September 1984
Vol. 25/9
Investigative Ophthalmology
& Visual Science
A Journal of Dosic and Clinical Research
Articles
Indoleamine-Accumulating Horizontal Cells
in the Squirrel Monkey Retina
Ingrid Floren* and Anita Hendricksonf
Horizontal cells capable of selectively accumulating indoleamines have been found in the retina of
Saimiri sciureus, the New World squirrel monkey. The uptake of 3H-serotonin (5HT) in vivo and in
vitro has been localized by light and electron microscopic (EM) autoradiography only to horizontal
cells. In the central retina up to 50% and in the periphery 15-20% of the somas lying at the scleral
border of the inner nuclear layer are labeled. In EM autoradiographs, labeled processes occur as lateral
elements in both cone and rod synaptic invaginations, but grain counts show four times as many rods
as cones are labeled. The light-dark adaptational state of the retina has no apparent effect on 3 H-5HT
uptake, but it is temperature dependent, and is saturable as indicated by its inhibition in the presence
of an excess of nonradioactive 5HT. The indoleamine-accumulating capacity of this population of
horizontal cells also was demonstrated by histofluorescence microscopy. A single row of fluorescent
cells with a round soma was found lying at the outer edge of the inner nuclear layer. These cells send
fluorescent processes only into the outer plexiform layer. The authors suggest that 5HT is not the
native transmitter of these squirrel monkey horizontal cells because they lack intrinsic fluorescence
and only could be visualized after uptake of extrinsically applied indoleamines. The true transmitter
is likely to be a closely related but nonfluorescent indole. Invest Ophthalmol Vis Sci 25:9971006, 1984
In the retina of many species,1 some neurons have
the ability to take up extrinsically applied indoleamines
(IA), rendering them visible by fluorescence or autoradiography. In cold-blooded vertebrates, and perhaps
in birds, serotonin is likely to be the transmitter of
these indoleamine-accumulating neurons (IA) on the
basis of biochemical and immunohistochemical evidence.1"3 The transmitter of these neurons is unknown
in mammals; although an indole or a closely related
substance is the most likely candidate, 134 it is unlikely
that serotonin, melatonin, or bufotonine is the transmitter.2-5-6
Previous descriptions1 place the IA cell bodies among
the amacrines in the innermost part of the inner nuclear
layer, although the distribution of their processes in
the inner plexiform layer varies in different species. In
this paper, we report the presence of an IA horizontal
cell in the squirrel monkey retina. The occurrence of
this IA horizontal cell is in marked contrast to another
New World monkey, the Cebus monkey, in which
only IA amacrine cells are reported,7 and to Old World
monkeys and to humans, in which IA neurons are
absent.8-9
Materials and Methods
Fluorescence Histochemistry
Fluorescence histochemistry was performed according to the Falck-Hillarp technique. 10 " Monkeys were
given an overdose of barbiturate and the eyes were
enucleated. The posterior half was frozen immediately
in Freon, kept in liquid nitrogen during freeze-drying,
and then exposed to formaldehyde at 80°C for 1 hr
to form the fluorophores. After this procedure, the
tissue was embedded in vacuo in paraffin wax and
sectioned at 10 nm. The sections were examined with
a Leitz Orthoplan microscope equipped with Ploem
illumination using a " C " cube. Indoleamines are characterized by their yellow and catecholamines by their
blue-green fluorescence with the filter system used.
All squirrel monkeys used for histofluorescence were
From the Department of Ophthalmology, University Hospital,
Lund, Sweden,* and the University of Washington School of Medicine, Seattle, Washington.f
Supported by National Institutes of Health Research Grants EY01208, EY-04536, RR00166, and EY-01730, by a Postdoctoral Research Fellowship funded by Fight for Sight, Inc. (IF), and in part
by an unrestricted departmental grant from Research to Prevent
Blindness, Inc.
Reprint requests: Anita Hendrickson, PhD, Department of Ophthalmology RJ-10, University of Washington, Seattle, WA 98195.
997
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
998
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1984
pretreated with a monoamine oxidase inhibitor to increase the fluorescence yield. Pargyline (Sigma Chemical Co.; St. Louis, MO) 30 mg/kg intraperitoneally
was given 4.5 hr before killing or 30 min before an
eye injection. Two retinas were examined without further treatment. In one eye, the fluorescence intensity
of the dopaminergic neurons of the retina was increased
by an intravitreal injection of 4 ng of a-methylnoradrenaline, a dopamine analogue resistant to degradation by monoamine oxidase,12 4 hr before the eye
was enucleated. A similar scheme had given good results in rabbits3 and Cebus monkeys.7
To induce the fluorescence of the IA neurons, 40
/xg of 5,6-dihydroxytryptamine (5,6-DHT) was injected
intravitreally, and 4 hr later the eye was enucleated.
This is essentially according to previous schemes used
for rabbits3 and Cebus monkeys.7 Two eyes were used
for this part of the study; several pieces from the posterior segment of each eye were examined.
Previous studies13 have shown that dihydroxylated
tryptamines like 5,6-DHT cause characteristic ultrastructural changes of the IA amacrine cells enabling
their identification. Additional pieces of fresh tissue
from the posterior segment of the two normal animals
and four eyes pretreated with 5,6-DHT were processed
for electron microscopy. Some tissue pieces were fixed
in 2% OsO4 in 0.056 M veronal acetate buffer (pH
7.6-7.8) containing 66 mM sucrose and 1.8 mM CaCl2
for 1 hr in an ice bath and for another half hour at
room temperature. 14 Other specimens were fixed in
1% OsO4 in 0.28 M veronal acetate buffer (pH 7.4)
with 93.7 mM NaCl, 3.8 mM KC1, and 1.1 mM CaCl2
(Zetterquist's fixative)15 following the same time
schedule as above. Other tissue pieces were fixed in
ice-cold 2.5% glutaraldehyde in 0.065 M cacodylate
buffer (pH 7.6) with 4.5 mM CaCl2 for 1 hr, briefly
washed and then treated with Zetterquist's fixative as
described above. After dehydration, the tissue was
embedded in Epon, sectioned, stained according to
conventional methods and analyzed in an AEI 801
electron microscope.
All eyes were anesthetized with 0.5% proparacaine
(Ophthaine) prior to intravitreal injections, which were
done under ketamine tranquilization. All drugs used
for intravitreal injection were dissolved in 40 nl saline
with ascorbic acid 1 mg/ml added as an antioxidant.
Hoechst Pharmaceuticals (Stockholm, Sweden) kindly
supplied a-methylnoradrenaline HC1 (Corbasil), and
5,6-DHT was obtained as creatinine sulphate from
Sigma Chemical Co. (St. Louis, MO). All weights given
refer to the salts.
Vol. 25
England Nuclear, Boston, MA; 28 Ci/mmol), and one
eye was injected with 25 ixd dopamine (DA) (Amersham, Arlington Heights, IL; 6.7 Ci/mmol). Saline was
added to make a total volume of 50 /xl. The eyes were
enucleated 4 hr later and immersion-fixed overnight
in 2% paraformaldehyde and 2% glutaraldehyde in 0.1
M phosphate buffer pH 7.3. Pieces of the posterior
segment were washed briefly in buffer, postfixed with
1% OsO4 in phosphate buffer for 1 hr and dehydrated
and embedded in Epon. 1 ^m sections on glass slides
were dipped in Kodak NTB-2 emulsion diluted 1:1,
exposed for 1-7 weeks, developed in D19, fixed in
Kodak Rapid Fix and stained with azure II-Methylene
blue. Thin sections of the same tissue were picked up
on carbon and parloidin-coated grids and coated with
Ilford L4 emulsion by means of a loop,16 exposed for
4-8 weeks, developed in Kodak D19 or Microdol X,
stained with uranyl acetate and lead citrate and analyzed in an AEI 801 electron microscope.
For in vitro labeling, five eyes were enucleated,
hemisected circumferentially at the equator, the vitreous humor removed and the posterior segment cut
into sectors, with care being taken not to detach the
retina. A medium prepared according to Ames17 was
used for incubation with 2 mg ascorbic acid and 1.8
mg Pargyline added per 100 ml of medium. 18 The
medium was gassed with 5% carbon dioxide in 95%
oxygen before and during the incubation. Each tissue
piece was placed in a bottle containing 2.5 ml of the
medium and preequilibrated for 10 min. The radioactive substance was added and the tissue was incubated for 15, 30, or 60 min at 37°C or 0°C. The
amount of serotonin used was 25, 50, or 125 fxCi yielding a final concentration range of 3.5 X 10~7 to
1.8 X 10~6 M, and 10 or 50 /id of dopamine was
used, giving a final concentration of 6 X 10~7 to 3
X 10~6 M. Most incubations were performed in ambient room light using eyes from light-adapted animals.
One eye was incubated in darkness using dim red illumination and foil-wrapped bottles. One eye was darkadapted for 1 hr and enucleated and dissected in darkness followed by incubation in light. The incubations
were terminated by rinsing in ice-cold unlabeled medium followed by fixation and processing for light and
electron microscope autoradiography as described
above for in vivo labeling.
All procedures involving animals were performed
in adherence to the ARVO Resolution on the Use of
Animals in Research.
Results
Autoradiography
Fluorescence Histochemistry
For in vivo labeling, two eyes from different animals
were injected intravitreally with 25 ^Ci serotonin (New
The normal squirrel monkey retina showed bluegreen fluorescent dopaminergic cell bodies among the
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
No. 9
SQUIRREL MONKEY HORIZONTAL CELLS / Floren ond Hendrickson
909
Fig. 1. A, OPL: Outer
plexiform layer; IPL: inner
plexiform layer. Normal
squirrel monkey processed
for fluorescence. There is one
fluorescent cell body in the
innermost part of the inner
nuclear layer, presumably an
amacrine cell. Two well-defined sublayers of fluorescent
terminals are found in the
IPL with fine twigs forming
a faint innermost third sublayer. Fibers also are directed
outward through the inner
nuclear layer but very few of
these reach the OPL (X400).
B; C, Retina from an eye injected in vivo 4 hr previously
with 40 ng 5,6-DHT and then
processed for fluorescence.
Many fluorescent cell bodies
now are found in the outermost part of the inner nuclear
layer at the level of the horizontal cells; these are more
numerous in the central retina (C) than in the peripheral
(B). The OPL also has become fluorescent due to the
large number of fine processes in it. Fluorescence micrograph (X400). D, Autoradiogram of peripheral retina incubated in vitro in 50
fid 3H-serotonin (7 X 10~7
M). Two labeled horizontal
cells (arrows) lie in the outermost part of the inner nuclear layer, with the labeled
OPL immediately scleral to
them (X650). E, Autoradiogram of the central retina
from an eye injected in vivo
4 hr previously with 25 ^Ci
3
H-serotonin. Arrows point
to labeled horizontal cells,
which are more numerous
here than in D (X65O). F,
Autoradiogram from the
same animal as in E. There
are two radioactive horizontal cells (small arrows) below
the labeled OPL. Heavily labeled horizontal cell processes pass through OPL and
end among the photoreceptor terminals (large arrows)
(X2100). G, Autoradiogram
of obliquely cut retina from
the same monkey as in D. There are numerous, heavily labeled horizontal cells in the outermost part of the inner nuclear layer (INL), and
the OPL is strongly radioactive. Clusters of radioactivity are seen among the photoreceptor terminals (arrows) (X65O).
amacrines at the junction between the inner nuclear
layer (INL) and inner plexiform layer (IPL) (Fig. 1A).
These cell bodies send their processes to two sublayers
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
in the IPL with the outermost sublayer at the border
of INL and IPL being most pronounced. Only a few
processes reach the innermost sublayer, which lies at
1000
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / September 1984
Vol. 25
the border of the IPL and ganglion cell (GC) layer.
Dopaminergic fibers also are sent sclerally through the
INL towards the outer plexiform layer (OPL), but only
occasionally do they reach it. Fluorescence was enhanced after the intravitreal injection of the catecholamine a-methylnoradrenaline, but no new labeled
structures were seen. No yellow indoleamine fluorescence was seen in normal monkeys. These results confirm earlier studies.19-20
After the injection of 5,6-DHT into the vitreous,
yellow fluorescent indoleamine-labeled cell bodies were
seen at the border between the OPL and INL throughout the retina, but more frequently in the central (Fig.
1C) than the peripheral retina (Fig. IB). Fine processes
extended to the OPL and distributed throughout its
entire width. In favorable sections from parafoveal regions, small round structures could be seen forming
a row at the distal edge of the OPL, whereas they were
not apparent in the periphery. The labeled cell bodies
did not send any processes through the INL toward
the inner retina. Thus, both the cell body location and
process distribution of the indoleamine-labeled neurons
indicate that these are horizontal cells.
through the INL, but rarely could any of these be seen
to reach the OPL. Very long exposure times (2-3
months) produced some faint labeling of the OPL,
indicating either uptake of the 3H-DA by the indoleamine-accumulating horizontal cells, or labeling of the
3
H-DA processes in the OPL.
EM autoradiography confirmed that the uptake of
3
H-serotonin was by horizontal cells (Figs. 2, 3). The
labeled cells bordered the OPL and often two or more
processes could be seen leaving the cell body and entering the OPL. In central retina, most of these swept
around the cone pedicles and invaginated the rod
spherules, which lie immediately distally to the densely
packed row of cone pedicles. Most often only one of
the rod lateral processes was labeled but occasionally
both were (Fig. 4). The horizontal cell processes contacting the rods were frequently very heavily labeled
(Figs. 4A, C), with clumps of grains overlying the invaginating processes within the spherule, explaining
the light microscopic distribution of grains in the OPL.
Infrequently, heavily labeled processes ended as one
of the lateral invaginations in the cone pedicles (Fig.
4E). In contrast to the rod spherules, both of the cone
lateral processes were never labeled.
Autoradiography
The number and type of photoreceptor synapses
involved with labeled horizontal cell processes was
counted in electron autoradiographs (Table 1). If one
grain is required for considering a process labeled, the
risk of spurious labeling is increased. Using the more
stringent criteria of two or more grains, the number
of labeled processes contacting cones is reduced 50%,
whereas the number of labeled processes to rods is
reduced only 20%. As seen in Table 1, using these
criteria rod synapses are involved four times more often
than cone synapses with labeled horizontal cell processes, even in the cone-rich central retina.
Both in vivo and in vitro studies showed uptake of
3
H-serotonin into horizontal cell bodies lying in the
most scleral row of neurons in the INL (Figs. 1D-G).
The distribution of labeled cell bodies varied in the
retina in that approximately 20% of the cell bodies in
the outermost cell row of the INL were labeled in the
peripheral retina (Fig. ID), whereas virtually every
other cell body (~50%) was labeled in the perifoveal
region (Fig. IE). In the periphery, these slightly oval
cells were dispersed widely allowing their long axis to
be oriented horizontally. They were packed more
densely in the central regions, and their long axis was
oriented more vertically. Labeled processes from the
cells extended throughout the OPL and often small
clusters of grains were seen at the level of the photoreceptor terminals (Fig. 1G). After very long exposure
times a faint band became apparent in the outermost
part of the IPL, representing uptake in dopaminergic
neurons. 4 The pattern of uptake for the IA neurons
was the same irrespective of the concentration of
3
H-serotonin or the length of in vitro incubation, but
the intensity of labeling was directly related to both.
No apparent differences were noted between light- or
dark-adapted retinas or retinas incubated under varying
light conditions. Incubation at 0°C or in solutions
containing an excessive amount of cold serotonin
(2 X 10"4 M) abolished 3H-serotonin uptake.
The uptake of 3H-DA both in vivo and in vitro was
localized to amacrine cells lying in the innermost part
of the INL whose processes run in the outermost sublayers in the IPL. Some processes were sent sclerally
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
Ultrastructural Labeling with 5,6-DHT
Labeling with 5,6-DHT produced characteristic ultrastructural changes in the IA horizontal cells (Fig.
5). These consisted of swelling and distortion of cell
organelles and increased electron density of membranes, in particular mitochondrial membranes. Labeled processes contained vesicular structures of varying electron density; presumably, some of these are
lysosomes, whereas others are synaptic vesicles. Multivesicular bodies were also common, but as they were
not confined to horizontal cells, they could not be used
for purposes of identifying the serotonin-uptake neurons.
In the OPL, 5,6-DHT-labeled processes invaginated
cone pedicles (Fig. 5A) and rod spherules (Fig. 5B) as
lateral elements, thus confirming the results obtained
with 3H-serotonin. In addition, the dopaminergic
amacrine cells showed similar ultrastructural changes
with labeled processes occurring mainly in the IPL,
SQUIRREL MONKEY HORIZONTAL CELLS / Floren and
No. 9
Hendrickson
1001
Fig. 2. A low-power electron auloradiograph showing the overall
distribution of 3H-serotonin-labeled
horizontal cell processes to the photoreceptors. One labeled horizontal
cell body is outlined as are the labeled processes in the outer plexiform layer. One cone pedicle (fine
hatching) contains a labeled invaginated horizontal cell process (*),
while six rod spherules {coarse dots)
contain one or two labeled processes
(*)(x6000).
but some also were found in the INL, and a few were
seen at the border between the INL and OPL.
Discussion
Previous work has shown that there is a group of
IA amacrine cells in the retina of many species in1
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
cluding teleosts, amphibians, birds, carnivores, and rodents.3'21"23 In primates, IA amacrine cells have been
found in only the New World Cebus monkey,7 while
similar techniques have failed to demonstrate them in
human and Old World monkey retinas.28 This lack
of IA neurons also was confirmed in screening exper-
1002
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Seprember 1984
X
•
c
Vol. 25
r *
r
t
••
Fig. 3. Low-power electron autoradiographs of the retina from an eye injected
in vivo 4 hr previously with 25 ^Ci serotonin. A, Labeled horizontal cell body (H)
underlies scattered labeled cone ribbon
synapses (C). Most of the labeled processes
swing around the cone pedicles and invaginate the more distal rod spherules (r), many
of which are labeled (X6250). B, Radioactively labeled horizontal cell (H) issues a
labeled process into the outer plexiform
layer. Only one cone synapse (C) is labeled,
whereas an array of rod spherules (r) above
the cone pedicles are contracted by labeled
lateral processes (XI 1,000).
H .
B
iments during this study using Macaca nemestrina retinas. Without preinjection of indoleamines to increase
indoleamine fluorescence,18 no IA neurons have been
found in the untreated retina of any species despite
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
extensive searching,3-5-24 with the single exception of
the chick retina during a short embryonic stage.2S On
the other hand, the adult retinas of all species investigated so far with histofluorescence show the presence
No. 9
1000
SQUIRREL MONKEY HORIZONTAL CELLS / Floren and Hendrickson
B
Fig. 4. Higher power electron autoradiographs from
the same retina as in Figure
3. Radioactive horizontal cell
processes commonly invaginate rod spherules (r) as lateral elements of the ribbon
synapses and often are as
heavily labeled as in A and
C. Usually only one of the
lateral processes is labeled (A
and C), but spherules with
both lateral processes labeled
also are found (B and D). E,
two cone (c) ribbon synapses
each contain one labeled lateral element in this cone
pedicle (c) (A: X26J00; B:
XI 5,200; C: x27,500; D:
X30,000; E: X25.OOO).
of dopaminergic neurons.3 In our squirrel monkey
studies, dopaminergic amacrine cells send their processes to two sublayers in the IPL as well as through
the INL to OPL> in agreement with earlier work.19-20
As well, some neurons take up extrinsically applied
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
indoleamines, but contrary to all other species investigated so far,3-21"23 the surprising result in this study
of squirrel monkey retina was that IA uptake was into
the horizontal cells rather than amacrine cells.
As demonstrated with both histofluorescence and
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1984
1004
Table 1. Photoreceptor synapses involved with
labeled horizontal cell processes in the perifoveal
region of the squirrel monkey 4 hr after the
intravitreal injection of 3H-5HT
One lateral
process labeled
Total
number of
synapses
One or
more
grains
Two or
more
grains
Rods 214
Cones 187
24%
12%
Both lateral processes
labeled
Each with
one or
more grains
Each with
two or
more grains
autoradiography, the location of the IA cells in the
outermost cell row of the INL is clearly that of the
horizontal cell bodies. As well, the EM distribution of
labeled processes in the OPL is consistent with that
described for horizontal cells. The lack of IA-processes
within the inner retina argues against the IA cells being
an undescribed form of interplexiform cell. Dopaminergic interplexiform neurons do exist in New World
monkeys,19i2a26<27 but their cell bodies are located
among the amacrine cells in the innermost part of the
INL. These neurons send processes to both plexiform
layers, but while the network in the IPL is very dense,
only a few fibers reach the OPL.
The Cebus monkey displays IA amacrine cells in
the inner part of the INL, which send processes to
three sublayers in the IPL.7 Using identical methods
in the squirrel monkey, no significant indoleamine
uptake could be seen in any neurons in the inner part
of the retina, indicating an interesting species difference
between these two New World monkeys. There is always some slight uptake of indoleamine into dopaminergic neurons,3 explaining the pale yellow fluorescence of the INL dopaminergic neurons and the
Vol. 25
light labeling pattern in these neurons after long exposure times for autoradiography.
The difference in the number of IA horizontal cells
between central and peripheral retina is striking with
both autoradiography and fluorescence. In peripheral
retina, the labeled cell bodies are clearly seen to have
their long axis positioned horizontally, probably due
to a reduction in the overall packing density in the
periphery. This overall lower cell density also may
account for the smaller number of IA horizontal cells
in the periphery, but would not account for the lower
percentage of IA horizontal cells.
Both electron microscopic autoradiography and labeling with 5,6-DHT have shown that the horizontal
cell processes invaginate both cone pedicles and rod
spherules as lateral elements. The neurotoxic property
of 5,6-DHT28 produces ultrastructural changes that
have been used successfully to identify IA amacrine
cells in the rabbit, cat, and goldfish.13-2930 The most
characteristic change is the increased density of synaptic
vesicles; however, as most of the horizontal cell processes in the primate OPL lack conventional synaptic
vesicles and synapses,3132 they have been identified by
their content of altered cell organelles like mitochondria, and round electron-dense structures, presumably
lysosomes. With both 3H-serotonin and 5,6-DHT, the
horizontal cell processes contacting rods more frequently contain several altered organelles or multiple
silver grains in contrast to those processes contacting
cones. Grain counts show that rod synaptic spherules
are involved with heavily labeled horizontal cell processes approximately four times as often as cone synaptic pedicles. In peripheral retina where IA horizontal
cells are less common, the rods contacted by labeled
processes always lie in a group immediately above the
labeled horizontal cell body. This suggests that the
labeled processes in OPL are relatively short, which
Fig, 5. Retina from an
eye injected in vivo 4 hr
previously with 40 ng
5,6-DHT and fixed in
Zetterquist's fixative. A,
Degenerating cell organelles are seen in both lateral elements in one rod
(r) ribbon synapse. In the
other rod synapse, there
is an electron dense round
structure in one of the
lateral elements, possibly
indicating that the process is labeled (x27,500).
B, A process containing
round electron dense structures passes through the outer plexiform layer and ends as the lateral element in a cone (c) ribbon synapse. Note
the suggestion of a horizontal-horizontal cell synapse in the outer plexiform layer (arrow) in which both processes contain electron-dense
granules (X30.000).
B
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
No. 9
SQUIRREL MONKEY HORIZONTAL CELLS / Floren and Hendrickson
would be more characteristic of dendrites. The less
frequent synapses with cones might represent contacts
made by axon terminals.
The primate retina was long considered to have only
one morphologic type of horizontal cell,33"38 thus differing from all other vertebrate retinas where at least
two—and as many as four—different horizontal cells
have been described.39 Recently Kolb et al39 used EM
Golgi impregnation to identify a second type of horizontal cell in the Rhesus monkey retina. The dendrites
of both cell types contact cone pedicles, whereas the
axon terminals of the first type (HI) contact rod spherules, and those of the second type (HII) contact cone
pedicles. In the New World monkey retina, only the
first type of horizontal cell (HI) has been observed37
(Ogden, personal communication). The present results
suggest that the squirrel monkey retina must have more
than one physiologic type of horizontal cell because
only about one-half of the horizontal cells accumulated
indoleamines. In our study, it is possible that the distribution of the cell processes is confounded by a differential indoleamine uptake into dendrites and axon.
If only the axon labeled heavily, that would explain
our EM and autoradiographic observations and suggest
that the IA horizontal cell was the HI type of Kolb,
which has an axon distributed only to rods. On the
other hand, the high concentration of IA cell bodies
in the central retinal indicates a close relationship of
the IA cell to cones. As well, in the periphery, the close
proximity of rod spherules containing labeled horizontal processes and a single labeled cell body suggests
that relatively short processes, likely to be dendrites,
are the ones related to rods. This distribution would
suggest a new type of horizontal cell in squirrel monkey
retina, but serial reconstruction of EM autoradiographs
will be required to document fully the distribution of
this IA cell.
While serotonin is now considered to be the likely
transmitter of the IA neurons in cold-blooded animals
and birds, 23 the present experiments cast little new
light on the search for the true IA transmitter in the
mammalian retina. The uptake of 3H-serotonin is an
active process as demonstrated by its inhibition at 0°C,
and the fact that excess amounts of cold serotonin
inhibited this uptake shows that the process is saturable.
No apparent difference in uptake could be seen between
retinas of different adaptional state incubated under
different light conditions. As in other mammals, serotonin is not likely to be the native transmitter for
the IA horizontal cells because neither we nor Tornquist (personal communication) could observe any
immunoreactivity in the squirrel monkey retina with
a specific antiserum against serotonin, while brain tissue from the same animals stained heavily. The situation is different in cold-blooded animals and birds
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
1005
where immunoreactive serotonergic cells are present,2
and quantitative assays for serotonin show significant
amounts in frog and goldfish2-40"43 and lower amounts
in bird retina. 26 ' 4445 Much lower concentrations of
serotonin were found in the retina of rabbit, guinea
pig, pig, and COW2-64143-46 and, consistent with this
finding, no immunoreactive cells can be shown in these
species.21 These results argue against serotonin as a
transmitter in the mammalian retina. In the IA horizontal neurons of squirrel monkey, serotonin might
be an intermediate in an indole pathway leading to a
nonfluorescent substance, which is the final transmitter.
Melatonin and bufotonine are such substances but are
not taken up by these neurons, 4 which makes them
unlikely to be the final transmitter. A number of other
similar substances also have been ruled out, either due
to low intrinsic concentration or lack of uptake.1 Although the true transmitter of the IA horizontal cell
and other IA neurons in mammalian retina remains
elusive, it is likely to be a nonfluorescent indole or
closely related substance.
Key words: Squirrel monkey, horizontal cells, indoleamine,
fluorescence microscopy, autoradiography
Acknowledgment
The assistance of Alan MacKenzie, PhD, is gratefully acknowledged.
References
1. Ehinger B and Floren I: Retinal indoleamine-accumulating neurons. Neurochem Intl 1:209, 1980.
2. Tornquist K, Hansson C, and Ehinger B: Immunohistochemical
and quantitative analysis of 5-hydroxytryptamine in the retina
of some vertebrates. Neurochem Intl 5:299, 1983.
3. Ehinger B and Floren I: Indoleamine accumulating neurons in
the retina of rabbit, cat, and goldfish. Cell Tissue Res 175:37,
1976.
4. Ehinger B and Floren I: Quantitation of the uptake of indoleamines and dopamine in the rabbit retina. Exp Eye Res 26:1,
1978.
5. Floren I: Arguments against 5-hydroxytryptamine as neurotransmitter in the rabbit retina. J Neurotransm 46:1, 1979.
6. Floren I and Hansson HC: Investigations into whether 5-hydroxytryptamine is a neurotransmitter in the retina of rabbit
and chicken. Invest Ophthalmol Vis Sci 19:117, 1980.
7. Dowling JE, Ehinger B, and Floren I: Fluorescence and electron
microscopical observation on the amine accumulating neurons
of the Cebus monkey retina. J Comp Neurol 192:665, 1980.
8. Ehinger B and Floren I: Absence of indoleamine accumulating
neurons in the retina of humans and Cynomolgus monkeys.
Albrecht Von Graefes Arch KJin Exp Ophthalmol 209:145, 1979.
9. Ehinger B, Floren I, and Tornquist K: Autoradiography of
[3H]-5-hydroxytryptamine uptake in the retina of some mammals. Albrecht Von Graefes Arch Klin Exp Ophthalmol 218:1,
1982.
10. Falck B and Owman JC: A detailed methodological description
of the fluorescence method for the cellular demonstration of
biogenic amines. Acta Univ Lund (Section II) 7:1, 1965.
11. Bjorklund A, Falck B, and Owman C: Fluorescence microscope
1006
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1984
and microspectra fluorometric techniques for the cellular localization and characterization of biogenic amines. In Methods
of Investigative and Diagnostic Endocrinology, Rail JE and Kapin
I, editors. Amsterdam, North Holland Publication Co, 1972,
pp. 318-368.
Iversen LL and Callingham BA: Adrenergic transmission. In
Fundamentals of Biochemical Pharmacology, Bacq ZM, editor.
New York, Pergamon Press, 1971, pp. 253-305.
Ehinger B and Holmgren I: Electron microscopy of the indoleamine accumulating neurons in the rabbit retina. Cell Tissue
Res 197:175, 1979.
West RW and Dowling JE: Synapses onto different morphological
types of retinal ganglion cells. Science 178:510, 1972.
Preece A: Special tissue processing methods. In A Manual for
Histologic Technicians, 3rd edition. Boston, Little, Brown &
Co, 1972, pp. 114-140.
Hendrickson AE: Technical modifications to facilitate tracing
synapses by electron microscopic autoradiography. Brain Res
85:241, 1975.
Ames A: Studies of morphology, chemistry, and function in
isolated retina. In Biochemistry of the Retina, Graymore CN,
editor. London, Academic Press, 1965, pp. 22-30.
Shaskan EG and Snyder SH: Kinetics of serotonin accumulation
into slices from rat brain: relationship to catecholamine uptake.
J Pharmacol Exp Ther 175:404, 1970.
Laties AM and Jacobowitz D: A comparative study of the autonomic innervation of the eye in monkey, cat, and rabbit. Anat
Rec 156:383, 1966.
Ehinger B and Falck B: Morphological and pharmacohistochemical characteristics of adrenergic retinal neurons of some
mammals. Albrecht Von Graefes Arch Klin Exp Ophthalmol
178:295, 1969.
Ehinger B, Holmberg K, and Ohman P: Aminergic and indoleamine-accumulating neurons in the retina of the river lamprey
{Lampetra fluviatilis). Acta Zool (Stockholm) 58:117, 1977.
Adolph A, Dowling JE, and Ehinger B: Monoaminergic neurons
of the mudpuppy retina. Cell Tissue Res 210:269, 1980.
Tornquist K: 5-hydroxytryptamine and neuropeptides in the
retina. Thesis, Lund, 1983.
Haggendal J and Malmfors T: Identification and cellular localization of the catecholamines in the retina and the choroid of
the rabbit. Acta Physiol Scand 64:58, 1965.
Hauschild DC and Laties AM: An indoleamine-containing cell
in chick retina. Invest Ophthalmol 12:537, 1973.
Dowling JE and Ehinger B: Synaptic organization of the aminecontaining interplexiform cells of the goldfish and cebus monkey
retinas. Science 188:270, 1975.
Dowling JE, Ehinger B, and Heddan W: The interplexiform
cell: a new type of retinal neuron. Invest Ophthalmol 15:916,
1976.
Downloaded From: http://iovs.arvojournals.org/ on 06/17/2017
Vol. 25
28. Baumgarten HG and Bjorklund A: Neurotoxic indoleamines
and monoamine neurons. Ann Rev Pharmacol Toxicol 16:101,
1976.
29. Holmgren-Taylor I: Electron microscopical observations on the
indoleamine-accumulating neurons and their synaptic connections in the retina of the cat. J Comp Neurol 208:144, 1982.
30. Holmgren-Taylor I: Synaptic organization of the indoleamineaccumulating neurons in the cyprimid retina. Cell Tissue Res
229:317, 1982.
31. Dowling JE and Boycott BB: Organization of the primate retina:
electron microscopy. Proc Roy Soc Lond (Biol) 166:80, 1966.
32. Boycott BB and Dowling JE: Organization of the primate retina:
light microscopy. Proc Roy Soc Lond (Biol) 225:109, 1969.
33. Polyak SL: The Retina. Chicago, University of Chicago Press,
1941, pp. 257-261.
34. Kolb H: Organization of the outer plexiform layer of the primate
retina: Electron microscopy of Golgi-impregnated horizontal
neurons. Proc Roy Soc Lond (Biol) 258:261, 1970.
35. Gallego A: Horizontal and amacrine cells in the mammal's retina.
Vision Res(Suppl) 11:33, 1971.
36. Boycott BB and Kolb H: The horizontal cell of the rhesus monkey
retina. J Comp Neurol 148:115, 1973.
37. Ogden TE: The morphology of retinal neurons of the owl monkey
aotes. J Comp Neurol 153:399, 1974.
38. Gallego A and Sobrino JA: Short-axon horizontal cells of the
monkey's retina. Vision Res 15:747, 1975.
39. Kolb H, Mariani A, and Gallego A: A second type of horizontal
cell in the monkey retina. J Comp Neurol 189:31, 1980.
40. Welsh JH: The quantitative distribution of 5-hydroxytryptamine
in the nervous system, eyes, and other organs of some vertebrates.
In Comparative Neurochemistry, Proceedings of the Fifth International Neurochemistry Symposium, Richter D, editor. New
York, Pergamon Press, 1964, pp. 355-366.
41. Osborne NN: Evidence for serotonin being a neurotransmitter
in the retina. In Biology of Serotonergic Transmission, Osborne
NN, editor. Sussex, England, John Wiley & Sons, 1982, pp.
401-430.
42. Osborne NN, Nesselhut T, Nicholas DH, and Cuello AC: Serotonin: A transmitter candidate in the vertebrate retina. Neurochemistry Intl 3:171, 1981.
43. Osborne NN, Nesselhut T, Nicholas DA, Patel S, and Cuello
AC: Serotonin-containing neurons in vertebrate retinas. J Neurochem 39:1519, 1982.
44. Suzuki O, Naguchi E, Miyake S, and Yagi K: Occurrence of
5-hydroxytryptamine in chick retina. Experientia 33:927, 1977.
45. Osborne NN: Uptake, localization and release of serotonin in
the chick retina. J Physiol 331:469, 1982.
46. Osborne NN: In vitro experiments on the metabolism, uptake,
and release of 5-hydroxytryptamine in bovine retina. Brain Res
184:283, 1980.