- Email: [email protected]
Vasoactive intestinal polypeptide (VIP) immunohistochemistry in the rabbit retina
Brain Research. 426 (1987) 157-163 Elsevier
157
BRE 22570
Vasoactive intestinal polypeptide (VIP) immunohistochemistry in the rabbit retina Stephen M. Sagar Neurology Service, Veterans Administration Medical Center, and Department of Neurology, University of Califi)rnia, San Francisco, CA 94143 (U.S.A.) (Accepted 28 July 1987) Key words: Vasoactive intestinal polypeptide; Retina; Amacrine cell; Immunohistochemistry
A commercially obtained antiserum to vasoactive intestinal polypeptide (VIP) was used with the avidin-biotin-peroxidaseimmunohistochemical method to localize VIP-like immunoreactivity in the adult rabbit retina. A population of cell bodies in the inner nuclear layer is specifically stained. The cells are most dense in the central retina (40-50 cells/mm2), and least numerous in the superior periphery (15-20 cells/mm2). The dendritic fields overlap extensively; cells in the periphery have dendritic fields up to 1 mm in diameter. The processes of the cells are in the inner plexiform layer, where they tend to form a tristratified pattern in sublayers 1, 3 and 5. This pattern of lamination is similar to that described by other laboratories for dopaminergic amacrine cells and provides further evidence for an interaction between these two neuromodulators.
Vasoactive intestinal p o l y p e p t i d e (VIP) is a 28amino acid p e p t i d e widely distributed in central and peripheral nervous systems and in the gastrointestinal tract 17. VIP-like immunoreactivity (VIP-LI) has been detected immunohistochemically in retinal neurons of many species 3'6,1°'2°. The physiologic function of these cells is unknown; but in the retina and the brain, VIP directly stimulates adenylate cyclase and glycogenolysis9,14.t~,19. Tornqvist et al. 2° r e p o r t e d the immunohistochemical demonstration of V I P - L I in fibers of the inner plexiform layer (IPL) of the rabbit retina, but did not localize immunoreactive cell bodies or report on the detailed m o r p h o l o g y of the stained processes. As part of our efforts to study interactions between neuropeptides in rabbit retina, we further investigated the morphology of V I P - i m m u n o r e a c t i v e neurons. Adult, male, New Z e a l a n d white rabbits were obtained from a local b r e e d e r and maintained on a 12:12 h light:dark cycle with free access to food and water. Some rabbits received intravitreal injections of colchicine. Forty-eight h prior to sacrifice, these animals were sedated with p e n t o b a r b i t a l 40 mg, i.v.,
and the cornea and sclera were topically anesthetized with 1% proparicaine. Colchicine, 100 ug in 50 gl sterile saline, was injected via a 30-gauge needle over several seconds while moving the needle tip to multiple sites in the vitreous cavity. All rabbits were sacrificed by an intravenous injection of 150-25(/mg pentobarbital followed by bilateral thoracotomy. Immunohistochemistry in whole retinas was performed as described previously ~6, with modifications. The eyes were removed and hemisected about 2 mm posterior to the limbus. The anterior chamber and vitreous were discarded; and the posterior eye cups were immersed in freshly p r e p a r e d 2% paraformaldehyde, 0.1 M k-lysine, 0.01 M sodium m - p e r i o d a t e , 0.05 M sodium phosphate buffer, p H 7.4 (ref. 12). After immersion fixation at 4 °C for 2 - 4 h, the eye cups were washed overnight in 0.1 M sodium phosphate buffer, p H 7.4 (PB). The retinas were dissected and reacted with primary antiserum diluted in 0,1 M sodium phosphate buffer, pH 7.4, 0.2% Triton X-100, 0.1% bovine serum albumin, 2% normal goat serum (Solution A ) at 4 °C for 48-72 h. The retinas were then washed 3 times for a total of 30 rain PB and
Correspondence: S.M. Sagar, Neurology Service (127), VA Medical Center, 4150 Clement Street, San Francisco, CA 94121, U.S.A. 0006-8993/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
158
g
Fig. 1. Low power micrograph of retinal flat mount showing VIP-immunoreactive cells (in the INL and slightly out of the plane of focus) and their processes ramifying the IPL. This field is in the central retina in the region of the visual streak showing the most dense packing of VIP-immunoreactive neurons observed. Colchicine pretreated retina. Bar = 200 pm.
processed using an avidin-biotin-peroxidase (ABC) kit (Vectastain Kit, Vector Laboratories, Belmont, MA). Both biotinylated second antibody and avidinperoxidase complex were dissolved in Solution A; retinas were incubated in each for 2 - 4 h at room temperature. Diaminobenzidine was used as the chromagen, Some stained retinas were reacted with 1% osmium tetroxide in PB for I0 rain to intensify staining. Whole retinas were flat-mounted in 70% glycerol/30% PB. To obtain cross-sections, stained retinas were embedded in 4% agarose (Type VII, Sigma Chemical Co., St. Louis, MO) in PB and sectioned at 20 #m intervals on a Vibratome. Sections were dehydrated and mounted in Permount. Rabbit antiserum to porcine VIP was obtained from Immunonuclear Corp., Stillwater, MN (Antibody E2261, Lot 8622013), and was used at dilutions
ranging from 1/500 to 1/3,000. A dilution of 1/1,000 gave the best staining with acceptable background. Inclusion of synthetic VIP (1 #g/ml)during incubation with the primary antiserum eliminated all neuronal staining. In whole mounts, a population of round cell bodies in the inner nuclear layer (INL) and their processes in the inner plexiform layer (IPL) are stained (Figs. l-3). The cells are most numerous in the central retina (Fig. 4) and are less numerous towards the periphery. Colchicine pretreatment increases the intensity of staining of both cells and processes, but the distribution of immunostained cells is the same in colchicine-treated and untreated retinas. The mean cell body area is 51.1 _+ 6.7 m m 2 (mean +_ S.D., n = 16) centrally and 63.7 + 8.7 mm 2 (n = 17) peripherally in a fixed, glycerol-mounted flat-mount. Each cell gives off 2-5 primary dendrites which
4
©
: ~i ¸ ¸
~ N
~ i¸:;~,:j
....
S ~ ~!ii~ iI ¸ f i
i~
•
~
~iii~~
E
~!Mi!~I~I~I!P:I ~ i~
ii~~ ~ i
/ ~i~~ ~7~
E
F
160
A
t \
¢
ramify at multiple levels of the IPL (Fig. 2). The primary dendrites are smooth and tapering near the perikaryon and became varicose distally. At the superior periphery of the retina, where the cells are least densely packed and hence the dendrites of individual cells can be traced, the dendritic fields are at least 0.7-1 mm in diameter (Fig. 3). The fields of adjacent cells therefore overlap extensively !Fig. 1). No immunostained processes suggestive of axons are seen. Because of the thinness of the rabbit IPL, we were unable to ascertain the lamination of the dendrites in flat mounts. Since it was never possible to have all processes of a cell in focus simultaneously, it was clear that the processes are not unistratified (Fig. 2). In cross-sections (Fig. 5), immunoreactive fibers are seen at all levels of the IPL. Howe'~er, stained processes are most consistently seen in the outermost sublayer (sublayer 1) of the 1PL, and in many fields, as shown in Fig. 5, a tri-stratified pattern is suggested, with fibers in sublayers 1,3 and 5. This study has demonstrated VIP-LI in a population of either tri-stratified or diffuse amacrine cells of the rabbit retina. Like many cell types, they are more numerous and somewhat smaller centrally than peripherally. Their overlapping dendritic fields cover the retina with a plexus of fibers. Fhese findings are in agreement with the report of 'lornqvist et al.:", who reported VIP-immunoreactive fibers primarily within the outermost sublamina of the IPL in the rabbit. The relation of the morphology of the VIP-Ll-containing cells and their physiologic function is a matter
F
(N
E E
70 I-
~.
60-
~-
.
.
.
.
.
.
o 6 m m nasa to disc , through disc ~ 6 m m t e m p o r a to disr
50~
,3
f /
Q
O
20 ~
/
- 12--10
Fig. 3. Camera lucida drawings of 3 representative cells from the superior periphery of the retina. Because all of the processes cannot be reliably traced through the plexus of immunostained fibers in the IPL, these drawings probably underestimate the extent and complexity of the dendritic arbors. Bars = 100#m.
ti':
,
-8
--6
-4
-2
0
2
4
6
8
VERTICAL DISTANCL, rnm
Fig. 4. Graph of the density of VlP-immunoreactive cell bodies along 3 vertical scans of the retina. Cells in 0.4-ram squares were counted at 10× magnification in i mm increments. The gaps in the lines at the level of the disc (0 mm) represents the myelinated fiber bundle, which obscures underlying retina.
161
Fig. 5, Cross-sections of retinas showing VIP-immunoreactive cells in the innermost row of perikarya of the INL and their processes ramifying in the IPL. The photoreceptors are at the top of each photograph and the ganglion cell layer at the bottom. Arrows indicate the layers of immunostained processes in sublayers 1, 3 and 5 of the IPL. The clark staining of photoreceptor outer segments and the mottled staining in the ganglion cell layer is non-specific. Bar = 50/~m.
I (~2
of speculation. V1P has been shown to stimulate cyclic adenosine monophosphate (cAMP) synthesis and glycogenolysis in the rabbit retina 14'Is't9. Since
have been extensively studied in rabbit retina. Dopamine-containing cells are of particular interest, since their processes, like those of the VIP immunoreactive neurons, are predominantly in sub!ayers 1, 3 and 5 of the IPL in the rabbit. This pattern has been described by Dowling and Ehinger 5 using histofluorescence, and by Brecha et ai. 4 using tyrosine hydroxylase immunohistochemistry. The lamination is distinctly different from that described for cholinergic 11 or G A B A e r g i c 2 amacrine cells. However, the overall appearance of the VIP-LI cells in whole mounts is different than that illustrated for tyrosine hydroxylase immunoreactive neurons in rabbit retina 5. In particular, the latter cells have fewer primary dendrites, all of which appear to lie in sublayer 1 of the IPL. The primary dendrites of the VIPLI cells are in multiple sublayers of the IPL (Fig. 2). Therefore, although no double label data is awfilable, the evidence is against co-localization of VIP and dopamine within the same population of neurons. Of note, Pachter and Lam ~4 have produced pharmacologic evidence that dopamine and VIP stimulate a c o m m o n adenylate cyclase in rabbit retina. This implies that both dopamine and VIP act on the same target cell. The results presented here provide a morphological substrate for such an interaction.
retinal glycogen is contained primarily within Muller cells s, VIP may act directly on Muller cells, as has been demonstrated in the chick retina 7. To the author's knowledge, ultrastructural studies have not demonstrated specialized contacts between amacrine cell processes and Muller cells, so this hypothesis implies that VIP acts in a paracrine fashion. A relatively diffuse projection of processes in the IPL is consistent with this hypothesis. However, c A M P is known to have a multitude of effects in neurons as well as gila ~3. Therefore, it is also possible that the V1P-immunoreactive cells modulate neuronal as well as glial function. Since there are immunoreactive processes in both inner and outer sublaminas of the IPL, one would predict that these are On-Off cells electrophysiologically ~. Many neurotransmitter candidates have been localized to neurons of mammalian retina 3. Immunohistochemical studies of neuropeptides in the rabbit retina have not been numerous, however. Somatostatin-like immunoreactivity is present in cells of the ganglion cell layer; these cells have intraretinal axons, so are classed as associational ganglion cells (ref. 16 and unpublished data). Also, Rickman et al. 15 have reported immunohistochemical evidence for substance P localization to classical ganglion cells, associational ganglion cells, and amacrine cells of the rabbit. The small molecular transmitters dopamine, ~-aminobutyric acid ( G A B A ) , and acetylcholine
The author thanks Ms: Karen Baner and Dr: Rao Pillarisetty for their expert technical assistance. This work was supported by the National Eye Institute (EY05721) and by the Veterans Administration,
i Bloomfield, S.A. and Miller, R.F,, A functional organization of On and Off pathways in the rabbit retina, J~ Neurosci., 6 (1986) 1-13. 2 Brandon, C., Lain, D.M.K. and Wu, J.-Y., The 7-aminobutyric acid system in the rabbit retina: localization by immunocytochemistry and autoradiography, Proc. Natl. Acad. Sci. (U,S.A.), 76 (1979) 3557-3561. 3 Brecha, N., Retinal neurotransmitters. In P.C. Emson (Ed.), Chemical Neuroanatomy, Raven, New York, 1983, pp. 85-130. 4 Brecha, N.C., Oyster, C.W. and Takahashi, E.S., Identification and characterization of tyrosine hydroxylase and immunoreactive amacrine cells, Invest. Ophthalmol. Vis. Sci., 25 (1984) 66-70. 5 Dowling, J.E. and Ehinger, B., Synaptic organization of the dopaminergic neurons in the rabbit retina, J. Comp.
Neurol., 180 (1978) 203-220. 6 Fukuda, M., Localization of neuropeptides in the avian retina; an immunohistochemical analysis, Cell. Mot. Biol., 28 (1982) 275-283. 7 Koh, S.-W.M., Kyritsis, A. and Chader, G.J., Interaction of neuropeptides and cultured glial (Muller) cells of the chick retina: elevation of intraceltular cyclic AMP by vasoactive intestinal peptide and glucagon, J. Neurochem., 43 (1984) 199-203. 8 Kuwabara, T. and Cogan, D.G., Retinal glycogen, Arch. Ophthalmpl., 66 (1961) 680-688. 9 Longshore, M.A. and Makman, M.H., Stimulation of retinal adenylate cyclase by vasoactive intestinal peptide (VIP), Eur. J. Biochem., 70 (1981) 237-240. 10 Loren, I., Tornqvist, K. and Alumets, J., VIP (vasoactive intestinal polypeptide)-immunoreactive neurons in the reti-
163 na of the rat, Cell Tissue Res., 210 (1980) 167-1270. 11 Masland, R.H. and Mills, J.W., Autoradiographic identification of acetylcholine in the rabbit retina, J. Cell Biol., 83 (1979) 159-178. 12 McLean, I.W. and Nakane, P.K., Periodate-lysine-paraformaldehyde fixative. A new fixative for immunoelectron microscopy, J. Histochem. Cytochem., 12 (1974) 1077-1083. 13 Nestler, E.J. and Greengard, P,, Protein Phosphorylation in the Nervous System, Wiley, New York, 1984. 14 Pachter, J.A. and Lain, D.M.-K., Interactions between vasoactive intestinal peptide and dopamine in the rabbit retina: stimulation of a common adenylate cyclase, J. Neurochem., 46 (1986) 257-264. 15 Rickman, D., Johnson, D., Sharva, S. and Brecha, N., Distribution of substance P and somatostatin immunoreactive
16
17 18
19
20
cells in the rabbit retina, Soc. Neurosci. (Abstr.) 16 (1986) 641. Sagar, S.M., Marshall, P.E., Onesti, S.T. and Landis, D.M.D., Somatostatin immunocytochemistry in the rabbit retina, Invest. Ophthalmol. Vis. Sci., 27 (1986) 316-322. Said, S.I., Vasoactive intestinal polypeptide (VIP); current status, Peptides, 5 (1984) 143-150. Schorderet, M., Hof, P. and Magistretti, P.J.. The effects of VIP on cyclic AMP and glycogen levels in vertebrate retina, Peptides, 5 (1984) 295-298. Schorderet, M., Sovilla, J.-Y. and Magistretti, P.J., VIPand glucagon-induced formation of cyclic AMP in intact retinae in vitro, Eur. J. Pharmacol., 71 (1981) 131-133. Tornqvist, K., Uddman, R., Sundler. F. and Ehinger, B., Somatostatin and VIP neurons in the retina of different species, Histochemistrv, 76 (1982) 137-152.
157
BRE 22570
Vasoactive intestinal polypeptide (VIP) immunohistochemistry in the rabbit retina Stephen M. Sagar Neurology Service, Veterans Administration Medical Center, and Department of Neurology, University of Califi)rnia, San Francisco, CA 94143 (U.S.A.) (Accepted 28 July 1987) Key words: Vasoactive intestinal polypeptide; Retina; Amacrine cell; Immunohistochemistry
A commercially obtained antiserum to vasoactive intestinal polypeptide (VIP) was used with the avidin-biotin-peroxidaseimmunohistochemical method to localize VIP-like immunoreactivity in the adult rabbit retina. A population of cell bodies in the inner nuclear layer is specifically stained. The cells are most dense in the central retina (40-50 cells/mm2), and least numerous in the superior periphery (15-20 cells/mm2). The dendritic fields overlap extensively; cells in the periphery have dendritic fields up to 1 mm in diameter. The processes of the cells are in the inner plexiform layer, where they tend to form a tristratified pattern in sublayers 1, 3 and 5. This pattern of lamination is similar to that described by other laboratories for dopaminergic amacrine cells and provides further evidence for an interaction between these two neuromodulators.
Vasoactive intestinal p o l y p e p t i d e (VIP) is a 28amino acid p e p t i d e widely distributed in central and peripheral nervous systems and in the gastrointestinal tract 17. VIP-like immunoreactivity (VIP-LI) has been detected immunohistochemically in retinal neurons of many species 3'6,1°'2°. The physiologic function of these cells is unknown; but in the retina and the brain, VIP directly stimulates adenylate cyclase and glycogenolysis9,14.t~,19. Tornqvist et al. 2° r e p o r t e d the immunohistochemical demonstration of V I P - L I in fibers of the inner plexiform layer (IPL) of the rabbit retina, but did not localize immunoreactive cell bodies or report on the detailed m o r p h o l o g y of the stained processes. As part of our efforts to study interactions between neuropeptides in rabbit retina, we further investigated the morphology of V I P - i m m u n o r e a c t i v e neurons. Adult, male, New Z e a l a n d white rabbits were obtained from a local b r e e d e r and maintained on a 12:12 h light:dark cycle with free access to food and water. Some rabbits received intravitreal injections of colchicine. Forty-eight h prior to sacrifice, these animals were sedated with p e n t o b a r b i t a l 40 mg, i.v.,
and the cornea and sclera were topically anesthetized with 1% proparicaine. Colchicine, 100 ug in 50 gl sterile saline, was injected via a 30-gauge needle over several seconds while moving the needle tip to multiple sites in the vitreous cavity. All rabbits were sacrificed by an intravenous injection of 150-25(/mg pentobarbital followed by bilateral thoracotomy. Immunohistochemistry in whole retinas was performed as described previously ~6, with modifications. The eyes were removed and hemisected about 2 mm posterior to the limbus. The anterior chamber and vitreous were discarded; and the posterior eye cups were immersed in freshly p r e p a r e d 2% paraformaldehyde, 0.1 M k-lysine, 0.01 M sodium m - p e r i o d a t e , 0.05 M sodium phosphate buffer, p H 7.4 (ref. 12). After immersion fixation at 4 °C for 2 - 4 h, the eye cups were washed overnight in 0.1 M sodium phosphate buffer, p H 7.4 (PB). The retinas were dissected and reacted with primary antiserum diluted in 0,1 M sodium phosphate buffer, pH 7.4, 0.2% Triton X-100, 0.1% bovine serum albumin, 2% normal goat serum (Solution A ) at 4 °C for 48-72 h. The retinas were then washed 3 times for a total of 30 rain PB and
Correspondence: S.M. Sagar, Neurology Service (127), VA Medical Center, 4150 Clement Street, San Francisco, CA 94121, U.S.A. 0006-8993/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
158
g
Fig. 1. Low power micrograph of retinal flat mount showing VIP-immunoreactive cells (in the INL and slightly out of the plane of focus) and their processes ramifying the IPL. This field is in the central retina in the region of the visual streak showing the most dense packing of VIP-immunoreactive neurons observed. Colchicine pretreated retina. Bar = 200 pm.
processed using an avidin-biotin-peroxidase (ABC) kit (Vectastain Kit, Vector Laboratories, Belmont, MA). Both biotinylated second antibody and avidinperoxidase complex were dissolved in Solution A; retinas were incubated in each for 2 - 4 h at room temperature. Diaminobenzidine was used as the chromagen, Some stained retinas were reacted with 1% osmium tetroxide in PB for I0 rain to intensify staining. Whole retinas were flat-mounted in 70% glycerol/30% PB. To obtain cross-sections, stained retinas were embedded in 4% agarose (Type VII, Sigma Chemical Co., St. Louis, MO) in PB and sectioned at 20 #m intervals on a Vibratome. Sections were dehydrated and mounted in Permount. Rabbit antiserum to porcine VIP was obtained from Immunonuclear Corp., Stillwater, MN (Antibody E2261, Lot 8622013), and was used at dilutions
ranging from 1/500 to 1/3,000. A dilution of 1/1,000 gave the best staining with acceptable background. Inclusion of synthetic VIP (1 #g/ml)during incubation with the primary antiserum eliminated all neuronal staining. In whole mounts, a population of round cell bodies in the inner nuclear layer (INL) and their processes in the inner plexiform layer (IPL) are stained (Figs. l-3). The cells are most numerous in the central retina (Fig. 4) and are less numerous towards the periphery. Colchicine pretreatment increases the intensity of staining of both cells and processes, but the distribution of immunostained cells is the same in colchicine-treated and untreated retinas. The mean cell body area is 51.1 _+ 6.7 m m 2 (mean +_ S.D., n = 16) centrally and 63.7 + 8.7 mm 2 (n = 17) peripherally in a fixed, glycerol-mounted flat-mount. Each cell gives off 2-5 primary dendrites which
4
©
: ~i ¸ ¸
~ N
~ i¸:;~,:j
....
S ~ ~!ii~ iI ¸ f i
i~
•
~
~iii~~
E
~!Mi!~I~I~I!P:I ~ i~
ii~~ ~ i
/ ~i~~ ~7~
E
F
160
A
t \
¢
ramify at multiple levels of the IPL (Fig. 2). The primary dendrites are smooth and tapering near the perikaryon and became varicose distally. At the superior periphery of the retina, where the cells are least densely packed and hence the dendrites of individual cells can be traced, the dendritic fields are at least 0.7-1 mm in diameter (Fig. 3). The fields of adjacent cells therefore overlap extensively !Fig. 1). No immunostained processes suggestive of axons are seen. Because of the thinness of the rabbit IPL, we were unable to ascertain the lamination of the dendrites in flat mounts. Since it was never possible to have all processes of a cell in focus simultaneously, it was clear that the processes are not unistratified (Fig. 2). In cross-sections (Fig. 5), immunoreactive fibers are seen at all levels of the IPL. Howe'~er, stained processes are most consistently seen in the outermost sublayer (sublayer 1) of the 1PL, and in many fields, as shown in Fig. 5, a tri-stratified pattern is suggested, with fibers in sublayers 1,3 and 5. This study has demonstrated VIP-LI in a population of either tri-stratified or diffuse amacrine cells of the rabbit retina. Like many cell types, they are more numerous and somewhat smaller centrally than peripherally. Their overlapping dendritic fields cover the retina with a plexus of fibers. Fhese findings are in agreement with the report of 'lornqvist et al.:", who reported VIP-immunoreactive fibers primarily within the outermost sublamina of the IPL in the rabbit. The relation of the morphology of the VIP-Ll-containing cells and their physiologic function is a matter
F
(N
E E
70 I-
~.
60-
~-
.
.
.
.
.
.
o 6 m m nasa to disc , through disc ~ 6 m m t e m p o r a to disr
50~
,3
f /
Q
O
20 ~
/
- 12--10
Fig. 3. Camera lucida drawings of 3 representative cells from the superior periphery of the retina. Because all of the processes cannot be reliably traced through the plexus of immunostained fibers in the IPL, these drawings probably underestimate the extent and complexity of the dendritic arbors. Bars = 100#m.
ti':
,
-8
--6
-4
-2
0
2
4
6
8
VERTICAL DISTANCL, rnm
Fig. 4. Graph of the density of VlP-immunoreactive cell bodies along 3 vertical scans of the retina. Cells in 0.4-ram squares were counted at 10× magnification in i mm increments. The gaps in the lines at the level of the disc (0 mm) represents the myelinated fiber bundle, which obscures underlying retina.
161
Fig. 5, Cross-sections of retinas showing VIP-immunoreactive cells in the innermost row of perikarya of the INL and their processes ramifying in the IPL. The photoreceptors are at the top of each photograph and the ganglion cell layer at the bottom. Arrows indicate the layers of immunostained processes in sublayers 1, 3 and 5 of the IPL. The clark staining of photoreceptor outer segments and the mottled staining in the ganglion cell layer is non-specific. Bar = 50/~m.
I (~2
of speculation. V1P has been shown to stimulate cyclic adenosine monophosphate (cAMP) synthesis and glycogenolysis in the rabbit retina 14'Is't9. Since
have been extensively studied in rabbit retina. Dopamine-containing cells are of particular interest, since their processes, like those of the VIP immunoreactive neurons, are predominantly in sub!ayers 1, 3 and 5 of the IPL in the rabbit. This pattern has been described by Dowling and Ehinger 5 using histofluorescence, and by Brecha et ai. 4 using tyrosine hydroxylase immunohistochemistry. The lamination is distinctly different from that described for cholinergic 11 or G A B A e r g i c 2 amacrine cells. However, the overall appearance of the VIP-LI cells in whole mounts is different than that illustrated for tyrosine hydroxylase immunoreactive neurons in rabbit retina 5. In particular, the latter cells have fewer primary dendrites, all of which appear to lie in sublayer 1 of the IPL. The primary dendrites of the VIPLI cells are in multiple sublayers of the IPL (Fig. 2). Therefore, although no double label data is awfilable, the evidence is against co-localization of VIP and dopamine within the same population of neurons. Of note, Pachter and Lam ~4 have produced pharmacologic evidence that dopamine and VIP stimulate a c o m m o n adenylate cyclase in rabbit retina. This implies that both dopamine and VIP act on the same target cell. The results presented here provide a morphological substrate for such an interaction.
retinal glycogen is contained primarily within Muller cells s, VIP may act directly on Muller cells, as has been demonstrated in the chick retina 7. To the author's knowledge, ultrastructural studies have not demonstrated specialized contacts between amacrine cell processes and Muller cells, so this hypothesis implies that VIP acts in a paracrine fashion. A relatively diffuse projection of processes in the IPL is consistent with this hypothesis. However, c A M P is known to have a multitude of effects in neurons as well as gila ~3. Therefore, it is also possible that the V1P-immunoreactive cells modulate neuronal as well as glial function. Since there are immunoreactive processes in both inner and outer sublaminas of the IPL, one would predict that these are On-Off cells electrophysiologically ~. Many neurotransmitter candidates have been localized to neurons of mammalian retina 3. Immunohistochemical studies of neuropeptides in the rabbit retina have not been numerous, however. Somatostatin-like immunoreactivity is present in cells of the ganglion cell layer; these cells have intraretinal axons, so are classed as associational ganglion cells (ref. 16 and unpublished data). Also, Rickman et al. 15 have reported immunohistochemical evidence for substance P localization to classical ganglion cells, associational ganglion cells, and amacrine cells of the rabbit. The small molecular transmitters dopamine, ~-aminobutyric acid ( G A B A ) , and acetylcholine
The author thanks Ms: Karen Baner and Dr: Rao Pillarisetty for their expert technical assistance. This work was supported by the National Eye Institute (EY05721) and by the Veterans Administration,
i Bloomfield, S.A. and Miller, R.F,, A functional organization of On and Off pathways in the rabbit retina, J~ Neurosci., 6 (1986) 1-13. 2 Brandon, C., Lain, D.M.K. and Wu, J.-Y., The 7-aminobutyric acid system in the rabbit retina: localization by immunocytochemistry and autoradiography, Proc. Natl. Acad. Sci. (U,S.A.), 76 (1979) 3557-3561. 3 Brecha, N., Retinal neurotransmitters. In P.C. Emson (Ed.), Chemical Neuroanatomy, Raven, New York, 1983, pp. 85-130. 4 Brecha, N.C., Oyster, C.W. and Takahashi, E.S., Identification and characterization of tyrosine hydroxylase and immunoreactive amacrine cells, Invest. Ophthalmol. Vis. Sci., 25 (1984) 66-70. 5 Dowling, J.E. and Ehinger, B., Synaptic organization of the dopaminergic neurons in the rabbit retina, J. Comp.
Neurol., 180 (1978) 203-220. 6 Fukuda, M., Localization of neuropeptides in the avian retina; an immunohistochemical analysis, Cell. Mot. Biol., 28 (1982) 275-283. 7 Koh, S.-W.M., Kyritsis, A. and Chader, G.J., Interaction of neuropeptides and cultured glial (Muller) cells of the chick retina: elevation of intraceltular cyclic AMP by vasoactive intestinal peptide and glucagon, J. Neurochem., 43 (1984) 199-203. 8 Kuwabara, T. and Cogan, D.G., Retinal glycogen, Arch. Ophthalmpl., 66 (1961) 680-688. 9 Longshore, M.A. and Makman, M.H., Stimulation of retinal adenylate cyclase by vasoactive intestinal peptide (VIP), Eur. J. Biochem., 70 (1981) 237-240. 10 Loren, I., Tornqvist, K. and Alumets, J., VIP (vasoactive intestinal polypeptide)-immunoreactive neurons in the reti-
163 na of the rat, Cell Tissue Res., 210 (1980) 167-1270. 11 Masland, R.H. and Mills, J.W., Autoradiographic identification of acetylcholine in the rabbit retina, J. Cell Biol., 83 (1979) 159-178. 12 McLean, I.W. and Nakane, P.K., Periodate-lysine-paraformaldehyde fixative. A new fixative for immunoelectron microscopy, J. Histochem. Cytochem., 12 (1974) 1077-1083. 13 Nestler, E.J. and Greengard, P,, Protein Phosphorylation in the Nervous System, Wiley, New York, 1984. 14 Pachter, J.A. and Lain, D.M.-K., Interactions between vasoactive intestinal peptide and dopamine in the rabbit retina: stimulation of a common adenylate cyclase, J. Neurochem., 46 (1986) 257-264. 15 Rickman, D., Johnson, D., Sharva, S. and Brecha, N., Distribution of substance P and somatostatin immunoreactive
16
17 18
19
20
cells in the rabbit retina, Soc. Neurosci. (Abstr.) 16 (1986) 641. Sagar, S.M., Marshall, P.E., Onesti, S.T. and Landis, D.M.D., Somatostatin immunocytochemistry in the rabbit retina, Invest. Ophthalmol. Vis. Sci., 27 (1986) 316-322. Said, S.I., Vasoactive intestinal polypeptide (VIP); current status, Peptides, 5 (1984) 143-150. Schorderet, M., Hof, P. and Magistretti, P.J.. The effects of VIP on cyclic AMP and glycogen levels in vertebrate retina, Peptides, 5 (1984) 295-298. Schorderet, M., Sovilla, J.-Y. and Magistretti, P.J., VIPand glucagon-induced formation of cyclic AMP in intact retinae in vitro, Eur. J. Pharmacol., 71 (1981) 131-133. Tornqvist, K., Uddman, R., Sundler. F. and Ehinger, B., Somatostatin and VIP neurons in the retina of different species, Histochemistrv, 76 (1982) 137-152.