FEBRUARY 1982 Vol. 22/2
Investigative Ophthalmology
& Visual Science
A Journal of Clinical and Basic Research
Articles
Distribution of melanosomes across the
retinal pigment epithelium of a hooded rat:
implications for light damage
W. L. Howell, L. M. Rapp,* and T. P. Williams
Distribution of melanosomes across the retinal pigment epithelium ofhoodedrats (Long-Evans)
is studied at the light microscopic and electron microscopic levels. This distribution is shown to
be nonuniform: more melanosomes exist in the periphery than elsewhere and, importantly,
there are very few melanosomes in a restricted area of the central portion of the superior
hemisphere compared with the corresponding part of the inferior hemisphere. The region with
fewest melanosomes is precisely the one that is highly susceptible to light damage. Because this
region is the same in both pigmented and albino eyes, the paucity of melanin in this region is
not the cause of its great sensitivity to light damage. Nor does light cause the nonuniform
distribution of melanin. A possible explanation, involving a proposed vestigial tapetum, is
given in order to explain the correlation of melanosome counts and sensitivity to light damage.
(INVEST OPHTHALMOL VIS SCI 22:139-144, 1982.)
Keywords: rat retina, retinal pigment epithelium, melanosome distribution, light
damage, photoreceptors, tapetum, environmental lighting
M
I elanin, an insoluble polymer produced
by the oxidation of tyrosine, 1 ' 2 is deposited in
macroscopic granules called melanosomes.
Tyrosinase activity is associated with melanin
From the Institute of Molecular Biophysics, Florida
State University, Tallahassee and * Department of
Ophthalmology, Baylor College of Medicine, Texas
Medical Center, Houston, Texas.
Supported by National Institutes of Health grant EY
02250 and the Department of Energy grant ORO
6021.
Submitted for publication June 15, 1980.
Reprint requests: Professor T. P. Williams, Institute of
Molecular Biophysics, Florida State University, Tallahassee, Fla. 32306.
synthesis, 2 and along with copious melanosomes, this activity is found in both the retinal pigment epithelium (RPE) and the choroid. 2 ' 3 The melanosomes in these structures
are densely packed and presumably act as a
"sun sink," the purpose of which seems to be
to prevent the reflection of light back through
the photoreceptors. This, in turn, reduces
spurious signals and permits good acuity.
The distribution of ocular melanin is not
uniform across the animal kingdom or even
within a given individual. Albino eyes, of
course, have no melanin. Even in heavily
pigmented eyes, there may be regions in the
RPE of low melanin concentration. For
0146-0404/82/020139+06$00.60/0 © 1982 Assoc. for Res. in Vis. and Ophthal., Inc.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
139
140
Hoivell et al.
example, in animals with tapeta, the RPE in
the tapetal area is almost devoid of melanin. 4
The obvious advantage of such an arrangement is that, if the tapetum is to be an effective mirror, the organism should not screen it
with a veil of melanin in the RPE.
In this article we report that there is a nonuniform distribution of melanosomes across
the RPE of the hooded rat (Long-Evans), as
previously suggested by La Vail.5 The rat has
no tapetum, 4 and consequently the reason for
the unusual distribution of melanin is not
clear. What is clear, however, is that the region with few melanosomes is the very area
that, an earlier paper 6 shows, is also highly
susceptible to light damage in this rat.
We also report here the distribution of
melanosomes in the RPE of rats raised in various photic environments, because we were
interested in the possibility that lighting
conditions might play a role in the melanosome distribution.
Materials and methods
Melanosome distribution. Male and female
Long-Evans rats were reared either in 12 hr cyclic
illumination of 5 to 10 lux or in completely dark
environments. Most, but not all, of the animals,
were sacrificed at 10 weeks of age for histological
observation. Several of the cyclic-light animals
(the ones not sacrificed at 10 weeks) were carried
over into light-damage experiments (vide infra).
The animals were perfused with 2% glutaraldehyde and 0.6% formaldehyde buffered to pH 7.35
with 0.1M sodium cacodylate. The right eyes were
removed and prepared for paraffin sectioning; the
left eyes were removed and fixed an additional 2
hr by being immersed in the perfusate, postfixed
in 1% O S O 4 for 2 hr, and then prepared for embedding in Araldite 506. Orientation of the eyeball
during processing was maintained with a suture
attached to extraocular muscle tissue on the temporal edge. Plastic sections 1 to 2 /xm thick were
prepared and stained with toluidine blue for qualitative examination of melanosome distribution by
light microscopy (LM). For observation of the
melanosomes by electron microscopy (EM), sections were made at 0.1 /xm in thickness and
stained with uranyl acetate and lead citrate for
examination in a Philips T.E. 201.
In order to actually count melanosomes at the
LM level; sections 0.5 /xm thick were made of the
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
Invest. Ophthalnwl. Vis. Sci.
February 1982
entire eyecup through the vertical meridian with a
10 mm glass-knife. A 2 mm micrometer was put
into the objective of the microscope, and calibration at 1000x magnification was used for counting
the melanosomes present at various intervals
along the RPE. The counting intervals (15 /xm in
length) were separated by 300 /xm, with the first
interval starting a distance 300 to 350 /xm from the
superior end (ora serrata). Counting continued
until a point 300 to 350 /xm from the opposing end
of the retina was reached. Melanosomes found
within the 15 /xm area were used for tabulating the
melanosome counts, and although the RPE contains both round and ovoid melanosomes, no distinction was made between these when counting. 5
In some instances, photographs were made of
the entire eyecup. Eyes were first fixed by immersion for 12 hr in a 1.25% glutaraldehyde and
3.2% formaldehyde solution buffered to pH 7.35
with 0.1M sodium cacodylate and 0.5% CaCl2.
They were then placed in buffer, and cornea, lens,
retina, and extraocular muscle tissue were removed. Photographs were made at low magnification with diffuse light passing through the eyecup
from behind. A large blood vessel running temporonasally was used to distinguish superiorinferior regions.
Retinal light damage. All rats were raised from
birth in a cyclic-light environment of 12 hr light,
12 hr dark, with a cage illumination of 5 to 10 lux.
Animals taken directly from this environment
served as controls. At the onset of the experiments
the rats ranged in age from 10 to 14 weeks. Both
male and female animals were used. To achieve
light damage at moderate intensities, dilation of
the pupils was necessary and was accomplished
with drops of 2% atropine sulfate.
The rats were exposed to constant light, 40 lux
for 6 days, in specially designed chambers where
the level of illumination in all directions varied by
no more than 20% from the mean value. These
chambers were constructed of sand-blasted plexiglass and were situated in a large wooden box with
white, inside walls. Fluorescent lamp fixtures
were mounted on the inside walls, floor, and ceiling of the box. Illumination was provided by "cool
white' fluorescent bulbs. Inside the cages, the rats
moved freely on a hardware cloth floor underneath
which a sheet of plexiglass was positioned at an
angle so that the animal's droppings were removed
from the rat's field of view. This arrangement kept
the field of view below the animal from becoming
dark with droppings. The plexiglass sheet and the
cage walls served as illuminated surfaces that
Volume 22
Number 2
Melanosomes and retinal light damage
141
Fig. 1. Top panels, Representative sections taken at equal distance {1.5 mm) on either side of
optic nerve head show less melanin in the superior retina (left) than in the inferior retina
(right). (X400.) Bottom panels: Electron micrographs of the superior and inferior retina RPE
cell in superior retina contains only two melanosomes (in this plane of section), whereas one in
the inferior contains about 20 times as many. (Araldite sections 0.1 ftm thick; X3000.)
completely surrounded the animal. Light intensity
in the chambers was controlled by metal screens
uiat were painted flat black and placed over the
fixtures. Illuminance was measured peqjendicularly to the walls, ceiling, and floor with a Tektronix J16 digital photometer and a J6501 illuminance probe. Fans circulated air into and out of
the box. This served to keep the temperature in
the cages within 2° C of room temperature and to
provide fresh air. Food and water were provided
ad lib. At the end of the exposure period, the rats
were dark-adapted for 18 to 24 hr. They were then
sacrificed and the eyes were enucleated.
Fixation and orientation of the eyecups were
carried out as described above. Eyecups, embedded in paraffin, were serially sectioned at 7 jam,
mounted on glass slides, and stained with hematoxylin-eosin. Measurements of outer nuclear
layer (ONL) thickness, which were used to indicate the number of photoreceptor cells present,
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
were made at a magnification of 400 X with an ocular micrometer calibrated in divisions of 4 fim.
Results
Examination of thick plastic sections taken
through the vertical meridian revealed a qualitative difference in the superior vs. inferior
RPE insofar as pigmentation was concerned.
The two panels at the top of Fig. 1 show this
at low magnification. It can be seen that the
melanosome concentration was greater in the
inferior hemisphere than in the superior.
However, these thick sections did not permit
a quantification of the melanosome concentration because the melanosome profiles severely overlapped each other. It was found
that the 0.5 /nm plastic sections permitted
clear counting of melanosomes, at leastin the
more central retinal regions, and such sec-
Invest. Ophthalmol. Vis. Sci.
February 1982
142 Howell et at.
4080 3400 2720 2040
INFERIOR
1360 680
0
680 1360 2040 2720 3400 4080
SUPERIOR
DISTANCE (fj.) FROM OPTIC NERVE HEAD
Fig. 2. Distribution of inelanosomes and ONL thickness across the pigmented rat retina.
Sections taken through the vertical meridian. Filled circles, Melanosomes; open circles, ONL
thickness; both from same (control, dark-reared) rat. Filled triangles, Melanosomes; open
triangles, ONL thickness; both from same (retina-damaged) animal. Note that light damage is
greatest in the region where melanosomes are fewest. Also, note that the retinas of both
control and light-exposed animals have very low melanosome counts in the central part of the
superior retina. Dashed lines show regions where melanosomes were too numerous to count
accurately.
tions are shown in the lower panels of Fig. 1.
Here the differences in melanosome distribution were striking.
A previous study6 on the distribution of
light damage revealed that the most severe
damage occurred in the superior retina, i.e.,
in a region 750 to 2400 /xm superior to the
optic nerve head. Fig. 2, which includes this
light-sensitive region, is a graphic representation of the ONL thickness and the tabulated melanosomes counts of a damaged and a
control retina. The correspondence between
these two measurements is clear: fewest
melanosomes were found in the region of
greatest damage. The damaged retina was
obtained from an animal that had been exposed to 40 lux lighting for 6 days; there was
more than 50% reduction of the ONL in the
region of most severe damage. Note, also,
that the distribution of melanosomes were
not caused by light damage because the control (dark-reared) animal (circles, Fig. 2)
showed the same distribution. The data of
Fig. 2 were obtained from just two animals.
Nevertheless, we have counted melanosomes
in more than 20 eyes and have measured
ONL thickness in more than 100 eyes. Aside
from slight individual differences, the distributions have always been as shown here.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
With photographic methods used on eyecup preparations, those areas which allowed
transmission of more light because they contained fewer melanosomes could be examined (Fig. 3). The regions of lower melanosome density were found in the superior half
of the eye, extending from nasal to temporal
ends. These findings corresponded fairly well
to the regions of low RPE melanosome concentration identified by histological sections.
Although qualitative in nature, Fig. 3 may be
helpful since it gives an overall view of the
melanosome distribution of the rat eyecup.
Discussion
This study has shown that fewer melanosomes exist in a central portion of the superior RPE of hooded rats. This has implication for the earlier report 6 that light damage
of photoreceptors was most severe in this
very region and has led us to consider reasons
for this correlation.
Superficially, it would seem that paucity of
melanosomes and susceptibility of damage
are causally related. Although we reject this
possibility below, it is instructive to consider
the evidence that appears to support it. For
example, Noell et al.7 found that pigmented
rats are very resistant to retinal light damage.
Volume 22
Number 2
However, when the pigmented rats' pupils
were chemically dilated, only about twice the
exposure needed for albinos was necessary
for damage. Thus pigmentation seems to protect against damage, but in this case, the
major protection lay in the pigmented iris not
in the RPE. In other light damage experiments LaVail5 found evidence that implicated melanin in the RPE as a protector and
suggested that the melanin may act as a
"free-radical sink." His suggestion stems
from the conclusions of McGinness and Proctor8 and perhaps could explain the protective
effect of melanin in the RPE if the mechanism of light damage were a free-radical
reaction (e.g., deriving from lipid peroxidation). Although we cannot exclude the possibility that melanosomes act as a free-radical
sink, we have now verified the result of Noell
et al.7 by showing that pigmented and albino
rats are different by only a factor of 2 in the
exposure time needed for criterion damage. 6
Since the pupils of our rats were dilated, this
means that melanin in RPE could contribute
only a limited amount (at most a factor of 2) of
protection. How does it offer this? We think
it is simply that the melanin prevents the
reflection of light back through the photoreceptors. Consider, for example, if (hypothetically) the reflection in albinos were total and
in the pigmented rats nil, the factor of 2
would be immediately realized. Even though
this idealized situation does not quite obtain,
describing the protective effect in these
terms emphasizes the minimal contribution
of the melanized RPE and shows that there is
no need to invoke a chemically active protective role such as a free-radical sink.
Now to the question of cause-and-effect
regarding melanosome distribution and light
damage. First, the light-damage did not
cause the nonuniform melanin distribution
(control animals, those raised in dim cyclic
light or total darkness, were identical to damaged ones). Second, albino rats, of course,
have no melanin, yet they suffer retinal damage in exactly the same, nonuniform way.6
Thus, although the distribution of melanosomes is highly correlated with the distribution of damage, the two must not be related
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
Melanosomes and retinal light damage
143
Fig. 3. Low-power micrograph of a pigmented rat
eyecup, with retina and extraocular muscle tissue
removed. The region of low pigmentation can be
visualized in the superior hemisphere (top half) by
passing diffuse light through the eyecup from behind. The bright spot in the center of the eyecup is
the optic disk, and it is prominent because, of
course, it is not a melanized structure.
in causal fashion. What other reason is there
for the correlation?
Tentatively, we propose that the region of
few melanosomes is a vestigial tapetum in the
rat. Only one rodent, Cuniculus paca, is
known to have a true tapetum. Walls4 indicates that this structure in C. paca subtends
nearly an entire hemisphere of the eye, and
significantly for our argument, this is the superior hemisphere. A prominent characteristic of tapeta is, as already mentioned, a scarcity of melanosomes in the associated RPE.
Hence, for reasons of its location in the superior hemisphere and its low melanosome
counts, we tender the proposition that this
region is a vestigial tapetum. Next, regarding
the possible consequence of this (if it is true):
Bernstein and Pease9 have suggested that the
tapetal area of the cat may be poorly vascularized by the choroid . . . perhaps, we suggest,
for the same reason that the area is poorly
melanized, that is, too many capillaries would
reduce tapetal reflection. If this is true for the
rat, then the photoreceptors and the RPE
cells of the tapetal area might be less well
Invest. Ophthalmol. Vis. Sci.
February 1982
144 Howell et al.
nourished as compared with their counterparts elsewhere in the eye. This, in turn,
might cause them to suffer disproportionately
when subjected to light stresses. It is known
that photoreceptors switch from predominantly aerobic metabolism to a mixture of
aerobic plus anaerobic metabolism when light
adaptation occurs. I0 Thus the availability of
glucose and oxygen could become crucial during the light stress, and poor blood supply
could readily compound the difficulties encountered during constant light.
We plan to examine directly the choroidal
circulation in the light-sensitive area with
angiography, but meanwhile we have made
two observations that lend support to the
above arguments: (1) perfusing animals with
toluidine blue stains the inferior hemisphere
more darkly than the superior and (2) unless
extra care is taken, the superior RPE and
choroid do notfixas well as the inferior regions. Both of these observations should be
pursued and quantified because they indicate
that perfusates infiltrate the superior hemisphere with more difficulty than the inferior
and they suggest that choroidal circulation
may, indeed, be poorer in the superior half of
the rat eye.
There are other anatomical peculiarities
associated with the area of high sensitivity to
light damage. Some of these are mentioned
in an earlier paper. 6 Another is the existence
of high ganglion cell counts in this region of
the superior retina. 11 In this regard, a similar
result was found in rabbit by Lawwill et al.12
These investigators discovered no correlation
of light damage with underlying pigmentation but a strong correlation with the position
of the visual streak—a region of high ganglion cell counts and acuity.
The results presented in Fig. 2 suggest
that there may be fewer melanosomes overall
in the light-damaged retina than in the control. Little attention was given to this because the traditional view is that melanization
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
of the RPE is a static, time-invariant condition. A very recent report 13 to the contrary
stimulates us to reinvestigate this matter.
REFERENCES
1. Mahler HR and Cordes EH: Metabolism of the
amino acids. In Biological Chemistry, New York,
1966, Harper & Row, Publishers, Inc., pp. 697-698.
2. Dryja TP, O'Neil-Dryja M, Pawelek JM, and Albert
DM: Demonstration of tyrosinase in the adult
bovine uveal tract and retinal pigment epithelium.
INVEST OPHTHALMOL Vis SCI 17:511, 1978.
3. Feeney Lynette: Lipofuscin and melanin of human
retinal pigment epithelium. INVEST OPHTHALMOL
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
VIS SCI 17:583, 1978.
Walls GL: In The Vertebrate Eye. New York, 1967,
Hafner Publishing Co., pp. 221, 231-233.
La Vail M: Eye Pigmentation and constant light
damage in the rat retina, /n The Effects of Constant
Light on Visual Processes, Williams TP and Baker
BN, editors. New York, 1980, Plenum Press, pp.
357-388.
Rapp LM and Williams TP: A parametric study of
retinal light-damage in albino and pigmented rats.
In The Effects of Constant Light on Visual Processes, Williams TP and Baker BN, editors. New York,
1980, Plenum Press, pp. 135-159.
Noell WK, Walker VS, Kang BS, and Berman S:
Retinal damage by light in rats. INVEST OPHTHALMOL 5:450, 1966.
McGinness J and Proctor P: The importance of the
fact that melanin is black. J Theor Biol 39:677, 1973.
Bernstein MH and Pease DC: Electron microscopy
of the tapetum lucidum of the cat. J Biophys Biochem Cytol 5:35, 1959.
Sickel W: Retinal Metabolism in dark and light. In
Handbook of Sensory Physiology, Vol. VII/2,
Fuortes MGF, editor. Berlin, 1972, SpringerVerlag, pp. 667-727.
Fukuda Y: A three group classification of rat retinal
ganglion cells; histological and physiological studies.
Brain Res 19:327, 1977.
Lawwill T, Crockett RS, and Currier G: The nature
of chronic light damage to the retina. In The Effects
of Constant Light on Visual Processes, Williams TP
and Baker BN, editors. New York, 1980, Plenum
Press, pp. 161-177.
Tabor GA, Fisher SK, and Anderson DH: Evidence
for melanin turnover in mammalian retinal pigment
epithelium. INVEST OPHTHALMOL VIS SCI 19 (ARVO
Suppl.):163, 1980.