Download Development of neonatal mouse retinal neurons and

Investigative Ophthalmology & Visual Science, Vol. 29, No. 4, April 1988
Copyright © Association for Research in Vision and Ophthalmology
Development of Neonatal Mouse Retinal Neurons
and Photoreceptors in Low Density Cell Culture
Luis E. Poliri, Mohamed Lehor, and Ruben Adler
We describe here a culture method which allows the growth of dissociated mouse retinal neurons and
photoreceptors in chemically defined medium. Neural retinas from 2-day-oId C57/BL mice were
dissected from other ocular tissues, including the pigment epithelium, and dissociated into a cell
suspension after brief trypsination. Most cells attached as single, unaggregated units to substrata
pretreated with polyornithine and the neurite-promoting factor (PNPF). The cells were cultured in
serum-free, high pyruvate Dulbecco's modified Eagle's medium containing chemically defined supplements. Under these conditions, onset of cell process development was rapid, giving rise to extensive
neurite networks. Three morphologically distinct cell types were apparent during the first week in
vitro. Some cells retained a circular outline and failed to produce processes, while 50-60% of the cells
developed as multipolar neurons showing a large cell body and several neurites. Approximately 90% of
these cells reacted with an amacrine cell-specific monoclonal antibody. Some 30% of the cultured cells
expressed phenotypic properties characteristic of rod photoreceptors, including a small cell body, an
apical cilium, a short neurite with a spherule-like terminal body, and immunoreactivity with antibodies
against opsin as well as a rod cell-specific monoclonal antibody. No further signs of outer segment
differentiation were observed in these cells. Non-neuronal "flat" cells, which represented less than
0.5% of the total cell number, reacted with an antibody against the glial fibrillary acidic protein. The
number of neurons and photoreceptors remained relatively stable during the first 4-7 days in vitro.
During the second week in culture, however, there was specific degeneration of greater than 90% of the
photoreceptor cells, while less than 20% of the multipolar neurons were similarly affected. Consequently, in addition to providing a system for studying the differentiation of retinal neurons and
photoreceptors, the specific degeneration of photoreceptors in these mouse retinal cell cultures makes
this system ideal for investigating factors influencing photoreceptor survival. Invest Ophthalmol Vis
Sci 29:534-543, 1988
Our knowledge of the cellular and molecular
mechanisms controlling photoreceptor development
and maintenance in normal animals is very limited,
as is our understanding of mechanisms through
which abnormal genes can lead to photoreceptor degeneration. Experience with other neural organs has
shown that the investigation of these questions can be
greatly enhanced by the availability of well defined in
vitro systems (review in ref. 1). The chick embryo is
the most frequently used source of retinal tissue for
these studies, and methods have been recently developed which allow growth of low density, clump- and
flat cell-free cultures of chick embryo neurons and
photoreceptors.2'4
The mouse retina offers an attractive system for
studies of photoreceptor cell survival and differentiation. For example, mouse strains have been described
in which single gene mutations cause photoreceptor
degenerations resembling human retinitis pigmentosa (review in ref. 5). Unfortunately, there is very
limited experience with cell culture systems for
mouse retina. Monolayer cultures have been described for retinal neurons and glial cells from other
mammals, such as rat6"9 and rabbit.10 The mouse retina has been studied in explant" and reaggregation
cultures,12 but low density monolayer cultures of
mouse retinal neurons and photoreceptors are not
available. We report here a new culture method in
which neonatal mouse retinal neurons and photoreceptors are grown at low density in serum-free, chemically defined media, and express differentiated properties which can be recognized with a variety of analytical techniques.
From the Retinal Degenerations Research Center, The Wilmer
Institute, The Johns Hopkins University School of Medicine, Baltimore, Maryland.
Supported by USPHS grant EY-05404, and core grant
EY-01765. Dr. Politi is a fellow from the Consejo Nacional de
Investigaciones Cientificas y Tecnicas of Argentina. Dr. Adler is a
William and Mary Greve Scholar from Research to Prevent Blindness.
Submitted for publication: August 17, 1987; accepted October
16, 1987.
Reprint requests: Ruben Adler, MD, Retinal Degenerations Research Center, The Wilmer Institute, The Johns Hopkins University School of Medicine, 600 North Wolfe Street, Baltimore, MD
21205.
504
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
MOUSE RETINA CELL CULTURES / Poliri er ol.
No. 4
Material and Methods
Materials
The following sources of supplies were used: Falcon (Oxnard, CA): tissue culture plastic, 35 mm
dishes; Miles Scientific (Naperville, IL): Lab Tek 8chamber culture slides; Gibco (Grand Island, NY):
Dulbecco's modified, high pyruvate Eagle's Medium;
Irvine Scientific (Santa Ana, CA): penicillin, trypsin,
Hank's balanced salt solution and calcium-, magnesium-free balanced salt solution; Calbiochem (San
Diego, CA): Hepes; Sigma (St. Louis, MO): polyornithine, bovine serum albumin, cytidine 5'-diphosphoethanolamine, cytidine 5'-diphosphocholine,
DNAase, glutamine, glutaraldehyde, insulin, transferrin, putrescine, progesterone, selenium, soybean
trypsin inhibitor, hydrocortisone; New England Nuclear (Boston, MA): 3H-thymidine (sp. activity 6.7
Ci/mmole); Kodak (Rochester, NY): Nuclear track
emulsion NTB2, Dektol, Fixer; EM Science (Fort
Washington, PA), Polaron (Hatfield, PA), American
Scientific (McGraw Park, IL): electron microscopy
supplies; Cooper Biochemical (Malvern, PA): fluorescein- and rhodamine-conjugated antibodies.
Experimental Animals
C57/BL mice were used for all these experiments.
The animals were either bred in our own colony or
purchased from Harlan-Sprague Dawley Industries
(Indianapolis, IN). In most experiments, mice were
used on the second day after birth, although other
stages were used as indicated in the text. Animals
were handled according to the PHS Policy on Humane Care and Use of Laboratory Animals and the
ARVO Resolution on the Use of Animals in Research.
Dissection Techniques
Two-day-old mice were sacrificed by decapitation,
and the heads were wrapped in alcohol-soaked pads
and kept on ice until dissected. Under the dissecting
microscope, the eyelids were removed and the eyes
were enucleated using different sets of forceps to preserve sterility. While immersed in a petri dish containing Hank's balanced salt solution (HBSS), the
eyes were freed of surrounding loose mesenchyme,
and the cornea, choroid, sclera and pigment epithelium were dissected in one step from the optic nerve
head towards the cornea using watchmaker's forceps.
After removing the vitreous and lens, the isolated
retinas were transferred to another petri dish with
fresh HBSS. With this dissection procedure, isolated
retinas are completely free of contamination with
other eye tissues, including pigment epithelium (see
535
histological section in Fig. 1). Retinas from four to six
eyes could be dissected in about 50 min.
Dissociation Procedures
Isolated retinas were transferred to calcium- and
magnesium-free HBSS (CMF), cut into small fragments using tungsten needles, and incubated at 37°C
for 17 min in 0.25% trypsin in CMF containing 100
ng DNAase/ml. The tissue was then rinsed twice in a
0.25% solution of soybean trypsin inhibitor in Eagle's
basal medium (EBM) with Hank's salts, further
rinsed twice in EBM containing 1% bovine serum
albumin (BSA), and triturated 10-12 times with a
narrow-tip Pasteur pipette in 1.5 ml of this solution.
The resulting suspension of single cells was diluted in
high-pyruvate, low-glucose Dulbecco's modified
Eagle's medium (DME), and aliquots were counted
with a hemocytometer. Ninety-five percent of the
cells were found to exclude Trypan blue.
Culture Conditions
The culture medium consisted of DME supplemented with penicillin (100,000 U/l), glutamine (2
mM), cytidine 5'-diphosphocholine (2.56 mg/1), cytidine 5'-diphosphoethanolamine (1.28 mg/1), hydrocortisone (100 nM), and the N! supplement at twice
the concentrations recommended by Bottenstein and
Sato,13 ie, insulin (16.6 X 10"7 M), progesterone (4
X 10~8 M), putrescin (2 X 10"4 M), selenium (6 X
10"8 M) and transferrin (12.5 X 10"8 M). Culture
containers were 35 mm tissue culture plastic dishes
(Falcon), or chambers mounted on tissue culture
plastic slides (Labtek, 8 chambers/slide), pretreated
for a minimum of 12 hr with polyornithine (50 Mg/ml
in borate buffer, pH 8.4), rinsed twice in DME, and
incubated for a similar period at 37°C in a 25% (V/V)
Schwannoma conditioned medium containing the
neurite promoting factor PNPF.14 Dishes received
600,000 cells in 2 ml of medium, and slide chambers
60,000 cells in 250 /A of medium. The cultures were
grown at 37°C in a humidified atmosphere of 5.5%
CO2 in air.
Microscopic Analysis of the Cultures
Live cultures were periodically examined by phase
contrast microscopy. For quantitative analysis, replicate cultures were rinsed in phosphate-buffered saline
(PBS), and fixed with 2% glutaraldehyde in PBS.
Total cell number was determined using an Artek
image analysis system (Artek Systems Co., Farmingdale, NY). The relative frequency of different cell
types was determined by visual analysis of at least 200
cells per dish. Quantitative determinations were
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
506
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / April 1988
made in triplicate dishes from at least two separate
experiments, and are expressed as mean ± standard
deviation.
Immunocytochemistry
For immunocytochemistry the cells were fixed
with freshly prepared 4% paraformaldehyde in PBS
for 45 min and rinsed in the same buffer. For intracellular antigens, the cells were exposed for 15 min to
0.25% Triton X-100 in Tris-saline. In all cases, the
cultures were further incubated overnight in a 2%
solution of unlabeled serum from the same species as
the corresponding secondary antibody to reduce nonspecific binding. Cultures were incubated with polyclonal primary antibodies at 1:100 dilution for 60
min at room temperature, or to monoclonal antibodies at 1:20 at 4°C, overnight. Primary polyclonal antibodies included sheep anti-bovine rhodopsin from D.
Papermaster15 and a rabbit anti-neurofilament antiserum from D. Dahl.16 Monoclonals included several
retinal cell-type-specific antibodies from C. Barnstable,617 and an anti-Thy 1.2 antibody from Mark
Soloski.18 Secondary antibodies were fluorescein- or
rhodamine-conjugated (Cooper Biochemical). The
slides were studied using an epifluorescence Nikon
microscope, and photographed using Kodak Ektachrome (400 ASA) pushed to 800 ASA. Black and
white prints were obtained from color slides using an
intermediate black and white negative.
Thymidine Incorporation and Autoradiography
Mice were injected subcutaneously in the neck region on the day of birth with 10 ^Ci tritiated thymidine (New England Nuclear; sp activity 6.7 Ci/
mmole) per gram of body weight. The following day,
retinas were dissected, dissociated and cultured as described above. After fixation in paraformaldehyde or
glutaraldehyde and extensive rinsing in PBS, the cultures were dehydrated using increasing ethanol concentrations, air-dried, coated with a 50% solution of
Kodak nuclear truck emulsion NTB2, exposed for
5-10 days at 4°C in the dark, developed in Dektol
and fixed in Kodak fixer. Autoradiograms were studied with phase contrast microscopy, or with bright
field microscopy after hematoxylin counterstaining.
Transmission Electron Microscopy
Cultures werefixedwith 2% paraformaldehyde and
2% glutaraldehyde in 0.15 M cocodylate buffer, pH
7.4. Fixation was carried out for 20 min at 22°C
followed by 40 min at 4°C. Cultures were rinsed in
cold buffer, postfixed for 7 min with cold 1% buffered
osmium tetroxide, dehydrated and flat-embedded in
Vol. 29
Epon. Cells to be sectioned were identified by phase
contrast microscopy and their position in the blocks
was scored with a diamond mounted in a "false"
objective. Ultrathin sections were gridstained with
uranyl acetate and lead citrate before examination.
Results
The Donor Retina
Retinas from 2-day-old mice were used for most
experiments. As shown in Figure 1A, mitotic figures
are very abundant at this stage, and the formation of
the characteristic retinal layers is still incomplete.
Retinal ganglion cells and inner plexiform layer are
present, and the inner nuclear layer already shows
some cellular heterogeneity. The outer plexiform
layer can not yet be recognized at this stage, and no
signs of photoreceptor differentiation are apparent by
light microscopy. The conspicuous cleavage plane between neural retina and pigment epithelium allows
for the easy isolation of the neural retina completely
free of pigment epithelial contamination (Fig. IB).
Retinal Dissociation
A variety of dissociation protocols were compared,
using cell yields, presence or absence of undissociated
clumps, and long-term cell survival in culture as criteria. Enzymatic treatments with papain or with collagenase followed by trypsin were less satisfactory
than the mild typsinization protocol finally adopted
(see Methods). With this protocol, the 2-day retina
yielded viable, clump-free suspensions (2.5 to 3.5
X 106 cells/retina on postnatal (PN) day 2; 4.0 to 4.5
X 106 cells/retina on PN day 5). Mouse retinal cells
appear to be extremely sensitive to proteases. Treatment with trypsin inhibitor was essential for cells to
be grown in serum-free, chemically defined medium;
cell survival was erratic in the absence of this treatment.
Overall Development of the Cultures:
Qualitative Observations
Cell attachment was essentially complete within
2-4 hr after seeding, when most of the cells appeared
attached as individual units; only a few cells showed
processes in these early cultures (Fig. 2A). Neurite
formation became more extensive by 6 hr, when
many cells established contacts with other cells
through their processes (Fig. 2B).
The heterogeneity of the cultures increased dramatically during the first week in vitro (Figs. 2C, 3).
Some of the cultured cells retained a circular outline
and failed to form processes (Fig. 3D); these cells will
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
No. 4
537
MOUSE RETINA CELL CULTURES / Poliri er ol.
^•i«
fcL&.*'*j
GC
#
IB
Fig. 1. Toluidine blue-stained, 1 ^m plastic sections of 2-day-old mouse retinas. (A) Partially dissected retina. The existence of a natural
cleavage plane between pigment epithelium (PE) and neural retina (NR) allows for the easy separation of these tissues. (B) isolated neural
retina. Note absence of pigment epithelial cell contamination. Retinal differentiation isstill incomplete at this stage. Although retinal ganglion
cells (GC) are already evident, the inner and outer nuclear layers have not yet become separated, and photoreceptor differentiation is not
obvious. Note presence of mitotic figures (arrows). Magnification bar: 20 jtm.
be referred to hereafter as "process-free cells." Other
cells developed the phenotype usually associated with
cultured neurons, including a relatively large cell
body (10 to 15 Mm in diameter) and extensive neurite
development (Fig. 3). These cells will be referred to as
"multipolar neurons." Figure 3 also illustrates the
broad diversity in the number, extension and
branching patterns of nerve fibers derived from different multipolar neurons. For example, some
neurons showed a circular meshwork of processes
concentric to the cell body (Fig. 3A, C), while other
neurons developed one or more long neurites (Fig.
3B, D). It was not always possible to trace the full
complement of neurites and branches belonging to
individual neurons because extensive neurite development resulted in the formation of complex networks. However, individual neurites as long as 150
/iin could be seen in some cases.
A third cell type which developed during the first
week in vitro was characterized by a small cell body
(4-6 jam in diameter), and by the presence of a single,
characteristically short process (5-15 jtm in length)
usually terminating in a spheroid-like structure (Fig.
3A, B). These cells were polarized in that there was an
apical process at the pole opposite the neurite (Fig.
3A, B). The process could be identified as a cilium by
electron microscopy (see below). Also described
below are immunocytochemical observations which
supported the identification of these cells as developing rod photoreceptors.
Non-neuronal, flat cells were frequently absent
from these cultures and, when present, did not represent more than 0.5-1,0% of the total cell population.
They showed an epitheloid-appearance, and were occasionally associated with small clusters of neuronal
cells. Immunocytochemical analysis suggested that
they are glial in origin (see Fig. 9).
Quantitative Analysis of the Cultures
The survival of different cell types during thefirst2
weeks in vitro is illustrated in Figure 4. Multipolar
neurons represented approximately 50-60% of the
total number of cells throughout the first week. The
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
538
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / April 1988
Vol. 29
number of small monopolar, photoreceptor-like cells
was low after 24 hr in vitro} but increased to approximately 30% of the total by in vitro day 7. This increase in photoreceptor-like cells was accompanied
by a decrease in the relative proportion of processfree cells. During the second week in vitro the number of multipolar neurons and process-free cells remained relatively constant. On the other hand, there
was an almost complete decline in the number of the
photoreceptor-like cells during the same period.
Immunocytochemical, Autoradiographic and
Electron Microscopical Characterization of
Photoreceptor-Like Cells
Opsin immunocytochemistry: The presence of the
visual pigment apoprotein, opsin, was investigated
immunocytochemically in cultured mouse retinal
cells using a sheep polyclonal antiserum against bovine rhodopsin.15 Positive immunostaining was observed in the majority of the small monopolar cells,
thus supporting their identification as photoreceptors. In most of these cells, immunoreactive materials
could be seen as a peripheral ring in the cell body, in
the neurite and its spherule-like terminal body, and
in the region occupied by the apical cilium (Fig. 5A).
This pattern was observed at all the in vitro stages
studied (3-7 days in culture). Opsin immunoreactivity was also observed in some small process-free
"round" cells but, very importantly, multipolar
neurons were consistently negative.
RET-P1 monoclonal antibody: This monoclonal
antibody has been shown to react selectively with rod
photoreceptors in the rat.6 The pattern of immunostaining obtained with this antibody in mouse retinal
cultures resembled that obtained with polyclonal antiopsin antibody (see above): multipolar neurons
were consistently negative, while many of the small,
monopolar photoreceptor-like cells were positive and
showed a relatively widespread distribution of immunoreactive materials (Fig. 5B).
Determination of the day ofphotoreceptor ''birth":
Autoradiographic studies have shown that many
mouse rod photoreceptor precursors undergo their
last mitotic division on thefirstday of postnatal life,19
and that rods represent almost 75% of the cells generated postnatally.20 To determine whether some of the
cells expressing a photoreceptor-like phenotype in
Fig, 2. Development of mouse retinal cells in dissociated cultures, shown after 1.5 hr (A), 6 hr (B) and 7 days (C) in vitro. Note
that most cells attach to the substratum as individual units. Neurite
development is already underway at 1.5 hr, and becomes fairly
extensive at later stages. At 7 days the cultures still remain essentially devoid of flat cells, and show multipolar neurons (long
arrows), photoreceptor-like cells (short arrow), and process-free
round cells. These cells are shown at higher magnification in Figure
3. Magnification bar: 40 ^m.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
MOUSE RETINA CELL CULTURES / Poliri er ol.
No. 4
539
9
3C
3D
Fig. 3. Mouse retinal cells after 7 days in vitro. As shown in A and B, photoreceptor-likc cells (PhR) show a short neurite ending in a
spherule-like body (short arrow) and an apical structure (long arrow) that EM studies show to be a cilium (sec also Fig. 7). Multipolar neurons
show different morphologies. In some cases the neurons show several long processes (B, D) while other cells appear surrounded by a circular
network of processes concentric to the cell body (A, arrowhead; C). A process-free, round cell is shown in D (arrow). Magnification bars: A, B:
: C , D: 5 pm.
culture were still able to divide in early postnatal life,
mice were injected on the day of birth with 10 ^Ci
3
H-thymidine per gram of body weight. On the next
day their retinas were dissected, dissociated, cultured
for 7 days as described and studied autoradiographically. As many as 5% of the cells were labeled with
tritiated thymidine in a 7 day culture. Labeling was
restricted to process-free, round cells and photoreceptor-like cells (Fig. 6). Although in some cases it was
possible to detect RET-P1 immunoreactivity in thymidine-labeled cells, the frequency of double-labeled
cells was not determined. No thymidine-labeled multipolar neurons were seen in the cultures.
Transmission electron microscopy: Electron microscopical preparations showed that in the cell body
of cultured photoreceptor cells the nucleus was dis-
placed towards the site of origin of the neurite, while
other organelles were segregated towards the opposite
cell pole (Fig. 7). Moreover, cultured photoreceptors
showed an apical cilium projecting from the pole of
the cell opposite to the nucleus (Fig. 7). No further
signs of outer segment development were observed in
these cells, such as distal dilation of the cilium or
formation of compacted membranous discs. The ultrastructure of the terminal spherule-like body of the
photoreceptor neurite was not investigated in this
study.
Analysis of Multipolar Neurons Using Cell-Specific
Monoclonal Antibodies
The "HPC-I" monoclonal antibody, which has
been shown to react specifically with amacrine
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
540
INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / April 1988
A : M u l t i p o l a r neurons
*
;
Small monopolar cells
•
:
Process- free cells
Vol. 29
The identity of the non-HPC-1 reactive multipolar
neurons has not been determined. No immunoreactivity was detected with antibodies specific for horizontal, bipolar or ganglion cells,617 but these negative
results cannot be considered conclusive. Also inconclusive are our observations with the thy-1 antigen,
which is selectively expressed by retinal ganglion cells
in rodents.21 Mouse retinal cultures reacted with a
monoclonal anti-Thy 1.2 antibody18 only rarely
showed immunoreactive cells; the few positive cells
that could be detected always showed some morphological signs of degeneration (not shown).
Non-Neuronal "Flat" Cells
100. .
Days in vitro
Fig. 4. Quantitative analysis of cell survival in mouse retinal
cultures. The small monopolar cells (•) can be identified as photoreceptor-like cells on the basis of immunocytochemical and electron microscopical criteria (see Figs. 5,7).
neurons in rodent retinas,17 reacted with up to 90% of
the multipolar neurons present in mouse retinal cultures studied immunocytochemically (Fig. 8). HPC-1
immunoreactivity could be detected without detergent pretreatment of the cultures, indicating that the
antigen was present at the cell surface. Although
some staining could be seen in nerve fibers, it was
particularly conspicuous in cell bodies (Fig. 8). Positive immunoreactivity was never seen in photoreceptors or in process-free round cells.
The expression of glial fibrillary acidic protein
(GFAP) was investigated immunocytochemically in
these cultures using a polyclonal antiserum.16 Immunoreactive cells were rare in these cultures. In all
cases, GFAP immunoreactivity was associated with
non-neuronal, flat cells (Fig. 9) which, as mentioned
above, represent less than 0.5% of the total cells
present in these preparations.
Cultures From Older Retinas
The success of growing neuronal cells in culture is
usually affected by the stage at which cells are isolated
from the donor animal. Therefore, we investigated
the behavior of retinal cells isolated from PN5 and
PN7 retinas, cultured with a protocol identical to the
one described in the preceding sections for PN2 retinal cells. Cell yields were approximately 4.0 to 4.5
X 106 cells/retina at both stages. By phase contrast
microscopy, the cultures were qualitatively similar to
those of PN2 retinas. However, the length of time
that viable cultures could be maintained decreased to
approximately 7-8 days in vitro for PN5 cultures,
and 4-5 days in vitro for PN7 cultures (not shown).
Fig. 5. Small monopolar,
photoreceptor-like cells
were immunoreactive with
antibodies against both
opsin (A) and the rod-specific antigen RET-P1 (B).
Note that, in both cases, immunoreactive materials can
be seen not only in the cell
body, but also in the ciliary
region (short arrows), and
the neurite with its terminal
spherule-like body (long
arrow). Multipolar neurons
were consistently negative
with both antibodies. Magnification bar: 20 jam.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
MOU5E RETINA CELL CULTURES / Poliri er ol.
No. 4
541
Thus, it appears that mouse retinal cells also show
important developmental stage-dependent changes in
their responses to culture.
Discussion
The studies reported here demonstrate the feasibility of growing mouse retinal neurons and photoreceptors in a serum-free, completely chemically denned
environment. Amacrine neurons and rod photoreceptor cells are the two most abundant cell types in
these cultures, but the identity of the other multipolar
neurons remains undetermined. Other striking features of these cultures include the virtual absence of
glial-like "flat" cells, and the presence of abundant
nerve processes which give rise to complex networks.
Little work has been done with stationary monolayer cultures of dissociated mouse retinal cells.
Serum-containing media has been used for retinal
monolayer cultures from other mammals,6"10 and it
has been reported that rat retinal cells failed to survive in a serum-free medium unless brain extracts
were used.8 The culture supplements used in our
Fig. 6. Autoradiogram of a retinal culture from mice injected in
vivo with 3 H-thymidine 24 hr before culture onset, and grown in
vitro for 7 days (see Methods for details). A labeled photoreceptor
can be seen (arrow). Magnification bar: 5 pun.
Fig. 7. Electron micrograph of a cultured photoreceptor-like cell identified by
phase contrast microscopy before sectioning as indicated in Methods. The cell nucleus (N) appears polarized towards one
end of the cell, while the opposite pole
shows an accumulation of organelles and
a cilium (Ci). Magnification bar: 1 ^m.
7
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
542
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / April 1988
Fig. 8. Fluorescence photomicrograph of a mouse retinal culture
reacted with the amacrine cell-specific monoclonal antibody
HPC-1. Approximately 90% of the multipolar neurons reacted with
this antibody, while photoreceptor-like cells were consistently negative. Magnification bar: 20 jim.
study were chosen in pilot experiments by systematic
trial-and-error, and their specific contributions to the
survival and differentiation of neurons and photoreceptors still await more detailed analysis. It is interesting, on the other hand, that mouse retinal cells failed
to survive in DME supplemented with different sera,
including batches of fetal calf serum routinely used in
our laboratory for culturing chick embryo neurons
and photoreceptors.3 It has been reported that some
animal sera contain inhibitory factors for mouse
cells.22
Vol. 29
Mouse photoreceptor cells cultured in chemically
defined medium appear similar to rat photoreceptor
cells grown in serum-supplemented cultures6 in their
morphology (by phase contrast microscopy) and their
immunoreactivity with RET-P1 and anti-opsin antibodies. These immunoreactive materials are distributed in a diffuse manner in both cases. Photoreceptor
cell differentiation has also been studied in dissociated cell culture from chick embryos in which the
majority of the photoreceptors are cones, rather than
rods. Under the conditions used in our laboratory,1"4
isolated chick photoreceptors develop a highly elongated, polarized phenotype. Opsin immunoreactive
materials are initially distributed in a diffuse manner
in these cells, but in older cultures opsin becomes
clearly polarized and accumulates in the apical region
of the cells which is occupied by a small, outer segment-like process.3 Considering that in vivo EM immunocytochemical studies have shown that polarized
opsin distribution is only observed in rat photoreceptors when an intact outer segment is present,23 the
failure of cultured mouse photoreceptors to achieve a
polarized pattern of opsin distribution may be due to
the lack of outer segment development beyond the
presence of an apical cilium. It is also noteworthy that
chick photoreceptors, when grown in vitro under
conditions different from those used in our studies,3
have been found to express a less polarized pattern of
morphological organization and opsin distribution.24
It is well established that the survival of many neuronal types is regulated by "trophic" factors produced
by postsynaptic target cells as well as by glia and other
"satellite" cells (review in ref. 1). Photoreceptor cells
have not been well studied in this regard, although
they are known to undergo developmental death,25
Fig. 9. Phase contrast (A) and fluorescence (B) photomicrographs of a flat cell reacted with an anti-GFAP antibody. Flat cells (which
represent less than 0.5% of the cells present in mouse retina cultures) were the only cells that reacted with this antibody. Magnification bar; 20
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017
MOUSE RETINA CELL CULTURES / Poliri er ol.
No. 4
and to degenerate in a variety of blinding retinal diseases, including human retinitis pigmentosa. In this
context, it is interesting that cultured mouse photoreceptors undergo extensive degeneration during the
second week in vitro, while multipolar neurons suffer
only minor losses during this same period. This selective degeneration of photoreceptor cells suggests
that they may have some special trophic requirements. The recent finding that interphotoreceptor
matrix preparations contain a survival-promoting
macromolecular activity for cultured chick photoreceptor cells may be relevant in this context.26 Possible
roles of this or similar factors can now be experimentally analyzed in cell culture, both from normal mice
(as described here) and from animals with retinal
degenerations of genetic origin, such as the rd
mouse.27'28
Key words: mouse retina, retinal cultures, neurons, photoreceptors, neural culture, retinal development
Acknowledgments
The authors are grateful to Dr. A. Tyl Hewitt for his
comments on the manuscript; to Drs. David Papermaster,
Colin Barnstable and Mark Soloski for generous antibody
gifts, and to Mrs. Doris Golembieski for secretarial help.
References
1. Adler R: Trophic interactions in retinal development and in
retinal degenerations: In vivo and in vitro studies. In The Retina: A Model for Cell Biology Studies, Part I, Adler R and
Farber D, editors. Orlando, Academic Press, 1986, pp.
112-150.
2. Adler R, Lindsey JD, and Eisner CL: Expression of cone-like
properties by chick embryo neural retina cells in glial-free
monolayer culture. J Cell Biol 99:1173, 1984.
3. Adler R: Developmental predetermination of the structural
and molecular polarization of photoreceptor cells. Dev Biol
117:520, 1986.
4. Adler R: The differentiation of retinal photoreceptors and
neurons in vitro. In Progress in Retinal Research, Osborne N
and Chader G, editors. London, Pergamon Press, 1987, pp.
1-27.
5. LaVail MM: Analysis of neurological mutants with inherited
retinal degeneration. Invest Ophthalmol Vis Sci 21:630, 1981.
6. Akagawa K. and Barnstable CJ: Identification and characterization of cell types in monolayer cultures of rat retina using
monoclonal antibodies. Brain Res 383:110, 1986.
7. Berg G and Schachner M: Electron microscopic localization of
A2B5 cell surface antigen in monolayer cultures of murine
cerebellum and retina. Cell Tissue Res 224:637, 1982.
8. Turner JE: Promotion of neurite outgrowth and cell survival in
dissociated fetal rat retinal cultures by a fraction derived from a
brain extract. Dev Brain Res 18:265, 1983.
9. Sarthy PV, Curtis BM, and Catterall WA: Retrograde labeling,
543
enrichment, and characterization of retinal ganglion cells from
the neonatal rat. J Neurosci 3:2532, 1983.
10. Osborne NN, Beaton DW, Vigny A, and Neuhoff V: Localization of tryosine-hydroxylase immunoreactive cells in rabbit
retinal cultures. Neurosci Lett 50:117, 1984.
11. Friedlander DR and Crain SM: Projections of growth conebearingfibersof retinal ganglion cells within co-cultured tectal
explants: Early branching depends on age of target tissue. Dev
Brain Res 78:231, 1985.
12. Moscona AA: Recombination of dissociated cells and the development of cell aggregates. In Cells and Tissues in Culture.
Wilmer E, editor. New York, Academic Press, 1965, pp.
489-529.
13. Bottenstein JE and Sato G: Growth of a rat neuroblastoma cell
line in serum-free supplemented medium. Proc Natl Acad Sci
USA 76:514, 1979.
14. Adler R: Regulation of neurite growth in purified retina cultures: Effects of PNPF, a substratum-bound neurite-promoting
factor. J Neurosci Res 8:165, 1982.
15. Papermaster DS and Schneider BG: Biosynthesis and morphogenesis of outer segment membranes in vertebrate photoreceptor cells. In Cell Biology of the Eye, McDevitt D, editor.
New York, Academic Press, 1982, pp. 475-531.
16. Dahl D and Bignami A: Immunogenic properties of the glial
fibrillary acidic protein. Brain Res 116:150, 1976.
17. Barnstable CJ, Hofstein R, and Akagawa K.: A marker of early
amacrine cell development in rat retina. Dev Brain Res 20:286,
1985.
18. Ledbetter JA and Herzenberg LA: Xenogenic monoclonal antibodies to mouse lymphoid differentiation antigens. Immunol
Rev 47:63, 1979.
19. Carter-Dawson LD and LaVail MM: Rods and cones in the
mouse retina: I. Autoradiographic analysis of cell generation
using tritiated thymidine. J Comp Neurol 188:263, 1979.
20. Young RW: Cell differentiation in the retina of the mouse.
AnatRec 212:199, 1985.
21. Leifer D, Lipton SA, Barnstable CJ, and Masland RH: Monoclonal antibody to Thy-1 enhances regeneration of processes
by rat retinal ganglion cells in culture. Science 224:303, 1984.
22. Loo DT, Fuquay JI, Rawson CL, and Barnes DW: Extended
culture of mouse embryo cells without senescence: Inhibition
by serum. Science 236:200, 1987.
23. Nir I, Sagie G, and Papermaster DS: Opsin accumulation in
photoreceptor inner segment plasma membrane of dystrophic
RCS rats. Invest Ophthalmol Vis Sci 28:62, 1987.
24. Araki M: Immunocytochemical study on photoreceptor cell
differentition in the cultured retina of the chick. Dev Biol
103:313, 1984.
25. Young RW: Cell death during differentiation of the retina in
the mouse. J Comp Neurol 229:362, 1984.
26. Lindsey JD, Hewitt AT, and Adler R: Interphotoreceptor matrix preparations promote photoreceptor survival in retinal
monolayer cultures. ARVO Abstracts. Invest Ophthalmol Vis
Sci 28(Suppl):254, 1987.
27. Politi LE and Adler R: Development of retina neurons and
photoreceptors from normal and rd mice in dissociated cell
culture. ARVO Abstracts. Invest Ophthalmol Vis Sci
28(Suppl):57, 1987.
28. Politi LE and Adler R: Selective failure of long term survival of
isolated photoreceptors from both homozygous and heterozygous rd (retinal degeneration) mice. Exp Eye Res, in press.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933139/ on 05/12/2017