|
|
||||||||
1 From the Department of Ophthalmology, University of Alabama at Birmingham, Birmingham, Alabama; and the 2 Retina Specialists of North Alabama, Huntsville, Alabama.
| Abstract |
|---|
|
|
|---|
smooth
muscle actin (
SMA), two cytoskeletal proteins associated with
phenotypic variation in culture.
METHODS. Six eyes with late ARMD and three age-matched control eyes were
preserved in buffered 4% paraformaldehyde and cryosectioned at 10
µm. Stages of RPE morphology and pigmentation were assessed by the
Alabama Age-Related Macular Degeneration Grading System. Vimentin,
SMA, and glial fibrillary acidic protein (GFAP) expression was
detected by indirect immunofluorescence. These results were compared
with regional variations in disease severity.
RESULTS. RPE changes in ARMD included acquired expression of vimentin, but
SMA-positive cells were rare. GFAP expression increased in
Müller cells in the neural retina in association with RPE changes
and photoreceptor degeneration.
CONCLUSIONS. The initial stages of RPE changes in eyes with ARMD mimic those
reported for cultured RPE cells. The absence of
SMA-positive cells
in regions of RPE atrophy suggests that RPE are lost rather than
persist in a dedifferentiated state.
| Introduction |
|---|
|
|
|---|
A pivotal role of the RPE in ARMD pathogenesis has been postulated for
almost three decades.6
Clinically visible hyper- and
hypopigmentation of macular RPE increase with age and are predictors of
late ARMD ranking in quantitative importance with
drusen.1
2
3
7
8
9
10
The functional status of RPE cells in
these states is incompletely understood. Pigmentary abnormalities are
transient with a time course of several years.11
12
13
Histopathologic studies reveal pigment clumping, loss of pigment,
heterogeneity of cell sizes and shapes, formation of multiple layers,
and presence of pigmented cells in the subretinal
space.5
14
Cells that are thought to be RPE in CNV
membranes surgically excised from patients with exudative
ARMD15
16
17
18
19
20
express the myofibroblast marker
smooth
muscle actin (
SMA), the intermediate filament protein cytokeratin,
and the angiogenic agent vascular endothelial growth factor.
Changes in RPE morphology associated with ARMD have been formalized by the Alabama Age-related Macular Degeneration Grading System (ALARMDGS21 ), summarized in Table 1 . RPE atrophy or absence of a pigmented layer (grade 4) is the logical end point of cellular degeneration, whether in large areas (geographic atrophy) or small (nongeographic atrophy). Heaping and sloughing of cells into the subretinal space (grade 2) and anterior migration of cells into the retina (grade 3) are logical intermediate stages for classifying disease-related RPE changes on a per-eye basis. However, it is not yet clear that these stages represent all the possible steps between stages 1 and 4 for individual RPE cells. Not only can RPE cells migrate anteriorly, but they may also die in situ or dedifferentiate into another phenotype unrecognizable as RPE.
|
SMA.24
25
Coincident with these phenotype
changes are the increased ability of the cells to adhere to certain
extracellular matrix molecules, generate tractional forces on
three-dimensional collagen matrices, and secrete growth factors, such
as insulin-like growth factor I and platelet-derived growth factor.
Although the functional significance of these RPE changes in
fibrocontractive diseases such as proliferative vitreoretinopathy is
apparent, their function in ARMD is much less clear. It is possible
that the increased risk of late ARMD signaled by a hyperpigmented RPE
may be indirect, through its association with diffuse deposits in
Bruchs membrane,26
27
28
or direct, through secretion of
growth factors or cytokines from RPE cells in an activated state. We hypothesized that ARMD-associated aberrations in RPE morphology represent definable stages of phenotypic modulation. We tested this hypothesis by examining RPE in the eyes of donors with ARMD and describing changes in cell morphology, pigmentation, and expression of the same cytoskeletal proteins known to vary in culture.
| Methods |
|---|
|
|
|---|
Tissue Preparation
Globes used for immunohistochemistry were preserved by immersion
in 4% paraformaldehyde in 0.1 M phosphate buffer (PB) for 6 hours
after removal of the anterior segment. Samples approximately 8 x
5 mm from the macular retina-RPE-choroid, which contained the fovea and
temporal half of the optic nerve head, were cut from each eye with a
razor blade. All samples were cryoprotected with successive solutions
of 10%, 20%, and 30% sucrose in PB, 4:1 30% sucrose-mounting
compound solution (Histo-Prep; Fisher Scientific, Fairlawn NJ), and 2:1
30% sucrose-mounting compound solution for at least 30 minutes each
and then frozen in -70°C isopentane.29
Specimens were
sectioned at 10 µm on a cryostat (CM3000; Leica, Deerfield, IL),
collected on gelatinized slides, dried on a hot plate at 40°C to
60°C, and stored at -20°C. Before immunohistochemical experiments,
the slides were warmed to room temperature, postfixed with acetone for
3 minutes, and heated to 50°C for 60 minutes. Some sections were
stained with periodic acid-Schiff and hematoxylin for histopathologic
evaluation.
Indirect Immunofluorescence Microscopy
Slides with cryosections were blocked with 20% nonimmune serum
in PBS from the same species as the secondary antibody for 60 minutes
at room temperature. Sections were then probed with primary and
secondary antibodies in 2% nonimmune serum in PBS for 1 hour each with
three 5-minute washes between probes. Photomicrographs were taken
(Optiphot; Nikon, Inc., Melville, NY) equipped with epifluorescence
illumination and phase-contrast optics using a 35-mm camera (Ilford
Delta 100 and SFX 200 film; Ilford Photograph Corp., Paramus, NJ).
Images were scanned from negatives (SprintScan 4000; Polaroid Corp,
Cambridge, MA) and assembled into composite photomicrographs with image
management software (Adobe Photodeluxe; Adobe Systems, Inc., San Jose,
CA).
Reagents
Primary antibodies used in this study included mouse anti-smooth
muscle actin (clone 1A4; Sigma Chemical Co., St. Louis, MO), mouse
anti-vimentin (clone V9; Dako A/S, Glostrup, Denmark), mouse
anti-cytokeratin 18 (clone CY-90; Sigma), mouse anti-glial fibrillary
acidic protein (GFAP; clone GA5; Sigma) and rabbit anti-GFAP (Dako).
The specificity of these antibodies has been previously verified by
Western blot analysis.25
30
Secondary antibodies included
allophycocyanin-conjugated goat anti-mouse IgG (Molecular Probes,
Eugene, OR), rhodamine-conjugated goat anti-mouse IgG (Jackson
ImmunoResearch Laboratories, West Grove, PA),and fluorescein-conjugated
donkey anti-rabbit IgG (Jackson ImmunoResearch). Normal goat and donkey
sera were obtained from The Binding Site, Ltd. (Birmingham, UK).
Analysis of Immunoreactivity in Cryosections
For each eye, we analyzed one slide containing two cryosections
for each primary antibody and one control slide. Each slide set was
examined in a masked fashion by two independent graders. The degree of
RPE change was assessed by bright-field microscopy using the five
grades of RPE morphology designated by the ALARMDGS 21
(Table 1)
. The entire 8-mm length of each section was divided into
different zones of RPE morphology that ranged in length from less than
100 µm for clumps of cells in ARMD eyes to several millimeters for
normal RPE in non-ARMD eyes. The mean number of graded zones in the two
sections was 15.6 in each ARMD eye. All RPE grades (04) were found in
all ARMD eyes. Sections were then viewed with epifluorescence
illumination to make binary judgments about immunoreactivity
("labeled" or "unlabeled") as a function of RPE grade. Data
from each grade were pooled across eyes. Choroidal fibroblasts and
vascular smooth muscle cells served as an internal positive control for
vimentin and
SMA labeling, respectively.
The relationship of GFAP immunoreactivity to RPE and photoreceptor changes was assessed using monoclonal anti-GFAP. Areas of positive and negative immunoreactivity were identified, and a grade for RPE and photoreceptor morphology assigned. For this analysis, the RPE grades provided by the ALARMDGS were collapsed into three categories: normal (0, 1), reactive (2, 3) and atrophic (4). Photoreceptors were assessed as normal, degenerating (shortened outer segments, broadened inner segments, and/or thinned outer nuclear layer), or absent.
| Results |
|---|
|
|
|---|
|
SMA Expression
SMA (Table 3)
. Each section from ARMD eyes analyzed for immunoreactivity typically
had two to three areas of each RPE grade, with the less-involved
tissues flanking more severe lesions. Staining results between the two
sections from each eye agreed closely, and the results across eyes were
consistent. Grade 0 RPE cells were uniformly negative for vimentin
content (Fig. 1A
, compared with 1B
for autofluorescence and 1C
for pigmentation), but a
variable percentage of cells at grades 1 and higher contained vimentin
immunoreactivity (Table 3)
. Cells at grade 1 contained a rim of
positive immunoreactivity at the basal margin (Figs. 1D
1E
1F
). At grade
2, the proportion of cells with vimentin expression increased, and the
extent of labeling within individual cells increased to include the
entire cell margin with faint fibrillar elements in the cytoplasm (Fig. 1G , compared with 1H
and 1I
). Of note, in regions of RPE atrophy (grade
4), the vimentin content of the layer immediately above Bruchs
membrane appeared to increase markedly (Fig. 1J , compared with 1K and
1L). However, because this labeling coincided with complete loss of
pigment and autofluorescent granules, it was not clear whether it
corresponded to the labeling of RPE cells undergoing phenotypic
modulation or to the encroachment of vimentin-positive Müller
cells from the overlying retina.
|
|
SMA.
When present,
SMA-positive cells were typically fibroblast-like in
morphology and had no autofluorescence and pigment granules (Figs. 2A
2B
2C
2G
2H
2I
, arrows). To distinguish between phenotypically modulated RPE
cells and Müller cells, we examined
SMA expression in
cryosections that were dual-labeled with a mouse monoclonal antibody
against
SMA and a rabbit polyclonal antibody against GFAP. This
analysis indicated that the
SMA-positive cells (Figs. 2A
2G)
also
had no GFAP immunoreactivity (Figs. 2B
2H)
and therefore were of
nonglial origin. The paucity of
SMA-positive cells (Figs. 2D 2E
2F)
persisted throughout the range of RPE changes through grade 4. From
this result we concluded that the
SMA-positive RPE cells of the type
observed in culture are rare in eyes with late ARMD (Table 3)
.
|
|
|
| Discussion |
|---|
|
|
|---|
Other investigators using the same antibody have reported that cells overlying choroidal melanomas also upregulate vimentin expression.22 34 In that study, vimentin-positive cells were considered hyperplastic RPE cells by morphologic criteria and the fact that other sections showed abundant cytokeratin 18positive cells of the same morphology.22 The appearance of RPE vimentin immunoreactivity also occurs in retinoblastoma34 and experimental retinal detachment in monkey.35 In all these conditions, including ARMD, the basal and basolateral aspects of RPE are labeled.
In contrast to vimentin,
SMA immunoreactivity was detectable in only
a few cells, even at the late, atrophic stage of ARMD-associated RPE
change. Others have found that
SMA expression is undetectable in
morphologically normal or atrophic RPE overlying choroidal
tumors.22
Further,
SMA-positive cells that are presumed
to be RPE by virtue of being labeled with a pancytokeratin antibody are
present in surgically excised CNV membranes.20
CNV is
hypothesized to be a process with dynamic initiation,
maintenance, and involutional stages.36
Because of
the long survival intervals between last clinical examination and donor
death (Table 2)
, CNV membranes in our donor eyes with late ARMD were at
postinvolutional stages of disease, considerably later than the
surgically excised CNV membranes examined by others.20
Thus, it is likely that the expression of
SMA exhibits several
different phases, including transient expression during early stage of
CNV maturation. This idea is strengthened by studies showing that
SMA-positive cells undergo apoptosis in CNV
membranes.37
Furthermore, these changes are not specific
to ARMD, because
SMA-positive, morphologically hyperplastic cells
also occur in subretinal membranes overlying choroidal
tumors.22
The functional consequences of increased vimentin immunoreactivity for RPE are currently unclear. Vimentin expression has long been associated with the onset of mitosis38 and may signal increased proliferative potential. However, recent studies indicate that vimentin expression and architecture are dynamic during other cellular activities. For example, during active cytoskeletal remodeling associated with cell spreading in culture, vimentin is initially found in nonfilamentous forms that are replaced first by short fibrous structures and then by extensive filamentous networks connected to other cytoskeletal systems.39 Further, a role for vimentin in cell motility associated with wound healing is implicated by studies in mammary epithelial cells and vimentin-deficient fibroblasts.40 41 Thus, it is possible that the reaction to local perturbation reflected by increased vimentin immunoreactivity in stage 2 RPE in ARMD includes enhanced potential for cell migration. This notion is consistent with the position of stage 3 RPE cells within the neurosensory retina.
Although involvement of Müller cells in ARMD-associated fibrovascular lesions has been shown,20 42 our observation of increased GFAP expression by Müller cells in eyes with late ARMD is novel and also warrants discussion. Dramatic changes in Müller cell GFAP expression occur in response to numerous retinal insults, including laser- or light-induced damage, diabetic retinopathy, retinal detachment, and inherited retinal degeneration.31 32 43 44 45 In the case of light or laser damage and diabetic retinopathy, increased GFAP expression may reflect a response to generally perturbed retinal physiology.31 46 However, in the case of retinal detachment, inherited degeneration, and ARMD, increased GFAP expression may reflect a secondary response to a primary insult to another cell type,43 44 although our results do not allow exclusion of the possibility of a primary role for Müller cells in this process. It is interesting that ARMD shares with retinal detachment not only upregulation of vimentin in RPE, but also the presence of Müller cell processes separating photoreceptors from RPE.42 47 In ARMD the most likely inductive cells are photoreceptors and/or RPE. These effects may be mediated by direct contacts between Müller cells and photoreceptors,44 Müller cell responses to humoral factors secreted by the RPE,25 or both. Understanding the relative contribution of these mechanisms to retinal degeneration requires a more detailed analysis of photoreceptor, RPE, and Müller cell changes at earlier stages of ARMD than was covered in this study.
| Footnotes |
|---|
Submitted for publication January 26, 2001; revised September 4, 2001; accepted September 14, 2001.
Commercial relationships policy: N.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be marked
"advertisement" in accordance with 18 U.S.C.
1734
solely to indicate this fact.
Corresponding author: Christine A. Curcio, Department of Ophthalmology, University of Alabama at Birmingham, 700 South 18th Street, Rm H020, Birmingham AL 35294-0009; curcio{at}uab.edu.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Ho, S. Kumar, X. S. Min, Y. L. Kng, M. Y. Loh, S. Gao, and L. Zhuo Molecular Imaging of Retinal Gliosis in Transgenic Mice Induced by Kainic Acid Neurotoxicity Invest. Ophthalmol. Vis. Sci., May 1, 2009; 50(5): 2459 - 2464. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Schmidt-Erfurth, M. Rudolf, M. Funk, C. Hofmann-Rummelt, N.-S. Franz-Haas, Z. Aherrahrou, and U. Schlotzer-Schrehardt Ultrastructural Changes in a Murine Model of Graded Bruch Membrane Lipoidal Degeneration and Corresponding VEGF164 Detection Invest. Ophthalmol. Vis. Sci., January 1, 2008; 49(1): 390 - 398. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Sarks, S. Cherepanoff, M. Killingsworth, and J. Sarks Relationship of Basal Laminar Deposit and Membranous Debris to the Clinical Presentation of Early Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., March 1, 2007; 48(3): 968 - 977. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Chowers, R. Wong, T. Dentchev, R. H. Farkas, J. Iacovelli, T. L. Gunatilaka, N. E. Medeiros, J. B. Presley, P. A. Campochiaro, C. A. Curcio, et al. The Iron Carrier Transferrin Is Upregulated in Retinas from Patients with Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., May 1, 2006; 47(5): 2135 - 2140. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. T. Johnson, M. N. Brown, B. C. Pulliam, D. H. Anderson, and L. V. Johnson Synaptic Pathology, Altered Gene Expression, and Degeneration in Photoreceptors Impacted by Drusen Invest. Ophthalmol. Vis. Sci., December 1, 2005; 46(12): 4788 - 4795. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. G. Marneros and B. R. Olsen Physiological role of collagen XVIII and endostatin FASEB J, May 1, 2005; 19(7): 716 - 728. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Howes, Y. Liu, J. L. Dunaief, A. Milam, J. M. Frederick, A. Marks, and W. Baehr Receptor for Advanced Glycation End Products and Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., October 1, 2004; 45(10): 3713 - 3720. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. T. Johnson, G. P. Lewis, K. C. Talaga, M. N. Brown, P. J. Kappel, S. K. Fisher, D. H. Anderson, and L. V. Johnson Drusen-Associated Degeneration in the Retina Invest. Ophthalmol. Vis. Sci., October 1, 2003; 44(10): 4481 - 4488. [Abstract] [Full Text] [PDF] |
||||
![]() |
K H C Wu, M C Madigan, F A Billson, and P L Penfold Differential expression of GFAP in early v late AMD: a quantitative analysis Br. J. Ophthalmol., September 1, 2003; 87(9): 1159 - 1166. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Guidry, K. M. Bradley, and J. L. King Tractional Force Generation by Human Muller Cells: Growth Factor Responsiveness and Integrin Receptor Involvement Invest. Ophthalmol. Vis. Sci., March 1, 2003; 44(3): 1355 - 1363. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sullivan, P. Penfold, and D. V. Pow Neuronal Migration and Glial Remodeling in Degenerating Retinas of Aged Rats and in Nonneovascular AMD Invest. Ophthalmol. Vis. Sci., February 1, 2003; 44(2): 856 - 865. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |