- Email: [email protected]
Receptors coupled to adenylate cyclase in isolated rabbit retina
Neurochem. Int. Vol. 14, No. 4, pp. 387-395, 1989
0197-0186/89 $3.00 + 0.00 Copyright © 1989 Pergamon Press plc
Printed in Great Britain. All rights reserved
RECEPTORS C O U P L E D TO ADENYLATE CYCLASE IN ISOLATED RABBIT RETINA* MICHEL SCHORDERETt Department of Pharmacology, University Medical Center (CMU), 1 rue Michel Server, Geneva and School of Pharmacy, University of Lausanne, Lausanne, Switzerland (Accepted 18 November 1988)
Abstract--Intact pieces or homogenates of rabbit retina were exposed to various established or putative retinal neurotransmitters for the study of receptors (or receptor-subtypes) linked to the production of cAMP. Since a dopamine-sensitive adenylate cyclase has been previously characterized in the retina of several species, the novel D~-agonist SKF 38393 was applied under similar experimental conditions. This compound was found to be more potent (although less efficacious) than dopamine, confirming the existence of a population of Drreceptors. On the other hand, the novel Dr-antagonist SCH 23390 was able to inhibit the stimulating effects of dopamine and of SKF 38393 in a concentration-dependent manner. Attempts to detect D2-receptors (negatively coupled with adenylate cyclase) were not conclusive, when a selective D2-agonist (quinpirole) was applied in the absence or presence of a D2-antagonist (sulpiride). A stimulation of cAMP production (mediated by A2-receptors ) was also detected in response to adenosine or adenosine-analogs, which was blocked by IBMX in a concentration-dependent manner. The effects of adenosine were potentiated in the presence of dipyridamole, an adenosine uptake inhibitor. Compared to the effects of dopamine and adenosine, the stimulation induced by VIP, a retinal neuropeptide, was found to be much more pronounced. These results indicate that retinal cAMP can be generated by three different neurotransmitters in an independent way via the stimulation of specific receptors.
Isolated mammalian and n o n m a m m a l i a n retinas have been used for more than 15 years for the study of dopamine-sensitive adenylate cyclase (see Schorderet and Magistretti, 1983 for a review). Initially, these studies were aimed at the possible screening of new drugs for the treatment of neurological or psychiatric diseases (e.g. Parkinsonism or schizophrenia). Still, they continue to be performed, taking into account the discovery of subtypes of dopamine-receptors (namely D~ and D2, Kebabian and Calne, 1979), which can be differentially or interdependently involved in physiological and/or pathologic conditions (Walters et al., 1987). On the
*Parts of the results of this refereed paper were presented at the European Society of Neurochemistry, in a workshop on Extracellular and 1ntracellular Messengers in the Retina held in G6teborg, Sweden on 12-17 June 1988. The workshop was organized by Dr N. N. Osborne, Oxford University, England. tAddress all correspondence to: Michel Schorderet, Department of Pharmacology, CMU, CH-1211 Geneva 4, Switzerland.
other hand, dopamine itself is actually considered to be a classical retinal neurotransmitter in most species. In teleost retina, dopamine modulates the responsiveness of horizontal cells to light and the strength of electrical coupling between them through the stimulation of D r r e c e p t o r s (Dowling, 1986). Other actions of dopamine at the level of photoreceptors (physical retraction of cones and elongation of rods) appear to be mediated through D2-receptors (Dearry and Burnside, 1986). Furthermore, retinal dopaminergic cells also seem to degenerate in human and monkey Parkinsonism, leading to E R G alterations (BodisWollner and Piccolino, 1988). Other stimulators of retinal adenylate cyclase were found to be adenosine (Paes de Carvalho and DeMello 1982; Schorderet, 1982) and VIP (Longshore and Makman, 1981; Schorderet et al., 1981). Much less is known as to the putative physiological role of these neurotransmitters in retina. Still, additional investigations aimed at the characterization of c A M P generation induced by these agents and/or of receptor-subtypes may help us to understand some of the neurophysiological action(s) of adenosine or VIP in other parts of the brain. 387
~tJ( '111i ~ tl4)t,tt)[:It]]
38h
EXPERIMENTAl.
PRO(:EI)I;RI,:b
Isolation m i d dissection ~?l rabbit retinas
Albino or pigmented rabbits of either sex were killed by a shot in the neck and the eyes quickly removed and placed on ice. Dissection of the retina was performed by cutting away the anterior portion of the eye, gently removing the vitreous, inverting the eyeball and then. ~ith a soft brush soaked in Krcbs Ringer buffer, detaching the retina and pushing it toward the middle of the disc. The retinal connection to the optic nerve was cut and the free retina was used either for experiments with intact cells or for experiments with tissue homogenates.
,4den),late cvelase actirit.r (endogenous cAMP) it1 intact relip~a After isolation, each retina (10 for each experiment) was immediately soaked in a beaker containing 20 ml of cold Krebs Ringer (pH 7.4, ice-temperature). All retinas were then incubated together for 40 min at 3 5 C under constant oxygenation (95% 02--5% CO.,) and slight shaking in a water-bath. At the end of the incubation, each retina was cut into four roughly equal pieces and the retinal quarters (40 randomized samples) successively distributed into assay tubes containing 250/~1 (final volume) Krebs-Ringer buffer containing 10mM theophylline or 0.5mM Ro 20-1724. The assay tubes were then placed in the shaking water-bath (35 C) and a 10 min incubation was started by adding 25/zl of drug solution or Krebs--Ringer buffer (controls) to each sample every 30 s. The assay tubes were removed and heated for 10rain in a boiling water-bath. The samples were sonicated, centrifuged and aliquots of the supernatants kept for cAMP measurements according to the saturation assay method of Brown et al. (1971l. The sediments were dissolved in NaOtt SDS for protein determination (Lowry et al., t951 ).
Adenylate ~3'clase activit)' (eAMP [~)rmed li'om exogenous A TP) in retinal homogenates The assay was essentially based on the procedure described by Kebabian et al. (1972) and Clement-Cormier
c"
,.t ~i. (1975) for striatal homogcnate~. Briell~, the t:,~tatcd retinas (6 for each experiment) were homogcniza! H~ ~ glass Teflon homogenizer containing 1.5ml ,~t 2ram Tris (hydroxymethylaminomethane) maleate. 2 mM I ( ; F A buffer, pit 7.4. Without further centrifugation and or wa:-;h ings, 40 ul of tissue suspensions were added into assay tubes containing 410/~l of a standard assay mixture (Kebabian and Saavedra. 1976). The tested drugs were pre,~iousl} dissolved in the same mixture. The assay tubes were allowed to sit lbr 30 min at ice-temperaturc, tJor the activati(m of the enzyme, 50pl of 15 mM ATP (i.e. final concentration 1.5 mM) were added and the assa) tubes were incubated at 30 C for 2.5 rain in a shaking water-bath. Following a 3 rain period of treatment in a boiling water-bath, the samples were centrifuged. Aliquots of the supernatants were then kept for cAMP measurements (Brown el ai., !~71) and the sediments dissolved in NaOH SDS for protein determination (Lowry et al., 19511
RESULTS
Retinal dopamine receptors and c A M P .[ormation: .li~rther pharmacological characterization o f receptorsubtypes W e have recently e x t e n d e d o u r p r e v i o u s results s u p p o r t i n g the view t h a t r a b b i t retinas c o n t a i n e d a s u b s t a n t i a l p o p u l a t i o n o f D l - t y p e o f d o p a m i n e receptors (positively c o u p l e d to a d e n y l a t e cyclase activity) ( M a g i s t r e t t i a n d S c h o r d e r e t , 1979; S c h o r d e r e t a n d Magistretti, 1980, 1983) by testing n e w D ~ - d o p a m i n e a g o n i s t or a n t a g o n i s t , such as S K F 38393 a n d S C H 23390 respectively, on retinal h o m o g e n a t e s . Figure 1 s h o w s a c o n c e n t r a t i o ~ r e s p o n s e g r a p h o f S K F 38393, w h i c h at 0.1 y M significantly s t i m u l a t e d thc activity o f a d e n y l a t e cyclase. At the s a m e c o n c e n t r a t i o n ,
200i
m
•
fir
160. 13n
11
E
rr
,~,$
#i.
e/j
~//
7V,' /
u
/L
Z ¢/.~
"5 E
/A
120. 13_
n
80"
0
BASAL ACTIVITY
I0 -8
10 -7
"/.4, ¢/J tO-S
S K F 36 393
,o
10-5
i0 -t.
// // // // // /7 // // /-/ // // // // // // // //
10-~
(M)
SKF 38393 plus lO'S(*) - butaclamol
Fig. I. SKF 38393-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): concentration-response graph. The stimulating effects of the highest dose of SKF 38393 (100#M) were also measured in the presence of 10~M (+)-butaclamol (right-hand bar). Results are the means + SEM of quintuplate determinations. Statistical significance was assessed by Student's l-test (*P < 0.001 vs basal activity).
cAMP generated by neurotransmitters in rabbit retina 24O C
'~ 200 O~
E
r/; '
I/A
¢1~
I/N F/A
¢/) ¢/,
160
¢IA FIA
¢1~
O.
¢IA
¢/# ¢/# f/J
120
IE
inhibit in a concentration-dependent manner the dopamine-stimulated activity of adenylate cyclase. Under similar experimental conditions, the stimulating effects of SKF 38393 were also inhibited (not shown).
¢IA
¢1,
Retinal adenosine receptors and c A M P formation: pharmacological characterization o f receptor -subtypes
csA ¢/A ¢/A
I/J g/J
r/A //A //A
f/J ¢ZJ
¢1A ¢1. ¢/A ¢IA
E
Q. 0
¢/A
# BASAL ACTIVITY
389
DA
10-4
(M )
BUTACI.AMOt.
,C+) 1o"5 (-),
Fig. 2. Dopamine(DA)-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): inhibition by 10#M (+)-butaclamol. Results are the means + SEM of quintuplate determinations.
dopamine was totally inactive (not shown). However, compared to dopamine tested simultaneously on the same retinal homogenate (Fig. 2), the efficacy of S K F 38393 appears to be less pronounced. Maximal stimulation was achieved in the presence of 1 #M, with no further stimulation at higher concentrations, indicating that the S K F compound is more potent (e.g. the affinity tbr Drreceptors is higher) but less efficacious (e.g. the intrinsic activity is lower) than dopamine itself. This feature was also supported by the fact that 10#M (+)-butaclamol was unable to reverse the stimulating effect of 100 # M S K F 38393, in contrast to a total inhibition of the dopaminesensitive adenylate cyclase, as measured under similar experimental conditions (Fig. 2). However, when applied at lower concentrations (0.01-10#M), the stimulating effects of S K F 38393 were also blocked by 10#M (+)-butaclamol (the biologically active isomer) and not by 10#M ( - ) - b u t a c l a m o l (the inactive isomer) (not shown). Note that in some experiments an apparent variability of the maximal response to dopamine can be detected (cf. Fig. 2 with Fig. 3). Still, a statistical difference between the maximal effects of S K F 38393 and those of dopamine was found when the t~vo agents were tested on the same preparation. Additional studies of retinal dopamine receptors were also performed with the novel antagonist SCH 23390. Figure 3 shows that this agent was able to
Studies on putative retinal adenosine receptors linked to cAMP generation (A2 type) were also performed by using adenosine itself in the absence or presence of adenosine uptake inhibitors (e.g. dipyridamote), various adenosine analogs as well as adenosine antagonists (methylxanthines). These experiments were carried out with intact rabbit retinas. Figure 4 shows a concentration-response graph of adenosine in the presence of a phosphodiesterase inhibitor (Ro 20-1724). The ECs0 was approx. 50 # M, maximal stimulation being achieved with 100~M adenosine. These values are more elevated than those observed for dopamine under similar experimental conditions, with the exception that theophylline (10 mM) was used in place of Ro 20-1724 (Schorderet and Magistretti, 1980). When 1-10/~M adenosine was applied in the absence of Ro 20-1724, cAMP elevation was hardly significant. However, in presence of 50/~ M dipyridamole, substantial increases in cAMP were restored (Fig. 5). Further experiments
240' C ".~
o_ 200'
o
77 160" // O.
120,
~
80"
~
40-
Hi //
ii II //
a BASAL ACTIVITY
--
10-4
DA
i
, (M)
10-7
IO-S +
i0-6
S C H 23 390
e
Fig. 3. Dopamine(DA)-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): concentration-dependent inhibition by SCH 23390. Results are the means + SEM of quintuplate determinations. Statistical significance was assessed by Student's t-test (° not different from dopamine alone; *P < 0.001 vs dopamine alone).
390
MI('HEI
S('HORI)ERI I
160c 6
260"
140-
E
i':2.°,
240-
CHI OROADENOSINE
120-
120"
100-
4- N 6- P H E N Y L I S O P R O P Y L
a..
80-
Ro 20-172t, 50jJM
80-
...
60-6 E
t~
60-
E 40-
**
o_
1"
40" 20-
-
ADENOSINE(L-PIA)
100-
in "
I
10
50 100 JM
ADENOSINE
Fig. 4. Adenosine-induced formation of c A M P in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentration-response graph in the presence of 50#M Ro 20-1724. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (*P < 0.05; ***P < 0.001 vs controls).
carried out with two adenosine analogs (i.e. 2-chloroadenosine and N6-phenylisopropyladenosine, L-PIA) showed that only the former analog was much more efficacious than adenosine (Figs 4 and 6).
2°LJL/
0
1o
mo ~M
Fig. 6. 2-Chloroadenosine- and NLphenylisopropyladenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10rain; for technical details, see under Experimental Procedures): concentration-response graphs in the presence of 50 #M Ro 20-1724. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (*P <0.05; **P <0,01; ***P < 0.001 vs controls).
Additional data, related to the stimulatory effects o f A2-receptors with adenosine-5'-cyclopropylcarboxamide, revealed that this analog was m o r e potent than 2-chloroadenosine (not shown). Finally, the effects o f adenosine were antagonized in a doserelated m a n n e r by I B M X (Fig. 7). A t t e m p t s to reveal an interaction o f adenosine or various adenosine analogs with A~-receptor subtypes (negatively
;so~
,°°l !'801
.E 140
120 pm
100
8O
~ -6 E
60
,I
°'t i i"
~
40 20
50)JM
60
o. ~E
Ro 2o-1724 l
Lrlrl I
4
10
40
/,'
10)JM
ADENOSINE DIPYRIDAMOLE 50;JM
Fig. 5. Adenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): potentiation by dipyridamole. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (**P < 0.01 vs controls).
10
ii
100
1000 )aM
IBMX
Fig. 7. Adenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentrationdependent inhibition by IBMX. Results are the means + SEM (% increases, N = 5). Statistical significance was assessed by Student's t-test (**P < 0.01; ***P < 0.001 vs adenosine alone).
cAMP generated by neurotransmitters in rabbit retina
391
900OO
100
tO t3.
700" ~s
600'
E t-1 Q.
500
400" 300.
o
E
200 100
I" I0-9
I0 -a
5x I0 "e
10-~
5 x I0 -7
M
Vl P
Fig. 8. VlP-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentration-response graph. The stimulating effects of the highest dose of VIP were also measured in the presence of 100 pM (+)-butaclamol or 10/~m somatostatin (SRIF) (the three right-hand bars). The inset shows for comparison the maximal increase induced by dopamine (10/aM) under identical experimental conditions. Results are the means + SEM (absolute values, N =5). Statistical significance was assessed by Student's t-test (*P <0.05; **P <0.001 vs controls).
coupled to adenylate cyclase) were not successful, as tested in a variety of experimental conditions (not shown).
Retinal VIP receptors and c A M P formation: pharmacological characterization Among several peptides which have been detected by various techniques in rabbit retina, VIP appears to be the most potent and the most efficient stimulator of adenylate cyclase in intact cells as well as in homogenates. Figure 8 shows that in intact rabbit retina 1 nM VIP is as efficient as 10pM dopamine, whereas in terms of potency, this neuropeptide is approx. 500 times more potent than dopamine tested under identical experimental conditions (see Schorderet and Magistretti, 1980). The effects of VIP were neither blocked by a specific dopamine antagonist [(+)-butaclamol], nor by somatostatin (Fig. 8). Other attempts to modulate the VIP-stimulated formation of cAMP with various pharmacological agents were not conclusive (not shown). PHI, a VIP-related peptide, was of one order of magnitude less potent than VIP (Schorderet et al., 1984), whereas secretin and glucagon were totally inactive in intact rabbit retina (Schorderet et al., 1981).
DISCUSSION A large body of experimental evidences indicates that dopamine and dopamine-related drugs are able to stimulate the formation of cAMP in intact as well as in homogenized rabbit retinas (see Schorderet and Magistretti, 1980, 1983 for extensive reviews). The effects of dopamine are concentration-dependent (ECs0 ~ l p M ) , a maximal accumulation of cAMP formation (,~2.5-fold over controls) being obtained with 10pM of this catecholamine. Comparable results were obtained with apomorphine and N-methyldopamine. The formation of cAMP induced by dopamine or dopamine-related drugs was blocked by specific antagonists [i.e. fluphenazine, haloperidol, cis-thioxanthenes or (+)-butaclamol] and lithium. On the other hand, noradrenaline and adrenaline were only effective at the highest concentration tested (100#M), their effects being blocked by dopamine antagonists and not by ~- or fl-antagonists. In addition, isoproterenol (a fl-agonist) was totally ineffective. Thus, these pharmacological data strongly suggest that dopamine receptors of D:type (positively coupled with adenylate cyclase, Kebabian and Calne, 1979) exist in rabbit retina as well as in
",'~2
M I ( H E I . ,'~ HORI]I Rill
a variety of other mammalian and nonmamnlalian species (see Schorderet and Magistretti, 1983). Other evidences arguing for the existence of dopaminergic cells and/or dopamine D~-receptors in rctina were provided by histochemical techniques (DoMing and Ehinger, 1975) or classical binding studies (Magistretti and Schorderet, 1979; Watling et al., 1979: Makman and Dvorkin, 1986; Gredal et al., 1987; De Keyser et al., 1988). Additional studies related to dopamine content, biosynthesis, metabolism and/or release in retina were undertaken with LC--EC techniques (Frederick et al., 1982; Proll and Morgan, 1982; Fernstrom et al., 1984; Otbri el al., 1986a,b; Brainard and Morgan, 1987: O'Connor et al., 1987; Gibson, 1988; Godley and Wurtman, 1988) whereas the most extensive analysis of putative physiological function(s) of retinal dopamine was beautifully achieved by DoMing and coworkers in teleost (DoMing, 1987; see also Bodis-Wollner and Piccolino, 1988 for another up-to-date overview). Interestingly, bovine retinal homogenates are now routinely used by Markstein and coworkers tbr the screening of potential drugs acting at D;-dopamine receptors (Markstein et al., 1987). A possible existence of Dz-type of dopamine receptors (negatively coupled to adenylate cyclase activity, Kebabian and Calne, 1979) was suggested and pharmacologically demonstrated in rabbit retina by Dubocovich and Weiner in studies devoted to the modulation of the stimulation-evoked release of [3H]dopamine in vitro (Dubocovich and Weiner, 1981, 1985; Dubocovich, 1984). Attempts to reveal such a population of D:-receptors by biochemical approach were not successful. For example, basal activity of retinal adenylate cyclase in the presence of 100ttM quinpirole (a D2-agonist) was only slightly decreased, whereas forskolin-stimulated activity (in the absence or presence of SCH 23390) was not inhibited by the same agent. In contrast, we have also found recently a concentration-dependent inhibitory effect of quinpirole on the synthesis of dopamine in citro (Ofori and Schorderet, 1988). Thus, dopamine receptors negatively coupled to adenylate cyclase have not been detected yet in retina, despite the fact that they seem to represent a subpopulation of D:-pre-synaptic or autoreceptors which are involved in the modulation of neurotransminer(s) release and/or synthesis (Starke, 1981; Chesselet, 1984: Rebec, 1984; Bell, 1988). Thus, we have no explanation at the present time for the discrepancy between the pharmacological results and our biochemical data, although it is known that the inhibition of cAMP formation through a receptor
Cii-subunit complex is not easily achieved m mlact nerve cells (Duman and Enna. 1986). Other considerations can be drawn for adenosine. tin ubiquitous neuromodulator capable of stimulating cAMP formation (via A:-type of receptors ,~i Iow affinity) or inhibiting its synthesis (via At-type of receptors of high affinity) (Daly et al., 1981 ). Adenosine receptors linked to the generation of cAMP (A2-type) were previously described in chick embryo retina (Paes de Carvalho and DeMello, 1982), chicken embryo retina cells in culture (DeMelto et al., 1982) and intact rabbit retina (Schorderet, abstract 1982: Blazynski et al., 1986). The present report confirms that retinal adenosine-receptors were pharmacologically distinct from dopamine receptors. However, in contrast to retinal dopamine-sensitive adenylate cyclase, which has been detected in almost every species (Schorderet and Magistretti, 1983), cAMP generation linked to adenosine-receptors stimulation did not occur in mouse, rat and guinea pig retina (Blazynski et al., 1986). Furthermore, the existence of A~-type of retinal adenosine receptors was apparently demonstrated in three different experimental approaches: (i) in modulating the dopamine-dependent cAMP accumulation in chick embryo retina (Paes de Carvalho and DeMello, 1985); (ii) in inhibiting the forskolin-stimulated adenylate cyclase in rabbit retina (Blazynski, 1987); (iii) in localizing binding sites matching Aradenosine-receptors and immunoreactivity for endogenous adenosine in ganglion cells of rat, guinea pig, monkey and human retina (Braas et al., 1987). In view of the fact that the two types of adenosine receptors may coexist in retina (at least in rabbit), definite answers as to the physiological functions of this putative retinal neurotransmitter require further investigations. A classical concept of its role in other parts of the brain is the well characterized capability of adenosine to inhibit the release of neurotransmitters, including neuropeptides, amino acids as well as biogenic amines (Fredholm and Dunwiddie, 1988). The production of cAMP is also achieved by VIP, a neuropeptide known to stimulate adenylate cyclase activity in non-nervous and nervous tissue (Said, 1980), including the retina (intact cells or homogenares) of several species (Longshore and Makman, 1981; Schorderet et al., 1981; Schorderet and Magistretti, 1983, 1984; Pachter and Lain, 1986). In intact rabbit retina, the elevation of cAMP is particularly pronounced, compared to that induced by dopamine or adenosine. This biochemical effect is presumably due to the interaction of VIP with specific receptors, although the possible physio-
cAMP generated by neurotransmitters in rabbit retina logical function(s) of VIP and of other retinal neuropeptides still remains a matter of speculation (Dowling, 1987). Interestingly, VIP and most other neuropeptides are localized to the morphologically distinct sub-class of amacrine cells, a cell type where dopamine has also been observed by histochemistry (Dowling and Ehinger, 1975; Dowling, 1987). Unfortunately, we have not been able to demonstrate a possible mutual interaction of VIP and dopamine neurons as assessed by adenylate cyclase stimulation. Furthermore, in order to investigate a possible link between cAMP generation and glycogenolysis, due to the activation of phosphorylase b (Sutherland and Robison, 1966), we have also tested the ability of VIP to trigger glycogenolysis in pieces of rabbit retina, in consideration of the fact that VIP and some other VIP-related peptides were found to induce glycogenolysis in mouse cerebral cortex (Magistretti et al., 1981). This possible relationship between the two biochemical events was not apparent, since the massive effect of VIP on retinal cAMP increase did not parallel the small effect of the neuropeptide on glycogenolysis (Schorderet et al., 1984). Plausible explanations related to this feature have been given elsewhere (Schorderet et al., 1984). Thus, cAMP generated by VIP in retina should be used as an intracellular messenger for some yet unknown physiological function(s), normally relayed by cAMP-dependent kinases, which may potentially alter a variety of cellular processes, by acting in the nucleus, in the cytoplasm or at the cell membrane. In summary, cAMP appears to be an unequivocal intracellular messenger in the retina of the rabbit (and possibly of several other species including man), which is generated when specific receptors for dopamine, adenosine and VIP are stimulated. For each neurotransmitter, a positive biochemical signal has been easily detected and quantified in a concentration-dependent manner, the potency and efficacy of VIP being considerably higher than those of dopamine and adenosine. For the two classical neurotransmitters, the positive signal is also indicative of a receptor sub-type, D~ and A 2, respectively. It is however not excluded that additional D 2- and Arreceptors (coupled to a negative signal) may exist, as suggested by a variety of other experimental approaches. The generation of cAMP induced by one or the other neurotransmitter is probably involved in some very fine physiological processes (Dowling, 1987; Bodis-Wollner and Piccolino, 1988), although the simple use of intact retina or retinal homogenates cannot give definite answers.
393
Acknowledgements--This work was supported by grants (Nos 3.969-0.84-1.84-2.84)from the "Fonds National Suisse de la Recherche Scientifique". The author is particularly grateful to P. J. Magistretti and N. C. Schaad for critical reading of the manuscript and wishes to thank Ms Gis61e Gilli6ron for technical help, Ms Sylvianne Bonnet for secretarial assistance and Mr Fred Pillonel for the graphs.
REFERENCES
Bell C. (1988) Dopamine release from sympathetic nerve terminals. Prog. Neurobiol. 30, 193-208. Blazynski C. (1987) Adenosine A I receptor-mediated inhibition of adenylate cyclase in rabbit retina. J. Neurosci. 7, 2522-2528. Blazynski C., Kinscherf D. A., Geary K. M. and Ferrendelli J. A. (1986) Adenosine-mediated regulation of cyclic AMP levels in isolated incubated retinas. Brain Res. 366, 224-229. Bodis-Wollner I. and Piccolino M. (eds) (1988) Dopaminergic Mechanisms in Vision. Liss, New York. Braas K. M., Zarbin M. A. and Snyder S. H. (1987) Endogenous adenosine and adenosine receptors localized to ganglion cells of the retina. Proc. natn. Acad. Sci. U.S.A. 84, 3906-3910. Brainard G. C. and Morgan W. W. (1987) Light-induced stimulation of retinal dopamine: a dose-response relationship. Brain Res. 424, 199-203. Brown B. L., Albano J. D. M., Ekins R. P. and Sgherzi A. M. (1971) A simple and sensitive saturation assay method for the measurement of adenosine 3':5'-cyclic monophosphate, Biochem. J. 121, 561 562. Chesselet M. F. (1984) Presynaptic regulation of neurotransmitter release in the brain: facts and hypotheses. Neuroscience 12, 347-375. Clement-Cormier Y. C., Parrish R. G., Petzold G. L. and Greengard P. (1975) Characterization of a dopaminesensitive adenylate cyclase in the rat caudate nucleus. J. Neurochem. 25, 143 149. Daly J. W., Bruns R. F. and Snyder S. H. (1981) Adenosine receptors in the central nervous system: relationship to the central actions of methylxanthines. Life Sci, 28, 2083-2097. Dearry A. and Burnside B. (1986) Dopaminergic regulation of cone retinomotor movement in isolated teleost retinas: I. Induction of cone contraction is mediated by D2 receptors. J. Neurochem. 46, 1006-1021. De Keyser J., Dierckx R., Vanderheyden P., Ebinger G. and Vauquelin G. (1988) D I dopamine receptors in human putamen, frontal cortex and calf retina: differences in guanine nucleotide regulation of agonist binding and adenylate cyclase stimulation. Brain Res. 443, 77 84. DeMello M. C. F., Ventura A. L. M., Paes de Carvalho R., Klein W. L. and DeMello F. G. (1982) Regulation of dopamine- and adenosine-dependent adenylate cyclase systems of chicken embryo retina cells in culture. Proc. nam. Acad. Sci. U.S.A. 79, 5708-5712. Dowling J. E. (1986) Dopamine: a retinal neuromodulator? Trend~ Neurosci. 9, 236240. Dowling J. E. (1987) The Retina, An Approachable Part o f the Brain. The Belknap Press of Harvard University Press, Cambridge, Mass. Dowling J. E. and Ehinger B. (1975) Synaptic organization
~94
MI(H1 I SCIIORI)IRI [
of the amine-containing interplexilorm cells of the goldfish and Cebus monkey retinas. Science 188, 270 273 Dubocovich M. L. (1984) Modulation of [~lt]dopamine release from rabbit retina. Fedn Proc. 43, 2714 2718. Dubocovich M. U and Weiner N. 11981) Modulation of the stimulation-evoked release of [aH]dopamine in the rabbit retina. J. Pharmac. exp. Ther. 219, 701 71t7 Dubocovich M. L. and Weiner N. 11985) Pharmacological differences between D-2 autoreccptor and the D-I dopamine receptor in rabbit retina..I. Pharmac. evp. The*'. 223, 747 754. Duman R. S. and Enna S. J. 11986) A procedure l\~r measuring cL,-adrenergic receptor-mediated inhibition of cyclic AMP accumulation in rat brain slices. Brain Res. 384, 391 394. Fernstrom M. H., Volk E. A. and Fernstrom J. D. (1984) In vivo tyrosine hydroxylation in the diabetic rat retina: effect of tyrosine administration. Brain Res. 298, 167 170. Frederick J. M., Rayborn M. E., Laties A. M., Lam D. M. K. and Hotlyfield J. G. (1982) Dopaminergic neurons in the human retina. J. comp. Neurol. 210, 65 79. Fredholm B. B. and Dunwiddie T. V. (1988) How does adenosine inhibit transmitter release'? Trends pharmac, Sci. 9, 130 134. Gibson C. J. (1988) Alterations in retinal tyrosine and dopamine levels in rats consuming protein or tyrosinesupplemented diets, J. Neuroehem. 50, 1769 1774. Godley B. F. and Wurtmann R. J. (1988) Release of endogenous dopamine from the superfused rabbit retina in vitro: effect of light stimulation. Brain Res. 452, 393 395. Gredal 0., Parkinson D. and Nielsen M. 11987) Binding of [~H]SCH 23390 to dopamine D-I receptors in rat retina in ~'itro. Eur, J. Pharmae. 137, 241 245. Kebabian J. W. and Calne D. 11979) Multiple receptors lbr dopamine. Nature 277, 93 96. Kebabian J. W. and Saavedra J. M. (1976) Dopaminesensitive adenylate cyclase occurs in a region of Substantia nigra containing dopaminergic dendrites. Science 193, 683-685. Kebabian J. W., Petzold G. L. and Greengard P. (1972) Dopamine-sensitive adenylate cyclase in caudate nucleus of rat brain, and its similarity to the "'dopamine receptors". Proe. natn. A~ad. Sci. U.S.A. 69, 2145 2149. Longshore M. A. and Makman M. 11981) Stimulation ot" retinal adenylate cyclase by vasoactive intestinal peptide (VIP). Eur. J. Pharmac. 70, 237~240. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. 11951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. Magistretti P. J. and Schorderet M. (1979) Dopamine receptors in bovine retina: characteristics of the 3H-spiroperidol binding and its use for screening dopamine receptor affinity of drugs. Life Sei. 25, 1675 1686. Magistretti P. J., Morrison J. H., Shoemaker W. J., Sapin V. and Bloom F. E. (1981) Vasoactive intestinal polypeptide induces glycogenolysis in mouse cortical slices: a possible regulatory mechanism for the local control of energy metabolism. Proc. nam. Aead. Sci. U.S.A. 78, 6535 6539. Makman M. H. and Dvorkin B. (1986) Binding sites for [3H]SCH 23390 in retina: properties and possible relationship to dopamine Drreceptors mediating stimu-
lation of adenylate cyclasc..~lotc,. /3rum Re,. i, ct,[ 271~. Markstein R.. Enz A.. Vigourel J. M...laton A_. (i~,sse \ , Briner U. and Gull P. (19871 Biochemical, behatioural. and endocrine efl'ects of CK 2tt4-933, a novel 8/~-crgolenc J. Neural 7)'ansmitler~ 69, 179 199 O'Connor P. M., Zucker C. 1,. and Dowting ,I. 1-7. 11%;7) Regulation of dopamine release from interplexitorm ceil processes in the outer plexiform layer of the carp retina J. Neuroehem. 49, 916 920. Ofori S. and Schorderet M. (1988) t h e rabbit retina m rmo: A pharmacological model to study the synaptic regulation of dopamine synthesis and release. In: Doparninergic .~techanism.s" in I'ision (Bodis-Wollner I. and Piccolmo M., eds), pp. 41 57. Liss, New York Ofori S., Bretton (., llof P. and Schorderet M. {1986a) Investigation of dopamine content, synthesis, and release in the rabbit retina #t viwo: 1. Effects of dopaminc precursors, reserpine, amphetamine, and L-dopa decarboxylase and monoamine oxydase inhibitors. J. N~'urochem. 47, 1199 1206. Ofori S., Magistretti P. J. and Schorderet M. (1986b) Investigation of dopamine content, synthesis, and release in the rabbit retina in vitro: II. Effects of high potassium, adenylate cyclase activators, and N-npropyl-3-(3-hydroxyphenyl)piperidine. J. Neurochem. 47, 1207 1213. Pachter J. A. and Lam D. M. K. 11986) Interactions between vasoactive intestinal peptide and dopamine in the rabbit retina: Stimulation of a common adenylate cyclase. ,I. Neurochem. 46, 257 264. Paes de Carvalho R. and DeMello E. G. 11982) Adenosineelicited accumulation of adenosine 3'-5'-cyclic AMP in the chick embryo retina. J. Neurochem. 38, 493 500. Paes de Carvalho R. and DeMello F. G. 11985) Expression of A~ adenosine receptors modulating dopaminedependent cyclic AMP accumulation in the chick embryo retina. J. Neuroehem. 44, 845 851. Proll M. A. and Morgan W. W. (19821) Adaptation of retinal dopamine neuron activity in light-adapted rats to darkness. Brain Res. 241, 359 361. Rebec G. V. 11984) Auto- and postsynaptic dopamine receptors in the central nervous system. In: Monograph), in Neural Seiem'e (Cohen M. M., ed.), Vol. 10. pp. 207 223. Karger AG, Basel. Said S. 1. (1980) Peptides common to the nervous system and the gastro-intestinal tract. In: Frontiers in Neuroendocrinology (Martini L. and Ganong W. F., eds), Vol. 6, pp. 293 331. Raven Press, New York. Schorderet M. 11982) Pharmacological characterization of adenosine-mediated increase of cyclic AMP in isolated rabbit retina. Fedn Proc. 41, 1707. Schorderet M. and Magistretti P. J. 11980) The isolated retina of mammals: a useful preparation for enzymatic(adenylyl cyclase) and/or binding studies of dopamine receptors. Neuroehem. Int. 1, 337 353, Schorderet M. and Magistretti P. J. 11983) Comparative aspects of the adenylate cyclase system in the retina. In: Progress in Nonrnammalian Brain Research (Nistico G. and Bolls L., eds), pp. 185 21 I. CRC Press, Boca Raton. Fla. Schorderet M. and Magistretti P. J. (1984) Neuropeptides in retina: morphological and biochemical aspects. In: Comparatit,e Physiology qf Senso O, Systems (Bolls L., Keynes R. D. and Maddrell H. P., eds), pp. 421 439. Cambridge University Press, Cambridge, Mass.
cAMP generated by neurotransmitters in rabbit retina Schorderet M., Hof P. and Magistretti P. J. (1984) The effects of VIP on cyclic AMP and glycogen levels in • ~vertebrate retina. Peptides 5, 295-298. Schorderet M., Sovilla J. Y. and Magistretti P. J. (1981) VIP- and glucagon-induced formation of cyclic AMP in intact retinae in vitro. Eur. J. Pharmac. 71, 131-133. Starke K. (1981) Presynaptic receptors. A. Rev: Pharmac. Toxic. 21, 7-30. Sutherland E. W. and Robison G. A. (1966) The role of
395
cyclic-Y-5'-AMP in responses to catecholamines and other hormones. Pharmac. Rev. 18, 145-161. Walters J. R., Bergstrom D. A., Carlson J. H., Chase T. N. and Braun A. R. (1987) D~ dopamine receptor activation required postsynaptic expression of D2 agonist effects. Science 236, 719 722. Watling K. J., Dowling J. E. and Iversen L. L. (1979) Dopamine receptors in the retina may all be linked to adenylate cyclase. Nature 281, 578-580.
0197-0186/89 $3.00 + 0.00 Copyright © 1989 Pergamon Press plc
Printed in Great Britain. All rights reserved
RECEPTORS C O U P L E D TO ADENYLATE CYCLASE IN ISOLATED RABBIT RETINA* MICHEL SCHORDERETt Department of Pharmacology, University Medical Center (CMU), 1 rue Michel Server, Geneva and School of Pharmacy, University of Lausanne, Lausanne, Switzerland (Accepted 18 November 1988)
Abstract--Intact pieces or homogenates of rabbit retina were exposed to various established or putative retinal neurotransmitters for the study of receptors (or receptor-subtypes) linked to the production of cAMP. Since a dopamine-sensitive adenylate cyclase has been previously characterized in the retina of several species, the novel D~-agonist SKF 38393 was applied under similar experimental conditions. This compound was found to be more potent (although less efficacious) than dopamine, confirming the existence of a population of Drreceptors. On the other hand, the novel Dr-antagonist SCH 23390 was able to inhibit the stimulating effects of dopamine and of SKF 38393 in a concentration-dependent manner. Attempts to detect D2-receptors (negatively coupled with adenylate cyclase) were not conclusive, when a selective D2-agonist (quinpirole) was applied in the absence or presence of a D2-antagonist (sulpiride). A stimulation of cAMP production (mediated by A2-receptors ) was also detected in response to adenosine or adenosine-analogs, which was blocked by IBMX in a concentration-dependent manner. The effects of adenosine were potentiated in the presence of dipyridamole, an adenosine uptake inhibitor. Compared to the effects of dopamine and adenosine, the stimulation induced by VIP, a retinal neuropeptide, was found to be much more pronounced. These results indicate that retinal cAMP can be generated by three different neurotransmitters in an independent way via the stimulation of specific receptors.
Isolated mammalian and n o n m a m m a l i a n retinas have been used for more than 15 years for the study of dopamine-sensitive adenylate cyclase (see Schorderet and Magistretti, 1983 for a review). Initially, these studies were aimed at the possible screening of new drugs for the treatment of neurological or psychiatric diseases (e.g. Parkinsonism or schizophrenia). Still, they continue to be performed, taking into account the discovery of subtypes of dopamine-receptors (namely D~ and D2, Kebabian and Calne, 1979), which can be differentially or interdependently involved in physiological and/or pathologic conditions (Walters et al., 1987). On the
*Parts of the results of this refereed paper were presented at the European Society of Neurochemistry, in a workshop on Extracellular and 1ntracellular Messengers in the Retina held in G6teborg, Sweden on 12-17 June 1988. The workshop was organized by Dr N. N. Osborne, Oxford University, England. tAddress all correspondence to: Michel Schorderet, Department of Pharmacology, CMU, CH-1211 Geneva 4, Switzerland.
other hand, dopamine itself is actually considered to be a classical retinal neurotransmitter in most species. In teleost retina, dopamine modulates the responsiveness of horizontal cells to light and the strength of electrical coupling between them through the stimulation of D r r e c e p t o r s (Dowling, 1986). Other actions of dopamine at the level of photoreceptors (physical retraction of cones and elongation of rods) appear to be mediated through D2-receptors (Dearry and Burnside, 1986). Furthermore, retinal dopaminergic cells also seem to degenerate in human and monkey Parkinsonism, leading to E R G alterations (BodisWollner and Piccolino, 1988). Other stimulators of retinal adenylate cyclase were found to be adenosine (Paes de Carvalho and DeMello 1982; Schorderet, 1982) and VIP (Longshore and Makman, 1981; Schorderet et al., 1981). Much less is known as to the putative physiological role of these neurotransmitters in retina. Still, additional investigations aimed at the characterization of c A M P generation induced by these agents and/or of receptor-subtypes may help us to understand some of the neurophysiological action(s) of adenosine or VIP in other parts of the brain. 387
~tJ( '111i ~ tl4)t,tt)[:It]]
38h
EXPERIMENTAl.
PRO(:EI)I;RI,:b
Isolation m i d dissection ~?l rabbit retinas
Albino or pigmented rabbits of either sex were killed by a shot in the neck and the eyes quickly removed and placed on ice. Dissection of the retina was performed by cutting away the anterior portion of the eye, gently removing the vitreous, inverting the eyeball and then. ~ith a soft brush soaked in Krcbs Ringer buffer, detaching the retina and pushing it toward the middle of the disc. The retinal connection to the optic nerve was cut and the free retina was used either for experiments with intact cells or for experiments with tissue homogenates.
,4den),late cvelase actirit.r (endogenous cAMP) it1 intact relip~a After isolation, each retina (10 for each experiment) was immediately soaked in a beaker containing 20 ml of cold Krebs Ringer (pH 7.4, ice-temperature). All retinas were then incubated together for 40 min at 3 5 C under constant oxygenation (95% 02--5% CO.,) and slight shaking in a water-bath. At the end of the incubation, each retina was cut into four roughly equal pieces and the retinal quarters (40 randomized samples) successively distributed into assay tubes containing 250/~1 (final volume) Krebs-Ringer buffer containing 10mM theophylline or 0.5mM Ro 20-1724. The assay tubes were then placed in the shaking water-bath (35 C) and a 10 min incubation was started by adding 25/zl of drug solution or Krebs--Ringer buffer (controls) to each sample every 30 s. The assay tubes were removed and heated for 10rain in a boiling water-bath. The samples were sonicated, centrifuged and aliquots of the supernatants kept for cAMP measurements according to the saturation assay method of Brown et al. (1971l. The sediments were dissolved in NaOtt SDS for protein determination (Lowry et al., t951 ).
Adenylate ~3'clase activit)' (eAMP [~)rmed li'om exogenous A TP) in retinal homogenates The assay was essentially based on the procedure described by Kebabian et al. (1972) and Clement-Cormier
c"
,.t ~i. (1975) for striatal homogcnate~. Briell~, the t:,~tatcd retinas (6 for each experiment) were homogcniza! H~ ~ glass Teflon homogenizer containing 1.5ml ,~t 2ram Tris (hydroxymethylaminomethane) maleate. 2 mM I ( ; F A buffer, pit 7.4. Without further centrifugation and or wa:-;h ings, 40 ul of tissue suspensions were added into assay tubes containing 410/~l of a standard assay mixture (Kebabian and Saavedra. 1976). The tested drugs were pre,~iousl} dissolved in the same mixture. The assay tubes were allowed to sit lbr 30 min at ice-temperaturc, tJor the activati(m of the enzyme, 50pl of 15 mM ATP (i.e. final concentration 1.5 mM) were added and the assa) tubes were incubated at 30 C for 2.5 rain in a shaking water-bath. Following a 3 rain period of treatment in a boiling water-bath, the samples were centrifuged. Aliquots of the supernatants were then kept for cAMP measurements (Brown el ai., !~71) and the sediments dissolved in NaOH SDS for protein determination (Lowry et al., 19511
RESULTS
Retinal dopamine receptors and c A M P .[ormation: .li~rther pharmacological characterization o f receptorsubtypes W e have recently e x t e n d e d o u r p r e v i o u s results s u p p o r t i n g the view t h a t r a b b i t retinas c o n t a i n e d a s u b s t a n t i a l p o p u l a t i o n o f D l - t y p e o f d o p a m i n e receptors (positively c o u p l e d to a d e n y l a t e cyclase activity) ( M a g i s t r e t t i a n d S c h o r d e r e t , 1979; S c h o r d e r e t a n d Magistretti, 1980, 1983) by testing n e w D ~ - d o p a m i n e a g o n i s t or a n t a g o n i s t , such as S K F 38393 a n d S C H 23390 respectively, on retinal h o m o g e n a t e s . Figure 1 s h o w s a c o n c e n t r a t i o ~ r e s p o n s e g r a p h o f S K F 38393, w h i c h at 0.1 y M significantly s t i m u l a t e d thc activity o f a d e n y l a t e cyclase. At the s a m e c o n c e n t r a t i o n ,
200i
m
•
fir
160. 13n
11
E
rr
,~,$
#i.
e/j
~//
7V,' /
u
/L
Z ¢/.~
"5 E
/A
120. 13_
n
80"
0
BASAL ACTIVITY
I0 -8
10 -7
"/.4, ¢/J tO-S
S K F 36 393
,o
10-5
i0 -t.
// // // // // /7 // // /-/ // // // // // // // //
10-~
(M)
SKF 38393 plus lO'S(*) - butaclamol
Fig. I. SKF 38393-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): concentration-response graph. The stimulating effects of the highest dose of SKF 38393 (100#M) were also measured in the presence of 10~M (+)-butaclamol (right-hand bar). Results are the means + SEM of quintuplate determinations. Statistical significance was assessed by Student's l-test (*P < 0.001 vs basal activity).
cAMP generated by neurotransmitters in rabbit retina 24O C
'~ 200 O~
E
r/; '
I/A
¢1~
I/N F/A
¢/) ¢/,
160
¢IA FIA
¢1~
O.
¢IA
¢/# ¢/# f/J
120
IE
inhibit in a concentration-dependent manner the dopamine-stimulated activity of adenylate cyclase. Under similar experimental conditions, the stimulating effects of SKF 38393 were also inhibited (not shown).
¢IA
¢1,
Retinal adenosine receptors and c A M P formation: pharmacological characterization o f receptor -subtypes
csA ¢/A ¢/A
I/J g/J
r/A //A //A
f/J ¢ZJ
¢1A ¢1. ¢/A ¢IA
E
Q. 0
¢/A
# BASAL ACTIVITY
389
DA
10-4
(M )
BUTACI.AMOt.
,C+) 1o"5 (-),
Fig. 2. Dopamine(DA)-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): inhibition by 10#M (+)-butaclamol. Results are the means + SEM of quintuplate determinations.
dopamine was totally inactive (not shown). However, compared to dopamine tested simultaneously on the same retinal homogenate (Fig. 2), the efficacy of S K F 38393 appears to be less pronounced. Maximal stimulation was achieved in the presence of 1 #M, with no further stimulation at higher concentrations, indicating that the S K F compound is more potent (e.g. the affinity tbr Drreceptors is higher) but less efficacious (e.g. the intrinsic activity is lower) than dopamine itself. This feature was also supported by the fact that 10#M (+)-butaclamol was unable to reverse the stimulating effect of 100 # M S K F 38393, in contrast to a total inhibition of the dopaminesensitive adenylate cyclase, as measured under similar experimental conditions (Fig. 2). However, when applied at lower concentrations (0.01-10#M), the stimulating effects of S K F 38393 were also blocked by 10#M (+)-butaclamol (the biologically active isomer) and not by 10#M ( - ) - b u t a c l a m o l (the inactive isomer) (not shown). Note that in some experiments an apparent variability of the maximal response to dopamine can be detected (cf. Fig. 2 with Fig. 3). Still, a statistical difference between the maximal effects of S K F 38393 and those of dopamine was found when the t~vo agents were tested on the same preparation. Additional studies of retinal dopamine receptors were also performed with the novel antagonist SCH 23390. Figure 3 shows that this agent was able to
Studies on putative retinal adenosine receptors linked to cAMP generation (A2 type) were also performed by using adenosine itself in the absence or presence of adenosine uptake inhibitors (e.g. dipyridamote), various adenosine analogs as well as adenosine antagonists (methylxanthines). These experiments were carried out with intact rabbit retinas. Figure 4 shows a concentration-response graph of adenosine in the presence of a phosphodiesterase inhibitor (Ro 20-1724). The ECs0 was approx. 50 # M, maximal stimulation being achieved with 100~M adenosine. These values are more elevated than those observed for dopamine under similar experimental conditions, with the exception that theophylline (10 mM) was used in place of Ro 20-1724 (Schorderet and Magistretti, 1980). When 1-10/~M adenosine was applied in the absence of Ro 20-1724, cAMP elevation was hardly significant. However, in presence of 50/~ M dipyridamole, substantial increases in cAMP were restored (Fig. 5). Further experiments
240' C ".~
o_ 200'
o
77 160" // O.
120,
~
80"
~
40-
Hi //
ii II //
a BASAL ACTIVITY
--
10-4
DA
i
, (M)
10-7
IO-S +
i0-6
S C H 23 390
e
Fig. 3. Dopamine(DA)-sensitive adenylate cyclase activity in homogenates of rabbit retina (incubation time: 2.5 min; for technical details, see under Experimental Procedures): concentration-dependent inhibition by SCH 23390. Results are the means + SEM of quintuplate determinations. Statistical significance was assessed by Student's t-test (° not different from dopamine alone; *P < 0.001 vs dopamine alone).
390
MI('HEI
S('HORI)ERI I
160c 6
260"
140-
E
i':2.°,
240-
CHI OROADENOSINE
120-
120"
100-
4- N 6- P H E N Y L I S O P R O P Y L
a..
80-
Ro 20-172t, 50jJM
80-
...
60-6 E
t~
60-
E 40-
**
o_
1"
40" 20-
-
ADENOSINE(L-PIA)
100-
in "
I
10
50 100 JM
ADENOSINE
Fig. 4. Adenosine-induced formation of c A M P in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentration-response graph in the presence of 50#M Ro 20-1724. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (*P < 0.05; ***P < 0.001 vs controls).
carried out with two adenosine analogs (i.e. 2-chloroadenosine and N6-phenylisopropyladenosine, L-PIA) showed that only the former analog was much more efficacious than adenosine (Figs 4 and 6).
2°LJL/
0
1o
mo ~M
Fig. 6. 2-Chloroadenosine- and NLphenylisopropyladenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10rain; for technical details, see under Experimental Procedures): concentration-response graphs in the presence of 50 #M Ro 20-1724. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (*P <0.05; **P <0,01; ***P < 0.001 vs controls).
Additional data, related to the stimulatory effects o f A2-receptors with adenosine-5'-cyclopropylcarboxamide, revealed that this analog was m o r e potent than 2-chloroadenosine (not shown). Finally, the effects o f adenosine were antagonized in a doserelated m a n n e r by I B M X (Fig. 7). A t t e m p t s to reveal an interaction o f adenosine or various adenosine analogs with A~-receptor subtypes (negatively
;so~
,°°l !'801
.E 140
120 pm
100
8O
~ -6 E
60
,I
°'t i i"
~
40 20
50)JM
60
o. ~E
Ro 2o-1724 l
Lrlrl I
4
10
40
/,'
10)JM
ADENOSINE DIPYRIDAMOLE 50;JM
Fig. 5. Adenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): potentiation by dipyridamole. Results are the means + SEM (absolute values, N = 5). Statistical significance was assessed by Student's t-test (**P < 0.01 vs controls).
10
ii
100
1000 )aM
IBMX
Fig. 7. Adenosine-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentrationdependent inhibition by IBMX. Results are the means + SEM (% increases, N = 5). Statistical significance was assessed by Student's t-test (**P < 0.01; ***P < 0.001 vs adenosine alone).
cAMP generated by neurotransmitters in rabbit retina
391
900OO
100
tO t3.
700" ~s
600'
E t-1 Q.
500
400" 300.
o
E
200 100
I" I0-9
I0 -a
5x I0 "e
10-~
5 x I0 -7
M
Vl P
Fig. 8. VlP-induced formation of cAMP in intact rabbit retina (incubation time: 10 min; for technical details, see under Experimental Procedures): concentration-response graph. The stimulating effects of the highest dose of VIP were also measured in the presence of 100 pM (+)-butaclamol or 10/~m somatostatin (SRIF) (the three right-hand bars). The inset shows for comparison the maximal increase induced by dopamine (10/aM) under identical experimental conditions. Results are the means + SEM (absolute values, N =5). Statistical significance was assessed by Student's t-test (*P <0.05; **P <0.001 vs controls).
coupled to adenylate cyclase) were not successful, as tested in a variety of experimental conditions (not shown).
Retinal VIP receptors and c A M P formation: pharmacological characterization Among several peptides which have been detected by various techniques in rabbit retina, VIP appears to be the most potent and the most efficient stimulator of adenylate cyclase in intact cells as well as in homogenates. Figure 8 shows that in intact rabbit retina 1 nM VIP is as efficient as 10pM dopamine, whereas in terms of potency, this neuropeptide is approx. 500 times more potent than dopamine tested under identical experimental conditions (see Schorderet and Magistretti, 1980). The effects of VIP were neither blocked by a specific dopamine antagonist [(+)-butaclamol], nor by somatostatin (Fig. 8). Other attempts to modulate the VIP-stimulated formation of cAMP with various pharmacological agents were not conclusive (not shown). PHI, a VIP-related peptide, was of one order of magnitude less potent than VIP (Schorderet et al., 1984), whereas secretin and glucagon were totally inactive in intact rabbit retina (Schorderet et al., 1981).
DISCUSSION A large body of experimental evidences indicates that dopamine and dopamine-related drugs are able to stimulate the formation of cAMP in intact as well as in homogenized rabbit retinas (see Schorderet and Magistretti, 1980, 1983 for extensive reviews). The effects of dopamine are concentration-dependent (ECs0 ~ l p M ) , a maximal accumulation of cAMP formation (,~2.5-fold over controls) being obtained with 10pM of this catecholamine. Comparable results were obtained with apomorphine and N-methyldopamine. The formation of cAMP induced by dopamine or dopamine-related drugs was blocked by specific antagonists [i.e. fluphenazine, haloperidol, cis-thioxanthenes or (+)-butaclamol] and lithium. On the other hand, noradrenaline and adrenaline were only effective at the highest concentration tested (100#M), their effects being blocked by dopamine antagonists and not by ~- or fl-antagonists. In addition, isoproterenol (a fl-agonist) was totally ineffective. Thus, these pharmacological data strongly suggest that dopamine receptors of D:type (positively coupled with adenylate cyclase, Kebabian and Calne, 1979) exist in rabbit retina as well as in
",'~2
M I ( H E I . ,'~ HORI]I Rill
a variety of other mammalian and nonmamnlalian species (see Schorderet and Magistretti, 1983). Other evidences arguing for the existence of dopaminergic cells and/or dopamine D~-receptors in rctina were provided by histochemical techniques (DoMing and Ehinger, 1975) or classical binding studies (Magistretti and Schorderet, 1979; Watling et al., 1979: Makman and Dvorkin, 1986; Gredal et al., 1987; De Keyser et al., 1988). Additional studies related to dopamine content, biosynthesis, metabolism and/or release in retina were undertaken with LC--EC techniques (Frederick et al., 1982; Proll and Morgan, 1982; Fernstrom et al., 1984; Otbri el al., 1986a,b; Brainard and Morgan, 1987: O'Connor et al., 1987; Gibson, 1988; Godley and Wurtman, 1988) whereas the most extensive analysis of putative physiological function(s) of retinal dopamine was beautifully achieved by DoMing and coworkers in teleost (DoMing, 1987; see also Bodis-Wollner and Piccolino, 1988 for another up-to-date overview). Interestingly, bovine retinal homogenates are now routinely used by Markstein and coworkers tbr the screening of potential drugs acting at D;-dopamine receptors (Markstein et al., 1987). A possible existence of Dz-type of dopamine receptors (negatively coupled to adenylate cyclase activity, Kebabian and Calne, 1979) was suggested and pharmacologically demonstrated in rabbit retina by Dubocovich and Weiner in studies devoted to the modulation of the stimulation-evoked release of [3H]dopamine in vitro (Dubocovich and Weiner, 1981, 1985; Dubocovich, 1984). Attempts to reveal such a population of D:-receptors by biochemical approach were not successful. For example, basal activity of retinal adenylate cyclase in the presence of 100ttM quinpirole (a D2-agonist) was only slightly decreased, whereas forskolin-stimulated activity (in the absence or presence of SCH 23390) was not inhibited by the same agent. In contrast, we have also found recently a concentration-dependent inhibitory effect of quinpirole on the synthesis of dopamine in citro (Ofori and Schorderet, 1988). Thus, dopamine receptors negatively coupled to adenylate cyclase have not been detected yet in retina, despite the fact that they seem to represent a subpopulation of D:-pre-synaptic or autoreceptors which are involved in the modulation of neurotransminer(s) release and/or synthesis (Starke, 1981; Chesselet, 1984: Rebec, 1984; Bell, 1988). Thus, we have no explanation at the present time for the discrepancy between the pharmacological results and our biochemical data, although it is known that the inhibition of cAMP formation through a receptor
Cii-subunit complex is not easily achieved m mlact nerve cells (Duman and Enna. 1986). Other considerations can be drawn for adenosine. tin ubiquitous neuromodulator capable of stimulating cAMP formation (via A:-type of receptors ,~i Iow affinity) or inhibiting its synthesis (via At-type of receptors of high affinity) (Daly et al., 1981 ). Adenosine receptors linked to the generation of cAMP (A2-type) were previously described in chick embryo retina (Paes de Carvalho and DeMello, 1982), chicken embryo retina cells in culture (DeMelto et al., 1982) and intact rabbit retina (Schorderet, abstract 1982: Blazynski et al., 1986). The present report confirms that retinal adenosine-receptors were pharmacologically distinct from dopamine receptors. However, in contrast to retinal dopamine-sensitive adenylate cyclase, which has been detected in almost every species (Schorderet and Magistretti, 1983), cAMP generation linked to adenosine-receptors stimulation did not occur in mouse, rat and guinea pig retina (Blazynski et al., 1986). Furthermore, the existence of A~-type of retinal adenosine receptors was apparently demonstrated in three different experimental approaches: (i) in modulating the dopamine-dependent cAMP accumulation in chick embryo retina (Paes de Carvalho and DeMello, 1985); (ii) in inhibiting the forskolin-stimulated adenylate cyclase in rabbit retina (Blazynski, 1987); (iii) in localizing binding sites matching Aradenosine-receptors and immunoreactivity for endogenous adenosine in ganglion cells of rat, guinea pig, monkey and human retina (Braas et al., 1987). In view of the fact that the two types of adenosine receptors may coexist in retina (at least in rabbit), definite answers as to the physiological functions of this putative retinal neurotransmitter require further investigations. A classical concept of its role in other parts of the brain is the well characterized capability of adenosine to inhibit the release of neurotransmitters, including neuropeptides, amino acids as well as biogenic amines (Fredholm and Dunwiddie, 1988). The production of cAMP is also achieved by VIP, a neuropeptide known to stimulate adenylate cyclase activity in non-nervous and nervous tissue (Said, 1980), including the retina (intact cells or homogenares) of several species (Longshore and Makman, 1981; Schorderet et al., 1981; Schorderet and Magistretti, 1983, 1984; Pachter and Lain, 1986). In intact rabbit retina, the elevation of cAMP is particularly pronounced, compared to that induced by dopamine or adenosine. This biochemical effect is presumably due to the interaction of VIP with specific receptors, although the possible physio-
cAMP generated by neurotransmitters in rabbit retina logical function(s) of VIP and of other retinal neuropeptides still remains a matter of speculation (Dowling, 1987). Interestingly, VIP and most other neuropeptides are localized to the morphologically distinct sub-class of amacrine cells, a cell type where dopamine has also been observed by histochemistry (Dowling and Ehinger, 1975; Dowling, 1987). Unfortunately, we have not been able to demonstrate a possible mutual interaction of VIP and dopamine neurons as assessed by adenylate cyclase stimulation. Furthermore, in order to investigate a possible link between cAMP generation and glycogenolysis, due to the activation of phosphorylase b (Sutherland and Robison, 1966), we have also tested the ability of VIP to trigger glycogenolysis in pieces of rabbit retina, in consideration of the fact that VIP and some other VIP-related peptides were found to induce glycogenolysis in mouse cerebral cortex (Magistretti et al., 1981). This possible relationship between the two biochemical events was not apparent, since the massive effect of VIP on retinal cAMP increase did not parallel the small effect of the neuropeptide on glycogenolysis (Schorderet et al., 1984). Plausible explanations related to this feature have been given elsewhere (Schorderet et al., 1984). Thus, cAMP generated by VIP in retina should be used as an intracellular messenger for some yet unknown physiological function(s), normally relayed by cAMP-dependent kinases, which may potentially alter a variety of cellular processes, by acting in the nucleus, in the cytoplasm or at the cell membrane. In summary, cAMP appears to be an unequivocal intracellular messenger in the retina of the rabbit (and possibly of several other species including man), which is generated when specific receptors for dopamine, adenosine and VIP are stimulated. For each neurotransmitter, a positive biochemical signal has been easily detected and quantified in a concentration-dependent manner, the potency and efficacy of VIP being considerably higher than those of dopamine and adenosine. For the two classical neurotransmitters, the positive signal is also indicative of a receptor sub-type, D~ and A 2, respectively. It is however not excluded that additional D 2- and Arreceptors (coupled to a negative signal) may exist, as suggested by a variety of other experimental approaches. The generation of cAMP induced by one or the other neurotransmitter is probably involved in some very fine physiological processes (Dowling, 1987; Bodis-Wollner and Piccolino, 1988), although the simple use of intact retina or retinal homogenates cannot give definite answers.
393
Acknowledgements--This work was supported by grants (Nos 3.969-0.84-1.84-2.84)from the "Fonds National Suisse de la Recherche Scientifique". The author is particularly grateful to P. J. Magistretti and N. C. Schaad for critical reading of the manuscript and wishes to thank Ms Gis61e Gilli6ron for technical help, Ms Sylvianne Bonnet for secretarial assistance and Mr Fred Pillonel for the graphs.
REFERENCES
Bell C. (1988) Dopamine release from sympathetic nerve terminals. Prog. Neurobiol. 30, 193-208. Blazynski C. (1987) Adenosine A I receptor-mediated inhibition of adenylate cyclase in rabbit retina. J. Neurosci. 7, 2522-2528. Blazynski C., Kinscherf D. A., Geary K. M. and Ferrendelli J. A. (1986) Adenosine-mediated regulation of cyclic AMP levels in isolated incubated retinas. Brain Res. 366, 224-229. Bodis-Wollner I. and Piccolino M. (eds) (1988) Dopaminergic Mechanisms in Vision. Liss, New York. Braas K. M., Zarbin M. A. and Snyder S. H. (1987) Endogenous adenosine and adenosine receptors localized to ganglion cells of the retina. Proc. natn. Acad. Sci. U.S.A. 84, 3906-3910. Brainard G. C. and Morgan W. W. (1987) Light-induced stimulation of retinal dopamine: a dose-response relationship. Brain Res. 424, 199-203. Brown B. L., Albano J. D. M., Ekins R. P. and Sgherzi A. M. (1971) A simple and sensitive saturation assay method for the measurement of adenosine 3':5'-cyclic monophosphate, Biochem. J. 121, 561 562. Chesselet M. F. (1984) Presynaptic regulation of neurotransmitter release in the brain: facts and hypotheses. Neuroscience 12, 347-375. Clement-Cormier Y. C., Parrish R. G., Petzold G. L. and Greengard P. (1975) Characterization of a dopaminesensitive adenylate cyclase in the rat caudate nucleus. J. Neurochem. 25, 143 149. Daly J. W., Bruns R. F. and Snyder S. H. (1981) Adenosine receptors in the central nervous system: relationship to the central actions of methylxanthines. Life Sci, 28, 2083-2097. Dearry A. and Burnside B. (1986) Dopaminergic regulation of cone retinomotor movement in isolated teleost retinas: I. Induction of cone contraction is mediated by D2 receptors. J. Neurochem. 46, 1006-1021. De Keyser J., Dierckx R., Vanderheyden P., Ebinger G. and Vauquelin G. (1988) D I dopamine receptors in human putamen, frontal cortex and calf retina: differences in guanine nucleotide regulation of agonist binding and adenylate cyclase stimulation. Brain Res. 443, 77 84. DeMello M. C. F., Ventura A. L. M., Paes de Carvalho R., Klein W. L. and DeMello F. G. (1982) Regulation of dopamine- and adenosine-dependent adenylate cyclase systems of chicken embryo retina cells in culture. Proc. nam. Acad. Sci. U.S.A. 79, 5708-5712. Dowling J. E. (1986) Dopamine: a retinal neuromodulator? Trend~ Neurosci. 9, 236240. Dowling J. E. (1987) The Retina, An Approachable Part o f the Brain. The Belknap Press of Harvard University Press, Cambridge, Mass. Dowling J. E. and Ehinger B. (1975) Synaptic organization
~94
MI(H1 I SCIIORI)IRI [
of the amine-containing interplexilorm cells of the goldfish and Cebus monkey retinas. Science 188, 270 273 Dubocovich M. L. (1984) Modulation of [~lt]dopamine release from rabbit retina. Fedn Proc. 43, 2714 2718. Dubocovich M. U and Weiner N. 11981) Modulation of the stimulation-evoked release of [aH]dopamine in the rabbit retina. J. Pharmac. exp. Ther. 219, 701 71t7 Dubocovich M. L. and Weiner N. 11985) Pharmacological differences between D-2 autoreccptor and the D-I dopamine receptor in rabbit retina..I. Pharmac. evp. The*'. 223, 747 754. Duman R. S. and Enna S. J. 11986) A procedure l\~r measuring cL,-adrenergic receptor-mediated inhibition of cyclic AMP accumulation in rat brain slices. Brain Res. 384, 391 394. Fernstrom M. H., Volk E. A. and Fernstrom J. D. (1984) In vivo tyrosine hydroxylation in the diabetic rat retina: effect of tyrosine administration. Brain Res. 298, 167 170. Frederick J. M., Rayborn M. E., Laties A. M., Lam D. M. K. and Hotlyfield J. G. (1982) Dopaminergic neurons in the human retina. J. comp. Neurol. 210, 65 79. Fredholm B. B. and Dunwiddie T. V. (1988) How does adenosine inhibit transmitter release'? Trends pharmac, Sci. 9, 130 134. Gibson C. J. (1988) Alterations in retinal tyrosine and dopamine levels in rats consuming protein or tyrosinesupplemented diets, J. Neuroehem. 50, 1769 1774. Godley B. F. and Wurtmann R. J. (1988) Release of endogenous dopamine from the superfused rabbit retina in vitro: effect of light stimulation. Brain Res. 452, 393 395. Gredal 0., Parkinson D. and Nielsen M. 11987) Binding of [~H]SCH 23390 to dopamine D-I receptors in rat retina in ~'itro. Eur, J. Pharmae. 137, 241 245. Kebabian J. W. and Calne D. 11979) Multiple receptors lbr dopamine. Nature 277, 93 96. Kebabian J. W. and Saavedra J. M. (1976) Dopaminesensitive adenylate cyclase occurs in a region of Substantia nigra containing dopaminergic dendrites. Science 193, 683-685. Kebabian J. W., Petzold G. L. and Greengard P. (1972) Dopamine-sensitive adenylate cyclase in caudate nucleus of rat brain, and its similarity to the "'dopamine receptors". Proe. natn. A~ad. Sci. U.S.A. 69, 2145 2149. Longshore M. A. and Makman M. 11981) Stimulation ot" retinal adenylate cyclase by vasoactive intestinal peptide (VIP). Eur. J. Pharmac. 70, 237~240. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. 11951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. Magistretti P. J. and Schorderet M. (1979) Dopamine receptors in bovine retina: characteristics of the 3H-spiroperidol binding and its use for screening dopamine receptor affinity of drugs. Life Sei. 25, 1675 1686. Magistretti P. J., Morrison J. H., Shoemaker W. J., Sapin V. and Bloom F. E. (1981) Vasoactive intestinal polypeptide induces glycogenolysis in mouse cortical slices: a possible regulatory mechanism for the local control of energy metabolism. Proc. nam. Aead. Sci. U.S.A. 78, 6535 6539. Makman M. H. and Dvorkin B. (1986) Binding sites for [3H]SCH 23390 in retina: properties and possible relationship to dopamine Drreceptors mediating stimu-
lation of adenylate cyclasc..~lotc,. /3rum Re,. i, ct,[ 271~. Markstein R.. Enz A.. Vigourel J. M...laton A_. (i~,sse \ , Briner U. and Gull P. (19871 Biochemical, behatioural. and endocrine efl'ects of CK 2tt4-933, a novel 8/~-crgolenc J. Neural 7)'ansmitler~ 69, 179 199 O'Connor P. M., Zucker C. 1,. and Dowting ,I. 1-7. 11%;7) Regulation of dopamine release from interplexitorm ceil processes in the outer plexiform layer of the carp retina J. Neuroehem. 49, 916 920. Ofori S. and Schorderet M. (1988) t h e rabbit retina m rmo: A pharmacological model to study the synaptic regulation of dopamine synthesis and release. In: Doparninergic .~techanism.s" in I'ision (Bodis-Wollner I. and Piccolmo M., eds), pp. 41 57. Liss, New York Ofori S., Bretton (., llof P. and Schorderet M. {1986a) Investigation of dopamine content, synthesis, and release in the rabbit retina #t viwo: 1. Effects of dopaminc precursors, reserpine, amphetamine, and L-dopa decarboxylase and monoamine oxydase inhibitors. J. N~'urochem. 47, 1199 1206. Ofori S., Magistretti P. J. and Schorderet M. (1986b) Investigation of dopamine content, synthesis, and release in the rabbit retina in vitro: II. Effects of high potassium, adenylate cyclase activators, and N-npropyl-3-(3-hydroxyphenyl)piperidine. J. Neurochem. 47, 1207 1213. Pachter J. A. and Lam D. M. K. 11986) Interactions between vasoactive intestinal peptide and dopamine in the rabbit retina: Stimulation of a common adenylate cyclase. ,I. Neurochem. 46, 257 264. Paes de Carvalho R. and DeMello E. G. 11982) Adenosineelicited accumulation of adenosine 3'-5'-cyclic AMP in the chick embryo retina. J. Neurochem. 38, 493 500. Paes de Carvalho R. and DeMello F. G. 11985) Expression of A~ adenosine receptors modulating dopaminedependent cyclic AMP accumulation in the chick embryo retina. J. Neuroehem. 44, 845 851. Proll M. A. and Morgan W. W. (19821) Adaptation of retinal dopamine neuron activity in light-adapted rats to darkness. Brain Res. 241, 359 361. Rebec G. V. 11984) Auto- and postsynaptic dopamine receptors in the central nervous system. In: Monograph), in Neural Seiem'e (Cohen M. M., ed.), Vol. 10. pp. 207 223. Karger AG, Basel. Said S. 1. (1980) Peptides common to the nervous system and the gastro-intestinal tract. In: Frontiers in Neuroendocrinology (Martini L. and Ganong W. F., eds), Vol. 6, pp. 293 331. Raven Press, New York. Schorderet M. 11982) Pharmacological characterization of adenosine-mediated increase of cyclic AMP in isolated rabbit retina. Fedn Proc. 41, 1707. Schorderet M. and Magistretti P. J. 11980) The isolated retina of mammals: a useful preparation for enzymatic(adenylyl cyclase) and/or binding studies of dopamine receptors. Neuroehem. Int. 1, 337 353, Schorderet M. and Magistretti P. J. 11983) Comparative aspects of the adenylate cyclase system in the retina. In: Progress in Nonrnammalian Brain Research (Nistico G. and Bolls L., eds), pp. 185 21 I. CRC Press, Boca Raton. Fla. Schorderet M. and Magistretti P. J. (1984) Neuropeptides in retina: morphological and biochemical aspects. In: Comparatit,e Physiology qf Senso O, Systems (Bolls L., Keynes R. D. and Maddrell H. P., eds), pp. 421 439. Cambridge University Press, Cambridge, Mass.
cAMP generated by neurotransmitters in rabbit retina Schorderet M., Hof P. and Magistretti P. J. (1984) The effects of VIP on cyclic AMP and glycogen levels in • ~vertebrate retina. Peptides 5, 295-298. Schorderet M., Sovilla J. Y. and Magistretti P. J. (1981) VIP- and glucagon-induced formation of cyclic AMP in intact retinae in vitro. Eur. J. Pharmac. 71, 131-133. Starke K. (1981) Presynaptic receptors. A. Rev: Pharmac. Toxic. 21, 7-30. Sutherland E. W. and Robison G. A. (1966) The role of
395
cyclic-Y-5'-AMP in responses to catecholamines and other hormones. Pharmac. Rev. 18, 145-161. Walters J. R., Bergstrom D. A., Carlson J. H., Chase T. N. and Braun A. R. (1987) D~ dopamine receptor activation required postsynaptic expression of D2 agonist effects. Science 236, 719 722. Watling K. J., Dowling J. E. and Iversen L. L. (1979) Dopamine receptors in the retina may all be linked to adenylate cyclase. Nature 281, 578-580.