768
Invest. Ophthalmol. Visual Sci.
August 1977
Reports
alterations in pH 4 were brought about by appropriate changes in the concentration of bicarbonate.
As regards the primary action of bicarbonatefree medium, a good case can be made that the
selective loss of the b wave and the decline in
the amplitude of the a wave in the rat retina results from a decrease in utilization of ATP below
critical levels at critical sites. Substitution of a
phosphate for a bicarbonate buffer decreases
aerobic and anaerobic glycolysis (Table I) and
respiration.2- 3 It was expected that this decreased
metabolism would probably result in low ATP levels in the tissue. However, under aerobic conditions this was not found, suggesting that lack of
bicarbonate affects ATP utilization as well as its
production. That absence of bicarbonate appears
to affect both the synthesis and utilization of ATP
does not permit us to assess the significance of
the differences in anaerobic ATP content of retinas
incubated in bicarbonate or phosphate (Table II).
Additional experiments will be required to determine the relative importance of changes in the
synthesis and utilization of ATP in relation to the
observed alterations in the ERG potentials in bicarbonate-free medium. Furthermore, the cellular
sites of action of bicarbonate and the mechanisms
which regulate these changes in metabolic and
electrical activities remain the most provocative
issues raised by these experiments. One possible
candidate for this regulatory role is ATP, a wellknown feedback inhibitor of glycolysis and respiration.8 A second candidate is calcium, since changes
in its concentration affect the rates of glucose
utilization and respiration in both bicarbonate- and
phosphate-buffered media.3' °
Finally, an intriguing speculation is that bicarbonate may be specifically involved in the function
of intracellular compartments. This is suggested
by the fact that a bicarbonate-stimulated ATPase
was found to represent approximately 15 percent
of total ATPase activity in retinal homogenates.111
From the Institute of Biological Sciences, Oakland University, Rochester, Mich. This work was
supported in part by research grant EY 01219
from the National Eye Institute, United States
Public Health Service. "Valjean Simson was a participant in an undergraduate research program at
Oakland University. Submitted for publication
March 30, 1977. Reprint requests: Dr. Barry S.
Winkler, Institute of Biological Sciences, Oakland
University, Rochester, Mich. 48063.
Key words: retina, bicarbonate, phosphate, buffers, electroretinogram, metabolism, lactate, ATP.
REFERENCES
1. Cohen, L. H., and Noell, W. K.: Glucose
catabolism of the rabbit retina before and
after development of visual function, J. Neurochem. 5:253, 1960.
2. Riley, M. V.: The effect of sodium ions on
glucose metabolism of ciliary body and retina.
In Graymore, C. N., editor: Biochemistry of
the Retina, New York, 1965, Academic Press,
Inc.
3. Riley, M. V., and Voaden, M. J.: The metabolism of the isolated retina, Ophthalmol.
Res. 1:58, 1970.
4. Winkler, B. S.: The electroretinogram of the
isolated rat retina, Vision Res. 12:1183, 1972.
5. Hagins, W. A., Penn, R. D., and Yoshikami,
S.: Dark current and photocurrent in retinal
rods, Biophys. J. 10:380, 1970.
6. Merriam, W. A., and Kinsey, V. E.: Studies
on the crystalline lens: technic for in vitro
culture of crystalline lenses and observations
on metabolism of the lens, Arch. Ophthalmol.
43:979, 1950.
7. Barker, S. B., and Summerson, W. H.:
Colorimetric determination of lactic acid in
biological material, J. Biol. Chem. 138:535,
1941.
8. Lehninger, A. L.: Biochemistry, ed. 2, New
York, 1975, Worth Publishers, Inc.
9. Kornbleuth, W., Yardeni-Yaron, E., and Wertheimer, W.: Glucose utilization of the retina,
Arch. Ophthalmol. 50:45, 1953.
10. Winkler, B. S., and Riley, M. V.: Na-K and
HCO3 ATPase activity in retina:dependence
on calcium and sodium (submitted for publication ).
Occurrence of 11-cis-retinal-binding protein
restricted to the retina. SIDNEY FUTTERMAN AND JOHN C. SAARI.
A binding protein for retinal (vitamin A aldehyde),
the 11-cis-retinalr-binding protein, was present in
the soluble protein fraction of rat retina but was
absent from brain, lung, heart, skeletal muscle,
liver, kidney, spleen, small intestine, and testes.
The binding protein was also found in human
retina but not human liver. The binding protein
for retinal from human retina did not bind retinol
and was larger than, and readily separable from,
the cellular retinol-binding protein which did not
bind retinal. It would appear that the occurrence
of a soluble binding protein for retinal, unlike the
more widely distributed binding proteins for
retinol and retinoic acid, may be unique to the
retina.
A protein that binds cis-trans isomers of retinal
(vitamin A aldehyde), with preference for 11cis-retinal, occurs in the soluble protein fraction
of bovine retina.1 Its properties suggest a possible
role in the regeneration of bleached rhodopsin.
However, a more general role might also be
inferred because retinal is also an intermediate in
the oxidation of retinol to retinoic acid and cellular
binding proteins for retinol and retinoic acid are
found in a variety of tissues2' ;i in addition to
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933075/ on 06/17/2017
Volume 16
Number 8
retina.1"11 The present study was undertaken to
determine whether 11-cis-retinal-binding protein
could be found in organs known to contain either
retinol- or retinoic acid—binding proteins.
Materials and methods
Tissue preparations. Tissues obtained from male,
250 gm. Sprague-Dawley rats, human autopsy
liver, and human retinas from Lions Eye Bank
eyes were stored frozen. Frozen bovine retinas
were purchased from Hormel Co.
The soluble protein fraction was prepared from
20 rat retinas or five human retinas by homogenizing in 2 ml. of buffer (50 mM Tris • Cl, pH 7.5/
200 mM NaCl) and centrifuging at about 150,000
x g for 2 hr. Supernatants were prepared similarly
from other tissues after homogenizing in three
volumes of buffer. The supernatant from human
retina was dialyzed overnight against buffer and
clarified by centrifugation. Protein content was
estimated,7 and the preparations were stored
frozen.
An ll-cis-retinal-binding protein fraction was
prepared by gel filtration on Sephadex G-100 of
the soluble protein extracted from 50 bovine
retinas (22 gm.) 1 and for comparison Sephadex
G-100 column fractions were obtained with the
same procedure from 22 gm. of rat liver and
brain. These preparations were concentrated to
5 ml. with an Amicon UM-10 membrane filter and
stored frozen.
Dialyzed supernatants were prepared from 22
gm. of rat brain and liver and human livers. The
tissues were homogenized in 5 ml. of buffer and
centrifuged as described above, and the supernatants were dialyzed overnight against buffer and
clarified by centrifuging again.
Incubation and analysis. Samples of tissue
supernatant, 11-cis-retinal-binding protein fraction or dialyzed supernatant were incubated for
30 min. at room temperature with 1 ^Ci of 3 Hretinal, specific activity 1.33 Ci./mmol., prepared
from 3H-retinol, or 2 <uCi of 3H-retinol, specific
activity 2.66 Ci./mmol., purchased from New
England Nuclear Corp. (Boston, Mass.). After
addition of 100 mg. of sucrose to increase the
density, samples were analyzed by gel filtration
through Sephadex G-100 columns of 1.5 by 143
cm. Following gel filtration 1 ml. samples of each
3.5 ml. fraction were counted in 10 ml. of scintillation fluid.
Results. Both the low molecular weight cellular
retinol-binding protein and the 11-cw-retinal—
binding protein were present in rat retina (Fig. 1,
A). The binding proteins in rat and bovine
retina also appeared to be of similar molecular
weight on the basis of elution volume in gel
filtration.
Although all the other tissues examined contained the cellular retinol-binding protein (Fig.
1, B to D), no ll-c/.9-retinal-binding protein could
Reports 769
Vo
A
°1
Retinol
binding
protein
Retinol
binding
protein
Vi
86-
Bovine ,' i
retina " 1 J
4-
Rat
retina
' <
~n\\
V
2J
10
iB
8-
Kidney
A
—Spleen
j 1
/ 1
6Small
4*
/
I
intestine/
\
2-
V.
i
10 C - .
8
6
42-
D
i
i
86Liver
A
Brain—v 1 \
SkeletalV \
k muscle K \
42-
J
20
30 40
50
60
70
FRACTION NUMBER
Fig. 1. Gel filtration on Sephadex G-100 of the
soluble protein from retina and other tissues following incubation with both 3H-retinal and 3 Hretinol. Reaction mixtures contained 10 mg. of
protein in 1.6 ml. of buffer and about 0.77 nmol.
of each ligand. Vo is the void volume, Vi the
included volume of the columns, and fractions
32-38 contain the 11-cw-retinal-binding protein.
The peaks occurring at Vo and Vi are also
present when 3H-retinal or 3H-retinol are chromatographed in the absence of protein.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933075/ on 06/17/2017
770
Invest. Ophthalmol. Visual Sci.
August 1977
Reports
20 1
Vo
HUMAN
RETINA
8-
16-1
Retinal
binding
protein
Retinol
binding
protein
RETINA
64-
* \2-\
2-
81
LIVER
HUMAN
8n LIVER
4-
BRAIN
35
45
FRACTION NUMBER
Fig. 2. Gel filtration of the 11-cts-retinal-binding
protein fraction from bovine retinas and comparable preparations made from liver and brain following incubation of 1 ml. samples with 1 ^Ci of 3 Hretinal.
be detected in testes, lung, heart, kidney, spleen,
small intestines, or skeletal muscle. Some binding
of tritiated ligand was observed in the region of
the 11-cis-retinal-binding protein in the case of
liver and brain (Fig. 1, D).
The possibility was considered that failure to
find ll-ci$-retinal-binding protein in various tissues might be due to reduction of the bulk of the
added 3H-retinal to 3H-retinol by reduced pyridine
nucleotides generated from endogenous substrates
and alcohol dehydrogenase activity in the supernatants. This question was examined by incubating
the tissue supernatant samples with added 3 Hretinal as in Fig. 1 but omitting the 3H-retinol.
Reaction mixtures were then extracted in the
presence of 0.1 mg. of added unlabeled retinal
and retinol, and after thin-layer chromatography
the retinol and retinal zones were recovered and
counted. The retinol spot contained no more than
1% of the total radioactivity except in the case of
retina (2%) and liver (20%), clearly insufficient conversion to prevent detection of 11-ctsretinal—binding protein.
The possible occurrence of 11-cw-retinal-binding protein in liver and brain was examined
further in comparable Sephadex G-100 fractions
prepared from rat liver and brain and bovine
retina in which the binding protein should be
found if present in the tissue (Fig. 2). Only the
protein fraction derived from retina showed unequivocal binding of 3H-retinal.
To confirm and extend this observation, dialyzed
supernatants of rat brain and liver and human
4-
20
30
40
50
60
FRACTION NUMBER
70
Fig. 3. Presence of 11-cis-retinal-binding protein
in human retina and absence from human liver.
Dialyzed soluble protein from human retina, 15
mg. in 1 ml. of buffer, or dialyzed supernatant
from human liver, 90 mg. of protein in 1 ml. of
buffer, were incubated with 3H-retinal (
) or
3
H-retinol (- - -) and then subjected to gel
filtration on Sephadex G-100.
liver and retina were incubated with 3H-retinal
or 3H-retinol and analyzed by gel filtration. With
the use of 3H-retinol, only the cellular retinolbinding protein from each tissue sample was
labeled, and no peak of radioactivity was observed
in the region of the 11-cis-retinal-binding protein,
as shown for human retina and liver (Fig. 3).
After incubation with 3H-retinal no peak was
detected in the region of the cellular retinol-binding protein nor in the region of the 11-cis-retinalbinding protein in the case of liver or brain. However human retina, like bovine and rat retina, did
contain the 11-ew-retinal-binding protein (Fig.
3).
Discussion. The occurrence in rat and human
retina of a soluble protein capable of binding 3 Hretinal is demonstrated in this study. This binding
protein does not bind 3H-retinol, is of a higher
molecular weight than the cellular binding proteins for retinol and retinoic acid, and is of
comparable size and presumably similar in its
properties to the 11-cis-retinal-binding protein
recently found in bovine retina.1
Although initial results with undialyzed preparations of the soluble proteins of brain and liver
suggested that these tissues might bind 3H-retinal
like retina, further experiments involving either
dialyzed tissue preparations or Sephadex G-100
column fractions clearly demonstrated the absence
of significant binding of 3H-retinal.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933075/ on 06/17/2017
Volume 16
Number 8
Reports 111
Unlike the binding proteins for retinol and
retinoic acid,2' 3 the binding protein for retinal
appears to be restricted to the eye. This indicates
that it is not an obligatory participant in the
sequential oxidation of retinol to retinal and
retinal to retinoic acid. It now seems likely that
the binding protein for retinal has a specific role
in, and perhaps limited to, the photoreceptor cells
of the retina.
Technical assistance by Brian Bowe and Lucille
Bredberg is gratefully acknowledged.
From the Department of Ophthalmology, University of Washington School of Medicine, Seattle.
This study was supported by United States Public
Health Service Grants EY 00343, EY 00529, and
EY 1730 from the National Eye Institute. Submitted for publication Feb. 23, 1977. Reprint
requests: Sidney Futterman, Ph.D., Department of
Ophthalmology, RJ-10, University of Washington
School of Medicine, Seattle, Wash. 98195.
Key words: retinal, 11-cis-retinal, 11-cts-retinalbinding protein, retina, vitamin A.
REFERENCES
1. Futterman, S., Saari, J. C , and Blair, S.:
Occurrence of a binding protein for 11-cisretinal in the retina, J. Biol. Chem. 252:3267,
1977.
2. Ong, D. E., and Chytil, F.: Retinoic acidbinding protein in rat tissues, J. Biol. Chem.
250:6113, 1975.
3. Ong, D. E., and Chytil, F.: Changes in levels
of cellular retinol- and retinoic acid-binding
proteins of liver and lung during perinatal
development of rat, Proc. Natl. Acad. Sci. 73:
3976, 1976.
4. Wiggert, B., and Chader, G.: A receptor for
retinol in the developing retina and pigment
epithelium, Exp. Eye Res. 21:143, 1975.
5. Saari, J. C , and Futterman, S.: Separable
binding proteins for retinoic acid and retinol
in bovine retina, Biochim. Biophys. Acta 444:
789, 1976.
6. Swanson, D., Futterman, S., and Saari, J. C :
Retinol- and retinoic acid-binding proteins:
occurrence in human retina and absence from
human cultured fibroblasts, INVEST. OPHTHAL-
MOL. 15:1017, 1976.
7. Lowry, O. H., Rosebrough, N. J., Farr, A. L.,
and Randall, R. J.: Protein measurement with
the Folin phenol reagent, J. Biol. Chem. 193:
265, 1951.
The in vitro frog pigment epithelial cell
hyperpolarization in response to light.
BURKS OAKLEY, II,
SHELDON
S.
ROY H.
MILLER, 0
STEINBERG,
AND SVEN
ERIK
NILSSON. 0 0
Pigment epithelial cell membrane potentials were
measured in an in vitro frog retina-pigment
epithelium-choroid preparation. Light stimuli
hyperpolarized the apical membrane of the pigment epithelium. This hyperpolarization was the
source of the c-wave of the electroretinogram.
Light stimuli also decreased retinal extracellular
potassium ion concentration, [K+]o, with the same
time course as the apical hyperpolarization. It is
suggested that the pigment epithelial hyperpolarization, which causes the c-wave, results directly
from the light-evoked decrease in retinal [K+]o.
The c-wave of the electroretinogram (ERG)
is thought to be produced by a hyperpolarization
of the pigment epithelial apical membrane.1 This
hyperpolarization must result from the direct
absorption of light by the photoreceptors, since
the only light-evoked responses in the isolated
pigment epithelium are fast photovoltages.2 There
is no anatomical evidence for synaptic or electrical interactions between the neural retina and
the pigment epithelium.3 It has been hypothesized
that the hyperpolarization of the apical membrane
is caused directly by a light-evoked decrease in
retinal extracellular potassium ion concentration,
[K+]o.4 This hypothesis is supported by the findings
that in the frog, the apical membrane has a large
relative potassium conductance5 and there is a
light-evoked decrease in [K+](, in the distal retina,
having the same time course as the c-wave.6
There are, however, several other ionic mechanisms which might also contribute to the generation of the apical membrane hyperpolarization.
The membrane potential could be altered by
changes in extracellular bicarbonate concentration,
since the membrane has a large relative bicarbonate conductance.5 The membrane potential
could also be affected by changes in the concentration of any substance that would alter the
relative conductances of potassium or bicarbonate
or alter the rate of an electrogenic pump on the
apical membrane.
In this paper, results of preliminary experiments are reported which support the hypothesis
that the light-evoked pigment epithelial hyperpolarization is the source of the c-wave and is
most likely caused by changes in [K+]o in the
distal retina. There do not seem to be any
changes in membrane conductance associated with
the hyperpolarization. During the course of these
experiments, a perfused frog retina-pigment epithelium-choroid preparation has been developed,
which has many advantages over an eyecup
preparation and should prove useful in the further
study of retinal and pigment epithelial potentials.
Methods. Large bullfrogs (Rana catesbeiana)
were used in these experiments. A procedure for
preparing a pigment epithelium-choroid preparation was described in detail previously.5 The
procedure used for the retina-pigment epitheliumchoroid preparation was similar, except that the
frogs were light-adapted before the dissection.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933075/ on 06/17/2017