A Method for the Isolation of Retinal Pigment
Epithelial Cells From Adult Rats
Nan Wang,* Cynthia A. Koutz,-\ and Robert E. Anderson*];
Purpose. Isolating retinal pigment epithelium (RPE) cells from adult rat was the goal of
this study.
Method. A modification of the procedure of Mayerson et al was the method employed. After
the cornea and lens were removed, the eyecup was treated with hyaluronidase plus collagenase.
The neural retina was carefully removed with little RPE attached. The eyecup, free of retina,
was treated with the same enzyme cocktail that loosens any attached rod outer segments as well
as the attachment of the RPE to Bruch's membrane. RPE cells were isolated in calcium-free
medium as sheets of cells.
Results. Light and electron microscopy revealed good cell morphology. Purity of the RPE
preparation was established by microscopy, polyacrylamide gel electrophoresis, and lipid analysis. The viability of the isolated RPE was about 74% by trypan blue exclusion test.
Conclusions. This technique enables RPE to be isolated from adult rat in quantities sufficient
for biochemical analysis. Invest Ophthalmol Vis Sci. 1993;34:101-107.
1 he retinal pigment epithelium (RPE) is a single layer
of cuboidal epithelial cells located between the choriocapillaris and photoreceptor cells. Rod outer segments (ROS) of photoreceptor cells and apical processes of RPE interdigitate with each other. This close
contact appears to be essential for photoreceptor cell
development and function.1 Tight junctions between
RPE cells form part of the blood-retinal barrier and
block direct diffusion of blood constituents from the
choriocapillaris to photoreceptor cells. Therefore, the
RPE not only provides the retina with nutrients from
the circulation but also protects the retina from other
From the * Department of Biochemistry, and fCullen Eye Institute, Baylor
College of Medicine, Houston, Texas.
This research was supported by grants from NIH/NEI (EY04149,
EY0087I, and EY02520), the RP Foundation Fighting Blindness,
Research to Prevent Blindness, Inc., New York, New York, and the Retina
Research Foundation, Houston, Texas.
Robert E. Anderson is the recipient of a Senior Investigator Award from
Research to Prevent Blindness, Inc.
Submitted for publication: December 17, 1991; accepted June 13, 1992.
Proprietary interest category: N.
Reprint requests: Nan Wang, Department of Ophthalmology, Baylor
College of Medicine, One Baylor Plaza, Houston, TX 77030.
Investigative Ophthalmology & Visual Science, January 1993, Vol. 34, No. 1
Copyright © Association for Research in Vision and Ophthalmology
chemicals in the blood. RPE cells phagocytize the tips
of photoreceptor cells daily. In rat, it is estimated that
each RPE cell must phagocytize 25,000-30,000 ROS
disks (250-300 ROS tips) each day.2 The RPE cell
must rapidly digest these membranes and clear or recycle the products. Failure of the RPE to phagocytize
ROS tips leads to degeneration of photoreceptor cells,
as demonstrated in the RCS rat.3"5.
To investigate the role of the RPE in photoreceptor cell metabolism, it is essential to isolate RPE intact
and free from contamination by other cells. Although
RPE cells have been isolated from human,6"8 bovine,910 frog,11 etc., success in rat and mouse is limited
to neonatal animals.1213 The strong attachment of rat
RPE to Bruch's membrane makes it impossible to
brush off the RPE (as is done in bovine and frog) without damaging the cells. Also, the strong interdigitation
of RPE apical microvilli with the ROS makes the separation of these two structures difficult in adult rodents. By using young animals (less than 15 days old),
when ROS are not fully developed, RPE cells can
be isolated without tearing away their apical microvilli.12"14 However, no satisfactory procedure exists to
101
102
Investigative Ophthalmology & Visual Science, January 1993, Vol. 34, No. 1
isolate RPE from adult rats. We present here a modification of the method of Mayerson et al12 for isolating
RPE from adult rats with good preservation of morphology and viability.
MATERIALS AND METHODS
Isolation of Adult Rat RPE
Long Evans rats (Charles River, Wilmington, MA)
were maintained under cyclic fluorescent light (lights
on from 7:00 AM to 7:00 PM) and killed between 1000
and 1400 hr. All procedures were conducted in accordance with the NIH Guide for the Care and Use of
Laboratory Animals and the ARVO Statement for the
Use of Animals in Ophthalmic and Vision Research.
Rats were killed between 4 and 14 weeks of age.
Under deep anesthesia, rats were perfused from left
ventricle to right auricle with about 4 blood volumes
of buffer to remove red blood cells and plasma from
the choriocapillaris. Eyes were enucleated and submerged in ice cold calcium-free Hank's EDTA
(CFHE), which contained (in mmol/1): 5.4 KC1, 0.4
KH2PO4, 0.8 MgSO4, 137 NaCl, 0.34 Na2HPO4, 2.0
ethylenediamine tetracetate, 5.5 D-glucose, 10 N-2hydroxyethylpiperazine-N'-2-ethanesulfonic acid, and
0.03 phenol red at pH 7.4.10 The connective tissues
were trimmed and the globe was rinsed twice in CFHE.
Cornea, lens, and vitreous humor were removed by a
circumferential cut just above the ora serrata, leaving
the neural retina still attached to the eyecup. Eyecups
were rinsed three times in ice cold CFHE and incubated at 37°C for 12 min in CFHE containing 220
U/ml hyaluronidase type IV (Sigma, St. Louis, MO)
and 65 U/ml collagenase (CLS 1; Worthington, Freehold, NJ). Eyecups then were transferred to ice cold
CFHE and, under the dissection microscope, were cut
in half through the optic disk. The neural retina was
carefully peeled away from the eyecup. Very little RPE
was attached to the retina. The rim of the eyecup
halves was trimmed to just below the ora serrata; the
optic disk area, where some retina still was attached,
was removed with a pair of iris scissors. The trimmed
eyecup halves were incubated at 37°C for 8 min in the
same enzyme solution, washed two times in ice cold
CFHE, and kept in CFHE at room temperature for 30
min to facilitate the dissociation of RPE from Bruch's
membrane. Under the dissection microscope, the RPE
sheets were carefully lifted from the remaining eyecup
using jeweler's forceps and were collected using a 200
ix\ pipette. Care was taken to leave the underlying choroid undisturbed. No effort was made to maximize the
yield of RPE cells.
heparinized tubes. Blood cells and plasma were separated by centrifugation. Approximately 0.5 ml of
blood cells were washed three times with 10 ml each of
Tris-buffered saline (TBS; 20 mmol/1 Tris HC1, 137
mmol/1 NaCl, 2 mmol/1 MgCl2, pH 7.4). After RPE
sheets were lifted off the eyecups, choroid was separated from the sclera and collected using jeweler's forceps. All tissues were kept at —20°C until they were
used. ROS were prepared from retina by discontinuous sucrose gradient centrifugation, as described
previously.15
Sodium Dodecyl Sulfate-Polyacrylamide Gel
Electrophoresis (SDS-PAGE)
SDS-PAGE of tissue total proteins was performed according to the method of Laemmli.16 Molecular
weight standards and SDS-PAGE reagents were from
Bio-Rad (Richmond, CA). ROS, RPE, choroid, blood
cells, and plasma were resuspended and diluted in
TBS, homogenized, and mixed well with half their volume of 3X loading buffer before being loaded on a
12% gel. The gel was stained with coomassie blue.
Lipid Analysis
Lipids were extracted according to the procedure of
Bligh and Dyer.17 Phospholipids were separated from
other lipid classes by thin layer chromatography on
silica gel 60 (EM Science, Gibbstown, NJ) plates using
a solvent system of hexane/ether/acetic acid in a volumetric ratio of 75/25/1. Diacylglycerol benzoate derivatives of phospholipids were prepared and analyzed
according to Louie et al.18
Morphologic Analysis
Isolated retinal pigment epithelial cells were fixed
overnight in 2.5% glutaraldehyde and 1% formaldehyde in 0.125 mol/1 sodium cacodylate buffer (pH 7.4)
at 4°C. The cells were washed in cacodylate buffer
with sucrose, post-fixed in 1% osmium tetroxide, dehydrated through a graded ethanol series (one time
50% ethanol, one time 70% ethanol, two times 95%
ethanol, three times 100% ethanol), and embedded in
an Epon-Araldite mixture (Electron Microscopy
Sciences, Fort Washington, PA). Low speed centrifugation (100 X g, 5 min) was required at each step of
processing. Sections approximately 0.5 fim thick were
cut using a Dupont Sorvall Porter-Blum (Wilmington,
DE) MT2-B Ultramicrotome. Thick sections were
stained with 0.5% toluidine blue and viewed on a Zeiss
(Thornwood, NY) Photomicroscope III. Ultrathin sections of approximately 600 A also were taken, stained
with uranyl acetate, and lead citrate, and viewed on a
JEOL (Peabody, MA) 100CX Electron Microscope.
Collection and Preparation of Other Tissues
In addition to RPE cells, blood, choroid, and ROS also
were collected from the same animal for study. Blood
was drawn before perfusion via cardiac puncture into
Cell Viability Assay
A stock solution of trypan blue (4 mg/ml; Gibco,
Grand Island, NY) was diluted 20 times with TBS and
Isolation of Rat Retinal Pigment Epithelial Cells
mixed with an equal volume of an aliquot of RPE cell
suspension. After 5 min at room temperature, the
number of total cells and the number of cells that took
up dye were counted. Cell viability was expressed as
the percentage of RPE cells that excluded dye. Three
preparations were obtained from three different rats.
In each preparation, two to four aliquots of cells were
counted. Between 150 and 650 cells were counted in
each preparation.
RESULTS
RPE Morphology
The morphology of the isolated RPE was examined by
light microscopy (Fig. 1) and transmission electron microscopy (Fig. 2). The RPE cells were pigmented, and
the majority of them also were binucleated. Their hexagonal shape was maintained and the cells still were
attached to each other. Both the apical microvilli (Fig.
2, double arrows) and the basal infoldings (Fig. 2, single arrows) were observed. ROS or large phagosomes
rarely were seen (Fig. 2). Tight junctions between adjacent RPE cells are indicated by open arrows (Fig. 2).
Occasionally, some choroidal membrane and red
blood cells could be seen (data not shown).
RPE Purity
SDS-PAGE analysis of total protein from ROS, RPE,
choroid, blood cells, and plasma is shown in Figure B.
Contamination from plasma components (lane 2) was
minimal, as judged by the lack of the two higher molecular weight bands (between 45 and 92 kD) that were
enriched in plasma. Most blood cells were red blood
cells that gave a heavy band of subunits of hemoglobin,
migrating around 14 kD on the gel (lane 3). In the RPE
preparation (lane 5), this region was clear. Compared
to ROS (lanes 6 and 7), the 35 kD region corresponding to opsin was almost clear in RPE. No band was
found in RPE that corresponded to transducin a subunit (40 kD) in ROS. There were some similarities between RPE and choroid (lane 4), especially between
the 40-66 kD region. However, this may not indicate
significant contamination of choroid in the RPE preparation, because other proteins prominent in the choroid are not present in the RPE preparations (eg, approximately 16 kD).
Molecular species analysis of phospholipids from
ROS and RPE showed that they were quite different
(Fig. 4). 22:6w3-containing molecular species (22:6o>322:6w3 and 18:0-22:6w3) were highly enriched in ROS
except for 16:0-22:6co3, which is the primary form of
22:6co3 found in RPE. In contrast, 20:4co6-containing
molecular species (16:0-20:4^6, 18:0-20:4co6) were
the major form of polyunsaturaled fatty acids in RPE.
The 22:6w3-22:6co3 was enriched in ROS (11.9%)
compared to RPE (0.8%). Assuming that all of the
22:6co3-22:6u3 in RPE came from ROS contamination, the maximum contamination would be 6.8%
(0.8% H- 11.9% X 100).
l. Light microscopy of isolated RPE. (Original magnification X1040.) (A) Longitudinal section. (B) Cross section.
FIGURE
Investigative Ophthalmology & Visual Science, January 1993, Vol. 34, No. 1
2. Electron microscopy
of isolated RPE. Single arrows
point to the basal infoldings
and double arrows point to the
apical microvilli. Open arrows
indicate junctional complexes
between adjacent cells. Panel
B is the enlargement of panel
A area enclosed by black
frame. (A, original magnification X4140; B, XI6000.)
FIGURE
RPE Viability
Trypan blue exclusion test revealed that 74% ± 2% of
isolated RPE cells excluded dye.
DISCUSSION
A number of investigators have isolated RPE from rodents. Edwards13 first isolated and cultured RPE from
neonatal rats by soaking the globes overnight (6-24
hr) and treating them with trypsin and collagenase.
Later, Mayerson et al12 used a procedure of hyaluronidase plus collagenase followed by trypsin. The latter
procedure proved successful for young rats up to 15
days old and has been widely used for culturing RPE
from neonatal rat, as well as for RPE transplantation.
However, it is not effective for older rats. Mayerson et
al12 reported that the time required for trypsin treatment (72 min) increased significantly at 14 days of age
and, after post-natal day 15, retinal contamination was
difficult to remove. This probably results from the for-
Isolation of Rat Retinal Pigment Epithelial Cells
66
—
45
••
3 1
—
2 1 W
14
**
FIGURE 3. Coomassie blue-stained 12% SDS-PAGE of proleins from plasma (lane 2), blood cells (lane 3), choroid (lane
4), RPE (lane 5), and ROS (lanes 6 and 7). Molecular weight
standards (lanes 1 and 8) from top to bottom are rabbit muscle phosphorylase B (97 kD), bovine serum albumin (66 kD),
hen egg while ovalbumin (45 kD), bovine carbonic anhydrase (31 kD), soybean trypsin inhibitor (21 kD), and hen
egg white lysozyme (14 kD).
mation of photoreceptor outer segments at this time
that begin to interdigitate with RPE apical microvilli.19
Based on these previous studies, our strategy to
isolate RPE from adult rats was to loosen the RPEROS interdigitation first, so the retina could be lifted
from the RPE without tearing off their apical microvilli. Then, the attachment of RPE to Bruch's membrane was loosened by the enzyme treatment, but
RPE-RPE junctions were maintained so the RPE could
be peeled off the choroid as sheets. To do this, we used
an open globe instead of whole globe for the enzyme
treatment. This procedure allowed the enzymes to diffuse to the RPE from retina side and attack the interphotoreceptor matrix first. Therefore, during the first
treatment, the RPE-ROS adherence was disrupted
and allowed the separation of RPE from ROS with
some of their apical microvilli still preserved. The enzymatic digestion of the RPE-Bruch's membrane occurred mainly during the second enzyme treatment
through some "holes" left in the RPE sheet as a result
PHOSPHOLIPID
MOLECULAR SPECIES
•
1-
30
ROS
D RPE
-
CC
LLJ
Q_
20
LU
t
,
d
0
-™
•
h
„c c
CO
CD CD to
CM CM CM CM
CM CM CNI CM
CM
CM
c
CO
R
CM
CD
1
1
•
I
1
TT
CO CD CC> o
o o o
co
CO CO
c CM
c ^S
c-: ^co 1to
c CO
c COc CCc:1 c - dc- CD
— ^~ *^"
CO
CO
CO
f1' d
1 T1 1
CO
•
16:
97 ^ £ l
of RPE being pulled away with the retina. Because no
trypsin was used, the tight junctions between the RPE
were preserved (Fig. 2) and the RPE cells could be
removed from the choroid as sheets.
The first enzyme treatment is important for separating ROS from RPE. If this treatment was too long,
the RPE would come off with the retina as the retina
was pulled away, probably because of the destruction
of RPE-Bruch's membrane adherence to an extent
that it could no longer hold the RPE in place. On the
other hand, if the treatment was not enough, removal
of the retina resulted in some of the apical microvilli of
the RPE being torn away. In our hands, we tested enzyme treatment between 5 and 20 min with hyaluronidase concentration ranging from 50-300 U/ml in
combination with collagenase concentrations ranging
from 50-130 U/ml. We found that 220 U/ml hyaluronidase plus 65 U/ml collagenase for 12 min
worked best. Another mechanism that prevented loss
of RPE apical microvilli was the circumferential cut
just above the ora serrata. Had the cut been made
below the ora serrata, the retina would have immediately curled inward, tearing the RPE apical microvilli
away from the RPE. Although most (74%) of the isolated RPE cells excluded trypan blue, those cells that
20:^
1 2 3 4 5 6 7 8
105
»
r) (^
•*)•
>
CM
CM
CM
C\1 O
C\1 CM
O CM CO
CM
) -^- O
R ^
c *! c JE
CM
i-
C> ( T * C3
C\
i h- to
6• • t9>
?
o cc
CO ^_
•
40
c ^z
! m
CO
CO
00
a>
?M
—"
to"
c
jo
CO
FIGURE 4. Composition of phospholipid molecular species
from ROS (closed bar) and RPE (open bar).
106
Investigative Ophthalmology & Visual Science, January 1993, Vol. 34, No. 1
did take up the dye probably had lost some of their
apical microvilli during the isolation procedure. We
have tested trypsin alone, collagenase alone, combination of collagenase and trypsin, or combination of collagenase and hyaluronidase during the second enzyme
treatment to release RPE from Bruch's membrane.
We found that using the same cocktail used in the first
enzyme treatment for 8 min followed by a short period
of incubation in CFHE at room temperature gave the
most satisfactory results. Longer enzyme treatment
released large amounts of RPE into the enzyme solution.
contaminants. 2223 Our approach is gentler for handling the cells and therefore should maintain better
morphology and viability. Furthermore, because contamination is minimal during the isolation procedure,
no further purification of the cells is necessary. The
technique we have described here has been used successfully in our laboratory to investigate the role of
RPE cells in selective uptake of polyunsaturated fatty
acids from the circulation. It provides a simple and
efficient method for isolating RPE from adult rats for
biochemical studies.
This technique was used for Long Evans rats 4-14
weeks old. Variance in cell recovery existed to a significant extent between animals. The effective enzymatic
treatment depended on the storage condition of the
enzymes. Even under frozen desiccated storage, the
activity of the enzymes deteriorated over a 1- to 2month period. Although we harvested RPE by increasing concentration of enzyme and time of treatment
with some old enzymes, no effort was made to determine such increases in time and concentration in
correlation to enzyme storage time. Consistent results
were obtained by following the described condition
using enzyme within 1 mo of purchase.
Key Words
Purity of the RPE was established by microscopy,
SDS-PAGE, and phospholipid molecular species analysis. Although no ROS were observed sticking to RPE
apical microvilli, the maximum possible contamination of ROS in the RPE was calculated to be 6.8%
based on lipid molecular species analysis, which assumed that all the di-22:6a>3 in the RPE was the result
of ROS contamination. However, 22:6co3-22:6a>3
could be a component of RPE, because previous experiments20-21 have shown that 22:6w3 is present in
ROS phagocytized by the RPE. The absence of large
phagosomes in RPE cells (Fig. 2) indicated that by the
time RPE cells were isolated, the shed ROS tips already had undergone lysosomal digestion. Although
red blood cells occasionally could be seen, they did not
pose a serious contamination problem because of their
small number. The SDS-PAGE profile also indicated
there was little contamination of RPE preparation
from plasma or blood cells.
This technique enables isolation of RPE from
adult rats with good morphology, fairly good viability,
and little contamination. Whereas Mayerson's 12
method is for culturing RPE, this method is designed
for isolating RPE for biochemical analysis. The biggest
advantage of this method is that it allows the use of
adult rats. Furthermore, it uses only one mixture of
enzymes, which makes the procedure simpler and
faster. Other approaches are to first prepare a crude
RPE sample by peeling off the retina, followed by
brushing off the RPE with a camel's hair brush. Glass
beads then are used to purify the RPE from other
photoreceptor cells, rat retina, retinal pigment epithelium,
RPE isolation.
Acknowledgments
The authors thank Dr. Thomas W. Call for sharing his experience in RPE isolation and for assisting in our attempts to
culture isolated RPE cells.
References
1. HollyfieldJG, Witkovsky P. Pigmented retinal epithelium involvement in photoreceptor development and
function. J Exp Zool. 1974; 189:357-378.
2. Bok D, Young RW. Phagocytic properties of the retinal pigment epithelium. In: Zinn KM, Marmor MF,
eds. The Retinal Pigment Epithelium. Cambridge, MA:
Harvard University Press; 1979:148-174.
3. Herron WL, Riegel BW, Myers OE, Rubin ML. Retinal dystrophy in the rat: A pigment epithelial disease.
Invest Ophthalmol Vis Set. 1969;8:595-604.
4. Bok D, Hall MO. The role of the pigment epithelium
in the etiology of inherited retinal dystrophy in the
rat. J Cell Biol. 197l;49:664-682.
5. Mullen RJ, LaVail MM. Inherited retinal dystrophy:
Primary defect in pigment epithelium determined
with experimental rat chimeras. Science. 1976;
192:799-801.
6. Mannagh J, Ayra DV, Irvine AR. Tissue culture of human retinal pigment epithelium. Invest Ophthalmol Vis
Sci. 1973; 12:52-64.
7. Flood MT, Gouras P, Kjeldbye H. Growth characteristics and ultrastructure of human retinal pigment epithelium in vitro. Invest Ophthalmol Vis Sci. 1980;
19:1309-1320.
8. Edwards RB. Glycosaminoglycan synthesis by cultured
human retinal pigmented epithelium from normal
postmortem donors and a postmortem donor with retinitis pigmentosa. Invest Ophthalmol Vis Sci. 1982;
23:435-446.
9. Anderson RE, Oissandrello PM, Maude MB, Mathes
MT. Lipids of bovine retinal pigment epithelium. Exp
Eye Res. 1976;23:149-157.
10. Heller J, Jones P. Purification of bovine retinal pigment epithelial cells by dissociation in calcium free
buffers and centrifugation in Ficoll density gradients
followed by "recovery" in tissue culture. Exp Eye Res.
1980;30:481-487.
Isolation of Rat Retinal Pigment Epithelial Cells
11. Salceda R. Isolation and biochemical characterization
of frog retinal pigment epithelium cells. Invest Ophthalmol VisSci. 1986; 27:1172-1176.
12. Mayerson PL, Hall MO, Clark V, Abrams T. An improved method for isolation and culture of rat retinal
pigment epithelial cells. Invest Ophthalmol Vis Sci.
1985;26:1599-1609.
13. Edwards RB. The isolation and culturing of retinal
pigment epithelium of the rat. Vision Res. 1981;
21:147-150.
14. Bonting SL, Caravaggio LL, Gouras P. The rhodopsin
cycle in the developing vertebrate retina. I. Relation
of rhodopsin content, electroretinogram and rod
structure in the rat. Exp Eye Res. 1961; 1:14-24.
15. Stinson AM, Wiegand RD, Anderson RE. Fatty acid
and molecular species compositions of phospholipids
and diacylglycerols from rat retinal membranes. Exp
Eye Res. 1991;52:213-218.
16. Laemmli UK. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature.
1970;227:680-685.
17. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology. 1959; 37:911-917.
107
18. Louie K, Wiegand RD, Anderson RE. Docosahexaenoate-containing molecular species of glycerophospholipids from frog retinal rod outer segments show different rates of biosynthesis and turnover. Biochemistry.
1988;27:9014-9020.
19. Tamai M, Chader GJ. The early appearance of disc
shedding in the rat retina. .Invest Ophthalmol Vis Sci.
1979;18:913-917.
20. Gordon WC, Bazan NG. Docosahexaenoic acid utilization during rod photoreceptor cell renewal. J Neurochem. 1990; 10:2190-2202.
21. Chen H, Wiegand RD, Anderson RE. Docosahexaenoic acid increases in frog retinal pigment epithelium following rod photoreceptor shedding. Exp Eye
Res. 1992; 54:885-892.
22. Siakotos AN, Aguirre G, Aders L, Schnell RJ. Column
Method for the isolation of pigment epithelium. In:
Packer L, ed. Methods in Enzymology. vol 81. London:
Academic Press; 1982:90-94.
23. Braunagel SC, Organisciak DT, WangH. Characterization of pigment epithelial cell plasma membranes
from normal and dystrophic rats. Invest Ophthalmol Vis
Sci. 1988;29:1066-1075.