voronoi domain distributions for RP
group was also determined. All the measurements were expressed as mean ± standard errors.
Student‘s t -tests were used to examine the difference between two different means. The tests
were performed by MATLAB version 7.4.0 (The MathWorks Inc., Natick, MA, USA) and all
graphs were generated by Microsoft Excel spreadsheet, 2010 (Microsoft Corporation, Redmond,
WA, USA). A difference between the means of separate experimental conditions was considered
statistically significant at P < 0.05.
125
RESULTS
Disturbance of the M-opsin-immunoreactive cones mosaic in normal rat retinas with
TIMP-1
It was previously reported that the MMP-TIMP balance, which is important for proper cell-ECM
interaction (Ahuja et al., 2006), also modulates the migration of neuronal cells, including
photoreceptors (Reed et al., 2003; Akahane et al., 2004; German et al., 2008). To test our
hypothesis that cell-ECM interactions is important for controlling the precise spacing of cones,
we applied TIMP-1 first to normal retinas and examined the changes in the M-opsin-
immunoreactive cone mosaic. The M-opsin-immunoreactive cones were labeled in the whole
mount retinas in all groups – 1 hour, 2 weeks and 6 weeks post application with saline (the
control groups; Figs. 1A, B, C) and TIMP-1 (Figs. 1G, H, I). The images of M-opsin-
immunoreactive cone cell bodies marked gave their position maps (Figs. 1D-F, J-L). The M-
opsin-immunoreactive cones in control groups showed homogenous distribution patterns (Figs.
1A, B, C) that were similar to that seen in the normal retinas with no treatment (data not shown –
please refer to our previous paper for reference; Ji et al., 2012; Fig. 2A). The position maps
revealed this more clearly (Figs. 1D, E, F). In contrast, the M-opsin-immunoreactive cones
mosaic showed some changes with TIMP-1. First, the orientation of array of the outer segments
were disturbed in some regions (Figs. 1G, H, I). Rather than showing steady orientation as in
control groups (Figs. 1A, B, C), varying orientation was observed in retinas with TIMP-1 (Figs.
1G, H, I). Second, TIMP-1 induced some irregularity in the distribution pattern of M-opsin-
immunoreactive cones – not so clearly after 1hour (Fig. 1J) but especially after 2 weeks and 6
weeks (Figs. 1K, L). Some cells were seemingly more-or-less ‗bunched‘ up in groups (example
126
enclosed by an ellipse) of which their shapes and sizes seem random. For more details, please
refer to Supplementary Results.
Fig. 1 Disturbance of the M-opsin-immunoreactive cones mosaic
in normal rat retinas with TIMP-1
127
M-opsin-immunoreactive cone spacings in normal retinas with TIMP-1 loses regularity and
becomes close to random
Nearest-neighbor analysis was used to assess the distribution of nuclei of M-opsin
immunoreactive cells quantitatively (Wä ssle and Riemann, 1978). The distance of each nucleus
to its nearest neighbor was measured in a field of the mid-peripheral retina 1x1 mm2 in size,
which includes the area shown in Figures 1D-F, J-L. These distributions were plotted in a
histogram and were fit with the prediction of our random-position model (line) for both the
control (Figs. 2A-C) and TIMP-1 treated groups (Figs. 2D-F). The histograms for the control
groups (n = 3-5); 1 hour (1HR; Fig. 2A), 2 weeks (2WK; Fig. 2B), and 6 weeks (6WK; Fig. 2C)
post treatment showed near-Gaussian distributions that do not fit well with the prediction from
the random-distribution model (line). The mean total number (n) of the M-opsin-immunoreactive
cones within the evaluated 1mm2 retinal areas showed a decreasing trend with increasing
survival period – 1HR group (5189 ± 130; Fig. 2A), 2WK group (4278 ± 22; Fig. 2B) and 6WK
group (3942 ± 40; Fig. 2C). The mean ND of all the M-opsin-immunoreactive cones for 1HR
group was 9.53 ± 0.05 µ m (Fig. 2A), for 2WK group, 10.29 ± 0.08 µ m (Fig. 2B); and for 6WK
group, 10.88 ± 0.07 µ m (Fig. 2C). The mean RI for the M-opsin-immunoreactive cone mosaic for
the 1HR group was 4.21 ± 0.23 (Fig. 2A); for the 2WK group, 4.03 ± 0.05 (Fig. 2B); and for the
6WK group, 4.29 ± 0.26 (Fig. 2C).
However, the distribution changed with TIMP-1 (n = 3-5). They all showed extension of
range into the lower values of ND (Figs. 2D-F). Also, with longer survival periods (2 weeks and
6 weeks), the distribution became skewed to the lower range of ND and better-fitting with the
random-distribution model compared to their corresponding controls (Figs. 2D-F). The mean
128
total number of M-opsin-immunoreactive cones within the tested areas again showed a
decreasing trend from the three groups; 1HR (5390 ± 363; Fig. 2D), 2WK (4261 ± 106; Fig. 2E)
and 6WK (3949 ± 97; Fig. 2F). The mean ND for the M-opsin-immunoreactive cones for 1HR
group was 8.94 ± 0.46 µ m , for 2WK group, 9.86 ± 0.24 µ m; and for 6WK group, 9.93 ± 0.21
µ m. The mean RI for the M-opsin-immunoreactive cone mosaic was 3.62 ± 0.39 for the 1HR
group (Fig. 2D); 3.61 ± 0.11, for the 2WK group (Fig. 2E) and 3.32 ± 0.12 for the 6WK group
(Fig. 2F). The density (cells/1mm2 area at the superior mid-periphery retinal region), the ND and
the RI were compared between all six groups (Figs. 2G-I). The mean cone density graphs
revealed that there were significant decrease from 1HR to 2WK control group (p < 0.0041, one-
tailed Student‘s t-test) and between the 2WK and the 6WK control groups (p < 0.0165, one-
tailed Student‘s t-test). Also, among the TIMP-1 groups, the decrease was significant between
the 1HR and 2WK (p < 0.0261, one-tailed Student‘s t-test), and between 2WK and 6WK
conditions (p < 0.0475, one-tailed Student‘s t-test) (Fig. 2G). However, there were no significant
differences between the controls and the TIMP-1 groups. The mean ND graph indicated
significant decrease in the TIMP-1 group compared to the control group at 6 weeks post
treatment (p < 0.0181, one-tailed Student‘s t -test; Fig. 6D) (Fig. 2H). Within the control groups,
a significant increase in the mean ND was observed in the 2WK compared to 1HR condition (p <
0.0037, one-tailed Student‘s t-test). Within the TIMP-1 group, a significant increase was
observed in the 2WK compared to 1HR condition (p < 0.0277, one-tailed Student‘s t-test). The
mean RI graph showed significant difference decrease in the TIMP-1 group from the control
group at 2WK (p < 0.0261, one-tailed Student‘s t -test; Fig. 6I) and 6WK (p < 0.0143, one-tailed
Student‘s t -test; Fig. 6I) after treatment. In addition, significant decrease of the mean RI was
129
observed in the 6WK TIMP-1 group compare to the 2WK TIMP-1 group (p < 0.0017, one-tailed
Student‘s t-test).
Fig. 2 ND analysis distributions of control and TIMP-1 normal retinas
130
Voronoi domain analysis was also implemented as to further assess the distribution of
nuclei of M-opsin-immunoreactive cells quantitatively. This analysis generated a Voronoi
polygon for each nucleus, which included region of space closest to itself than any other
neighboring nucleus (Voronoi, 1907). The areas of the resulting polygons were measured in a
field of the mid-peripheral retina 1x1 mm2 in size, which includes the area shown in Figures 1D-
F, J-L. These distributions were plotted in a histogram and were fit with the prediction of the
random-distribution model (line) for both the control (Figs. 3A-C) and TIMP-1 treated groups
(Figs. 3D-F). Examples of the resulting Voronoi domains are shown shown inside a square
beside the each histogram (Figs. 3A-F). The histograms for the control groups (n = 3-5) – 1HR
(Fig. 3A), 2WK (Fig. 3B), and 6WK (Fig. 3C) showed distributions that are away from the fit
generated from the random-distribution model (line). However, such distribution changed with
TIMP-1 (n = 3-5) at 1 hour (Fig. 3D), 2 weeks (Fig. 3E), and 6 weeks (Fig. 3F) post treatment.
With increasing survival time periods, the distribution became wider, showed sknewness and fit
better with the random-distribution model (Figs. 2D-F).
131
Fig. 3 Voronoi domain analysis distribution for control and TIMP-1 normal retinas
No glial activation in TIMP-1 treated retina
Neurodegeneration and injury is associated with gliosis, which can be seen with increased GFAP
expression (Eng et al., 1994; Tani et al., 1996). Thus, in order to check if TIMP-1 was toxic to
the retinal cells, cryostat sections of normal retinas from the control group (Fig. 4A) and the
TIMP-1 treated group (Figs. 4B-D) were immunostained with GFAP. The control groups showed
no upregulation of GFAP expression at 1 hour (data not shown), 2 weeks and 6 weeks (data not
shown) after saline injection (Fig. 4A). The GFAP expression is seen predominantly in the nerve
fiber layer (NFL) and ganglion cell layer (GCL) (inner retina). The TIMP-1 group retinas also
showed no upregulation of GFAP, 1 hour (Fig. 3B), 2 weeks (Fig. 3C) and 6 weeks (Fig. 3D)
after treatment.
132
Fig. 4 No glial activation in TIMP-1 treated retina
Mosaic of M-opsin-immunoreactive cone can be manipulated with TIMP-1
Migration of M-opsin-immunoreactive cones and thus change in their mosaic upon TIMP-1
treatment was observed in normal retinas (Figs. 1-3). Subsequently, RP retinas were also tested
to see if cones in rings also changed in their mosaic with TIMP-1. The RP retinas of the control
groups (Figs. 5A-C) and the TIMP-1 treated groups (Figs. 5G-I) immunostained with M-opsin
showed the entire parts of the M-opsin-immunoreactive. The images M-opsin-immunoreactive
cone cell bodies marked gave their position maps (Figs. 1D- F, J- L). The control groups showed
their cell bodies forming the rims of the rings and processes pointing towards the center of the
rings (Figs. 5A-C). Also, these rings increased in their size with progression of the disease. The
position maps revealed this more clearly (Figs. 1D, E, F). Such M-opsin-immunoreactive cones
mosaic was manipulated remarkably with TIMP-1. At 1 hour post TIMP-1 treatment, the rings
that were already considerably loosened (Fig. 5G). The position map revealed such change more
clearly (Fig. 5J). The rings became less defined and smaller compared to the control group with
same survival period (Fig. 5D). At 2 weeks post treatment, the rings were close to gone as the M-
133
opsin-immunoreactive cones distribution became almost homogenous (Fig. 5H). Such striking
change was maintained, even 6 weeks after the treatment (Fig. 5I). The position maps revealed
this homogeneity of the spacings more clearly (Figs. 5K, L). The cells, however, were not spaced
in crystal-regularity. In fact, some cells were more closely spaced than the others, and showed
some degree (Figs. 5H,I).
134
Fig. 5 Mosaic of M-opsin-immunoreactive cone can be manipulated with TIMP-1
135
Mosaic of Mü ller cell processes can be manipulated with TIMP-1
Close association between the M-opsin-immunoreactive cones and the Mü ller cells (MCs) were
observed with mosaic change with TIMP-1 in normal retinas (refer to Supplementary Results). In
order to see if MCs in RP retinas are also involved during the movements of the M-opsin-
immunoreactive cones, the control and the TIMP-1 group retinas were immunostained with M-
opsin and GS. The results from the whole mount (Figs. 6A, B, E, F, I, J) and the cryostat-retinal
sections (Figs. 6C, D, G, H, K, L) are shown. The whole mount retinas of the control groups
showed the remodeled processes of the MCs filling the insides of each M-opsin-immunoreactive
cone rings at 1 hour (data not shown), 2 weeks and 6 weeks (data not shown) post injection (Fig.
6A). A high-magnification view of a ring marked by the inset rectangle revealed this more
closely (Fig. 6B). The cryostat section showed alternating regions rich in cell bodies and regions
showing primarily the processes of the M-opsin-immunoreactive cones at the upper level near
the outer plexiform layer (OPL; Fig. 6C). A higher-power micrograph of area marked by the
inset rectangle showed close association between the MC processes and the M-opsin-
immunoreactive cones (Fig. 6D). The whole mount RP retinas at 1 hour after TIMP-1 treatment
showed greater retinal spaces covered by the M-opsin-immunoreactive cone cell bodies, which
became more widely scattered and the rings became loosened (Fig. 6E). The MCs were still
filling inside the center of the shrinking rings but subsequently became distinct from the large,
neat bulbs observed in the control group (Figs. 6A, E). A higher-power micrograph revealed that
the MC processes were also covering regions outside the rings, becoming more homogenous (Fig.
6F). The cryostat section showed alternating regions rich in cell bodies and regions showing
primarily the processes of the M-opsin-immunoreactive cones at the outer part of the OPL (Fig.
6G). A higher-power micrograph showed close association between the MC processes and the
136
M-opsin-immunoreactive cones (Fig. 6H). The whole mount RP retinas 2 weeks and 6 weeks
(data not shown) post TIMP-1 treatment showed M-opsin-immunoreactive cones more
homogenously spaced and their rings no longer obvious (Fig. 6I). The MC apical processes also
showed homogenous pattern (Fig. 6I). A higher-magnification view also revealed small degree
of irregularity in the spacing of the M-opsin-immunoreactive cones in that their nuclei showed
some tendency to ‗stick‘ beside each other (Fig. 6J). The cryostat section showed M-opsin-
immunoreactive cones that are more evenly scattered across the length of the retina at the outer
part of the OPL (Fig. 6K). A higher-power micrograph showed close association between the
MC processes and the M-opsin-immunoreactive cones (Fig. 6L).
Fig. 6 Close association between cones and Mü ller cell processes during mosaic change
137
M-opsin-immunoreactive cone spacings in RP retinas with TIMP-1 gains homogeneity,
loses regularity, and becomes close to random
Again, the nearest-neighbor analysis was used to assess the distribution of nuclei of M-opsin
immunoreactive cells quantitatively. The histograms generated were fit with the prediction of our
random-distribution model (line) for both the control (Figs. 7A-C) and TIMP-1 treated groups
(Figs. 7D-F). The distributions for control groups (n = 3-5) of different survival periods – 1 hour
(Fig. 7A), 2 weeks (Fig. 7B), and 6 weeks (Fig. 7C) – showed distributions that were skewed at
the low ND range with a small peak at a high ND range. Also, the distributions became
increasingly away from the prediction generated from the random-distribution model. The mean
total number (n) of M-opsin-immunoreactive cones within the 1mm2 retinal areas tested showed
a decreasing trend from 1HR group (5402 ± 17; Fig. 7A) to 2WK group (4252 ± 33; Fig. 7B) and
to 6WK group (4037 ± 216; Fig. 7C). The mean ND for the M-opsin-immunoreactive cones for
1HR group was 7.31 ± 0.16 µ m (Fig. 7A), for 2WK group, 7.99 ± 0.33 µ m (Fig. 7B); and for
6WK group, 7.18 ± 0.09 µ m (Fig. 7C). The mean RI for the M-opsin-immunoreactive cone
mosaic was 4.55 ± 0.08 for the 1HR group (Fig. 7A); for the 2WK group, 4.74 ± 0.54 (Fig. 7B);
and for the 6WK group, 4.60 ± 0.16 (Fig. 7C).
However, the distribution changed with TIMP-1 (n = 3-5). For all groups of different
survival periods, the distributions were better fitting with the random-distribution model (Figs.
7D-F). The mean total number (n) of M-opsin-immunoreactive cones within the 1mm2 retinal
areas tested showed a decreasing trend from 1 HR group (5388 ± 28; Fig. 7D) to 2WK group
(4217 ± 54; Fig. 7E), and to 6WK group (3693 ± 27; Fig. 7F). The mean ND for the M-opsin-
immunoreactive cones for 1HR group was 7.72 ± 0.19 µ m (Fig. 7D), for 2WK group, 8.95 ±
138
0.04 µ m (Fig. 7E); and for 6WK group, 9.15 ± 0.31 µ m (Fig. 7F). The mean RI for the M-opsin-
immunoreactive cone mosaic was 4.85 ± 0.18 for the 1 HR group (Fig. 7D); for the 2WK group,
3.93 ± 0.22 (Fig. 7E); and for the 6WK group, 3.90 ± 0.13 (Fig. 7F). The cone density
(cells/1mm2 area at the superior mid-periphery retinal region), the ND and the RI were
compared between all six groups (Figs. 7G-I). The mean cone density graphs revealed that there
that there was a significant decrease in the 2WK control group compared to 1WK control group
(p < 0.0019, one-tailed Student‘s t-test). Among the TIMP-1 groups, significant reduction of the
mean density was observed at 2WK compared to 1HR (p < 0.0001, one-tailed Student‘s t-test),
and at 6WK compared to 2WK (p < 0.0017, one-tailed Student‘s t-test) (Fig. 7G). However,
there were no significant differences between the controls and the TIMP-1 treated groups. The
mean ND graph indicated significant increase in the TIMP-1 groups compared to the control
groups at all tested stages; at 1 hour (p < 0.0245, one-tailed Student‘s t -test), at 2 weeks (p <
0.0088, one-tailed Student‘s t -test) and 6 weeks (p < 0.00079, one-tailed Student‘s t -test) post
treatment (Fig. 7H). Among the TIMP-1 groups, there was a significant increase in the mean ND
in 2WK compared to 1HR (p < 0.0066, one-tailed Student‘s t-test). The mean RI graph showed
that there was a significant decrease in the 6WK TIMP-1 group compared to 6WK control group
(p < 0.0190, one-tailed Student‘s t -test) (Fig. 7I). Within the TIMP-1 groups of different testing
stages, a significant reduction in the mean RI was observed at 2WK compared to 1HR (p <
0.0056, one-tailed Student‘s t-test) post treatment.
139
Fig. 7 ND analysis distributions of control and TIMP-1 RP retinas
Voronoi domain analysis was implemented as another mean to assess the spacings of
nuclei of M-opsin-immunoreactive cells quantitatively. The areas of the resulting polygons were
measured in a field of the mid-peripheral retina 1x1 mm2 in size, which includes the area shown
in Figures 5D-F, J-L. These distributions were plotted in a histogram and were fit with the
140
prediction of the random-distribution model (line) for both the control (Figs. 8A-C) and TIMP-1
treated groups (Figs. 8D-F). Examples of the resulting Voronoi domains are shown inside a
square beside the each histogram (Figs. 8A-F). The histograms for the control groups (n = 3-5)
after 1 hour (Fig. 8A), 2 weeks (Fig. 8B), and 6 weeks (Fig. 8C) all showed distribution skewed
to the low Voronoi domain area range with a single large peak at a large range. These
distributions were all far away from the fit generated from the random-distribution model.
However, the distribution changed with TIMP-1 (n = 3-5). With increasing survival time
periods, the distribution became wider with much less pronounced peak at a large value range
(Figs. 8D-F). Also, they showed close fit with the random-distribution model prediction. The
sknewness of each of the distributions (for both the data and the model simulation; Figs. 8A-F)
were plotted in a bar graph (Fig. 8G). The results showed that the sknewness of the simulated
random-distribution for all the control groups was significantly higher compared to the
distribution from the actual data collected; 1HR (p < 0.0036, one-tailed Student‘s t -test) , 2WK
(p < 0.0186, two-tailed Student‘s t -test) and 6WK (p < 0.0471, two-tailed Student‘s t -test). Same
was also true for the 1HR TIMP-1 group (p < 0.0285, two-tailed Student‘s t -test). However, in
2WK and 6WK TIMP-1 groups, no significant differences of sknewness were observed between
the distribution from the actual data and the model data.
141
Fig. 8 Voronoi domain analysis distribution for control and TIMP-1 RP retinas
142
DISCUSSION
In the present study, two mosaic measures were taken into account – regularity and homogeneity.
A mosaic was considered regular if local pattern of certain cell spacings were repeated
throughout the entire area. Thus, the cone mosaics in normal retinas and rings in RP retinas were
considered regular as same spacing patterns were repeatedly observed throughout the retina. In
contrast, homogeneity considered whether or not the local statistical measures were even
throughout the retina. Thus, the cone mosaic found in normal retinas was be considered
homogenous. However, rings were considered not homogenous as the statistical measures
outside the rings and inside the rings are different.
Saline does not change M-opsin-immunoreactive cone mosaic
We observed a regular and homogenous mosaic in the control groups of normal retinas (Figs.
1A-F). Such distribution pattern showed no difference to that one usually sees in normal retinas
(Ji et al., 2012). The ND and Voronoi domain analysis histograms revealed distributions that did
not fit well with the prediction generated from our random-distribution model (Figs. 2A-C, 3A-
C). This indicated that the mosaic in the control retinas were not random but were under precise
control. Also, the histograms were near-Gaussian, which indicated that the mosaic is
homogenous. The mean M-opsin-immunoreactive cone density graph revealed significant drop
(Fig. 2G). However, the GFAP immunostain of the control retina revealed absence of glial
activation (Fig. 4A). It is well known that glial activation is associated with retinal injuries
(Harada et al., 1995; Wen et al., 1995; Wen et al., 1998). This is also the reason why the GFAP
143
immunoreactivity was examined only in the normal retinas in the present study, as it is highly
upregulated in RP retinas (Lee et et., 2011). Thus, the drop in the mean M-opsin-immunoreactive
cone density could be explained by the growth of the retinal area involved with increasing
postnatal stage (McCall et al., 1987; Harman et al., 2003; Ji et al., 2012). Moreover, there is
lacking evidence of new cones born by development or transdifferentiation in normal adult
mammalian retinas. Thus, it is likely that the decline in the cone density occurred as retinas grew
in their area. This could also explain the significant increase in the mean ND observed in the
1HR and 2WK groups (Fig. 2H). The cells may have become more distantly spaced to maintain
the mosaic despite the growth of the retina. There were no significant changes in the RI in the
control groups of normal retinas, which suggests that the regularity of the mosaic was not disrupt.
The typical RI for a regular mosaic is in range 3 – 9 (Wassle & Riemann, 1978).
Saline also did not disrupt the M-opsin-immunoreactive cone mosaic in RP retinas (Fig.
5A-F). The M-opsin-immunoreactive cones were arranged side by side forming the rim of the
rings with their cell bodies and the outer segments, and extending their processes into the center
of the rings (Figs. 5A-C). The rings were left unaltered from what one would expect to observe
from the RP retinas (Lee et al., 2011; Ji et al., 2012). The ND and the Voronoi domain area
histograms all showed a skewed distribution with a peak at a large value range (Figs. 7A-C, 8A-
C). The Voronoi domain analysis proved to be a more sensitive means of test compared to the
ND analysis for RP retinas. The peaks at a large value range, more pronounced in the Voronoi
domain analayses, represent large domains caused by cells forming the edge of the rings and a
very few cells found inside the rings (Figs. 8A-C). In all, these distributions became increasing
away from the random-distribution model prediction as the rings became larger with the
progression of the disease (Ji et al., 2012). Thus, the mosaics of rings were not homogenous and
144
not random. Again, the mean cone density showed significant drops with increasing postnatal
stages (Fig. 7G), which we believe is produced by the growth of the retinal area. The RI for all
control groups showed no significant differences from each other (Fig. 7I). The value range of RI
indicated that the rings were highly regular.
In conclusion, neither saline nor injection itself had any effect on the mosaic of M-opsin-
immunoreactive cones.
M-opsin-immunoreactive cone mosaic can be manipulated by TIMP-1
Contrary to saline, TIMP-1 treatment led to significant changes in the M-opsin-immunoreactive
cone mosaics in both normal and RP retinas. In normal retinas, TIMP-1 led some cones to bunch
up (Figs. 1G-L). Such observation is supported by a significantly less mean ND measured in the
TIMP-1 group compared to the normal group by 6 weeks post treatment (Fig. 2H). In addition,
histograms from the ND and the Voronoi domain analysis became skewed towards the lower ND
and Voronoi domain area range, especially after 2 and 6 weeks (Figs. 2D-F, 3D-F). These results
suggest that the M-opsin-immunoreactive cones became more closely spaced or bunched up, and
the mosaic became less homogenous. With time, the histograms became better fitting with the
prediction from the random-distribution model (Figs. 2D-F, 3D-F). Thus, the mosaic became
close to random distribution, but not completely. The RI was significantly less than the control
group after 2 weeks and 6 weeks (Fig. 2I). The lower the RI, the less regular the mosaic is
(Wassle & Riemann, 1978). Thus, TIMP-1 reduced the regularity of the mosaic in normal retinas.
145
In RP retinas, TIMP-1 led to striking changes to the M-opsin-immunoreactive cone
mosaics. Even only 1 hour after the treatment, the rings became visibly smaller in size and less
distinct (Figs. 5G, J). By 2 weeks, the cells were no longer in obvious rings but were scattered
across the retina and showed some tendency of the nuclei to clump (Figs. 5H, K, 6J). And such
mosaic was maintained at even much later stage (Figs. 5I, L). The ND analysis results showed
increase in the number of cells with greater ND, especially in 2WK and 6WK groups (Fig. 7E, F).
This indicates that the spacings of the M-opsin-immunoreactive cone became further apart
compared to the usual RP retinas. In other words, the cells that were spaced side by side around
rings became scattered as they no longer maintained rings. This is supported by significantly
larger mean ND in TIMP-1 groups at all stages compared to the control groups (Fig. 7H). The
significant reduction of the RI in the TIMP-1 group 6 weeks post treatment indicates that TIMP-
1 disrupted the regularity of the mosaic. Both the ND and the Voronoi domain analysis
histograms became much better fitting with the random-distribution model (Figs. 7D-F, 8E, F),
revealing that the mosaics became close to a predicted random distribution with TIMP-1. In
addition, the less sknewness of the ND histogram compared to the control group suggests that
TIMP-1 induced homogeneity in the RP retinas (Figs. 7E, F). Moreover, the graph of sknewness
for the Voronoi distributions from the actual data collected in 2WK and 6WK retinas showed no
significant differences to that of the simulation-distribution (Fig. 8G). This result strengthens our
evidence that TIMP-1 restores the homogeneity of the M-opsin-immunoreactive cone mosaic.
In summary, TIMP-1 can modulate the M-opsin-immunoreactive cone mosaics
significantly both in the normal and in the RP retinas by disrupting their regularity and restoring
their homogeneity.
146
Possible mechanisms underlying modulation of M-opsin-immunoreactive cone mosaic with
TIMP-1
In normal retinas, TIMP-1 also disrupt the orientation of the outer segments of M-opsin-
immunoreactive cones (Figs. 1G-I). The arrays of the outer segments were no longer in regular
orientation. This may be due to the some disruption made in the interphotoreceptor matrix (IPM)
surrounding the photoreceptor outer segments (Johnson et al., 1986). TIMP-1 and MMPs are
found in the IPM among various ocular structures and are suggested to play an important role in
the turnover of the IPM component (Padgett et al., 1997; Ahuja et al., 2006). Disruption of the
IPM may have resulted in improper hold of the cone outer segments.
The exact mechanism(s) through which TIMP-1 induced striking change in the mosaic is
subject to further study. However, in our study, TIMP-1 induced the migration of M-opsin-
immunoreactive cones and thus their mosaic change in both normal and in R retinas (Figs. 1G-L,
5G-L). The close association observed between the M-opsin-immunoreactive cones and MC (Fig.
6) may reflect MC giving aid to cones for their migration (Sullivan et al., 2003; Lee et al., 2011).
For movement of cells to occur, their interaction with the surrounding ECM is necessary (Raines,
2000; Streuli and Akhtar 2009). And since, TIMP and MMP are the key enzymes that modulate
the degradation of the ECM (Murphy et al., 1991; Matrisian, 1992), delicate balance in their
level is bound to be crucial. Indeed, the imbalance in the level of MMP and TIMP was reported
to modify the cell-ECM interactions (Ahuja et al., 2006), and also to affect the migration of
neuronal cells, including photoreceptors (Akahane et al., 2004; German et al., 2008; Hansson et
al., 2011). Taken together, we propose that TIMP-1 modulated the cone-ECM interaction and
thus led to migration of cones necessary for mosaic change.
147
Possible mechanisms controlling the regularity of cone mosaics in the retina
The current study revealed that TIMP-1 application disrupted much of the regularity of M-opsin-
immunoreactive cone mosaics (Figs 2I, 7I). The mosaic also became very similar to simulated
random distribution (Figs. 2A-F, 7A-F). TIMP-1 is one of the key enzymes associated with
modulation of ECM breakdown, turnover and its metabolism (Alexander and Web, 1989;
Matrisian, 1992; Woessner, 1991; Jones et al., 1994; Padgett et al., 1997), Taken together, the
current study brings out that proper cone-ECM interaction is necessary for the regularity of cone
mosaics.
The current study also showed that TIMP-1 treatment on RP retinas successfully restored
the homogeneity of the M-opsin-immunoreactive cone mosaics. The resulting mosaic had similar
Voronoid domain statistics to that of the normal control retinas (Figs. 3, 8). Thus, TIMP-1 may
be useful for when restoration of homogeneity of photoreceptor mosaics rings is concerned – for
example, in RP retinas and human patients with some eye diseases that show similar ring
mosaics (Carroll et al., 2004; Choi et al., 2006; Duncan et al., 2007; Joeres et al., 2008; Rossi et
al., 2011). However, a drawback to TIMP-1 is that it also lessens the regularity of the mosaic,
which will disrupt the equal sampling of the visual space (French et al., 1997; Manning and
Brainard, 2009).
Further understandings of what cellular and molecular factors can manipulate the cone
mosaics will enlighten therapeutic methods for treatment of various eye diseases, where
remodeling of photoreceptors is necessary in order to achieve efficient sampling of the visual
space.
148
REFERENCES
Akahane, T., et al. (2004). "TIMP-1 inhibits microvascular endothelial cell migration by MMP-
dependent and MMP-independent mechanisms." Exp Cell Res. 301(2): 158-167.
Alexander CM, W.Z. (1989). Proteinases and extracellular matrix remodeling. Curr Opin Cell
Biol 1, 974-982.
Ahuja S, A.P., Caffé AR, Ekstrom P, Abrahamson M, van Veen T. (2006). rd1 mouse retina
shows imbalance in cellular distribution and levels of TIMP-1/MMP-9, TIMP-2/MMP-2 and
sulfated glycosaminoglycans. Ophthalmic Res 38, 125-136.
BreMiller, R. and K. Larison (1990). "Early onset of phenoype and cell patterning in the
embryonic zebrafish retina." Development 109: 567-576.
Bruhn, S. and C. Cepko (1996). "Development of the Pattern of Photoreceptors in the Chick
Retina." The Journal of Neuroscience 16(41430-1439).
Carroll J, Neitz M, Hofer H, Neitz J, Williams DR. 2004. Functional photoreceptor loss revealed
with adaptive optics: An alternate cause for color blindness. Proc Natl Acad Sci U S
A. 101(22):8461-6.
149
Choi SS, Doble N, Hardy JL, Jones SM, Keltner JL, Olivier SS, Werner JS. 2006. In vivo
imaging of the photoreceptor mosaic in retinal dystrophies and correlations with visual function.
Invest Ophthalmol Vis Sci. 47(5):2080-92.
Cook, J. E. and L. M. Chalupa (2000). "Retinal mosaics: new insights into an old concept."
Trends Neurosci. 23: 26-34.
Duncan JL, Zhang Y, Gandhi J, Nakanishi C, Othman M, Branham KE, Swaroop A, Roorda A.
2007. High-resolution imaging with adaptive optics in patients with inherited retinal
degeneration. Invest Ophthalmol Vis Sci. 2007 Jul;48(7):3283-91.
Eng, L. and R. Ghirnikar (1994). "GFAP and astrogliosis." Brain Pathol. 4(3): 229-237.
French, A., et al. (1997). "Image degradation by an irregular retinal mosaic." Biol Cybern 27:
229-233.
Galli-Resta, L., et al. (1999). "Modelling the mosaic organization of rod and cone photoreceptors
with a minimal-spacing rule." European Neuroscience Association 11: 1461-1469.
150
German, O., et al. (2008). "Retinal Pigment Epithelial Cells Promote spatial reorganization and
differentiation of Retina Photoreceptors." J of Neuro Res 86: 3503-3514.
Hansson, J., et al. (2011). "Biomarkers of extracellular matrix metabolism (MMP-9 and TIMP-1)
and risk of stroke, myocardial infarction, and cause-specific mortality: cohort study." PLoS ONE
19(6): e16185.
Harada, T., et al. (1995). "Phosphorylation of CREB in rat retinal cells after focal retinal injury."
Exp Eye Res 61: 769-771.
Harman, A., et al. (2003). "Numbers of neurons in the retinal ganglion cell layer of the rat do not
change throughout life." Gerontology 49: 350-355.
Ji Y, Zhu CL, Grzywacz NM, Lee EJ. Rearrangement of the cone mosaic in the retina of the rat
model of Retinitis Pigmentosa. J Comp Neurol. 2012. 520(4):874-88.
Jones BE, M.P., Gallo S, Tombran-Tink J, Arand G, Reid DA, Thompson EW, Chader GJ,
Waldbillig RJ. (1994). Characterization and novel activation of 72-kDa metalloproteinase in
retinal interphotoreceptor matrix and Y-79 cell culture medium. Exp Eye Res 59, 257-269.
151
Johnson L, Hageman GS, Blanks JC. (1986) Interphotoreceptor Matrix Domains Ensheath
Vertebrate Cone Photoreceptor Cells. Inv Ophthal & Vis Sci. 27(2):129-135.
Kram, Y., et al. (2010). "Avian cone Photoreceptors Tile the Retina as Five Independent, Self-
Organizing Mosaics." PLoS ONE 5(2): e8992.
Lee EJ, Ji Y, Zhu CL, Grzywacz NM. Role of Mü ller cells in cone mosaic rearrangement in a rat
model of Retinitis Pigmentosa. Glia. 2011. 59(7):1107-17.
Lin, B., et al. (2004). "Retinal Ganglion Cell Type, Size, and Spacing can be Specified
Independent of Homotypic Dendritic Contacts." Neuron 43: 475-485.
Manning, J. and D. Brainard (2009). "Optimal design of photoreceptor mosaics: why we do not
see color at night." Vis Neurosic 26: 5-19.
Matrisian, L. (1992). "The matrix-degrading metalloproteinases." Bioessays 14(7): 455-463.
McCabe, K. L., et al. (1999). "The development of the pattern of retinal ganglion cells in the
chick retina: mechanisms that control differentiation." Development 126: 5713-5724.
152
McCall, M., et al. (1987). "Differential retinal growth appears to be the primary factor producing
the ganglion cell density gradient in the rat." Neurosci Lett 79: 78-84.
Murphy, G., et al. (1991). "The N-terminal domain of tissue inhibitor of metalloproteinases
retains metalloproteinase in-hibitory activity." Biochemistry 30: 8097-8102.
Raines, E. (2000). "The extracellular matrix can regulate vascular cell migration, proliferation,
and survival: relationships to vascular disease." Int. J. Exp. Pathol. 81: 173-182.
Padgett LC, L.G., Werb Z, LaVail MM. (1997). Matrix metalloproteinase-2 and tissue inhibitor
of metalloproteinase-1 in the retinal pigment epithelium and interphotoreceptor matrix: vectorial
secretion and regulation. Exp Eye Res 64, 927-938.
Raven, M. A., et al. (2003). "Determinants of the exclusion zone in dopaminergic amacrine cell
mosaics." J Comp Neurol 461: 123-136.
Raven, M. A., et al. (2003). "Determinants of the exclusion zone in dopaminergic amacrine cell
mosaics." J. Comp. Neurol. 461: 123-136.
153
Raymond, P., et al. (1995). "Developmental Patterning of Rod and Cone Photoreceptors in
Embryonic Zebrafish." The Journal of Comparative Neurology 359: 537-550.
Reese, B. E. and L. Galli-Resta (2002). "The role of tangential dispersion in retinal mosaic
formation." Prog. Retin. Eye Res. 21: 153-168.
Reese, B. E., et al. (1999). "Clonal expansion and cell dispersion in the developing mouse
retina." Eur. J. Neurosci. 11: 2965-2978.
Rockhill, R. L., et al. (2000). "Spatial order within but not between types of retinal neurons."
Proc. Natl. Acad. Sci. USA 97: 2303-2307.
Rossi EA, Chung M, Dubra A, Hunter JJ, Merigan WH, Williams DR. 2011.
Imaging retinal mosaics in the living eye. Eye (Lond). 25(3):301-8.
Streuli, C. and N. Akhtar (2009). "Signal co-operation between integrins and other receptor
systems." Biochemistry 418: 491-506.
Tani, M., et al. (1996). "In situ hybridization analysis of glial fibrillary acidic protein mRNA
reveals evidence of biphasic astrocyte activation during acute experimental autoimmune
encephalomyelitis." Am J Pathol 148(3): 889-896.
154
Voronoi, G. (1907). "Nouvelles applications des param`etres continus `a la th´ eorie des formes
quadratiques. Primiere M´ emoire: Sur quelques prepri´ et´ es des formes quadratiques positives
parfaites. ." J Reine Angew Math 133: 97-178.
Wä ssle, H. and B. Boycott (1991). "Functional architecture of the mammalian retina." Physiol
Rev. 71(2): 447-480.
Wassle, H. and H. Riemann (1978). "The mosaic of nerve cells in the mammalian retina."
Proceedings of the Royal Society of London Series B 200: 441-461.
Wen, R., et al. (1998). "Continuous exposure to bright light upregulates bFGF and CNTF
expression in the rat retina." Curr Eye Res 17: 494-500.
Wen, R., et al. (1995). "Injury-induced up-regulation of bFGF and CNTF messenger-RNAS in
the rat retina." J Neurosci 15: 7377-7385.
Wikler, K. and P. Rakic (1991). "Relation of an array of early-differentianting cones to the
photoreceptor mosaic in the primate retina.
." Nature 351(6325): 397-400.
155
Wikler, K. and P. Rakic (1994). "An array of early differentiating cones precedes the emergence
of the photoreceptor mosaic in the fetal monkey retina." Proc. Natl. Acad. Sci. USA 91: 6534-
6538.
Williams, D. and N. Coletta (1987). "Cone spacing and the visual resolution limit." J. Opt. Soc.
Am. A 4(8): 1514-1523.
Woessner JF Jr. (1991). Matrix metalloproteinases and their inhibitors in connective tissue
remodeling. FASEB J 5, 2145-2154.
Zhu CL, Ji Y, Grzywacz NM, Lee EJ. Spatiotemporal Pattern of Rods in Rat Model of Retinitis
Pigmentosa. Cell Tissue Res. 2013. 351:29-40
156
FIGURE LEGENDS
Figure 1. Confocal micrographs taken from whole mount normal retinas processed for M-opsin
immunoreactivity (A, B, C, G, H, I) and nuclei position maps (D, E, F, J, K, L). The micrographs
for control groups show P45 N (A), P59 N (B), P87 N (C) rat retinas 1 hour, 2 weeks, and 6
weeks post saline injection, respectively. Each dot represents a M-opsin-immunoreactive cone
nucleus (D, E, F). The mosaic is regular and homogenous. The micrographs for TIMP-1-injected
groups show P45 N (G), P59 N (H), P87 N (I) rat retinas 1 hour, 2 weeks, and 6 weeks post
injection, respectively. The position maps (J, K, L) reveal that, not so much after 1 hour, but 2
weeks and 6 weeks after TIMP-1 intravitreal injection, some M-opsin-immunoreactive cone cells
show ‗groupings‘ in hard-to-define shapes and sizes (example in eclipse). HR, hour; WK, week;
N, normal; RP, retinitis Pigmentosa; TIMP-1, Tissue inhibitor of metalloproteinase 1. Scale bars
= 500 µ m.
Figure 2. Histogram for number of M-opsin-immunoreactive cells (within the 1mm
2
in the
dorsal mid-peripheral area) versus the ND analyzed in the normal retinas. Control groups (n = 3-
5) with survival period of 1 hour (A), 2 weeks (B), 6 weeks (C), and TIMP- groups (n = 3-5)
with survival period of 1 hour (D), 2 weeks (E), and 6 weeks (F) are shown. These histograms
are overlaid with distribution generated from the random-distribution model (line). The summary
graphs for mean cone density (G), mean ND (H) and the mean RI (I) for all six groups are
illustrated. With TIMP-1 treatment, the ND distributions gradually become better fitting with the
prediction of random-distribution model and the RI is reduced significantly. Data presented as
157
mean ± standard error. The symbol * represents p < 0.05 or better. n, mean total number of cells;
ND, Nearest-neighbor distance; RI, Regularity index.
Figure 3. Histogram generated from Voronoi domain analysis for all normal retinal groups are
illustrated. Control groups (n = 3-5) with survival period of 1 hour (A), 2 weeks (B), 6 weeks (C),
and TIMP- groups (n = 3-5) with survival period of 1 hour (D), 2 weeks (E), and 6 weeks (F) are
shown. These histograms are overlaid with distribution generated from the random-distribution
model (line). Voronoi domains within the test areas (1x1 mm
2
of dorsal mid-peripheral retina)
are shown as examples (~ 170 µ m x 170 µ m) for each group. The Voronoi domain analysis
distribution become better fitting with the random-distribution model upon TIMP-1 treatment.
Scale bar = 60 µ m.
Figure 4. Confocal micrographs taken from cryostat sections of normal retinas processed for
GFAP immunoreactivities are shown for 2WK control (A), 1 HR (B), 2 WK (C), and 6 WK
TIMP-1 groups. Small extent of immunoreactivity is observed at the ganglion cell layer and in
the nerve fiber layer. ONL, outer nuclear layer; OPL, outer plexiform layer; INL, Inner nuclear
layer; IPL, Inner plexiform layer; GCL, Ganglion cell layer; NFL, Nerve Fiber Layer. Scale bar
= 100 µ m.
Figure 5. Confocal micrographs for control groups show P45 RP (A), P59 RP (B), P87 RP (C)
retinas, 1 hour, 2 weeks, and 6 weeks post saline treatment, respectively. The corresponding
158
position maps illustrate dots for M-opsin-immunoreactive cone nuclei (D, E, F). Rings are
observed in the M-opsin-immunoreactive cone mosaics. Confocal micrographs for TIMP-1-
injected groups are shown for P45 RP (G, J), P59 RP (H, K), and P87 RP (I, L) retinas, 1 hour, 2
weeks, and 6 weeks post treatment, respectively. The corresponding position maps (J, K, L)
reveal loosening of rings and increased homogeneity of M-opsin-immunoreactive cone mosaic
with TIMP-1. RP, Retinitis Pigmentosa. Scale bars = 500 µ m.
Figure 6. Confocal micrographs taken from whole mounts and vertical sections of saline or
TIMP-1-injected RP retinas processed for GS (green) and M-opsin (red) immunoreactivities.
Double exposure (A) show M-opsin-immunoreactive cone rings around the remodeled Mü ller
cell processes in brocolli-like shapes (A). A higher-power micrograph (B) shows that the Mü ller
cell apical processes are filling inside the ring. The cryostat section show alternating regions rich
in cell bodies and regions showing primarily the processes of the M-opsin-immunoreactive cones
at the upper level of the OPL (C). A higher-power micrograph reveals close association between
the Mü ller cell processes and the M-opsin-immunoreactive cones (D). TIMP-1-injected RP
retinas 1 hour post treatment show more scattered M-opsin-immunoreactive cones and Mü ller
cell processes losing their neat broccoli-like shapes (E). A higher-power micrograph show this
more clearly (F). The cryostat section show alternating regions rich in cell bodies and regions
showing primarily the processes of the M-opsin-immunoreactive cones at the OPL (Fig. G). A
higher-power micrograph show close association between the Mü ller cell processes and the M-
opsin-immunoreactive cones (H). TIMP-1-injected RP retinas 2 weeks post treatment show
more-or-less homogenous mosaic of M-opsin-immunoreactive cone mosaic and Mü ller cell
159
processes (I). A higher-magnification reveal that M-opsin-immunoreactive cones show some
tendency to cluster in a group of few (J). The cryostat section showed M-opsin-immunoreactive
cones more evenly scattered across the length of the retina at the OPL (K). A higher-power
micrograph show close association between the Mü ller cell processes and the M-opsin-
immunoreactive cones (L). All scale bars = 100 µ m.
Figure 7. Histogram of number of M-opsin-immunoreactive cells (within the 1mm
2
area tested)
versus the ND obtained using the control RP retinal groups (n = 3-5) with survival periods of 1
hour (A), 2 weeks (B), 6 weeks (C), and for TIMP-1 RP retinal groups (n = 3-5) with survival
period of 1 hour (D), 2 weeks (E), and 6 weeks (F). These histograms are overlaid with the
prediction of the random-distribution model (line). The summary graphs for mean cone density
(G), mean ND (H) and the RI (I) for all groups are illustrated. The TIMP-1 groups show
widening in their ND distribution, which gradually become better fitting with the random-
distribution model. The RI is also significantly reduced with TIMP-1 by 6 weeks after treatment.
Data presented as mean ± standard error. The symbol * represents p < 0.05 or better. n, mean
total number of cells; ND, Nearest-neighbor distance; RI, Regularity index.
Figure 8. Histogram generated from Voronoi domain analysis for all RP retinal groups are
illustrated. Control groups (n = 3-5) with survival period of 1 hour (A), 2 weeks (B), 6 weeks (C),
and TIMP- groups (n = 3-5) with survival period of 1 hour (D), 2 weeks (E), and 6 weeks (F) are
shown. These histograms are overlaid with distribution generated from the random-distribution
model (line). Voronoi domains within the test areas (1x1 mm
2
of dorsal mid-peripheral retina)
160
are shown as examples (~ 170 µ m x 170 µ m) for each group. The Voronoi domain analysis
distribution becomes well-fitting with the random-distribution model upon TIMP-1 treatment.
The mean sknewness of the Voronoi domain distributions are plotted for all groups from both
their collection of real data and from their simulation of the random-distribution model (G). The
sknewness of the actual data distribution gradually lessen and become similar to that of the
simulated prediction of the random-distribution model. VD, Voronoi; simu, simulation. Scale bar
= 60 µ m.
161
SUMMARY
_____________________________________________________________________________________
In my research project, there were four main parts. Scientists have for long tried to
understand the fate of cones in RP retinas in determination to rescue them from degeneration at
later stages of the disease following rod degeneration. However, no past studies have examined
cone mosaic in depth in the whole mount of RP retinas. In the first part, the extensive remodeling
of cone mosaic in the S334ter-line-3 rat retinas was examined in depth. We have found that
cones in these RP retinas rearranged themselves in a regular array of rings and survived for a
long time after rod deaths. The rings continued to remodel through later stage of life as their
mean size and quantity increased with the progression of disease before the eventual cone deaths.
Similar photoreceptor distribution pattern of rings are also observed in human patients with some
eye diseases, and this makes the current study even more significant (Carroll et al., 2004; Choi et
al., 2006; Duncan et al., 2007; Joeres et al., 2008; Rossi et al., 2011). In the second part of the
research project, we examined our hypothesis that rod deaths would have some relationship to
the cone rings. We have found that rods degenerated initially in random scatter. Gradually, these
dying rods seemed to affect the neighboring rods as they led to clusters of deaths, which then
radially propagated outward in a circular wave fashion to create growing rings of dying rods
leaving growing holes in their mosaic. These holes were always accompanied by cone rings that
occurred exactly at the same locations. Thus, we have shown that rod deaths triggered the
remodeling of cone mosaic as they migrated away from regions where rods were unhealthy or
dead. It was previously reported that Mü ller cells aided the migration of neurons (Sullivan et al.,
2003). In the third part of the study, we showed that cones and Mü ller cells in our RP retinas
were indeed in close association. Moreover, we found that Mü ller cells remodeled remarkably in
162
their apical processes to fill inside each cone ring. We then aimed to examine whether or not the
Mü ller cells were necessary for cone rings. We performed the experiment this by poisoning the
Mü ller cells and showed ring deformation. In conclusion, we have illustrated that proper
metabolism of Mü ller cells and their interaction with cones were necessary for the maintenance
of cone rings. In the final part of the study, a broader and more fundamental question of what
determines and modulates the regularity of the cone mosaics in normal and in RP rat retinas was
addressed. For this investigation, tissue inhibitor of metalloproteinase (TIMP-1) known to
modulate the degradation and turnover of the extracellular matrix, and also affect the ability of
photoreceptors to migrate, was administered into the eyes of both normal and RP rats. The
results indicated that TIMP-1 led to striking change in the cone mosaic in both normal and in RP
retinas. Upon TIMP-1 application, cones in normal retinas showed tendency to clump while
cones in RP retinas lost their ring arrangement and regained homogeneity of their mosaic. Most
important of all, the TIMP-1in normal and RP retinas disrupt the regularity of cone mosaics as
they became close to that of random distribution. Therefore, we conclude that the cone-ECM
interaction is necessary for controlling of regular cone spacing. In addition, we have shown that
TIMP-1 is one of the key factors that take part in modulation of such process. We hope that the
addition to knowledge harvested from the current research would eventually lead to better
approach for therapy against Retinitis Pigmentosa and other above-mentioned eye diseases,
where retinal transplantation or reengineering of cone mosaic is necessary.
163
Supplementary Results
__________________________________________________________________
Mü ller cells are closely associated with cones during movement their movement (Lee et al.,
2012). Thus, we examined to see if cones were migrating along with Mü ller cell processes. The
double labeling with M-opsin and GS in the control group showed homogenous pattern of Mü ller
cell apical processes (Supp. Fig. A) and position of M-opsin-immunoreactive cones (Supp. Figs.
A, B) in the control group. Such result showed similar pattern as in normal retinas with no
treatment (data not shown, please refer to Lee et al., 2011, Fig. 5C). The cell bodies of the M-
opsin-immunoreactive cones were also spaced in a homogenous arrangement (Supp. Fig. C).
Cryostat sections of the retinas showed close association between the processes of the Mü ller
cells and the M-opsin-immunoreactive cones (Supp. Fig. D). The whole mount retinas with
TIMP-1 treatment showed homogenous pattern of Mü ller cell apical processes (Supp. Fig. E).
However, the array of the M-opsin-immunoreactive cone outer segments showed some
disruption from regularity (Figs. 2E, F). The outer segments were not always pointing towards
the same orientation. Also, the cell bodies of the M-opsin-immunoreactive cones seemed to have
moved; some were associated in loose groups (example circled; Supp. Fig. G). The cryostat
sections showed close association between the Mü ller cells and the M-opsin-immunoreactive
cones (Supp. Fig. H)
164
Supplementary Figure. Confocal micrographs taken from whole mounts and vertical
sections of saline or TIMP-1 injected normal retinas processed for GS and M-opsin
immunoreactivities. Double exposure (A) show homogenous pattern of GS (green) and M-opsin
(red) immunoreactivity in P59 normal retina 2 weeks after saline application. Both the outer
segments (B) and the cell body positions (B) of the M-opsin-immunoreactive cones are
homogenous. The cryostat section of the retina show close association between the processes of
the Mü ller cells (green) and the M-opsin-immunoreactive cones (red; D). The TIMP-1-injected
whole mount retinas (2 weeks) show homogenous pattern of GS immunoreactivity (E). The
orientation of the array of outer segments of M-opsin-immunoreactive cones show irregularity (E,
F). The cell bodies of these cones seem to have moved slightly (G). The cryostat section of the
retina show close association between the processes of the Mü ller cells and the M-opsin-
immunoreactive cones (H) that not always showed less regular spacing. ONL, outer nuclear
layer; OPL, outer plexiform layer. All scale bars = 100 µ m.
Abstract (if available)
Abstract
In this research project, there are four main parts. Scientists have for long tried to understand the fate of cones in RP retinas in determination to rescue them from degeneration at later stages of the disease following rod degeneration. However, no past studies have examined cone mosaic in depth in the whole mount of RP retinas. In the first part, the extensive remodeling of cone mosaic in the S334ter-line-3 rat retinas was examined in depth. We have found that cones in these RP retinas rearranged themselves in a regular array of rings and survived for a long time after rod deaths. The rings continued to remodel through later stage of life as their mean size and quantity increased with the progression of disease before the eventual cone deaths. Similar photoreceptor distribution pattern of rings are also observed in human patients with some eye diseases, and this makes the current study even more significant (Carroll et al., 2004
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Spatial anomalies of visual processing in retinitis pigmentosa and a potential treatment
PDF
Towards a high resolution retinal implant
PDF
Elements of photoreceptor homeostasis: investigating phenotypic manifestations and susceptibility to photoreceptor degeneration in genetic knockout models for retinal disease
PDF
Visual arrestin interactions with clathrin adaptor AP-2 regulate photoreceptor survival
PDF
Cone arrestin 4 contributes to vision, cone health, and desensitization of the dopamine receptor D4
PDF
Characterization of human embryonic stem cell derived retinal pigment epithelial cells for age-related macular degeneration
PDF
Understanding the degenerate retina's response to electrical stimulation: an in vitro approach
PDF
RPE secretome for the treatment of retinal degeneration in the RCS rat
PDF
Morphological and functional evaluation of hESC-RPE cell transplantation in RCS rats
PDF
Use of remotely sensed imagery to map sudden oak death (Phytophthora ramorum) in the Santa Cruz Mountains
PDF
The effect of tumor-mediated immune suppression on prostate cancer immunotherapy
PDF
The roles of endoplasmic reticulum chaperones in regulating liver homeostasis and tumorigenesis
PDF
Mediating the homeostatic response during light & cytotoxic induced retinal damage
PDF
Age related macular degeneration in Latinos: risk factors and impact on quality of life
Asset Metadata
Creator
Ji, Yerina
(author)
Core Title
Cellular mechanisms controlling the mosaic of surviving cones in retinitis pigmentosa retinas
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Neuroscience
Publication Date
05/06/2013
Defense Date
01/09/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cone mosaic,glia cells,mosaic regularity,OAI-PMH Harvest,reorganization,retina,retinitis pigmentosa,tissue inhibitor of metalloproteinase
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Grzywacz, Norberto M. (
committee chair
), Lee, Eun-Jin (
committee chair
), Craft, Cheryl M. (
committee member
), Hinton, David R. (
committee member
)
Creator Email
yerinaji85@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-251251
Unique identifier
UC11288430
Identifier
etd-JiYerina-1653.pdf (filename),usctheses-c3-251251 (legacy record id)
Legacy Identifier
etd-JiYerina-1653.pdf
Dmrecord
251251
Document Type
Dissertation
Rights
Ji, Yerina
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
cone mosaic
glia cells
mosaic regularity
reorganization
retina
retinitis pigmentosa
tissue inhibitor of metalloproteinase