Muscarinic cholinergic receptors in the retina of the larval tiger salamander

Brain Research, 340 (1985) 355-362 Elsevier

355

BRE 10908

Muscarinic Cholinergic Receptors in the Retina of the Larval Tiger Salamander ARTHUR S. POLANS, JAMES B. HUTCHINS and FRANK S. WERBLIN Electronics" Research Laboratory, Graduate Group in Neurobiology and the Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA 94720 (U.S.A.) (Accepted November 6th, 1984) Key words: muscarinic cholinergic receptors - - retinal transmitter - - retinal biochemistry and histochemistry

The pharmacology and autoradiographic localization of muscarinic cholinergic receptors in retinal slices of the larval tiger salamander have been examined using the muscarinic antagonist [3H]propylbenzilylcholine mustard. Under the conditions of these experiments the binding of this ligand is irreversible. Saturation and maximum specific binding of 270 pM of ligand per gram protein are observed after an incubation of 1 h, and autoradiographic studies show that this does not reflect surface binding alone. Muscarinic but not nicotinic drugs suppress the binding of propylbenzilylcholine mustard at physiologically relevant concentrations: half-maximal suppression of binding by the muscarinic antagonists atropine sulfate and quinuclidinyl benzilate occurs, respectively, at 9.0 and 7.5 x 10-HJ M. Light microscopic autoradiography reveals the discrete localization of the ligand to the sites of synaptic contact, the retinal plexiform layers, predominantly the inner plexiform layer. The implications of the present study on current theories of cholinergic function in the vertebrate retina are discussed. INTRODUCTION

probably in association with a subset of cholinergic amacrine cells 19,22.

The examination of chemical transmitters has become a useful aid in identifying connections between cells in complex nervous tissue. Owing to the variety of specific probes available, acetylcholine ( A C h ) especially has been evaluated as a n e u r o t r a n s m i t t e r and n e u r o m o d u l a t o r in the p e r i p h e r a l and central nervous systems, allowing details of synaptic connections to be established. The suggestion that A C h acts as a n e u r o t r a n s m i t t e r in the v e r t e b r a t e retina has been derived from an assortment of electrophysiological 1.3,~, biochemical 5,14.23, histochemical 2 and autoradiographic2S, 35 experiments. The retina contains mechanisms for the high-affinity u p t a k e of choline 5,14,23, its incorporation into A C h 5,14,23 and its subsequent release 5,18. This l a b o r a t o r y has demonstrated that application of A C h appears to interfere with coupling between horizontal cells in the salam a n d e r retina r , and these cells are the p r e d o m i n a n t site of acetylcholinesterase (ACHE) activity in the outer plexiform layer ~1. A C h E also has been localized to the inner plexiform layer of most retinas 2,s,zo,

Cholinergic receptors have been localized at the light and electron microscopic levels in both retinal plexiform layers 24,26,28,30.31,34,35. Virtually all of these studies have e x a m i n e d the distribution of nicotinic cholinergic receptors using a snake toxin, a - b u n g a r o toxin. With the exception of a single study in chick retina 28, little is known about the distribution of muscarinic cholinergic receptors, although their presence in a variety of retinae has been ascertained from binding studieslO,21. In these studies the muscarinic antagonist quinuclidinyl benzilate ( Q N B ) was used: unfortunately this ligand binds reversibly and cannot be fixed by conventional means, which necessitates the use of frozen tissue and more e l a b o r a t e sectioning techniques in o r d e r to limit diffusion artifacts. These difficulties have been circumvented in the present study by using propylbenzilylcholine mustard (PrBCM), a specific, essentially irreversible muscarinic antagonist which can be used with tissue subsequently fixed with glutaraldehyde 7,~3,25. Determining the distribution of muscarinic recep-

Correspondence: F. S. Werblin. Present address: Graduate Group in Neurobiology, 253 Cory Hall, University of California, Berkeley, CA 94720, U.S.A. 0006-8993/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

356 tors is important, since muscarinic antagonists are known to block cone to horizontal cell transmission in turtle 9 and salamander 27 retinae. The decrease in membrane conductance of horizontal cells in salamander retina associated with the uncoupling action of ACh 17 also is consistent with the presence of muscarinic rather than nicotinic cholinergic receptors. The outer plexiform layer, however, may not be the only site of ACh action. Electrotonic and dye coupling of the sort observed with horizontal cells also occur between amacrine cells ~2. Gap junctions, which form the morPhological basis for such coupling, have been found between amacrine cells in the salamander retina 33. The physiological effect of ACh upon the syncytium of amacrine cells has not been examined. Continuing our studies on the actions of ACh in the salamander retina, we now have confirmed the presence of muscarinic receptors and have examined their distribution by light microscopy and autoradiography. Further studies of cholinergic receptor localization at the ultrastructural level will help verify connections between subsets of cells in the salamander retina. MATERIALS AND METHODS

Preparation of retinal slices. Larval tiger salamanders (Ambystoma tigrinium) were kept in holding tanks at 4 °C and exposed to a 12 h light-dark cycle. Animals were sacrificed during the dark cycle; this and the subsequent dissection were performed under dim red illumination. Retinal slices were prepared as described previously 32. Briefly, after enucleation, the cornea, iris and lens were removed in succession. A small portion of the remaining eyecup was cut and placed retina side down on a piece of Millipore filter paper (type HA) which had been attached to a glass microscope slide. The sclera was removed, leaving the retina with its ganglion cell layer adhering to the filter paper. The retina was covered with a preoxygenated Ringer's solution containing (in mM): 104 NaCI, 2.1 KCI, 3.6 CaCI 2, 5 Na2HPO4, 4 MgCI2, 3 MgSO 4, 5 glucose and 5 HEPES, adjusted to pH 7.5. Slices of retina and filter paper were made with a razor blade mounted in a tissue chopper. Individual slices could be moved under solution by manipulating the adherent filter paper with fine forceps.

Slices were anchored between two rows ~1 Vaseline and rotated 90° so that the retinal layers could be visualized in a light microscope. In this fashion, slices 100-150/~m thick and 1-2 mm long were prepared reproducibly. Such slices contain physiologically viable cells4,32 and provide a preparation essentially free of surface diffusion barriers. Activation ofpH]PrBCM. [3H]PrBCM is supplied as an inactive species in an ethanolic solution. The active species is the aziridinium ion formed in aqueous solution near pH 7. In the present experiments [3H]PrBCM was cyclized at a concentration of 2.9/~M in 10 mM sodium phosphate buffer, pH 7.4, at room temperature for 1 h prior to incubation with retinal slices. An aliquot of this solution was added to Ringer's solution to yield the appropriate concentration. [3H]PrBCM binding to retinal slices. In each experiment approximately 6 retinal slices were incubated in a Ringer's solution containing [3H]PrBCM to yield a final concentration of 2.9 nM. Incubations were terminated by the rapid exchange of normal Ringer's solution with one containing 10 mM sodium thiosulfate, pH 7.4. which quenches the active, aziridinium form of PrBCM. Slices were perfused with this solution for 30 rain to remove unbound ligand. Each retinal slice was removed from its filter paper support and dissolved overnight at room temperature in 1 ml Soluene 350. The following day 10 ml of 0.7°A 2,5-diphenyloxazole (PPO) -0.06% 2.2'(1,4-phenylene)bis(4-methyl)5-phenytoxazole t DmPOPOP) -10% Soluene 350 in toluene were added to each sample and counted in a Searle Mark IlI scintillation counter. In experiments with cholinerglc competitors, the various drugs in Ringer's solution were added to the retinal slices 30 min prior to the addition of [3H]PrBCM and remained throughoul the incubation. Protein determination. In a separate series of experiments retinal slices were viewed with a light microscope, and the dimensions of each slice measured with a calibrated micrometer eyepiece. The volume of each slice thus could be determined. Slices then were pooled, and their protein content measured after alkaline hydrolysis. Protein determinations were made either by the method of Lowry 16 or using a dye-binding assay6 after readjusting the pH with

357 acid. Both m e t h o d s gave c o m p a r a b l e results using ovalbumin as a standard. F r o m these experiments a p r o t e i n : v o l u m e ratio was obtained. In subsequent binding experiments the volume of each retinal slice was ascertained and t h e r e b y converted into an estimate of protein concentration.

Light microscopy and autoradiography of [3H]PrBCM binding. Incubation conditions of retinal slices for light microscopy and a u t o r a d i o g r a p h y were the same as described for binding experiments through the perfusion step with sodium thiosulfate. Incubation with [3H]PrBCM was for 1 h. Retinal slices were fixed for 1 - 2 h at 4 °C with 2% glutarald e h y d e - l % OsO~ in 0.1 M sodium p h o s p h a t e buffer, pH 7.4. A f t e r washing with p h o s p h a t e buffer, samples were d e h y d r a t e d in an ethanol series, cleared in p r o p y l e n e oxide and e m b e d d e d in A r a l d i t e - P o l y b e d 812 resin. Sections 2 g m thick were dried onto ethanol-clean-

ed glass slides and coated in total darkness at 40 °C with K o d a k NTB3 diluted 1:1 with filtered, distilled water. A f t e r drying in a humid c h a m b e r for 2 h, slides were sealed with dessicant in containers and exposed at 5 °C for 1 - 4 weeks. Slides were develo p e d for 4 min at 16 °C with K o d a k D-19 (1:1, v/v). stopped with distilled water and fixed with a nonrapid K o d a k fixer for 5 min. Tissue was stained with toluidene blue and viewed for p h o t o g r a p h y with an Olympus light microscope. [3H]PrBCM was purchased from New England Nuclear, Boston, M A . QNB was a generous gift from H o f f m a n - L a R o c h e , Nutley, NJ. Resin c o m p o n e n t s were obtained from Polysciences, Warrington, PA. ct-Bungarotoxin was purchased from Boehringer Mannheim Biochemicals, Indianapolis, IN. Soluene 350 was obtained from Mallinckrodt, Los Angeles, C A . The remaining chemicals were o b t a i n e d from Sigma, St. Louis, MO.

I

!

OO i,,¢

I "¢- /

I-2mm Fig. 1. Retinal slice preparation. As described in Materials and Methods, retinal slices 1-2 mm long and 100-150 ~tm wide were used throughout this study. Depicted in this figure is a portion of a living retinal slice viewed with a 40x water immersion objective modified with Hoffman modulation contrast optics. The schematic illustrates more clearly the various retinal layers and cell types. OS and IS, outer segment and inner segment; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer: IPL, inner plexiform layer; GCL, ganglion cell layer; R, rod photoreceptor; C, cone photoreceptor', H, horizontal cell; B, bipolar cell: A, amacrine cell; G ganglion cell. x 182.

358 RESULTS

A 400

Fig. 1 illustrates a portion of a living retinal slice and a corresponding diagram depicting the various retinal layers and the cell types which comprise them. Slices are viewed with a 40x water immersion objective modified with Hoffman modulation contrast optics in order to accentuate cell edges for electrophysiological recordings 4,32. Unlike eyecup preparations 21, retinal slices present no discernable surface diffusion barriers (as will be demonstrated); therefore, pharmacological agents are effective at low, physiologically relevant concentrations. Retinal slices incubated with [3H]PrBCM demonstrate atropine-sensitive binding (Fig. 2). In these experiments retinal slices were incubated for various periods of time with 2.9 nM [3H]PrBCM either in the absence or presence of 10-5 M atropine sulfate, in the absence of atropine, retinal slices bound 94 pM [3H]PrBCM/g protein after a 5-rain incubation, and binding increased to 368 pM [3H]PrBCM/g protein after 2 h (Fig. 2A). In the presence of atropine, the initial binding of [3H]PrBCM was reduced to 61 pM/g protein and a maximum binding after 2 h of 103 pM/g protein. This represents a 72% reduction of binding in the presence of atropine. The difference between these two binding curves operationally defines the specific, atropine-sensitive, binding of PrBCM (Fig. 2B). Also plotted in Fig. 2B is the ratio of specific to non-specific binding. Saturation of specific binding and the maximum ratio of specific to nonspecific binding both occur after a l-h incubatton. The maximum specific binding of [3H]PrBCM in retina is approximately 12% of the values obtained from sections of whole rat brain 25 and roughly twenty-fold greater than values from intestinal smooth muscle 7. The inhibition of specific [3H]PrBCM binding as a function of atropine concentration is shown in Fig. 3. Half-maximal inhibition was obtained at an atropine concentration of approximately 10-v M. which is consistent with previously reported values for brain25 and muscle 7. Construction of a Hill plot from the inhibition curve rendered a Hill coefficient of 0.9. Values near one are characteristic of most ACh antagonistslO. In order to confirm that the atropine-sensitive binding was indicative of muscarinic cholinergic receptors, various nicotinic and muscarinic cholinergtc

360 520~ c

o¢'

280

cx

/

cm 2 4 0 <) 123 n~

2O0 16o

40

B c

'~ 280[

.

~-

2°° I

£

co

/ /

36~, ~

~2 8

4O .04 c

" ~-o-~b-~,b

~ - 8 o - - ~6o ~ ~2o

o

Fig. 2. Binding of pH]PrBCM to retinal slices as a function of time in the presence and absence of atropine sulfate. A illustrates the binding of 2.9 nM [3H]PrBCM to retina/slices in the presence (O) and absence (D) of t0 -5 M atr01~ne sulfate. The specific (atropine-sensitive) binding is the difference between these two curves and is shown in B (i)~ The ratio of specific to non-specific binding also is illustrated (A). Saturation of binding along with the highest ratio of specific to nonspecific binding occurs after 1 h. Each point represents the mean + S.D. for n = 6.

competitors were incubated with retinal slices prior to the addition of [3H]PrBCM. As with atropine, inhibition curves were constructed for each competitor. and the concentrations producing hag-maximal inhibition (IC50) are presented in T~tble I. T h e two muscarinic antagonists, atropine sulfate and quinuclidinyl benzilate, produced half-maximal inhibition of [3H]PrBCM binding at low, "physiologically relevant' concentrations, while D-tubocurarine. a nicotinic antagonist, was effective only at significantly higher concentrations. The A C h agonist carbachol also was capable of blocking PrBCM binding b m only

359

I001

50 =

9 x IO-'°M

501 _c

0

I0 -t2

I0 -'°

IO-B

I0-"

IO-4

Atropine cone. (M) Fig, 3. Binding of [3H]PrBCM to retinal slices as a function of atropine sulfate concentration. Slices were preincubated in various concentrations of atropine sulfate followed by a 15-20 rain incubation with 2.9 nM [3HIPrBCM. The incubation was terminated and samples processed as described in Materials and Methods. In comparison to control values obtained in the absence of atropine sulfate, 50% inhibition (ICs0) of [3HIPrBCM binding was achieved by incubation with approximately 9 × I0 m M atropine sulfate. Each point represents the mean _+ S.D. for at least triplicate determinations.

at concentrations similar to tubocurarine, a-Bungarotoxin, specific for nicotinic cholinergic receptors, binds in the salamander retina (data not shown) but at concentrations comparable to tubocurarine had little effect on the binding of [3H]PrBCM; the IC50 for a-bungarotoxin, if one can be reached, lies above 10 5 M but was not determined in these experiments. By pharmacological definition both muscarinic and

TABLE I Effect of cholinergic competitors on the binding of PrBCM

Binding of pH]PrBCM to retinal slices in the presence of cholinergic competitors. The ICs~~for each drug was obtained as described in Fig. 3 and Materials and Methods. Muscarinic antagonists (atropine sulfate and QNB) effectively blocked [3H]PrBCM binding at low, physiologically relevant concentrations with Hill coefficients of approximately 1. Nicotinic drugs (D-tubocurarine chloride and c~-bungarotoxin) were effective only at much higher concentrations, and the ACh agonist carbamylcholine chloride was an effective blocker at intermediate concentrations. Cholinergic agent

1Cso (M)

Atropine sulfate Quinuclidinyl benzilate-HCI Carbamylcholine chloride D-Tubocurarine chloride c~-Bungarotoxin

9.0 x 10-m 7.5 x 10-10 6.5 x 10-7 2 x 10-a > 10 5

nicotinic cholinergic receptors are present in salamander retina, and PrBCM binds specifically to the muscarinic receptors. The distribution of muscarinic receptors in salamander retina was analyzed by light microscopy and autoradiography after incubation of retinal slices with [3H]PrBCM. As shown in Fig. 4A, binding of the ligand is unevenly distributed between the retinal layers. Most apparent is the dense labeling of the inner plexiform layer, the site of synaptic contact between bipolar, amacrine and ganglion cells. This contrasts with the negligible labeling of the photoreceptor and nuclear layers, which contain few, if any, synaptic processes. More difficult to assess is the degree of labeling of the outer plexiform layer. A low but reproducible pattern of labeling in the outer plexiform layer is observed. This may be due to uptake bv Mfiller cells, although labeling of the Mfiller cell cytoplasm in the inner nuclear layer rarely is observed. The question of outer plexiform labeling is best addressed by ultrastructural analysis, and such studies are underway. Atropine suppresses the labeling profile of [3H]PrBCM (Fig. 4B). Again, there is no discernible binding of the ligand in the photoreceptor or nuclear layers, but atropine dramatically reduces labeling of the inner plexiform layer and eliminates the low density of labeling previously observed in the outer plexiform layer. Therefore, as in the preceding binding studies, PrBCM labeling of the retina reflects specifically the localization of muscarinic receptors. The thickness of the retinal slice does not appear to limit accessibility of PrBCM to receptors. This is demonstrated by cross-sections through the thickness of the slice (Fig. 5). A uniform pattern of dense labeling of the inner plexiform layer can be seen, and the distribution of ligand is the same in the center of the slice as at the edges. DISCUSSION Retinal slices have been used in a variety of studies aimed at describing the electrical properties of individual neurons and neurons comprising networks 4.32. More recently, histological experiments have been performed with retinal slices, thus overcoming diffusion barriers and enzymatic activities associated with other preparations 1~,35. In the present study we have

360 begun to characterize cholinergic receptors in the salamander retinal slice by binding experiments and light microscopy and autoradiography. [3H]PrBCM is a specific muscarinic antagonist which under the conditions of these experiments binds irreversibly to the muscarinic cholinergic receptor. Specificity can be demonstrated by the effective inhibition of [3H]PrBCM binding in the presence of the muscarinic drugs atropine and QNB. Nicotinic drugs such as D-tubocurarine are effective inhibitors only at concentrations 5 orders of magnitude greater than atropine. The atropine-sensitive binding of [3H]PrBCM is maximum after an incubation of 1 h and represents approximately 270 pM of bound ligand per gram protein. This compares to 320 pM of [3H]QNB bound per gram protein in 13-day chicken embryo retina 28, although this quantity decreases in the adult. Bovine retinal homogenate contains approximately 148 pM of [3H]QNB bound per gram protein 10 and 253 pM of [3H]QNB bound per gram protein in a synaptosomal fraction from bovine retina. [3H]PrBCM binds predominantly and uniformly in the inner plexiform layer of salamander retina. Confirmation of binding in the outer ptexiform layer requires ultrastructural analysis, which we currently

A

are performing with antibodies directed against the acetylcholine receptor. Like salamandc~ rctim~, QNB binding in the adult chicken retina is localized almost exclusively to the inner plexiform layer:L although it is distributed between 3 discrete bands within the inner plexiform layer. One function of ACh action not often considered in the retina might be to alter the receptive field size of cell networks mediating lateral interactions. As mentioned earlier, ACh reduces the receptive field size of broad-field horizontal cells in the salamander retina 11,17. ACh also might reduce coupling between cells in the inner plexiform layer, perhaps a network of amacrine ceils 12connected by gap junctions 33. The multistratified amacrine cells have been considered to mediate lateral interactions in the salamander retina and are the only amacrine cell type to have processes in the distal as well as proximal portions of the inner plexiform layer 29. As demonstrated in this paper, muscarinic receptors also are present in the distal as well as proximal portions of the inner plexiform layer, thereby making the multistratified amacrine cell an interesting candidate for muscarinic cholinergic action. Further ultrastructural and electrophysi0logical studies in our laboratory are aimed at identify-

II

Fig. 4. Distribution of [3H]PrBCM binding in retinal slices in the presence and absence of atropine sulfate. Retinal slices were incubated with 2.9 nM [3H]PrBCMfor 1 h either in the absence (A) or presence (B) of 10-6 M atropine sulfate and processed for light microscopic autoradiography as descibed in Materials and Methods. A uniform, dense labeling of the inner plexiform layer is seen after incubation with [3H]PrBCM(A) which can be suppressed by preincubation with atropine sulfate (B). A small but reproducible pattern of labeling also occurs in the outer plexiform layer, while the photoreceptor and nuclear layers are mostlydevoid of label. Autoradiographic exposure was 3 weeks. × 180.

361

Y

Fig. 5. Penetration of [3H]PrBCM through the retinal slice. A cross-section through the 150 ~m thickness of the retinal slice demonstrates that [3H]PrBCM has access to receptors in the interior, and therefore binding does not represent solely a surface phenomenon. Slices were incubated and processed for light microscopy autoradiography as described in Fig. 4 and Materials and Methods. Autoradiographic exposure was 3 weeks, x 192.

ing cells c o n t a i n i n g c h o l i n e r g i c m a r k e r s , thus d e t e r -

E l e c t r o n M i c r o s c o p e L a b o r a t o r y at the U n i v e r s i t y of

m i n i n g h o w cells are c o n n e c t e d in the r e t i n a and per-

California,

haps the role of c h o l i n e r g i c t r a n s m i s s i o n in this tis-

e q u i p m e n t . This w o r k was s u p p o r t e d by N . I . H . post-

sue.

doctoral fellowship E Y 0 5 4 9 2

Berkeley,

for p r o v i d i n g

(to A . S . P . ) ,

T r a i n i n g G r a n t G M 0 7 0 4 8 (to J . B . H . ) ACKNOWLEDGEMENTS

training

and

N.I.H.

and N . I . H .

G r a n t EY00561 (to F . S . W . ) .

T h e a u t h o r s w o u l d like to t h a n k the staff of the

REFERENCES 1 Ames, A., III and Pollen, D. A., Neurotransmission in central nervous tissue: a study of isolated rabbit retina, J. Neurophysiol., 32 (1969) 424-442. 2 Araki, M., Ide, C. and Saito, T., Ultrastructural localization of acetylcholinesterase activity in the developing chick retina, Acta histochem, cytochem., 15 (1982) 242-255. 3 Ariel, M. and Daw, N. W., Effects of cholinergic drugs on receptive field properties of rabbit retinal ganglion cells, J. Physiol. (Lond.), 324 (1982) 135-16(/.

4 Attwell, D. and Wilson, M., Behaviour of the rod network in the tiger salamander retina mediated by membrane properties of individual rods, J. Physiol. (Lond.). 309 (1980) 287-315. 5 Baughman, R. W. and Bader, C. R., Biochemical characterization and cellular localization of the cholinergic system in the chicken retina, Brain Research, 138 (1977) 469-485. 6 Bradford, M., A rapid and sensitive method for the quantiration of microgram quantities of protein utilizing the principle of dye-binding, Anal. Biochem., 72 (1976) 248-254. 7 Burgen, A. S. V., Hiley, C. R. and Young, J. M., The bind-

3~2

8 9

10

11

12

ing of 3H-propylbenzilylcholine mustard by longitudinal muscle strips from guinea-pig small intestine, Brit. ,I. Pharmacol.. 50 ( 19741 145-15 I. Frances, C. M., Cholinesterase in the retina, J. Physiol. (Lond,), 120 (1953) 435-439. Gerschenfeld, H. M. and Piccolino, M., Muscarinic antagonists block cone to horizontal cell transmission in turtle retina, Nature (Lond.), 268 ( 19771 257-259. Hruska. R. E., White. R., Azari, J. and Yamamura, H'. I., Muscarinic cholinergic receptors in mammalian retina, Brain Research, 148 (1978) 493-498. ttutchins, J. B., Polans, A. S. and Werblin, F. S., Localization of cholinesterase activity in the outer plexiform layer of the larval tiger salamander retina, Brain Research, 292 ( 19841 303-315. Jensen, R. J. and DeVoe, R. D., Ganglion cells and (dyecoupled) amacrine cells in the turtle retina that have possible synaptic connection, Brain Research. 240 (1982) 146-150.

13 Kuhar. M. J., Taylor, N., Wamsley, J. K., Hulme, E. C. and Birdsatl, N. J. M., Muscarinic cholinergic receptor localization in brain by electron microscopic autoradiography, Brain Research, 216 (1981) 1-9. 14 Lam, D., Biosynthesis of acetylcholine in turtle photoreceptors, Proc. nat. Acad. Sci. U.S.A., 69 (1972) 1987-1991. 15 Lasansky. A., Contacts between receptors and electrophysiologically identified neurons in the retina of the larval tiger salamander. J. Physiol. (Lond.). 285 (1978) 531-542. 16 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the folin phenol reagent, J. biol. Chem.. 193 (1951) 265-275. 17 Marshall, L. M. and Werblin, F. S., Synaptic transmission to the horizontal cells in the retina of the larval tiger salamander. J. Physiol. (Lond.). 279 (1978) 321-346. 18 Masland, R. H. and Livingstone, C. J., Effect of activity on the synthesis and release of acetylcholine by an isolated mammalian retina, J. Neurophysiol., 39 (1976) 1210-1219. 19 Masland, R. H., Acetylcholine in the retina, Neurochem. Int., 1 (1980) 501-518. 20 Miyata, M. and lshikawa, S., Acetylchotinesterase in the outer plexiform layer of the retina, Jap. J. Ophthal., 20 ( 19761487-496. 21 Moreno-Yanes, J. A. and Mahler. H. R., Subcellular distribution of )H-quinuclidinyl benzilate binding activity in vertebrate retina and its relationship to other cholinergic markers, J. Neurochem., 33 (1979) 5(/5-516. 22 Morgan, 1. G., Oliver, J. and Chubb. 1. W.. Discrete distri-

23 24

25

26

27

28

29

30

31

32

33

34

35

butions of putative cholinergic and somatostatmcrglc an> acrine cell dendrites in chicken retina. Neurc,.wi. l.ett.. 27 (1981) 55-60. Neal, M. J. and Gilroy, J., High-affinity chohnc transport in the isolated retina, Brain Research, 93 (1975) 548-551. Pourcho, R. G., Localization of cholinergic synapses in mammalian retina with peroxidase-conjugated atpha-bungarotoxin, Vision Res., 19 ( 19791 287-292. Rotter. A., Birdsall, N. J. M.. Burgen, A S. V.. Field. P. M., Hulme, E. C. and Raisman. G.. Muscarinic receptors in the central nervous system of the rat. I Technique for autoradiographic localization of the binding of [3H]propylbenzilylcholine mustard and its distribution in the forebrain, Brain Res. Rev., 1 (19791 141-165. Schwartz, I. R. and Bok, D., Electron m,croscopic localization of 12q-alpha-bungarotoxin binding sites in the outer plexiform layer of the goldfish retina. J Neurocvtol.. g (1979) 53-66. Skrzypek, J. and Werblin, F. S.. Rods increase and cones decrease the conductance of the horizontal cell membrane in the tiger salamander retina. Invest. Ophthal.. 20 Suppl. 3 (1981) 183. Sugiyama, H., Daniels, M. P. and Nirenberg, M., Muscarinic acetylcholine receptors of the developing retina. Proc. nat. Acad. Sci. U.S.A., 12 (1977) 5524-5528. Vallerga, S., Physiological and morphological identification of amacrine cells in the retina of the larval tiger salamander, Vision Res., 21 (1981) 1307-1317. Vogel, Z. and Nirenberg, M., Localization of acetvlcholinc receptors during synaptogenesls in retina. Proc. nat. Acad. Sci. U.S.A.. 73 (t976) 1806-1810. Vogel, Z., Maloncy, G. J., Ling A. and Daniels. M. P., Identification of synaptic acetylcholine receptor sites m retina with peroxidase-labeled alpha-bungarotoxin, Proc. nat. Acad. Sci. U.S.A., 74 (1977) 3268-3272. Werblin, F. S., Transmission along and between rods in the tiger salamander retina, J. Physiol. tLond. L 280 (1978) 449-470. Wong-Riley, M. T. T., Synaptic organization of the inner plexiform layer in the retina of the tiger salamander. J. Neurocytol., 3 (1974) 1-33. Yazulla, S. and Schmidt, J., Radioautographic localization of ~25I-alpha-bungarotoxin binding sites in the retinas of goldfish and turtle, Vision Res. 16 ( 19761 878-880. Zucker, C. and Yazulla. S., Localization of synaptic and nonsynaptic nicotinic acetylcholine receptors m the goldfish retina, J. comp. Neurol.. 204 f 19821 188-195.