THE JOURNAL OF COMPARATIVE NEUROLOGY 405:394–405 (1999)

Retinal Mitosis is Regulated by Dopa,

a Melanin Precursor That May Influence
the Time at Which Cells Exit
the Cell Cycle: Analysis of Patterns
of Cell Production in Pigmented
and Albino Retinae
MARIA ILIA AND GLEN JEFFERY*
University College London, Institute of Ophthalmology, London, United Kingdom

ABSTRACT
A melanin-associated agent seems to play a role in regulating retinal development. When
absent, diverse deficits occur. There is evidence that this agent regulates patterns of mitosis.
This study examines retinal development in pigmented and albino rats to identify the
regulating agent and its mode of action. Throughout neurogenesis, many more mitotic profiles
are found in albinos than pigmented animals. At the peak of retinal neurogenesis, approxi-
mately 50% more mitotic profiles are found in albinos than in matched pigmented animals,
resulting in abnormal retinal thickening. Concurrently, increasing numbers of pyknotic nuclei
are identified, such that later in development retinal thickness normalises. However, the
crude centre-to-periphery pattern of cell production is preserved. Abnormal cell proliferation
is found in a range of albino rat strains, but it is not present in their brains, confirming that
the abnormality is ocular and melanin related. Dopa is a critical element in initial stages of
melanin synthesis and is present in abnormally low levels in developing albino retinae.
Furthermore, it is an antimitotic agent. Addition of dopa to albino eyes in vitro normalises
patterns of cell production. These results are consistent with the hypothesis that dopa is a
major regulator of retinal cell production and that it influences the capacity of cells to exit the
cell cycle. J. Comp. Neurol. 405:394–405, 1999. r 1999 Wiley-Liss, Inc.

Indexing terms: development; cell death; retinal pigment epithelium

Elements associated with melanin in the mammalian
retinal pigment epithelium (RPE) are critical regulators
for normal development of the neural retina. When mela-
nin is absent or significantly reduced, the central retina
remains underdeveloped (Elschnig, 1913; Stone et al.,
1978; Oyster et al., 1987; Jeffery and Kinsella, 1992), and
many ganglion cells located in temporal regions that
should remain uncrossed at the optic chiasm cross the
chiasmatic midline inappropriately to innervate the contra-
lateral hemisphere (Lund, 1965; Cooper and Pettigrew,
1979; Leventhal, 1982). Furthermore, there is a rod deficit
across the retina (Jeffery et al., 1994a). Each of these
deficits are regulated by the tyrosinase gene, because its
introduction in transgenic albino mice and rabbits rescues
the deficits (Jeffery et al., 1994b, 1997).
Few studies have compared retinal development be-
tween pigmented and albino mammals to determine how
these abnormalities arise and to attempt to identify the
critical agent that regulates the normal development of
these features. However, it is known that there is a
temporal delay in the development of some events
in the visual system of hypopigmented animals. The
centre-to-periphery pattern of retinal maturation is de-
layed in albinos (Webster and Rowe, 1991). This includes
patterns of cell production in the ganglion cell layer
(Ilia and Jeffery, 1996). There is also evidence for a
delay in the retinal innervation of primary visual struc-
Grant sponsor: Medical Research Council; Grant sponsor: Wellcome
Trust.
*Correspondence to: Glen Jeffery, Institute of Ophthalmology, University
College London, 11–43 Bath Street, London EC1V 9EL, United Kingdom.
E-mail: g.jeffery@ucl.ac.uk
Received 10 March 1998; Revised 14 October 1998; Accepted 20 October
1998

r 1999 WILEY-LISS, INC.

RETINAL MITOSIS IN PIGMENTED AND ALBINO RETINAE
tures in the brains of these animals (Berman and Payne,
1985).
Clearly, an important regulator of normal development
is missing in hypopigmentation. The aim of this study is to
analyse patterns of retinal cell addition and cell death in
pigmented and albino animals and to relate these data to
elements in the synthetic pathway of melanin. The cascade
of reactions resulting in melanin synthesis is initiated
when tyrosinase converts tyrosine to dopa and then dopa
to dopaquinone (Garcia et al., 1979). Consequently, it is
possible that dopa concentrations might be abnormally
low in albino retinae. This possibility could be very signifi-
cant, because dopa is a biologically active agent that is
known to be an important moderator of the cell cycle
(Wick, 1977, 1980; Akeo et al., 1994). In this study, dopa is
measured in developing pigmented and albino animals,
and its presence augmented in vitro, to determine whether
it can influence any abnormalities found in developing
albino retinae in vivo.
MATERIALS AND METHODS
Animals and histology: In vivo experiments
All procedures on animals were undertaken under United
Kingdom Home Office guidelines. To produce litters con-
taining both pigmented and albino phenotypes, hence,
reducing the variations in maturation between the strains
at defined postconception days (PCD), female rats were
bred to produce litters containing both pigmented and
albino offspring. Pigmented DA rats (outbred) were crossed
with albino Lewis rats (Inbred; Both Olac, UK). Then the
male DAL F1 offspring were crossed with female Sprague
Dawley rats (SD, Outbred; Olac, United Kingdom), which
resulted in the production of mixed litters. The progeny
produced by this back-crossing are heterozygotes, and as
such it is possible that the pigmented animals could
display some features found in albinism (Leventhal et al.,
1985). However, in studies not reported here, the pig-
mented progeny were shown to display none of these
features (Ilia, 1998).
Males were put with females from 4:00 PM until 9:00 AM
the following morning. Day 0 was taken to be the first 24
hours after the removal of the female if a plug was found.
Parturition occurred on approximately PCD 22. Litters
were harvested for in vivo analysis of patterns of mitosis at
PCD 12 (n ⫽ 2), 13 (n ⫽ 1), 14 (n ⫽ 1), 15 (n ⫽ 1), 17 (n ⫽ 2),
19 (n ⫽ 2), 21 (n ⫽ 1), 23 (n ⫽ 1), 25 (n ⫽ 2), 28 (n ⫽ 2), 30
(n ⫽ 2), and 33 (n ⫽ 1), where n is the number of litters
collected. Specific analyses were confined to one litter at
each stage. Because retinal melanin was not apparent in
the rat before PCD 13, it was not possible to distinguish
pigmented from albino fetuses before this stage. Conse-
quently, for PCD 12 animals, the two strains were bred
independently, and comparisons were confined to litters
from mothers found to plug at the same time.
Prenatal fetuses were obtained by injecting mothers
with a lethal dose of sodium pentobarbitone (1.0 ml/kg).
Postnatal pups were killed the same way. All animals were
fixed in Carnoy’s fluid (Bancroft and Stevens, 1990) for 2–4
hours, depending on the stage of development. The heads
of PCD 12 and the PCD 13 animals were processed whole.
All others were decapitated, and the heads were placed in
fixative. The left eye was removed after an orienting mark
was made on its dorsal aspect. The anterior chamber and
lens were removed in animals older than PCD 15. Eyes
395
were then dehydrated and embedded in paraffin wax,
which were serially sectioned at 3 µm in the horizontal
plane. Sections were mounted onto gelatinised slides,
dewaxed, and stained with cresyl violet before being
dehydrated and cover-slipped with DPX.
Because wax histology can generate variable results
when used on retinae, strict criteria were applied to
determine which eyes in each group were used for quanti-
tative analysis. Eyes that were distorted or appeared to
have shrunk abnormally were not included in the analysis.
With the exception of PCD 17 albinos, each group used for
quantitative analysis at each stage of development contained
six eyes, i.e., three pigmented and three albino. Additional
material was not obtained from other litters to increase
the numbers because this may have resulted in increased
variability due to differences in developmental stages.
To determine whether the albinos progeny of the DA ⫻
SD crosses were tyrosinase negative, the dopa-oxidase test
(Chayen et al., 1969) was used. This test is a standard
histochemical test for tyrosinase activity undertaken on
skin sections. However, all tests for tyrosinase activity
actually reveal the presence of dopa, which is associated
with tyrosinase, rather than tyrosinase itself. The skin
sections came from mature animals killed as above for the
removal of fetuses. They were fixed in 10% formalin and
cut at 12 µm on a cryostat.
Rat strains
To determine whether any differences in patterns of cell
production were phenotypic or tissue specific, two controls
were undertaken. First, mitotic figures in the retina were
counted in the progeny of a range of pigmented and albino
rat phenotypes. Along with the standard cross (DA ⫻ SD)
used throughout the experiments, two additional groups of
animals were crossed in a similar method to produce
litters containing both pigmented and albino animals. The
first of these came from crossing Lister Hooded (Outbred)
male pigmented rats with Wistar female albinos (Out-
bred). The male offspring were crossed with female albino
Wistar rats. The second involved crossing CD (Hsd-
Sprague Dawley-CD; Outbred) male pigmented rats with
PVG female albinos. The male offspring were crossed with
albino PVG rats (Inbred; all animals supplied by Olac).
Procedure of mating were as above. One litter from both
additional groups were harvested at PCD 17, 19, and 28.
Second, to determine whether any differences in retinal
mitosis between pigmented and albino animals were spe-
cific to the eye, and hence, potentially related to retinal
melanin, or were present throughout the central nervous
system, mitotic profiles were also counted in the brains.
Foetuses from five litters of DA ⫻ SD crosses were
removed at PCD 19 and fixed as above. The brains were
dissected free and embedded in wax. These brains were cut
horizontally at 5 µm. A continuous series was collected.
Sections were dewaxed and stained with cresyl violet.
Mitotic figures were counted unilaterally in one in three
sections through the depth of the lateral ventricle. The
total length of the ventricular margins along which mitotic
figures were counted was measured. Counts were made
over similar lengths in all animals.
High performance liquid chromatography
analysis of retinal dopa concentrations
Dopa is present in the early stages of melanin synthesis.
Tyrosinase converts tyrosine to dopa and then dopa to
396
dopaquinone. Consequently, it is probable that dopa levels
are lower in albinos than pigmented animals. High perfor-
mance liquid chromatography (HPLC) was used to deter-
mine relative dopa levels between developing pigmented
and albino eyes at fixed stages of development. These came
from crossing DA ⫻ SD rats. Tissue was taken at PCD 15
(n ⫽ 30), 17 (n ⫽ 30), 19 (n ⫽ 30), 21 (n ⫽ 20), 30 (n ⫽ 8), 35
(n ⫽ 8), and adult (n ⫽ 2), where n is the number of eyes
collected from each strain. At younger stages (PCD 15–19),
eyes within each group were pooled from three litters to
produce a significant tissue volume. The eyes were re-
moved after deep anaesthesia as above, and when devel-
oped, all muscle and attached tissue was removed from the
eye so that its surface appeared clean, before being placed
in liquid nitrogen. They were subsequently sonicated for
approximately 45 minutes and centrifuged. The phenolic
compounds were purified by using the method of alumina
extraction followed by acid elution described by Ofori et al.
(1986). The HPLC detector was operated at 0.8 V. Stan-
dard solutions of dopa in water were made at a concentra-
tion of 0.256 mg/ml, and dilutions were prepared at 1/100,
1/1,000, and 1/10,000.
Organ cultures
Dopa is an antimitotic agent (Wick, 1977) that is known
to influence retinal cell cycle rates (Akeo et al., 1994). To
determine whether any differences in patterns of mitosis
between pigmented and albino animals are dopa medi-
ated, embryonic eyes from mixed (DA ⫻ SD) litters were
placed in organ culture to which dopa was added. Litters
were removed at PCD 19 as above. Fetal eyes were
removed and divided into pigmented and albino, and
equally into experimental and control groups. The control
solution consisted of 0.2% sodium bicarbonate, 0.18%
sucrose, and 0.89% Ames medium (Sigma, UK), saturated
continuously with 95% oxygen and 5% carbon dioxide.
L-Dopa (Sigma) was added to this mixture in the experimen-
tal solution. The concentration of dopa was varied in
separate experiments. Experiments were run at the follow-
ing dopa concentrations: 0.1 mM, 0.5 mM, 1.0 mM, 2.0
mM, 2.5 mM, 3.0 mM, and 4.0 mM. One litter was used at
each concentration, except 2.5 mM for which three litters
were used. Both solutions were incubated at 34°C for 7
hours. The eyes were then fixed as above, and the anterior
chamber and lens were removed. They were then dehy-
drated, embedded in paraffin wax, and processed as for in
vivo preparations.
Analysis
Mitotic and pyknotic retinal profiles in vitro and in
vivo. Every 8th or 16th retinal section was analysed, and
the number of mitotic and pyknotic profiles counted at
⫻400 or ⫻600, depending on retinal size. For animals
between PCD 12 and 15, 15 sections were analysed per
eye, whereas for older animals, 20 sections per eye were
examined. The proportion of the retina examined varied
between animals depending on eye size. For the in vivo
preparations, the length of the retina was measured in the
central region at each PCD by drawing the sections with
the aid of a drawing tube at magnifications of ⫻40, ⫻100,
or ⫻200 depending on retinal size. This approach was used
to locate the section corresponding to the horizontal merid-
ian. Retinal thickness was measured in this section. Three
perpendicular measurements were taken from the edge of
the RPE to the vitreal margin in a region spanning the
posterior pole of the eye. Their mean was calculated. If the
M. ILIA AND G. JEFFERY
meridional section contained the optic nerve head (ONH),
an adjacent section was used.
The location of mitotic and pyknotic profiles in the retina
were plotted onto outline drawings of the retina produced
with the aid of a drawing tube attached to the microscope
at ⫻400 or ⫻600 according to retinal size. To create retinal
maps on which the relative location of profiles could be
marked, a thread was laid over the outline drawings and
the location of these marked on it. The threads were
straightened and aligned with those from sections taken
from more dorsal and ventral locations. This representa-
tion was transferred to graph paper, and the location of
cells was digitised by using a graphics tablet connected to a
personal computer. Measurements of retinal size were
generated from these representations.
RESULTS
The dopa-oxidase test for tyrosinase activity was posi-
tive for the pigmented animals but negative for the
albinos, confirming that the albinos used were tyrosinase
negative.
Patterns of retinal mitosis in vivo
Cell production in the RPE was complete before that in
the neural retina. Mitotic profiles in the RPE could be
identified from PCD 12–15 (Fig. 1). None were identified at
or after PCD 17. In both phenotypes, the maximal number
of mitotic profiles was found at PCD 14 (Fig. 2). At each
stage, more profiles were consistently identified in the
albinos. Between PCD 13 and 14, this difference was
between 40 and 45%. To determine whether there was a
significant effect on mitotic profile numbers between phe-
notypes across the age range examined, a simple factorial
(two-way interaction) analysis of variance (ANOVA) was
applied to the data. This analysis revealed a significant
difference (F ⫽ 0.000, P ⬍ 0.01). This difference was not
the result of pigment granules obscuring mitotic profiles in
pigmented animals, because granule density was not high
enough to hide them. Even by PCD 17, melanin content of
the RPE is still well below that found in the adult. Because
of the relative sparsity of mitotic profiles in the RPE, a
meaningful picture of their distribution could not be
obtained from plots derived from any single retina. Conse-
quently, they have not been presented here. However, by
examining all of the retinae, it was clear that in pigmented
animals there was a crude centre-to-periphery gradient
with mitotic figures mainly confined to peripheral regions
by PCD 15. This finding was not the case in albinos, in
which mitotic figures in the RPE remained scattered at all
stages. No pyknotic figures were seen in the RPE at any
stage in either phenotype.
Mitotic profiles could be identified in the neural retina of
both pigmented and albino animals from PCD 12 to 30
(Fig. 1). There was no obvious difference in the period over
which they were present between the pigmentation pheno-
types, or the time of peak mitotic activity. But, as with the
RPE, there were marked differences in their number
between pigmented and albino animals.
In both phenotypes, the number of mitotic profiles
increased gradually from PCD 12 to 19, with a sharp peak
in their number at PCD 23 (Fig. 3). From this point, the
number of mitotic profiles declined steeply. Mitosis ap-
peared to have ceased between PCD 30 and 33, as few
profiles were found at PCD 30 and none at PCD 33. From
PCD 17 to 28, there were many more profiles in albinos
RETINAL MITOSIS IN PIGMENTED AND ALBINO RETINAE
Fig. 1. A: Low-power photomicrograph of a postconception day–14
pigmented eye, which has been sectioned horizontally. Temporal is
toward the lower half of the section. B: High-power micrograph of the
encircled region shown in A. Mitotic profiles are present in both the
neural retina (NR, arrows pointing left) and the retinal pigment
epithelium (RPE, arrow pointing right). Scale bars ⫽ 250 µm in A, 25
µm in B.
compared with their pigmented litter mates. Mitotic pro-
files were counted in the section corresponding to the
horizontal meridian (Fig. 3) and in a series of sections
throughout each retina (Fig. 4). Between PCD 19 and 23,
which is around the period of peak of mitotic activity in
both phenotypes, the magnitude of this difference was of
the order of 30–65% (Figs. 3, 4). However, from the more
detailed sampling shown in Figure 4, it is clear that the
differences between the groups are larger between PCD 17
397
Fig. 2. The number of mitotic profiles in the retinal pigment
epithelium (RPE) for postconception days (PCDs) 12–15. Open bars
represent data of the mean number of mitotic profiles obtained from
three albinos, whereas filled bars represent data obtained from three
pigmented animals sampled throughout the retina. Standard devia-
tions are given for each group. There were many more mitotic profiles
in the albinos than the pigmented animals. These differences were
statistically significant. In both groups, there was a progressive
increase in the number of mitotic profiles up to PCD 14 at which time
they reach a maximum, whilst after PCD 14, the number of mitotic
profiles declined steeply. In each case, data were derived from three
eyes from three separate animals in each group.
Fig. 3. The number of mitotic profiles in the neural retina in the
section corresponding to the horizontal meridian from postconception
days (PCDs) 12 to 30. In each case, they represent data of the mean
number of mitotic profiles obtained from three albino animals (open
circles) and three pigmented animals (filled circles). Standard devia-
tions are given for each group. From PCD 14 to 28, there were many
more mitotic profiles in the albinos compared with the pigmented
animals. In both groups, the number of mitotic profiles increased
gradually from PCD 12 to 19, with a sharp peak in their number at
PCD 23. At this point the difference between the groups was approxi-
mately 40%. From this point, the number of mitotic profiles was
steeply reduced. By PCD 30–33, mitosis appeared to be ceased as very
few mitotic figures were identified. In each case, data were derived
from three eyes from three separate animals in each group. B, day of
birth.
and 19, being approximately 50% greater in the albinos. To
determine whether there was a significant difference in
mitotic numbers between the two phenotypes across the
range examined (PCD 12–33), a simple factorial (two-way
398
Fig. 4. The number of mitotic profiles in the neural retina of
pigmented and albino animals from detailed sampling of sections
throughout the retina at postconception days (PCDs) 12–30. Open
bars represent data of the mean number of mitotic profiles obtained
from three albinos, whereas filled bars represent data obtained from
three pigmented animals. There were many more mitotic profiles in
the albinos compared with pigmented animals. Standard deviations
are given for each group. These differences were statistically signifi-
cant. The number of mitotic profiles followed a similar pattern as
shown in Figure 3; however, it is more clear that the difference in the
number of mitotic profiles between the groups were greater between
PCD 17 and 19, being approximately 50%. In each case, data were
derived from three eyes from three separate animals in each group. B,
day of birth.
interaction) ANOVA was applied, which revealed a signifi-
cant result (F ⫽ 0.000, P ⬍ 0.01). The vast majority of
mitotic profiles were located in the outer retina adjacent to
the RPE. However, they were rarely encountered at other
locations (Robinson et al., 1985). There were no phenotypic
differences in respect of this feature. At each stage exam-
ined, the retinae from both groups were of comparable
sizes.
Retinal mitosis follows a rough centre-to-periphery gra-
dient focussed around the optic nerve head (Young, 1983;
Reese et al., 1996). If there are more mitotic profiles in
albinos, are they still constrained by this pattern? Mitotic
figures were distributed throughout the entire outer reti-
nal surface until approximately PCD 21 in both pigmented
and albino rats. From this stage onward, a crude cold spot
developed around the optic nerve head where mitotic
profile density declined. This cold spot expanded at progres-
sive stages of development. Although more mitotic figures
were apparent in albinos, there were no obvious differ-
ences in their distribution with respect to the development
of this senescent region. But from PCD 28 onward, more
mitotic profiles were found in albinos at locations closer to
the nerve head than in their pigmented litter mates (Fig.
5), consistent with the notion that there is a temporal lag
at these stages in the centre-to-periphery pattern of cell
production in albinos.
If there are more mitotic profiles in albinos, and hence,
possibly more cells, is this reflected by an increase in
retinal size? There were no obvious differences in retinal
area between pigmented and albino groups at any stage.
However, albino retinae were consistently thicker than
those from their pigmented litter mates from PCD 17 to 23
M. ILIA AND G. JEFFERY
(Fig. 6). These differences mirrored those found in the
number of mitotic figures (Figs. 3, 4). In both groups,
retinal thickness increased until PCD 23 and then de-
clined gradually to PCD 33. At PCD 23, when retinal
mitosis peaks, differences in retinal thickness are maximal
at 0.07 mm, which represents 20% difference between the
phenotypes. Between PCD 17 and 19, the difference is
approximately 15–20%. There were no differences in the
length of the meridional sections used to estimate retinal
thickness between the two animal types. To determine the
statistical significance of these thickness differences over
the age range that was examined (PCD 12–33), a simple
factorial (two-way interaction) ANOVA was applied to the
data. This analysis proved to be significant (F ⫽ 0.001, P ⬍
0.01).
If differences in retinal thickness decline, could this
decline be due to an excessive wave of cell death, reducing
exuberant cell numbers generated in the albino? Pyknotic
nuclei were identified from PCD 14 onward throughout the
depth of the neural retina of both phenotypes. From PCD
17 to 28, there were consistently more pyknotic nuclei in
the albino retinae compared with those of pigmented
animals (Fig. 7). From PCD 17 to 19, there were approxi-
mately 40% more pyknotic profiles in the albinos. At PCD
23, when the greatest number of pyknotic nuclei were
found in each phenotype, the difference in their number
between the groups was 30%. At this stage of development,
the ganglion cell layer has differentiated, but the other
layers have not. The majority of pyknotic profiles at this
stage were located in the ganglion cell layer. The number
of pyknotic nuclei and the difference in their population
between the two groups declined gradually from PCD 25 to
33. There were no obvious differences in the distribution of
pyknotic figures between the groups at the stages exam-
ined. A simple factorial (two-way interactive) ANOVA
applied across the age range examined (PCD 14–33)
revealed that these differences were significant (F ⫽ 0.000,
P ⬍ 0.01).
It is possible that differences in levels of retinal mitosis
between pigmented and albino animals were not eye
specific, and hence related to melanin, but were common
throughout the CNS. Analysis of mitotic numbers at the
ventricular zone in the brains of the two animal types
failed to reveal any significant differences in the number of
mitotic figures (t ⫽ 0.642, not significant) or their distribu-
tion in these central regions. Consequently, differences in
levels of mitosis were specific to the retina.
It is possible that differences in the levels of mitosis were
strain specific and independent of melanin production. To
determine whether this was correct, patterns of retinal
mitosis were examined at PCDs 17, 19, and 28 in a range of
pigmented and albino animals crossed to produce mixed
litters. Analysis of all three animal type crosses (SD x DA,
LH x Wistar, and CD x PVG) at each stage showed that in
every case albino progeny had a significantly higher
numbers of mitotic profiles than their pigmented litter
mates (Fig. 8). Consequently, the differences found in the
number of mitotic figures between the two phenotypes are
unlikely to be the result of differences that are indepen-
dent of melanin.
Relative concentration of dopa as
determined with HPLC
Dopa was present in both pigmented and albino strains
when measured with HPLC at a retention time of 6.8
RETINAL MITOSIS IN PIGMENTED AND ALBINO RETINAE
Fig. 5. Retinal maps reconstructed from sections showing the
distribution of mitotic profiles at postconception days (PCDs) 25 and
28. There was no obvious difference in the distribution of mitotic
profiles between the groups. However, the density of mitotic figures
was lower around the optic nerve head at PCD 28, resulting in the
minutes. This was confirmed by comparison with standard
solutions and by spiking the column with dopa, which
elevated the peak identified as dopa in the samples. At
PCD 15, dopa concentrations in both strains were at trace
level, around 16 ng/ml. However, at PCD 17 levels in
pigmented animals had increased to 650 ng/ml, whereas in
albinos they were only 25 ng/ml. Large differences in dopa
concentrations between the phenotypes were a consistent
feature of samples taken at later stages. At PCD19, which
is around the time of peak rod production, concentrations
in pigmented eyes were 950 ng/ml, whereas in albinos they
399
formation of a crude mitotic cold spot. The cold spot was consistently
larger in pigmented (P) than albino (A) animals from PCD 28 onward,
consistent with the notion of there being a delay in centre-to-periphery
gradient in albinos. ● ⫽ optic nerve head. In each case, the left eye is
shown. Temporal is to the right and dorsal up. Scale bars ⫽ 0.1 mm.
were only 298 ng/ml. At PCD 35, dopa concentrations in
pigmented animals were 145 ng/ml and 120 ng/ml in
albinos. In adult eyes dopa levels were 430 ng/ml in
pigmented and 120 ng/ml in albinos. Although these
differences are accurate measures of relative dopa concen-
trations, revealing major deficits in albinos, no inference
can be made for measurements between age groups,
because tissue volumes could not be standardised across
them. Consequently, these data have not been presented
graphically. Despite this, it is clear that from PCD 17 to 19
when significant differences exit in the magnitude of
400
Fig. 6. Retinal thickness from postconception day (PCD) 12 to 33.
In each case, they represent data of the mean number of mitotic
profiles obtained from three albino animals (open circles) and three
pigmented animals (solid circles). Standard deviations are given
for each group. From PCD 17 to 23, there were clear differences
between the groups with the retina being thicker in the albinos than
in the pigmented animals. In both groups, retinal thickness reached
its maximum at PCD 23, which is consistent with the peaks in Fig-
ures 3 and 4. At this point, the difference in retinal thickness
between the groups is approximately 20%. Retinal thickness
in both groups declined progressively from PCD 23 to 33. These
differences were statistically different. In each case, data were derived
from three eyes from three separate animals in each group. B, day of
birth.
Fig. 7. The number of pyknotic nuclei in the neural retina of both
pigmented (filled bars) and albino (open bars) animals from postconcep-
tion day (PCD) 14 to 33. From PCD 17 to 28, there were many more
pyknotic profiles in the albinos compared with the pigmented retinae.
In both groups, the number of pyknotic nuclei reached its maximum at
PCD 23, with a difference between the strains at this point approxi-
mately up to 30%. From PCD 25 to 33, the number of dying cells as
well as the difference in their population in both groups declined
gradually. These differences were statistically significant. In each
case, data were derived from three eyes from three separate animals
in each group. B, day of birth.
retinal mitosis between the two phenotypes, dopa concen-
trations in albino retina are approximately 20% of those
found in normally pigmented animals, and even at matu-
rity, they never reach normal levels.
M. ILIA AND G. JEFFERY
Fig. 8. The number of mitotic profiles in retinae of different
pigmented and albino animals at postconception days (PCDs) 17, 19,
and 28. Animals were crossed to produce litters containing both
phenotypes (pigmented: DA, PVG, LH; albino: SD, CD, W). In each
case, there are many more mitotic figures in albino retinae than in
their pigmented litter mates. The average percentage increase in the
number of mitotic profiles for the SD ⫻ DA group over the three stages
examined was approximately 79%, whereas for the LH ⫻ W cross, the
average increase was approximately 31%. For the CD ⫻ PVG crosses,
the average increase over the stages was 60%. When analysed
statistically with an independent (two-tailed) t-test, all of the compari-
sons at each stage were statistically significant at P ⬍ 0.01. In each
case, data were derived from three eyes from three separate animals
in each group.
Patterns of retinal mitosis
in vitro: Organ cultures
Dopa is a critical element in the early stages of the
melanin synthesis. Furthermore, it is a known antimitotic
agent (Wick, 1977) and has been demonstrated to influence
the pace of the retinal cell cycle (Akeo et al., 1989, 1994). As
ocular dopa levels have been shown to be abnormally low
in albinos, it is possible that this relates to the increased
mitotic/pyknotic activity found in these animals. To test
this hypothesis, PCD 19 retinae were maintained in organ
culture to which dopa was added to determine whether it
was possible to regulate patterns of mitosis with this
agent.
Mitotic and pyknotic profiles could be identified in the
retinae used in both control and experimental organ
cultures. In each control group, many more mitotic and
pyknotic profiles were present in albinos compared with
the pigmented eyes, in a proportion similar to that found
in the in vivo preparations (Fig. 9). The number of mitotic
and pyknotic figures in the experimental group, which
contained dopa, were significantly lower than in the
 Fig. 9. The number of mitotic (left side) and pyknotic (right side)
profiles in pigmented (filled bars) and albino (open bars) eyes from
sampled sections of postconception day–19 animals maintained in
organ cultures for 7 hours. In the experimental group, dopa was added
at three separate concentrations, 0.5 mM (top row), 1.0 mM (middle
row), and 2.5 mM (lower row). The number of mitotic and pyknotic
profiles in control groups were similar to those found in vitro.
However, addition of dopa significantly reduced the number of mitotic
profiles in both types of animal. Over this concentration range, the
reduction in pigmented eyes was dose dependent, whereas in albinos,
it was dose independent. There was also a significant reduction in
pyknotic profiles in the experimental groups, which was mainly due to
reduced cell death in the ganglion cell layer, which at this stage had
just differentiated. Tables 1 and 2 show the statistical significance of
these results.
402 M. ILIA AND G. JEFFERY

TABLE 1. Statistical Significance of the Number of Mitotic Profiles Table 1). There was also a further reduction in the number
For Control and Experimental Groups Between
the Pigmentation Phenotypes at Different Dopa Concentrations1
of mitotic profiles in the pigmented animals. At this
concentration, unlike the 0.5 mM, the difference between
Mitosis
the experimental pigmented and albino groups was not
PCAC PEAE PCPE ACAE significant, because the number of mitotic profiles in the
[dopa] S S S S pigmented eyes fell to a level comparable to that in the
(mM) 0t0 (%) 0t0 (%) 0t0 (%) 0t0 (%) albinos. The reduction in the number of pyknotic profiles
0.1 0.000 1 0.001 1 0.437 NS 0.328 NS was significant for both pigmented and albino groups
0.5 0.000 1 0.001 1 0.566 NS 0.000 1 (Table 2) and was roughly similar to that seen at the 0.5
1.0 0.000 1 0.444 NS 0.015 5 0.000 1
2.0 0.000 1 0.593 NS 0.000 1 0.000 1 mM concentration.
2.5 0.000 1 0.154 NS 0.007 1 0.000 1 At the higher concentration of 2.0 mM, patterns of
1Differences in the number of mitotic profiles between control animals (P A ) were
C C
mitosis and pyknosis were similar to those observed at 1.0
significant at all seven dopa concentrations, with the albinos having many more mitotic mM (Fig. 9). That is to say, there was a significant
profiles than their pigmented litter mates. Mitotic profiles in experimental albinos were
fewer than their control group (ACAE ). This difference was significant except at the reduction in mitotic activity in the experimental albino
lowest concentration. There was also a reduction mitotic profile numbers in experimen- retinae compared with their control group (Table 1). There
tal pigmented animals (PCPE ), which was significant except at 0.1 and at 0.5 mM.
Differences in the mitotic profile numbers between the two experimental groups (PEAE ) was also a significant reduction in the number of mitotic
were not significant apart from at 0.1 mM and at 0.5 mM. P, pigmented; A, albino; C, profiles in the pigmented. At this concentration, like the
control; and E, experimental. 0 t 0 , t value in t test; S, level of significance; NS, not
significant. 0.1 mM, the difference between the experimental pig-
mented and albino groups was not significant. The reduc-
TABLE 2. Statistical Significance of Pyknotic Profile Numbers tion in the number of the pyknotic profiles was significant
For Control and Experimental Groups Between Pigmentation Phenotypes for both pigmented and albino groups (Table 2).
at Different Dopa Concentrations1 When the dopa concentration was increased to 2.5 mM,
Pyknosis the number of mitotic profiles in the albino and pigmented
PCAC PEAE PCPE ACAE experimental groups were again significantly reduced
compared with the control groups (Fig. 9; Table 1). As with
[dopa] S S S S
(mM) 0t0 (%) 0t0 (%) 0t0 (%) 0t0 (%)
the 1.0 and 2.0 mM concentrations, there was not a
significant difference between the two experimental groups
0.1 0.000 1 0.000 1 0.327 NS 0.439 NS
0.5 0.000 1 0.000 1 0.000 1 0.000 1
in terms of the number of mitotic figures. Reductions in the
1.0 0.000 1 0.001 1 0.000 1 0.000 1 number of pyknotic profiles were also significantly differ-
2.0 0.000 1 0.394 NS 0.000 1 0.000 1
2.5 0.000 1 0.000 1 0.000 1 0.000 1 ent for pigmented animals and albinos (Table 2).
1Differences in the number of pyknotic profiles between control animals (P A ) were
At concentrations spanning 0.5 to 2.5 mM, the effect on
C C
significant for dopa concentrations spanning the range used. From 0.5 to 2.5 mM, the the level of mitosis was dose dependent in the pigmented
number of pyknotic profiles was significantly reduced in both pigmented (PCPE ) and animals. That is to say that simple factorial (two-way
albino (ACAE ) animals. The difference between the two experimental groups (PEAE ) was
significant apart from the concentration of 2.0 mM. 0 t 0 , t value in t test; S, level of interaction) ANOVA revealed a statistically significant
significance; NS, not significant. difference between the eyes exposed to 0.5 mM and those
exposed to 2.5 mM (F ⫽ 0.041, significant at 5% level). This
was not the case for the albino animals. Here, the effects on
controls, confirming that dopa restricts mitosis. At each mitosis of exposure to 0.5 mM were no different than those
concentration, one-way ANOVAs consistently revealed sig- obtained with 2.5 mM (F ⫽ 0.126, not significant). Al-
nificant effects (F ⫽ 0.000, P ⬍ 0.01 in each case) of dopa on though there was no clear trend in the level of pyknosis in
mitotic and pykotic levels between pigmentation pheno- experimental pigmented eyes with increasing doses of
types. Subsequently, independent sample (two tailed) t dopa, the albino animals showed no dose-dependent effects
tests were used for comparisons in numbers of mitotic and here either, with 0.5 mM having the same effect as 2.5 mM
pyknotic profiles between pigmentation phenotypes at (F ⫽ 0.539, not significant).
different dopa concentrations. The overriding effect of dopa addition is a dramatic
At the lowest concentration used, of 0.1 mM, there were reduction in mitotic and pyknotic profiles in albino eyes,
no significant differences in the number of mitotic and bringing them to a level comparable to that found in the
pyknotic profiles between control and experimental ani- pigmented control group. However, the effects of this agent
mals (Tables 1, 2). At the higher concentration of 0.5 mM, differed depending on whether the eyes were pigmented or
there was a significant reduction in the number of mitotic albino. Presumably, this difference is partially due to the
profiles found in the albinos compared with their control
different endogenous dopa levels between the groups,
group (Fig. 9; Table 1). There was also a reduction in the
which from HPLC measurements at this stage are around
number of mitotic profiles in the experimental pigmented
animals, but this was not significant (Table 1). The differ- threefold.
ence between the two experimental groups was significant Figures 10 and 11 show the percentage reduction in
(Table 1), with the reduction in mitotic activity being more mitosis and pyknosis in albino eyes with increasing dopa
marked in albinos than pigmented eyes. At this concentra- concentration. In both cases, the maximal effect occurs
tion there was also a significant reduction in the number of over a very small range, between 0.1 and 0.5 mM and
pyknotic profiles in both pigmented and albino animals remains relatively constant between 0.5 and 2.5 mM. At
(Table 2). greater concentrations, the drug is toxic. Because the
At 1.0 mM concentration, there was again a significant maximal effect is reached so rapidly and is effective over
reduction in the number of mitotic profiles in the experi- such a short concentration range, these data have been
mental albino retinae than in their control group (Fig. 9; plotted on linear rather than logarithmic axes.
RETINAL MITOSIS IN PIGMENTED AND ALBINO RETINAE
Fig. 10. Percentage reduction in mitosis in albino retinae in vitro
as a function of increasing dopa concentration. The percentage reduc-
tion is against control albino eyes to which dopa was not added. At 1
mM, the drug had no significant effect, but at 0.5 mM, the effect was
maximal and remained so up to 2.5 mM. Higher concentrations
resulted in excess pyknosis. The y axis is linear because the range over
which there was an effect was so limited and because the drug reached
a maximal level over a very small concentration range.
Fig. 11. Percentage reduction in pyknosis in albino retinae in vitro
as a function of increasing dopa concentration. The percentage reduc-
tion is against control albino eyes to which dopa was not added. At 1
mM, the drug had no significant effect, but at 0.5 mM, the effect was
maximal and remained so up to 2.5 mM. Higher concentrations
resulted in excess pyknosis. The y axis is linear because the range over
which there was an effect was so limited and because the drug reached
a maximal level over a very small concentration range.
DISCUSSION
Synopsis of results
A substantial elevation in the number of mitotic profiles
has been demonstrated in the RPE and neural retina in
developing albino eyes. During the main phase of cell
production, the increase in mitotic profile numbers in the
neural retinae of albinos was approximately 50%. This
increase was accompanied by a significant transitory
increase in retinal thickness. However, because there were
also many more pyknotic profiles in albino retina, differ-
403
ences in retinal thickness between the groups declined.
Dopa is a regulator of retinal mitosis and is a key element
in the synthetic pathway of melanin. HPLC measure-
ments revealed that it is present in abnormally low levels
in developing albino eyes, whereas organ culture experi-
ments demonstrate that its addition to the ocular environ-
ment regulates levels of mitosis, closing the gap between
pigmented and albino animals.
Patterns of cell production found in this study are
similar to those previously reported for rodents (Young,
1983, 1985). There is evidence that mitosis in the RPE is
more extensive than that found here, in that it continues
at the retinal margin after PCD 17 and that binucleated
cells are produced postnatally (Bodenstein and Sidman,
1987). These data were obtained from whole-mounts rather
than sampling retinal sections, which probably accounts
for the difference. No pyknotic profiles were found in the
developing RPE. Similarly, Bodenstein and Sidman (1987)
also failed to identify these in the developing mouse RPE.
There are at least two possible explanations for the
elevation of mitosis in albino retinae. First, there may be
differences in cell cycle rates between the two pigmenta-
tion phenotypes. Second, cells may be unable to leave the
cell cycle at appropriate times. Neither of these explana-
tions is exclusive, because the two factors may be linked,
with cell cycle rates influencing the probability of cell cycle
exit. During the cell cycle, the nuclei of retinal neurons
move between the vitreal margin and the ventricular
surface adjacent to the RPE. The pace of this cycle
lengthens from 10 hours at PCD 10, to 30 hours at PCD 21
in normally pigmented animals, mainly as a result of an
increase in G1 and S phases (Young, 1983). Because albino
retinae were thicker than the pigmented retinae, for the
cycle rate to remain the same in both strains, nuclei in
albinos would have to move faster than in pigmented
animals. Given the data presented here, it is not parsimo-
nious to assume simply that they move even faster to
account for differences in levels of mitosis. Rather, it is
probable that excessive mitosis in albinos arises because
cells remain in the cell cycle beyond the stage at which
they would normally exit and differentiate. This process
probably results in retinal congestion producing an eleva-
tion in the number of dying cells. In support of this theory,
there is evidence that when cells are forced to remain in
the cell cycle for abnormally long periods they die (Heintz,
1993), which is consistent with the excess cell death in the
albino retinae. Despite this, cell cycle rates and exit points
may be related. There is evidence that cycle rates between
the two animal types used here do differ, because more
cells are consistently labelled with pulsed 3H-thymidine in
albinos than in their pigmented litter mates, suggesting
that albinos have faster cycles rates (Ilia and Jeffery,
unpublished observations). Current studies are examining
cell cycle rates and exit points in these animals.
Tyrosinase triggers the cascade of reactions producing
melanin by means of dopa (Garcia et al., 1979). The albino
animals used in this study were tyrosinase negative. As
such, they were likely to have reduced retinal dopa levels
during development, which was confirmed with HPLC.
Data from in vitro studies on mitosis in the RPE are
significant with respect to this. Dopa has profound effects
on the cell cycle of RPE cells (Akeo et al., 1989), and the
effects differ, depending on whether the cell is pigmented
or nonpigmented. The cell cycle rate of RPE cells in vitro is
dose-dependently arrested in the S phase by small amounts
404
of dopa (100–250 M), and there is a decrease in the G1
phase (Akeo et al., 1994). When dopa is applied to RPE
cultures, the time required for the cell populations to
double lengthens from 19 to 27 hours (Akeo et al., 1989).
Furthermore, it has been established that dopa can be
used as an antimitotic agent in the treatment of melanoma
(Wick, 1977, 1980). Hence, a possible explanation of the
results obtained here, is that dopa acts to restrict retinal
cell division, and/or to signal cell cycle exit. When dopa
concentration is reduced, as in albinos, cell proliferation is
not regulated appropriately and cells remain in the cell
cycle beyond their normal exit point. This is supported by
the effects of dopa addition in organ culture presented in
this study.
By what route could dopa regulate mitosis/cell cycle
exit? The RPE is developmentally precocious in relation to
the neural retina (Young, 1983; Bodenstein and Sidman,
1987). Retinal mitosis occurs when cells are adjacent to the
ventricular margin, which is the border between the
neural retina and the RPE. There is evidence that at this
stage they form transitory gap junctions with RPE cells,
which are not found later (Hayes, 1976; Fujisawa et al.,
1976; Townes-Anderson and Raviola, 1981). In the mature
retina, many gap junctions are gated by dopamine, the
application of which restricts junction permeability (Vaney,
1994). However, by using HPLC analysis, only trace
amounts of dopamine can be found at PCD 19 in either
pigmentation phenotype (Heals, personal communica-
tion). It is possible that dopa may play a role in gating
junctional connections between mitotic profiles and RPE
cells, hence, influencing the cell cycle. Perhaps the reduced
dopa levels result in junctions remaining permeable in
albinos at a time when they are becoming increasingly
restrictive in pigmented animals. This permeability may
fail to restrict mitosis to a level appropriate for normal
development. These possibilities await experimental inves-
tigation.
Dopa is clearly critical in regulating an aspect of retinal
mitosis. It is less clear how its absence might produce the
albino abnormalities. Cell production in the neural retina
is biphasic. Cells produced in the first phase include cones
and ganglion cells, whereas those in the second phase
include the more numerous rods and bipolar cells (Harman
and Beazley, 1989). The excessive period of cell production
and death in albinos is focussed relatively late in retinal
development. It is highly probable that the majority of
cells produced at this stage are rods and bipolar cells,
which are known to be effected in albinism (Jeffery and
Kinsella, 1992; Jeffery et al., 1994). In birds and squirrels,
which have cone-dominated retinae (Walls, 1942), there
are little or no deficits in cell density at the area centralis
in albinos, suggesting that there is a critical relationship
between rods and the development of this aspect of the
abnormality (Jeffery and Williams, 1994; Esteve and Jef-
fery, 1998). Deficits found in the ONL that arise from
excessive mitosis throughout the period of rod production
are not only reflected in a delay in the centre-to-periphery
gradient of cell production but also in the distribution of
3H-thymidine labelled cells through the depth of the ONL.
Patterns of cell addition in the ONL follow a gradient in
pigmented animals (Carter-Dawson and LaVail, 1979), but
there is a major disruption to this pattern in albinos
focussed around the time of peak rod production (Ilia and
Jeffery, unpublished observations). Hence, excessive cell
M. ILIA AND G. JEFFERY
production and pyknosis seem to be associated with re-
duced rod numbers in albinos (Jeffery et al., 1994a, 1997)
and a disruption in the normal spatial distribution of these
cells through the depth of the ONL. With respect to this
finding, an additional facet of our results is that they
critically undermine the established use of albinos to study
normal retinal development and factors influencing rod
production (Watanabe and Raff, 1990, 1992; Alexiades and
Cepko, 1997; Ezzeddine et al. 1997).
Although the majority of cells affected in albinism are
generated in the second phase of retinal neurogenesis,
ganglion cells also show central retinal abnormalities, and
some have aberrant chiasmatic pathways but are pro-
duced in the first phase. It is possible that central retinal
ganglion cell abnormalities occur due to transneuronal
affects, but the problem of chiasmatic misrouting remains
more problematic. One factor influencing chiasmatic path-
way selection is the cells birth date. Uncrossed cells in the
temporal retina are born before those with a crossed
projection (Drager, 1985; Reese and Colello, 1991). There is
evidence of a delay in spatiotemporal patterns of cell
production in the ganglion cell layer in albinos (Ilia and
Jeffery, 1996). This could explain the reduced uncrossed
pathway in these animals. However, cells giving rise to the
rat uncrossed pathway are generated between PCD 14 and
17 (Reese and Colello, 1991). If dopa were to play a key role
in regulating the temporal pace of the centre-to-periphery
gradient, and hence, the possible rerouting of chiasmatic
pathways, it would have to do so even though it were
present at very low concentrations, because until PCD 15
it is only present in trace quantities. Experiments are in
progress to determine whether there is a relationship
between chiasmatic pathways and dopa by administering
it systemically during pregnancy. Although it appears that
dopa can regulate retinal mitosis in vivo, preliminary
results show that it has no significant effect on mitosis in
the brain.
ACKNOWLEDGMENTS
We thank Ben Reese and Gary Baker for their comments
on the manuscript. We also thank Simon Grant, Alison
Harman, and Alan Whitmore for discussions and com-
ments pertaining to the experiments that were under-
taken in this study and their interpretation. We are
particularly grateful to Brian Wheals for assistance with
HPLC.
LITERATURE CITED
Akeo K, Ebenstein DB, Dorey CK. 1989. Dopa and oxygen inhibit prolifera-
tion of retinal pigment epithelial cells, fibroblasts and endothelial cells
in vitro. Exp Eye Res 49:335–346.
Akeo K, Tanaka Y, Okisaka S. 1994. A comparison between melanotic and
amelanotic retinal pigment epithelial cells in vitro concerning the
effects of L-dopa and oxygen on cell cycle. Pigment Cell Res 7:145–141.
Alexiades MR, Cepko CL. 1997. Subsets of retinal progenitors display
temporally regulated and distinct biases in the fates of their progeny.
Development 124:1119–1131.
Bancroft JD, Stevens A. 1990. Theory and practice of histological tech-
niques. Edinburgh: Churchill Livingstone. p 144–145.
Berman NEJ, Payne BR. 1985. An exuberant retinocollicular pathway in
Siamese kittens: Effects of abnormal activity on its maturation. Dev
Brain Res 22:197–209.
Bodenstein L, Sidman RL. 1987. Growth and development of the mouse
retinal pigment epithelium. Dev Biol 121:192–204.
RETINAL MITOSIS IN PIGMENTED AND ALBINO RETINAE
Carter-Dawson LD, LaVail M. 1979. Rods and cones in the mouse retina: II
Autoradiographic analysis of cell generation using tritiated thymidine.
J Comp Neurol 188:263–272.
Chayen J, Butcher RG, Bitensk L, Poulter LW, editors. 1969. A guide to
practical histochemistry. Edinburgh: Oliver & Boyd.
Cooper ML, Pettigrew JD. 1979. The retinothalamic pathway in Siamese
cats. J Comp Neurol 187:313–348.
Drager UC. 1985. Birthdates of retinal ganglion cells giving rise to the
crossed and the uncrossed optic projection in the mouse. Proc R Soc
Lond Biol 224:57–77.
Elschnig A. 1913. Zur Anatomie des Manschlichen Albinoanges. Graefes
Arch Ophthalmol 84:401–419.
Esteve J, Jeffery G. 1998. Reduced retinal deficits in an albino mammal
with a cone rich retina: a study of the area centralis of pigmented and
albino gray squirrels. Vision Res 38:937–940.
Ezzeddine ZD, Yang X, DeChiara T, Yancopoulos G, Cepko CL. 1997.
Postmitotic cells fated to become rod photoreceptors can be respecified
by CNTF treatment of the retina. Development 124:1055–1067.
Fujisawa H, Morioka H, Watanabe K, Nakamura H. 1976. A decay of gap
junctions in association with cell differentiation of neural retina in
chick embryonic development. J Cell Sci 22:585–596.
Garcia RI, Szabo G, Fitzpatrick TB. 1979. Molecular and cell biology of
melanin. In: Zinn KM & Marmor MF, editors. The retinal pigment
epithelium. Cambridge MA: Harvard University Press.
Harman AM, Beazley LD. 1989. Generation of retinal cells in the wallaby,
Setonix brachyurus. Neuroscience 28:219–232.
Hayes BP. 1976. The distribution of intercellular gap junctions in the
developing retina and pigment epithelium of Xenopus laevis. Anat
Embryol 150: 99–111.
Heintz N. 1993. Cell death and the cell cycle: a relationship between
transformation and neurodegeneration? Trends Biochem Sci 18:157–
159.
Ilia M. 1998. Melanin: its role in retinal development. Ph.D. thesis,
University College, London.
Ilia M, Jeffery G. 1996. Delayed neurogenesis in the albino retina: evidence
of a role for melanin in regulating the pace of cell generation. Dev Brain
Res 95:176–183.
Jeffery G, Kinsella B. 1992. Translaminar deficits in the retinae of albinos.
J Comp Neurol 326:637–644.
Jeffery G, Williams AL. 1994. Is abnormal retinal development in albinism
only a mammalian problem? Normality of a hypopigmented avian
retina. Exp Brain Res 100:47–57.
Jeffery G, Darling K, Whitmore A. 1994a. Melanin and the regulation of
mammalian photoreceptor topography. Eur J Neurosci 6:657–667.
Jeffery G, Schutz G, Montoliu L. 1994b. Correction of abnormal retinal
pathways found in albinism by introduction of a functional tyrosinase
gene in transgenic mice. Dev Biol 166:460–464.
Jeffery G, Brem G, Montoliu L. 1997. Correction of retinal abnormalities
found in albinism by introduction of a functional tyrosinase gene in
transgenic mice and rabbits. Dev Brain Res 99:95–102.
405
Leventhal AG. 1982. Morphology and distribution of retinal ganglion cells
projecting to different layers of the dorsal lateral geniculate nucleus in
normal and Siamese cats. J Neurosci 2:1024–1042.
Leventhal AG, Vitek DJ, Creel DJ. 1985. Abnormal visual pathways in
normally pigmented cats that are heterozygous for albinism. Science
229:1395–1397.
Lund RD. 1965. Uncrossed visual pathways of hooded and albino rats.
Science 149:1506–1507.
Ofori S, Bretton C, Hof P, Schorderet M. 1986. Investigation of dopamine
content, synthesis, and release in the rabbit retina in vitro: I Effects of
dopamine precursors, reserpine, amphetamine, and L-DOPA decarbox-
ylase and monoamine oxidase inhibitors. J Neurochem 47:1199–1206.
Oyster CW, Takahashiu ES, Fry KR, Lamb DMK. 1987. Ganglion cell
density in albino and pigmented rabbit retinas labelled with a ganglion
cell-specific monoclonal antibody. Brain Res 425:25–33.
Reese BE, Colello RJ. 1991. Neurogenesis in the retinal ganglion cell layer
of the rat. Neuroscience 46:419–429.
Reese BE, Johnson PT, Baker GE. 1996. Maturational gradients in the
retina of the ferret. J Comp Neurol 375:252–273.
Robinson SR, Rapaport DH, Stone J. 1985. Cell division in the developing
cat retina occurs in two zones. Dev Brain Res 19:101–109.
Stone J, Rowe MH, Campion JE. 1978. Retinal abnormalities in the
Siamese cat. J Comp Neurol 180:773–782.
Townes-Anderson E, Raviola G. 1981. The formation and distribution of
intercellular junctions in the rhesus monkey optic cup: the early
development of the cilio-iridic and sensory retinas. Dev Biol 85:209–
232.
Vaney DI. 1994. Patterns of neuronal coupling in the retina. Prog Retinal
Eye Res 13:301–355.
Walls GL. 1942. The vertebrate eye and its adaptive radiation. Cranbrook
Institute of Science. Michigan: Bloomfield Hills.
Watanabe T, Raff MC. 1990. Rod photoreceptor development in vitro:
intrinsic properties of proliferating neuroepithelial cell changes as
development proceeds in the rat retina. Neuron 2:461–467.
Watanabe T, Raff MC. 1992. Diffusible rod-promoting signals in the
developing rat retina. Development 114:899–906.
Webster ME, Row MH. 1991. Disruption of developmental timing in the
albino rat retina. J Comp Neurol 307:460–474.
Wick MM. 1977. I-Dopa methyl ester as a new antitumour agent. Nature
269:512–513.
Wick MM. 1980. Levadopa and dopamine analogs as DNA polymerase
inhibitors and antitumour agents in human melanoma. Cancer Res
40:1414–1418.
Young RW. 1983. The life history of retinal cells. Tr Am Eighth Soc
51:193–228.
Young RW. 1985. Cell proliferation during postnatal development of the
retina in the mouse. Dev Brain Res 21:229–239.