In vitro studies on shedding and
phagocytosis of rod outer segments in the
rat retina: effect of oxygen concentration
Makoto Tamai, Katsuyoshi Mizuno, and Gerald J. Chader*
The processes of outer segment shedding and phagocytosis were found to be significantly
enhanced in explants of rat retina and pigment epithelium (PE) newly placed in culture. The
number of phagosomes reached a maximum at 2 hr of incubation and then rapidly decreased,
as is observed in rats entrained in a normal light-dark schedule when the lights are turned on in
the morning. This pattern was not qualitatively affected by the time of day the tissues were
excised or by the presence of a high concentration of melatonin in the culture medium. The
shedding peak in vitro was observed only when the retina-PE unit was incubated in an atmosphere composed of 45% oxygen, 5% C02, and 50% air. When the atmosphere contained lower
(95% air and 5% CO2) or higher (95% 0 2 and 5% CO J oxygen, PE cells contained only a few
scattered phagosomes. These functional differences may be caused by different arrangement of
PE cell microvilli at a "moderate" (45%) O 2 concentration vs. "low" or "high" O 2 concentrations. A balanced 02 level thus appears to be critical for phagocytosis in explant culture.
(INVEST OPHTHALMOL VIS SCI 22:439-448, 1982.)
Key words: retina, pigment epithelium, photoreceptor disc shedding,
phagocytosis, oxygen concentration, organ culture
Shedding of retinal outer segments and
phagocytosis of the shed discs by the pigment
epithelium (PE) are important steps in the
process by which the retinal photoreceptor
cell undergoes constant renewal.x Although it
is known that outer segment membranes are
From the Department of Ophthalmology, Tohoku University School of Medicine, Sendai, Japan, and the
^Laboratory of Vision Research, National Eye Institute, National Institutes of Health, Bethesda, Md.
This experiment was begun during the stay of Dr. M.
Tamai in the National Eye Institute, Bethesda, Md.,
as a grantee of the U.S.-Japan Exchange Program in
Vision Research, and was completed after his return to
the Department of Ophthalmology, Tohoku University, School of Medicine, Sendai, Japan.
Submitted for publication Feb. 8, 1980.
Reprint requests: Gerald J. Chader, Laboratory of Vision
Research, National Eye Institute, National Institutes
of Health, Bethesda, Md.
shed and phagocytized in a circadian manner,2 little is known about the actual control
of these processes or in fact whether there
are separate signals for disc shedding and for
phagocytosis.
Shedding and phagocytosis have recently
been reported in vitro in tissue culture,3"5
eyecup preparations,6 and organ culture of
retinal and PE tissue.7* 8 Since the signals for
disc shedding and phagocytosis may be intrinsic to the eye and not directly controlled
from an extraocular source,9"11 it is reasonable to assume that significant information
about these processes can be gained in such
in vitro experiments. In the present study,
we use our previously described culture
technique to give evidence that microenvironmental conditions such as O2 concentration may play an important role in controlling
the functional expression of phagocytosis.
439
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
440
Invest. Ophthaltnol. Vis. Set.
April 1982
Tamai et al.
Materials and methods
Animals. Sprague-Dawley albino rats were obtained from the NIH animal colony. They were
maintained under a cyclic lighting condition (i.e.,
lights were turned on between 6:00 A.M. and 6:00
P.M.). The intensity of the fluorescent light was 40
to 50 foot candles at the top of the cages. Animals
were sacrificed at three different times during the
cyclic light-dark schedule: (1) 6:15 A.M., (2) 11:00
A.M., and (3) 9:00 P.M. In the experiments at 9:00
P.M., animals were killed under dim red light,
with all the subsequent procedures (e.g., dissection) performed under an infrared image converter
(Noctovision). Animals were sacrificed with sodium
pentobarbital (Diamond Laboratory, Des Moines,
Iowa) or with ether, and the eyes were rapidly
enucleated and rinsed several times in Hanks' Minimal Essential Medium (MEM; GIBCO, Grand
Island, N.Y.).
Tissue culture. The anterior segments of the
globes were removed at the equator. The posterior retina, along with adherent PE, choroid, and
sclera, was dissected into quarters in Hanks'
MEM. Segments of the neural retina-PE-choroid-sclera complex were put on pieces of paper
(choroidal side down) with great care not to separate the PE from the retina and were transferred
to stainless steel wire grids (Falcon, Oxnard,
Calif.). Tissues were kept in culture dishes (35 by
10 mm, Falcon), supplied with 2 ml of culture
medium, and rested on the grid at the liquidatmosphere interface. The medium consisted of a
3:3:1 mixture of Hanks' MEM, fetal calf serum
(GIBCO), and chick embryo extract (GIBCO).
Cultures were kept in an incubator (Heinicke Instruments Co., Hollywood, Fla.) at 37° C with a
continuous flow of atmosphere composed of 5%
CO 2 , 45% O2, and 50% air unless otherwise noted.
In experiments designed to study the effects of
differing concentrations of oxygen, animals were
sacrificed at 11:00 A.M., and tissues were dissected
and immediately incubated in one of three different atmospheres: "A", 5% CO 2 ) 45% O 2 ,and 50%
air; "B", 5% CO 2 and 95% air; "C", 5% CO 2 and
95% O 2 for periods of time given in the figure
legends. When appropriate, melatonin (Sigma
Chemical Co., St. Louis, Mo.) was dissolved in
ethanol and added to the medium at a final concentration of 0.2 mM. Control cultures in this
series of experiments received the same amount of
ethanol and were incubated in atmosphere "A."
Microscopy. Tissues were sampled just before
incubation was started ("0" hr control) or after 1 to
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
4 hr of incubation and were fixed with 1% glutaraldehyde and 1% paraformaldehyde in 0.075M
phosphate buffer, pH 7.2 overnight. They were
then postfixed with 1% osmium tetroxide in 0.15M
phosphate buffer for 1 hr and dehydrated in a
graded ethanol series, which was exchanged with
propylene oxide, and were finally embedded in
epoxy resin. The number of phagosomes was
counted in a 200 /am field of PE cells with thick (1
/am) sections stained with 1% toluidine blue.
Criteria for phagosome counting was as previously
defined by LaVail2; that is, all inclusion bodies in
the PE soma as well as intensely staining structures among PE processes that were greater in any
dimension than 0.75 /am were defined as phagosomes and were counted. For each experimental
condition, tissue specimens were taken for three
animals and 10 sections from each were examined
for numbers of phagosomes. Each experiment was
repeated two times and the mean values and standard deviations were calculated. For transmission
electron microscopy, thin sections were stained
with uranyl acetate and lead citrate and observed
with a JEOL 100C electron microscope.
Results
Tissues dissected and fixed before in vitro
incubation ("0" hour control) are shown in
Fig. 1, a. Only occasional phagosomes are
seen. Fig. 1, b to d, are tissues incubated for
1, 2, and 3 hr, respectively, in atmosphere
"A." The number of phagosomes increased
remarkably after transferring in vitro. The
peak value was obtained at 2 hr of incubation;
subsequently, the number of phagosomes
sharply decreased (Fig. 1, d).
In vivo, a burst of shedding occurs in the
rat soon after lights are turned on in the
morning. The in vitro shedding sequence
demonstrated in Fig. 1, b to d, however, was
the same as that in tissues excised at different
times during the day (Fig. 2, a). The peak
number of phagosomes (40 to 50 phagosomes
per 200 /am of PE) was observed at 2 hr of
incubation in tissues excised at 6:15 A.M. (15
min after lights on), 11:00 A.M., and 9:00 P.M.
(dark period). The burst of shedding in vitro
was also not affected by the presence of 0.2
mM melatonin in the culture medium (Fig.
2, b). Shedding was neither enhanced nor
suppressed at this high concentration of the
Volume 22
Number 4
Shedding, phagocytosis of rod outer segments 441
Fig. 1. Light micrographs of retina-PE explants dissected approximately 5 hr after lights were
turned on. a, Rat tissue dissected andfixedjust before starting incubation ("0" hr control), b, c,
and d, Retina-PE explant incubated for 1, 2, and 3 hr, respectively. Incubations were in
atmosphere "A." (Bar = 100 ^,m.)
neurohormone. Melatonin also had no noticeable effect on retinal structure in the cultured explants (not shown).
Fig. 3 shows the effect of different compositions of atmosphere on disc shedding in culture. Control atmosphere "A" shows the typ-
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
ical pattern of shedding at 2 hr of incubation
(Fig. 3, a). When the explant was cultured in
atmosphere "B," large numbers of inclusion
bodies were present in the PE that stained
greenish yellow with toluidine blue (Fig. 3,
b). Photoreceptor nuclei showed some de-
442
50
•
- a
PIGIVIENT EP
40 --
20
10 -
o
•
30 -
LL
O
I
Invest. Ophthalmol. Vis. Set.
April 1982
Tamai et al.
o
1
-
o
•
*
•
o
•
•
o
-
LENC
m
0 -
•
8
1
1
I
|
1
50 - b
o
<N
CC
LU
^RGE PHA GOSOMEJ
a.
40
o
tr\
30
•
o
20
•
10 -
o
o
1
0
1
2
3
4
INCUBATION TIME (Hours)
Fig. 2. Changes in phagosome number in culture.
a, Number of phagosomes over a 4 hr culture period, o, Tissue excised at 6:15 A.M.; A, tissue excised at 11:00 A.M.; •, tissue excised at 9:00 P.M.
b, Number of phagosomes in the presence (o) or
absence (•) of 0.2 mM melatonin. Incubations
were in atmosphere "A." Data points are mean
values from three rats in each of two separate experiments.
generative changes. In atmosphere "C" the
neural retina and PE cells were well preserved but very few phagosomes were evident (Fig. 3, c). The phagosome number
under the three atmospheric conditions over
a 4 hr period in culture is shown in Fig. 4.
Only in atmosphere "A" was there a burst of
shedding comparable to the situation observed in vivo.
Ultrastructurally, retinas incubated in atmosphere "A" were found to be well preserved
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
with numerous phagosomes (2 hr of incubation) in the apical portion of PE cell soma (Fig.
5, a). Phagosomes were usually surrounded by
optically empty space. Phagocytized disc
membranes were in various degrees of degradation. Some phagosomes still contained several dozen well-defined discs; other phagosomes contained clumped, electron-dense
masses. PE cell microvilli were generally
found grouped together and elongated perpendicularly to cell soma, surrounding adjacent outer segment fragments. Microvilli expanded only around rod outer segments
(ROS) located near the PE cell. Primary and
secondary lysosomes were abundant in the
apical half of the PE cell and in the microvilli;
most mitochondria were located just above
the basal infoldings. The basal infoldings
were tall and were often surrounded by large
optically empty spaces. Outer segments were
well preserved (Fig. 5, h) but exhibited a
segmented appearance every few decades of
discs (see also Fig. 1, h to d), although this
could very well have been caused by the natural cleavage characteristics of fixed rat ROS.
Several ROS appeared to have recently shed
their terminal discs; shed packets of discs
were not of uniform size.
Retina and PE cultured in atmosphere "B"
showed a generally similar ultrastructure to
that discussed above, e.g., well-preserved
disc membranes and tall basal infoldings (Fig.
6). In general, preservation of organelle
structure actually seems to be somewhat
better than that seen in atmosphere "A." At
least three characteristic differences were observed, however. First, there was an obvious
paucity of phagocytic activity. Second, numerous sheets of microvilli elongated parallel
to PE soma and had somewhat less tendency
to envelop adjacent ROS fragments. Third,
large numbers of homogeneous inclusion
bodies were contained in the PE. They had
no limiting membrane and relatively high
electron density.
The tissues incubated in atmosphere "C"
showed distinctly different morphologic characteristics (Fig. 7, a). Microvilli were more
plentiful than in atmospheres "A" or "B";
Volume 22
Number 4
Shedding, phagocytosis of rod outer segments 443
Fig. 3. Effect of different compositions of atmosphere on disc shedding in culture, a, Tissue
incubated in atmosphere "A"; b, tissue incubated in atmosphere "B"; c, tissue incubated in
atmosphere "C." Tissues were excised at approximately 11:00 A.M.; incubations were for 2 hr.
(Bar = 10 /urn.)
they were elongated, multilayered, and
parallel to the apical surface of the PE soma.
Few phagosomes were found. Multilayered
cell processes containing cytoplasm and cell
organelles were observed; these often "sandwiched" microvilli between them and the PE
cell soma. Basal infoldings were elongated
but only small optically empty spaces existed
between them. Some microvilli were elongated and surrounded ROS fragments. These
fragments were usually very large and had
irregularly arranged, vacuolized discs.
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
Discussion
The retina-PE unit shows an enhanced sequence of disc shedding and phagocytosis
when placed in organ culture under proper
conditions. These phenomena occur independent of lighting conditions and of the time
of day the samples are taken, since the numbers of phagosomes in the control (0 hr) retinas are quite different from each other but
uniformly increase several times within 2 hr
in culture. It thus appears that the precipitation of the shedding sequence is not acutely
Invest. Ophthalmol. Vis. Sri.
April 1982
444 Tamai et al.
1
2
3
INCUBATION TIME (Hours)
Fig. 4. Number of phagosomes in tissues incubated under different atmospheric conditions, o,
Tissue incubated in atmosphere "A"; A, tissue incubated in atmosphere "B"; •, tissue incubated in atmosphere "C." Tissues were excised at approximately 11:00 A.M. Data points are
mean values ± S.D. from two experiments.
dependent on lighting conditions or on circadian rhythm.
The present results are in accord with data
reported by Goldman and O'Brien6 in an
eyecup preparation, although they used a different strain of rats, different dissection
technique, and different medium as well as a
different gassing atmosphere. Their incubation atmosphere was 5% CO2 and 95% O2, an
atmosphere similar to our atmosphere "C,"
where we observed no shedding. We assume, however, that the O2 tension at the
retinal surface within the eyecup would be
somewhat more comparable to our atmosphere "A" (i.e., 45% O 2 where shedding is
observed), since the entire eyecup was immersed in buffer under the Goldman-O'Brien
conditions, whereas the explant sits at the
atmosphere/liquid interface in the present
experiments. No attempt was made to actually measure O2 tension in the buffer under
our conditions, since only a thin film of liquid
separates the tissue explant from the gas
phase. Flannery and Fisher12 have also reported that disc shedding occurs in cultured
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
whole-eye explants of Xenopus laevis within
1 hr of light stimulus.
The ultrastructural changes and lack of
phagocytosis observed with atmosphere "B"
could be ascribed to pH differences or to
simple anoxia. These changes include the arrangement of PE microvilli parallel to the cell
body instead of perpendicular and their inability to fully engulf ROS fragments. These
possible explanations are simplistic, however, since the buffer capacity of the enriched medium used is high and the effects
are observed within 2 hr of incubation. Similarly, both low (atmosphere "B") and high
(atmosphere "C") oxygen concentrations result in changes in microvillus orientation and
inhibition of phagocytosis. It thus appears
that a relatively delicately balanced O2 concentration is necessary for ROS shedding and
phagocytosis at least in vitro.
A somewhat parallel situation is observed
in RCS rats that demonstrate an inherited
form of retinal degeneration.13' 14 In this animal, early development and elongation of
ROS appear normal, but PE cell microvilli
Volume 22
Number 4
Shedding, phagocytosis of rod outer segments
Fig. 5. Electron micrographs of retina-PE cultured for 2 hr in atmosphere "A." a, PE cell soma
containing many phagosomes surrounded by optically empty space. Some phagosomes are
clumped into electron-dense masses. (Bar = 1 /xm.) b, Outer segment area of retina. (Bar = 1
fim.)
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
445
446 Tamai et al.
Invest. Ophthalmol. Vis. Sci.
April 1982
Fig. 6. Electron micrographs of retina-PE explants cultured for 2 hr in atmosphere "B." PE
cell contains large numbers of inclusion bodies under these conditions. {Bar = 1 fim.)
are deranged and phagocytosis is minimal. It
is hoped that the presently described in vitro
system may ultimately yield information relevant to this disease.
The basic mechanism by which a cascade of
disc shedding and phagocytosis is triggered
in vitro is presently unknown, as are the signals that initiate the process in vivo. In vitro,
it appears that the simple act of enucleation
is not sufficient to precipitate the burst of
disc shedding and phagocytosis, since these
events can be inhibited by manipulation of
the O 2 concentration in a subsequent incubation period. Similarly, Besharse et al.15
have recently shown that the bicarbonate ion
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
concentration of the culture medium is of
critical importance in supporting disc shedding in incubated eyecup preparations of X.
laevis. The present and previous work16* 1T
appears to rule out the influence of circulating neurohormones such as melatonin. Teirstein et a l . n have recently presented in vivo
evidence for the presence of independent
oscillators regulating outer segment phagocytosis within each eye. It is clear from the present work that the supply of O 2 available to the
PE cells and ROS from the choroid are critical in maintaining a normal shedding cycle.
Moreover, it may be that a simple change in
O2 concentration may itself act as one of the
Volume 22
Number 4
Shedding, phagocytosis of rod outer segments
Fig. 7. Electron micrographs of retina-PE explants cultured for 2 hr in atmosphere "C." a, PE
cell soma and parallel arrays of cell processes, {Bar = 1 /Am.) b, Outer segment area of retina
and ROS fragments. {Bar = 1 ftm.)
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
447
Invest. Ophthalmol. Vis. Sci.
April 1982
448 Tamai et al.
signals in the processes of shedding and
phagocytosis.
REFERENCES
1. Young R: Visual cells and the concept of renewal.
INVEST OPHTHALMOL 15:700, 1978.
2. LaVail M: Rod outer segment disc shedding in rat
retina: relationship to cyclic lighting. Science 194:
1071, 1976.
3. Edwards R and Szamier R: Defective phagocytosis
of isolated rod outer segments by RCS rat retinal
pigment epithelium in culture. Science 197:1001,
1977.
4. Goldman A, O'Brien P, Masterson E, Israel P,
Teirstein P, and Chader G: A quantitative system
for studying phagocytosis in pigment epithelium tissue culture. Exp Eye Res 28:455, 1979.
5. Hall M: Phagocytosis of light- and dark-adapted rod
outer segments by cultured pigment epithelium.
Science 202:526, 1978.
6. Goldman A and O'Brien P: Phagocytosis in the retinal pigment epithelium of the RCS rat. Science
201:1023, 1978.
7. Tamai M, Takahashi J, Noji T, and Mizuno K: Development of photoreceptor cells in vitro. Influence
and phagocytic activity of homo and heterogenic
pigment epithelium. Exp Eye Res 26:581, 1978.
8. Tamai M and O'Brien P: Retinal dystrophy in the
RCS rat: in vivo and in vitro studies of phagocytic
action of the pigment epithelium on the shed rod
outer segments. Exp Eye Res 28:399, 1979.
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
9. Hollyfield J and Basinger S: Photoreceptor shedding
can be initiated within the eye. Nature 274:794,
1978.
10. Goldman A, Teirstein P, and O'Brien P: The role of
ambient lighting in circadian disc shedding in the
rod outer segment of the rat retina. INVEST OPHTHALMOL VIS SCI 19:1257, 1980.
11. Teirstein P, Goldman A, and O'Brien P: Evidence
for both local and central regulation of rat rod outer
segment disc shedding. INVEST OPHTHALMOL Vis Sci
19:1268, 1980.
12. Flannery J and Fisher S: Light-triggered rod disc
shedding in Xenopus retina in vitro. INVEST OPHTHALMOL VIS SCI 18:638, 1979.
13. Bourne M, Campbell D, and Tansley K: Hereditary
degeneration of the rat retina. Br J Ophthalmol
22:613, 1938.
14. Dowling J and Sidman R: Inherited retinal dystrophy in the rat. J Cell Biol 14:73, 1962.
15. Besharse J, Terrill R, and Dunis D: Light-evoked
disc shedding by rod photoreceptors in vitro: relationship to medium bicarbonate concentration. Invest Ophthalmol Vis Sci 19:1512, 1980.
16. Tamai M, Teirstein P, Goldman A, O'Brien P, and
Chader G: The pineal gland does not control rod
outer shedding and phagocytosis in the rat retina
and pigment epithelium. INVEST OPHTHALMOL VIS
SCI 17:558, 1978.
17. LaVail M and Ward P: Studies on the hormonal control of circadian outer segment disc shedding in the
rat retina. INVEST OPHTHALMOL VIS SCI 17:1189,
1978.