Vitamin A Utilization in Human Retinal
Pigment Epithelial Cells In Vitro
M. T. Flood,* C. D. D. Bridges,! R. A. Alvarez,t W. S. Blaner,* and P. Gouras*
Vitamin A (vit A) metabolism was studied in freshly isolated and cultured human retinal pigment
epithelial (RPE) cells obtained from postmortem donor eyes. Fluorometric determination of vit A in
human RPE cells demonstrated that freshly isolated cells contained approximately 1.0 to 4.0 pg vit
A/cell which decreased with increasing time in culture; after 48 hrs in culture cellular vit A was
reduced 80%. High performance liquid chromatography (hplc) profiles of the retinyl esters in freshly
isolated RPE cells showed the presence of 11-cis retinyl stearate and palmitate and all-trans retinyl
stearate, palmitate and oleate; all-trans palmitate was the major ester. Hplc analyses of cell cultures
supplemented with all-trans retinol, using fatty acid-free bovine serum albumin as a carrier, showed
that the cells in primary and subcultures took up all-trans retinol and esterified it to form palmitate,
stearate, and oleate. Palmitate was the major ester synthesized by the cells in primary cultures. In
the subcultures the esters synthesized differed from that found in freshly isolated cells and in the
cells in primary culture; in the subcultures, the overall synthesis of ester was reduced and oleate was
more prominent. The esters that were synthesized in culture were all-trans; the formation of 11-cis
isomers was not observed in human RPE cells in culture. Electron microscopy of retinol-supplemented
cultures indicated that vit A doses up to 1.0 Mg/ml had no obvious effects on the cells; at higher
doses the cells no longer adhered to the culture surface. Invest Ophthalmol Vis Sci 24:1227-1235,
1983
The retinal pigment epithelium (RPE) serves an
essential role in the storage, metabolism, and transport of vitamin A used in the visual cycle. In some
way, vitamin A is transported between the retinal
epithelium and the photoreceptor cells depending
upon the latter's need.1"3 Exactly what the role of the
RPE cells is in the vitamin A cycle and whether it is
similar in the retinas of different animals is not yet
clear. There is some evidence, however, that this operation is quantitatively and even qualitatively different in different vertebrate species.12'4"11 For example, frog RPE contains relatively large quantities
of vitamin A in both the light- and dark-adapted
states; the rat, in contrast, stores large amounts of
vitamin A in the RPE in the light-adapted state and
almost none in the dark-adapted state. Most species
store vitamin A in the ester form.1"3'5-71012
Because of the difficulty to obtain viable human
retinal tissue for biochemical studies, very little is
known about vitamin A use and metabolism in the
human retina. The postmortem viability of human
RPE cells, and the ability to maintain these cells in
culture13"16 have provided an opportunity to study
the cell biology and biochemistry of human RPE cells
under controlled conditions. This paper provides basic information on the way human retinal epithelial
cells maintained in culture use and metabolize vitamin A.
Materials and Methods
Cell Cultures
Postmortem eyes of varying chronologic ages were
received through the New York Eye Bank for Sight
Restoration, the Retinitis Pigmentosa Donor Eye
Program, and the National Diabetes Research Interchange. In most cases the eyes were received after the
corneas were removed for corneal transplants. The
cells were obtained by trypsin-dissociation and primary cultures were established and maintained according to methods previously described.15 Cells removed from postmortem eyes by trypsin-dissociation
and processed immediately for biochemical studies
are referred to as freshly isolated cells. In vitro time
was calculated as postmortem time plus time in culture.
Subculturing was done when the primary cultures
became confluent, generally 21 to 30 days after the
From the College of Physicians and Surgeons,* Columbia University, New York, New York and the Cullen Eye Institute^ Baylor
College of Medicine, Houston, Texas.
Supported in part by Retina Research Foundation of Houston,
National RP Foundation, and NEI Grants EY03854, EY02591,
EY02489, EY02520, and Grant AM05968 from the National Institutes of Health.
Submitted for publication: December 30, 1982.
Reprint requests: M. T. Flood, PhD, Department of Ophthalmology, Columbia University, 630 W. 168 Street, New York, NY
10032
0146-0404/83/0900/1227/$ 1.25 © Association for Research in Vision and Ophthalmology
1227
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
1228
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1983
primary culture was established. A solution of 0.05%
trypsin and Versene (1:1000) was added to the culture
dishes for 5 min at 37 C to release the cells. The
number of cells in the suspension was determined
with a hemocytometer and cultures were plated out
at 1.5-2.0 X 105 cells/ml. Experiments on the subcultures were done on the seventh to tenth day when
the subcultures were confluent, as determined by
daily microscopic observations.
Vitamin A Studies
Total vitamin A determinations: To determine the
total vitamin A content of freshly isolated and cultured human RPE cells, the vitamin A was extracted
from the cells with hexane after saponification, concentrated under nitrogen, and separated chromatographically on an alumina column as described by
Harrison et al.17 Each fraction was assayed fluorometrically according to the method of Thompson and
his colleagues18 with the modification of Muto et al.19
In order to determine the amount of vitamin A in
the fetal calf serum (Sterile Systems) used to supplement the culture medium, an equal volume of 100%
ethanol was added to 2 ml of serum and the mixture
was extracted with n-hexane as previously described.20 The extract was analyzed by high performance liquid chromatography (hplc) to identify its
isomer content.
Retinol supplementation: RPE cell cultures were
supplemented with all-trans retinol (Hoffman-LaRoche) for varying intervals of time from 30 min to
24 hrs. Retinol, in concentrations of 0.1 to 50.0 ng/
ml, was added to Eagle's minimum essential medium
(MEM) (Flow Laboratories) which contained 500
/ig/ml fatty acid free bovine serum albumin (BSA)
(Sigma) as the retinol carrier. In these studies the fetal
calf serum was removed from the incubation medium
18 hrs before retinol was added. After the desired
incubation time, the cultures were washed with
Hank's balanced salt solution, Ca++ and Mg++ free,
and removed from the culture-plate surface by mild
trypsinization. Vitamin A content per cell in the retinol-supplemented cultures was determined with respect to incubation time with retinol. The optimum
supplementation concentration of retinol was determined.
Twelve additional pairs of eyes were used to study
the rate of retinol uptake in RPE cell homogenates.
To obtain the RPE cell homogenates, the eyes were
dissected as described for cell culturing.21 After removal of the neural retina, the RPE was rinsed twice
with Puck's saline at pH 7.4. A small aliquot of Puck's
saline was then added to the eyecup and the RPE
cells were gently brushed off Bruch's membrane. The
Vol. 24
cell suspension was homogenized in a glass/glass homogenizer and centrifuged at 100,000 X g for 1 hr.
The supernatant was then removed. The pellet was
resuspended in Puck's saline at pH 7.4 and divided
into 0.5 ml aliquots each containing 204 ^g protein
(as measured by the method of Lowry22). To each
aliquot was added 0.8 ng [ll,12-3H]-retinol (either
11-cis or all-trans). The [ll,12-3H]-retinol was prepared from the corresponding retinoic acid20 and labeled 11-cis and all-trans isomers were separated on
a preparative scale by hplc. The retinol (1.8 X 106
dpm) was added in 20 //I of ethanol and incubations
were made at 37 C. The reaction was stopped at various times by the addition of 10 ml chloroform-methanol(2:l v/v).
Two milliliters of water were then added, the upper
phase removed and discarded. The lower (chloroform) phase was then washed with Folch theoretical
upper phase (methanol:H2O:CHCl3, approximately
48:48:1). The chloroform extract was then dried in
a stream of prepurified nitrogen and the residue dissolved in 10% dioxane in n-hexane. This solution was
then chromatographed on alumina as described previously.20
The eluent from the hplc columns was collected
into 7-ml polypropylene scintillation vials at 0.2- to
1.0-min intervals. Six milliliters of Scintilene (Fisher)
were then added to each vial and the radioactivity
was counted in a Packard Tricarb Liquid Scintillation
Spectrometer. The counts emerging in the peaks corresponding to retinyl stearate and palmitate were used
to measure the amount of retinol substrate esterified.
High performance liquid chromatography analysis:
Hplc was used to identify the specific retinyl esters
present in freshly isolated and cultured human RPE
cells and to identify the fatty acid moiety of the retinyl
esters synthesized in retinol-supplemented cultures.
In these studies the esters were extracted from the
cells with 2 X 4 ml volumes of acetone. The extract
was filtered through glass fiber, evaporated under a
stream of nitrogen, and redissolved in 1 ml of n-hexane containing 10% v/v dioxane. The extract was
purified on an alumina column (5% water deactivated) and eluted with 10% v/v dioxane in n-hexane.
A portion of the eluent was prepared for total vitamin
A measurement by the Carr-Price reaction as previously described20 or saponified and prepared for vitamin A determinations as described above in section
on total vitamin A determinations. For normal phase
chromatography, two 4.6 mm X 25 cm columns, an
Ultrasphere Si 60 and a Spherisorb CN, packed with
5 fim adsorbents were used separately or in series.
The mobile phase for retinyl ester analysis was 0.4%
ether/n-hexane at 1 ml/min. All solvents were hplc
grade (Burdick & Jackson). The equipment for hplc
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
1229
VITAMIN A UTILIZATION / Flood er ol.
No. 9
analysis has been previously described.4'23 Standardization and preparation of authentic retinoids is described in reference.20
ICH
Electron Microscopy
For ultrastructural studies, RPE cell cultures were
supplemented for 18 hours with all-trans retinol (0.5
and 1.0 Mg/ml) and then were fixed with 3% gluteraldehyde for 2 hrs at 4 C. The cultures were then
washed with Earle's buffer (pH 7.4) and postfixed in
1% osmium tetroxide for 1 hr. After ethanol dehydration, Epon was poured into the culture dish and
polymerized. Following polymerization, the culture
dish was cut into blocks and the sections were cut
perpendicular to the culture surface to study the apical and basal cell surfaces. Thin sections were stained
with uranyl acetate and lead citrate and studied on
a Siemens electron microscope.
10-
01-
O.OI-
8
DAYS
Results
Total Vitamin A in RPE Cells
In a previous paper24 we reported that the total
amount of vitamin A in human RPE cells, freshly
isolated from 24 pairs of donor eyes of varying chronologic ages ranged from approximately 1.0 to 4.0 pg/
cell. Data from 14 additional pairs of eyes confirmed
our initial report (Fig. 1).
Normal phase hplc analyses of the retinyl esters in
freshly isolated RPE cells showed that these cells
stored 11-cis retinyl palmitate and stearate and alltrans retinyl palmitate, stearate, and oleate (Fig. 2).
Occasionally, 13-cis retinyl stearate and palmitate
were present. However, there was great variability in
the presence and the content of the 13-cis esters in
these cells; the factors that control this variability are
unknown at this time.
Loss of Vitamin A in RPE Cells in Vitro
Analysis of the vitamin A content in RPE cells in
vitro demonstrated that vitamin A is rapidly lost from
these cells in culture even though the cells are maintained in medium supplemented with fetal calf serum
(FCS). Our data showed that within 2 to 4 days in
vitro, there is an 80% decrease in the total amount
of vitamin A per cell (Fig. 1; Table 1). Analysis of the
FCS used to supplement the culture medium showed
that the serum contained approximately 12 tig alltrans retinol per 100 ml serum. The amount of retinol
varied slightly in different serum lots. Over 90% of
this retinol was in the all-trans configuration.
Hplc analysis of the isomeric ester forms present
in freshly isolated and cultured RPE cells from the
same donor cell populations (Fig. 3) indicated that
10
12
14
50
150
250
POST MORTEM + IN VITRO
Fig. 1. Vitamin A content in human RPE cells in vitro. Closed
circles (•) represent freshly isolated cells; open circles (O) represent
cells in culture. Data points connected by a line indicate cell populations from the same donor.
there is no preferential loss of specific isomeric forms
of retinyl palmitate and stearate in vitro. Within 48
hours in culture there is a significant loss of the alltrans, 11-cis, and 13-cis palmitate and stearate in human RPE cells. After 48 hours in culture, all-trans
palmitate and stearate had decreased by 85%; 11-cis
palmitate and stearate by 75%, and 13-cis palmitate
and stearate by 65%. However, the initial level of all-
^325
0.005
10 min.
Fig. 2. Normal phase hplc separation of retinyl esters extracted
from freshly isolated human RPE cells. Mobile phase = 0.4% diethyl ether in n-hexane at 1.0 ml/min. Columns = Ultrasphere
Si60 and Spherisorb CN in series. Peak identification: 1 = 11-cis
retinyl stearate; 2 = 11-cis retinyl palmitate; 3 = unknown; 4 = alltrans retinyl stearate; 5 = all-trans retinyl palmitate; 6 = all-trans
retinyl oleate.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
1230
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1983
Table 1. Decrease in vitamin A content in retinal
epithelial cells in vitro
Vit. A content
In vitro time*
(days)
Freshly
isolated
2-4
8-12
+ SEM
No. of
donors
(pg/cell)
Percent
decrease
35
9
14
2.02 + 0.16
0.41 + 0 . 1 0
0.36 + 0.08
—
80%
82%
• In vitro time = postmortem time + time in culture.
trans palmitate and stearate was significantly greater
than the 11-cis and 13-cis forms. The ester profile
from the cells that did not attach to the culture surface
suggested that the loss of each isomeric form was
slightly less in the unattached cells than in the attached cells. The difference in ester loss between the
attached and unattached cells was minimal.
Retinol Supplementation
Supplementation of primary cell cultures for an
18-or 24-hr period with medium containing all-trans
retinol in concentrations of 0.10 to 50.0 ng retinol
per ml culture medium with BSA as the retinol carrier
demonstrated that concentrations of 0.25 to 0.50 /*g
retinol per ml culture medium could raise the vitamin
A content in the RPE cells in vitro to levels commensurate with those in freshly isolated cells (Fig. 4).
Retinol supplementation doses greater than 2.0 ng/
ml were found to affect cell attachment when presented to the cells in this way. At higher retinol con-
Total Vitamin A
Esters
300-
centrations the cells rounded up and detached from
the culture surface.
Hplc analysis of human RPE cells in primary and
subcultures showed that the cells were able to take
up all-trans retinol and esterify it to the major esters
found in these cells in vivo. Figure 5 is an hplc chromatogram of the esters extracted from cells maintained in primary culture for 11 days. It is clear that
these cells contained negligible amounts of retinyl
esters (Fig. 5A). After 18 hrs supplementation with
all-trans retinol (Fig. 5B), however, the cells synthesized all-trans retinyl palmitate, stearate, and oleate;
palmitate was the major ester synthesized. Similarly,
there was no evidence of retinyl esters in subcultured
cells after 50 days in culture (Fig. 6A). When supplemented with all-trans retinol for 24 hrs, the subcultured cells were also able to take up and esterify
all-trans retinol, forming retinyl palmitate, stearate
and oleate (Fig. 6B). Subcultured cells formed a
higher proportion of oleate compared with that present in freshly isolated cells or that formed by cells in
primary cultures. The total amount of retinyl esters
synthesized in each cell population studied is summarized in Table 2. There was no evidence of 11-cis
isomers in the cultured RPE cells.
The data also showed that in the retinol-supplemented cultures the amount of retinyl ester present
in primary culture of RPE cells from the same donor
increased gradually with increased supplementation
time over a 24-hr period (Fig. 7). After 24 hrs of
supplementation with all-trans retinol, the amount
of retinyl esters extracted per /*g protein for cells in
primary cultures was equivalent to the range of retinyl
All-Trans
Palmitate+Stearate
I-Freshly isolated cells
2-Cells in Primary Culture
(48 hrs. in Culture)
3-Unattached cells
(48hrs. in culture)
300-
200-
200-
100-
100-
50-
50-
\
I
2
3
Vol. 24
Il-Cis
Palmitate +Stearate
50-
50-
I
2
3
13-cis
Palmitate+Stearate
I
2
3
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
Fig. 3. Loss of vitamin A
esters in human RPE cells
in vitro. Analysis was done
on normal phase hplc using
a Ultrasphere Si60 and
Spherisorb CN columns in
series; mobile phase = 0.4%
diethyl ether in n-hexane at
1.0 ml/min.
No. 9
1231
VITAMIN A UTILIZATION / Flood er ol.
ester content in freshly isolated RPE cells of the various donors studied. In supplemented subcultures, the
amount of retinyl esters extracted from the cells increased with time over seven hours for each donor
cell population that was studied; however, the total
amount of retinyl esters synthesized per ^g protein
was not as high as in RPE cells in primary cultures
(Fig. 7).
The uptake and esterification of 11-cis, and alltrans retinol by the particulate fraction of RPE cell
homogenates is illustrated in Figure 8. The esterification process appeared to be rapid for both isomers
but the total amount esterified was 104-fold less (per
n% protein) than observed with whole cells in primary
or subculture. Part of this difference may be due to
the different modes of delivery. When an ethanol solution of retinol is added to an aqueous buffer, the
resulting dispersion is highly susceptible to destruction by oxidation. It is notable that both 13-cis and
all-trans retinyl esters are formed in addition to 11cis when the substrate is 11-cis retinol (Fig. 8). Thirteen-cis and all-trans esters are also formed when the
substrate is all-trans retinol.
The synthesis of retinyl esters from exogenous retinol by whole cells was considered to be a true indication of retinol uptake since an artifactually high
level of free retinol could occur in the cell suspensions
because of adsorption to surface membranes.
Electron Microscopy
Electron microscopic studies of the retinol supplemented cells showed intact cells with ultrastructural
features characteristic of RPE cells in vitro1516 and
similar to the control cells (Fig. 9A). The micrographs
of the human RPE cells supplemented for 18 hrs with
0.5 and 1.0 fig retinol per ml culture medium (Figs.
9B, C) demonstrated that these retinol concentrations
had no apparent effect on the cultured RPE cells. The
cells showed an apical-basal polarity with numerous
slender processes along the apical surface and dense
infoldings of the basal surface. Mitochondria with
well-defined cristae and dense concentrations of intramitochondrial granules in addition to channels of
rough endoplasmic reticulum and numerous ribosomal granules were scattered among the lipofuscin
granules in the cytoplasm of the cells. Although both
the retinol-supplemented and control cells had ultrastructural features suggestive of much activity on the
apical surface, there appeared to be a greater number
of pinocytotic vesicles along the apical surface of the
retinol-supplemented cells.
Discussion
Our data showed that human RPE cells contain
approximately 1.0 to 4.0 pg vitamin A per cell as
IOO-I
10-
1.0
O.I
1.0
2.0
3.0
25
50
RETINOL SUPPLEMENT (/jg/ml)
Fig. 4. Uptake of all-trans retinol at 37 C by RPE cells in primary
cultures.
A
A
325
0.0 02
12
W
5
6
B
12
10 min.
Fig. 5. Normal phase hplc analysis of the retinyl esters of human
RPE cells 11 days in primary culture. Mobile phase = 0.4% diethyl
ether in n-hexane at 1.0 ml/min. Columns = Ultrasphere Si60 and
Spherisorb CN in series. A, Unsupplemented cells; B, Cells supplemented with all-trans retinol (0.5 Mg/ml), for 18 hours. Peak
identification: 1 = 11-cis retinyl stearate; 2 = 11-cis retinyl palmitate; 3 = unknown; 4 = all-trans retinyl stearate; 5 = all-trans
retinyl palmitate; 6 = all-trans retinyl oleate.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
1232
Vol. 24
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1983
A
0.002
12
5 6
0
B
10 min.
Fig. 6. Normal phase hplc analysis of the retinyl esters in subcultured human RPE cells at 50 days in vitro. Mobile phase = 0.4%
diethyl either in n-hexane at 1.0 ml/min. Column = Ultrasphere
Si60 and Spherisorb CN in series. A, Unsupplemented cells, B,
Cells supplemented with all-trans retinol (0.5 /*g/ml) for 24 hrs.
Peak identification: 1 = 11-cis retinyl stearate; 2 = 11-cis retinyl
palmitate; 3 = unknown; 4 = all-trans retinyl stearate; 5 = all-trans
retinyl palmitate; 6 = all-trans retinyl oleate.
determined through fluorometric analysis. The actual
amount of vitamin A per cell in the freshly isolated
RPE cells is probably slightly higher since 11-cis retinol has less fluorescence than the all-trans form.25
Vitamin A is stored in the RPE cells of most species. Even when there is a tremendous reduction in
the liver store due to vitamin A deprivation, a large
store of vitamin A is generally retained in the retina.2
This fact insures that the photoreceptors are supplied
with vitamin A, which is essential for normal visual
function. Within 48 hrs in culture, however, the
amount of vitamin A in the human RPE cells is drastically reduced. This suggests that some factor that
influences the storage and release of vitamin A is
missing in human RPE cells in vitro. This factor may
be an abnormal concentration gradient for vitamin
A across the RPE cell membrane, drawing the intracellular vitamin A into the media. The concentration
of vitamin A in calf serum is about 25% of that in
normal human serum.26 The vitamin A remaining
in these cells is much less than 1% of their normal
levels, however, implying that it is not mass action
alone that is responsible for the vitamin A depletion.
It is possible that the extracellular vitamin A concentrations in the extracellular pool around the photoreceptors is higher than it is in serum. It cannot be
too high, however, because vitamin A as suggested
by both our results and those of others27 appears to
be toxic in concentrations much higher than those
in normal serum. Further experiments will be necessary to uncover the factors that cause this massive
and relatively sudden loss of vitamin A from cultured
RPE cells.
There is presently no evidence that the loss of photoreceptors will lead to a concomitant loss of vitamin
A from retinal epithelium. It is interesting that detachment of the neural retina in cats28 and monkeys
(our unpublished results) leads to proliferation of retinal epithelium. Normally retinal epithelial cells in
mammals do not enter mitosis after birth. Under
nomal conditions, however, RPE cells have relatively
high levels of vitamin A compared to other body cells
except the liver cells. The proliferation of retinal epithelium after retinal detachment, as well as the proliferation observed in tissue culture, may be related
to the loss of vitamin A from these cells.
Our studies showed that all-trans retinol added to
culture medium containing fatty acid free bovine
serum albumin as the retinol carrier was taken up by
the cells in vitro and esterified to form the major
esters found in freshly isolated human RPE cells.
Table 2. Composition of esters in retinol-supplemented cultures
Donor age
(yrs)
Culture
70
48
48
71
71
30
30
30
30
Freshly isolated
Primary
Primary
Primary
Primary
Subculture
Subculture
Subculture
Subculture
In vitro time*
(days)
Vi
11
11
11
11
50
50
51
51
* In vitro time = postmortem time + time in culture.
Retinol
supplementation
(hrs)
Unsupplemented
Unsupplemented
18
Unsupplemented
24
Unsupplemented
9
18
24
All-trans
All-trans retinyl esters
ng/10s cells
Stearate
Palmitate (%)
Oleate
448.7
21.7
73.7
4.6
t
t
38.2
t
36.3
t
47.7
t
45.0
13.7
29.6
36.1
33.0
*
39.7
34.5
38.2
*
25.0
30.7
29.1
869.8
t
138.5
t
44.3
55.4
44.3
•
t Neglible amount = could not be determined.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
t
t
18.1
No. 9
1233
VITAMIN A UTILIZATION / Flood er ol.
• Cells in Primory Culture
o Subcultured Cells
c
<u
2
Q.
CT>
1.0-
Fig. 7. Rate of retinyl ester synthesis in RPE cells in
primary and subcultures
supplemented with all-trans
retinol (0.5 fig/m\).
UJ
<
0.5o
3
4
5
6
7
10 15 20 25
Supplementation Time (hours)
Previous studies on human29-30 and bovine31 RPE that
reported that RPE cells do not take up retinol without
its being bound to serum retinol binding protein must
be incorrect. In human RPE, the highest concentration of retinol that could be added to the culture
medium without producing toxicity was about onefifth that customarily used to supplement liver cell
cultures with vitamin A.27 This implies that human
RPE is relatively sensitive to free retinol. With the
addition of SRBP, Saari et al32 demonstrated that the
uptake of retinol by retinoblastoma cells was much
slower compared with retinoblastoma cells that were
supplemented with free retinol. Thus, SRBP may
control the uptake of retinol by RPE cells through
its interaction with membrane receptors and thus
protect these cells from the toxic effect of free retinol.
The increased amount of oleate and the decreased
amounts of palmitate and stearate esters synthesized
by RPE cells in subculture is the first evidence of a
biochemical alterations in human RPE cells in culture. This change may be occurring early in the in
vitro life of human RPE cells since primary cells also
show a slight increase in the amount of oleate synthesized by these cells when compared to freshly isolated cells. This difference is probably not due to a
unique composition of acyl carriers provided to the
cells by the fetal calf serum since both the primary
and subcultures have been exposed to identical serum
and show very different ester profiles. The time for
turnover of endogenous pools, the size and stability
of which are unknown, may also be a contributing
factor to the differing ester profiles.
Human RPE cells in culture, under the conditions
studied, do not isomerize all-trans retinol to 11-cis,
a reaction that is critical to rhodopsin synthesis and
photoreceptor function. This reaction has been
thought to require the RPE,1021 although there has
been evidence that this may not be the case in rat
retina.33 The 11-cis esters found in human RPE cells
in vivo and in freshly isolated human RPE cells may
be due to the interaction of RPE cells with photoreceptors as suggested for the frog.' The isomerization
20.0
Total
ll-cis
Incubation Time (minutes)
Fig. 8. Retinol esterification in RPE cell homogenates. A, Substrate: 11-cis retinol (0.8 jtg)- Protein: 204 Mg/assay (B) Substrate:
all-trans (0.8 j*g) Protein 204 /tg/assay
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
Fig. 9. A, Electron micrograph of human RPE cells
from a 74-year-old donor maintained in primary culture
for 14 days with MEM plus 20% FCS. The cells maintain
an apical(a)-basal(b) polarity with microvilli projections
(m) on the apical surface (magnification = 6,000x). B,
and C, Electron micrographs of human RPE cells from
the same donor as in A supplemented with O.S ftg/ml
all-trans retinol for 18 hrs. Note the number of microvilli
projections (m) and pinocytottc vesicles ( s ) on the apical surface (B, magnification = 6.500X) (C, magnification = I2,500X).
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017
VITAMIN A UTILIZATION / Flood er ol.
No. 9
of all-trans retinol to 11-cis in the human retina may
depend upon the presence of the neural retina.
16.
Key words: vitamin A, RPE cells, human, in vitro
17.
References
18.
1. Bridges CDB: Vitamin A and the role of the pigment epithelium
during bleaching and regeneration of rhodopsin in the frog eye.
Exp Eye Res 22:435, 1976.
2. Dowling JE: Chemistry of visual adaptation in the rat. Nature
188:114, 1960.
3. Noell WK, Delmelle MC, and Albrecht R: Vitamin A deficiency
effect on retina: dependence on light. Science 172:72, 1971.
4. Alvarez RA, Bridges CDB, and Fong SL: High-pressure liquid
chromatography of fatty acid esters of retinol isomers. Analysis
of retinyl esters stored in the eye. Invest Ophthalmol Vis Sci
20:304, 1981.
5. Berman ER, Segal N, and Feeney L: Subcellular distribution of
free and esterified forms of vitamin A in the pigment epithelium
of the retina and in liver. Biochim Biophys Acta 572:167, 1979.
6. Bridges CDB: Storage, distribution and utilization of vitamins
A in the eyes of adult amphibians and their tadpoles. Vision
Res 15:1311, 1975.
7. Bridges CDB and Yoshikami S: The rhodopsin-porphyropsin
system in freshwater fishes-2. Turnover and interconversion of
visual pigment prosthetic groups in light and darkness: role of
the pigment epithelium. Vision Res 10:1333, 1970.
8. Bridges CDB, Fong SL, and Alvarez RA: Separation by programmed-gradient high-pressure liquid chromatography of vitamin A isomers, their esters, aldehydes, oximes and vitamin
A2: presence of retinyl ester in dark-adapted goldfish pigment
epithelium. Vision Res 20:355, 1980.
9. Hubbard R and Dowling JE: Formation and utilization of 11cis vitamin A by the eye tissues during light and dark adaptation.
Nature 193:341, 1962.
10. Krinsky N: The enzymatic esterification of vitamin A. J Biol
Chem 232:881, 1958.
11. Liou GI, Bridges CDB, and Fong SL: Vitamin A transport between retina and pigment epithelium—An interphotoreceptor
matrix protein carrying endogenous retinol (IRBP). ARVO Abstracts. Invest Ophthalmol Vis Sci 22(Suppl):65, 1982.
12. Zimmerman WF: The distribution and proportions of Vitamin
A compounds during the visual cycle in the rat. Vision Res
14:795, 1974.
13. DelMonte MA and Maumenee IH: Mucopolysaccharide metabolism of cultured human RPE. ARVO Abstracts. Invest
Ophthalmol Vis Sci 18(Suppl): 154, 1979.
14. Edwards RB: Synthesis of glycosaminoglycans by cultured human
retinal pigment epithelium. ARVO Abstracts. Invest Ophthalmol
Vis Sci 17(Suppl):124, 1978.
15. Flood MT, Gouras P, and Kjeldbye H: Growth characteristics
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
1235
and ultrastructure of human retinal pigment epithelium in vitro.
Invest Ophthalmol Vis Sci 19:1309, 1980.
Flood MT and Gouras P: V. Pigment Epithelium. The organization of human retinal pigment epithelium in vitro. Vision
Res 21:119, 1981.
Harrison EH, Smith JE, and Goodman DS: Unusual properties
of retinyl palmitate hydrolase activity in rat liver. J Lipid Res
20:760, 1979.
Thompson JN, Erdody P, Brien R, and Murray TK: Fluorometric
determination of vitamin A in human blood and liver. Biochem
Med5:67, 1971.
Muto Y, Smith JE, Milch PO, and Goodman DS: Regulation
of retinol-binding protein metabolism by vitamin A status in
the rat. J Biol Chem 247:2542, 1972.
Bridges CDB and Alvarez RA: Measurement of the vitamin A
cycle. Methods Enzymol 81:463, 1982.
Fong SL, Bridges CDB, and Alvarez RA: Utilization of exogenous
retinol by frog pigment epithelium. Vision Res 23:47, 1983.
Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ: Protein
measurement with the Folin phenol reagent. J Biol Chem
193:265, 1951.
Bridges CDB, Alvarez RA, and Fong SL: Vitamin A in human
eyes: amount, distribution, and composition. Invest Ophthalmol
Vis Sci 22:706, 1982.
Flood MT, Gouras P, Haley JE, and Blaner WS: Human retinal
pigment epithelium in vitro: organization, ultrastructure, and
biochemistry. In Genetic Eye Diseases: Retinitis pigmentosa and
other inherited eye disorders, Cotlier, Maumenee, Berman, editors. New York, Alan R. Liss, Inc., 1982, pp. 53-66.
Thomson AJ, Fluorescence spectra of some retinyl polyenes. J
Chem Phys 51:4106, 1969.
Smith FR and Goodman DS: The effects of diseases of the liver,
thyroid, and kidneys on the transport of vitamin A in human
plasma. J Clin Invest 50:2426, 1971.
Smith JE, Borek C, and Goodman DS: Regulation of retinolbinding protein metabolism in cultured rat liver cell lines. Cell
15:865, 1978.
Anderson DH, Stern WH, Fisher SK, Erickson PA, and Borgula
GA: The onset of pigment epithelial proliferation after retinal
detachment. Invest Ophthalmol Vis Sci 21:10, 1981.
Maraini G and Gozzoli F: Binding of retinol to isolated retinal
pigment epithelium in the presence and absence of retinol-binding
protein. Invest Ophthalmol 14:785, 1975.
Maraini G, Ottonello S, Gozzoli F, and Merli A: Identification
of a membrane protein binding the retinol in retinal pigment
epithelium. Nature 265:68, 1977.
Chen C-C and Heller J: Uptake of retinol and retinoic acid from
serum retinol-binding protein by retinal pigment epithelial cells.
J Biol Chem 252:5216, 1977.
Saari JC, Bredberg L, and Futterman S: Esterification of retinol
by retinoblastoma cell lines W E R l - R b l and Y79. Invest
Ophthalmol Vis Sci 19:1301, 1980.
Cone RA and Brown PK: Spontaneous regeneration of rhodopsin
in the isolated rat retina. Nature 221:818, 1969.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933116/ on 06/17/2017