Journal of Neuroscience Research 25236-242 (1990)

Development of A, Adenosine Receptors in the Chick Embryo Retina R. Paes de Carvalho Department of Neurobiology, Institute of Biology, Federal Fluminense University, Niteroi, Brazil

Adenosine inhibits cyclic AMP synthesis induced by dopamine in embryonic but not in post-hatched chick retinas. N6-Cyclohexyladenosine (CHA), which preferentially activates A, receptors as well as 2-chloroadenosine, inhibits cyclic AMP accumulation induced by dopamine in retinas from 10-day-old embryos (E10) with ICJo’s of 0.1 and 0.5 pM, respectively, but this effect is not detectable after hatching. In order to verify if this developmental change reflects variations in the number or affinity of A, adenosine receptors, their development during chick retina ontogeny was studied. Binding studies using 3(H)CHA revealed the presence of A, receptors at all stages of development examined, including the post-hatched retina. The number of binding sites increased between El0 and E17, and then decreased in post-hatched animals. In the latter, 3(H)CHA binding was to a single site with a B,, of 128.6 f 13.4 fmol/mg protein and a K, of 2.1 0.2 nM. Various ligands showed similar hierarchies of affinity for the A, receptor in embryonic and post-hatched retinas, namely, CHA> R-N6-phenylisopropyladenosine (l-PIA)> 5’-N-ethylcarboxamideadenosine (NECA)> isobuthylmethyl-xanthine (IBMX). Given that CHA inhibited forskolin-induced cyclic AMP production and Gpp(NH)p inhibited 3(H)CHA binding in both embryonic and posthatched retinas, it appears that receptor coupling to adenylate cyclase is present since early embryonic stages. The results suggest that the A, receptors may have different functions in the embryonic as compared to the mature chick retina.

+

Key words: dopamine, adenylate cyclase, adenosine binding sites, ontogenesis, G protein INTRODUCTION The regulation of adenylate cyclase activity by adenosine through A, and A, membrane receptors, whose activation inhibits or stimulates the enzyme, respectively (Van Calker et al., 1979), was previously observed in neuronal and non-neuronal tissues (Londos et a]. , 1980; Kenimer and Nirenberg, 1981; Anand-Srivastava and Cantin, 1983; dos Reis et al., 1986; Ventura and 0 1990 Wiley-Liss, Inc.

Paes de Carvalho, 1987). In recent years the presence of high-affinity adenosine receptors has also been detected in the CNS by radioligand binding techniques and quantitative autoradiography by using predominantly the A , selective ligand 3(H)CHA (Bruns et al., 1980; Barnes and Thampy, 1982; Marangos et al., 1982; Geiger et al., 1984; Lloyd and Stone, 1985; Erfurth and Reddington, 1986; Snowhill and Willians. 1986). The receptors detected by these methods are distributed in a localized manner in the CNS (Lewis et al., 1981; Fastbon et al., 1986). The regulation of adenylate cyclase by activation of A, and A, adenosine receptors was shown to be present in the retina of some vertebrate species like the chick and rabbit (Paes de Carvalho and de Mello, 1982; Schorderet, 1982; Paes de Carvalho and de Mello, 1985; Blazynski et al., 1986; Blazynski, 1987). Moreover, both A, receptors and endogenous adenosine were detected predominantly in ganglion cells of mammalian retinas (Braas et al., 1987). These studies, together with the release of purines from the rabbit retina induced by high potassium depolarization (Perez et al., 1986), indicate that adenosine may play a physiological role in this tissue. An adenosine-induced accumulation of cyclic AMP has been demonstrated in the chick retina (Paes de Carvalho and de Mello, 1982). This A,-mediated effect is absent at early developmental stages, appearing only after the embryonic day 13 (E13). The response is maximal on E17, but is greatly reduced in retinas from posthatched animals (Paes de Carvalho and de Mello, 1982). While dopamine also increases cyclic AMP levels in the chick retina (de Mello, 1978; de Mello and de Mello, 1980; de Mello et al., 1982), this effect can be inhibited in a dose-dependent fashion by adenosine and related analogs acting through A, receptors (Paes de Carvalho and de Mello, 1985). This inhibitory effect can already Received June 8, 1989; revised September 15, 1989; accepted September 18, 1989. Address reprint requests to Roberto Paes de Carvalho, Departamento de Neurobiologia, Instituto de Biologia, Universidade Federal Fluminense, Caixa Postal 100180, Niteroi, Rio de Janeiro 24000, B r a d .

A, Receptors in the Retina

be detected on E10, when nearly 70% of the dopamineinduced cyclic AMP accumulation is blocked by saturating concentrations of adenosine. On the other hand, apomorphine-induced cyclic AMP increases are not affected by the nucleoside. Remarkably, adenosine does not inhibit either dopamine- or apomorphine-dependent cyclic AMP accumulation in post-hatched retinas (Paes de Carvalho and de Mello, 198s). These developmental changes raise questions about the ontogeny of A , receptors, which has now been studied by radioligand binding methods. It is reported here that, although adenosine does not regulate dopamineinduced cyclic AMP accumulation in the post-hatched chick retina, A, receptors are present at high levels at this stage, when they can mediate inhibition of forskolininduced adenylate cyclase activity. The results indicate that the adenosine receptors present in the early retinal development may have different functions as {compared to the receptors detected in the mature tissue.

MATERIALS AND METHODS Materials '(H)CHA (25-34 Ci/mmol) was from New England Nuclear; 1-PIA, CHA, adenosine deaminase (E.C.3.5.4.4) type 11, and NECA were from Sigma; IBMX was from Aldrich Chem. Co.. All other reagents were of analytical grade. Fertilized White Leghorn eggs were obtained from a local hatchery. Methods Retina dissection and tissue preparation. The embryos were staged according to Hamburger and Hamilton (195 1) and the retinas were dissected according to the procedure of Piddington and Moscoria (1965). Post-hatched animals were kept under constant illumination and the materials were always harvested in the morning period. The eyes were removed and transferred to calcium- and magnesium-free Hank's solution (CMF). The retinas were then carefully dissected and processed to obtain a crude synaptosomal fraction (P2) by a modification of a method described for rat forebrain by Patel et al. (1982). The retinas were homogenized in 0.32M sucrose (2 to 4 retinadml) in a motor-driven Teflon glass homogenizer. After dilution with an equal amount of 0.32M sucrose, the homogenate was centrifuged at 1,OOOg for 10 min. The pellet was discarded and the supernatant was further centrifuged at 30,OOOg for 30 min. The pellet obtained was resuspended in 3 ml of SO mM Tris-HCL buffer pH 7.5 and centrifuged at 30,OOOg for 20 min. This latter procedure was repeated twice. The resulting pellet was resuspended in 50 mM Tris-HC1 containing 2 units/ml adenosine deaminase and incu-

237

bated for 30 min at 37°C. The suspension was then centrifuged at 30,OOOg for 1.5 min, resuspended in SO mM Tris-HCI, and stored at -70°C until used. This preparation was stable for at least 1 month. The protein content of the homogenates was approximately 2.5% of that of tissues before the homogenization procedure. Other methods were tested, but although they gave greater protein yields the same number of receptors were measured. Thus, the enriched homogenates obtained as described were used throughout this study. Binding assay procedure. '(H)CHA binding assays were performed essentially as described by Bruns et al. (1980). The incubations were at 25°C in a final volume of 0.2 ml containing approximately 0.1 mg protein, SO mM Tris-HCI buffer pH 7.5 and '(H)CHA at the indicated concentrations. The assessment of nonspecific binding was performed by using 10 pM unlabeled CHA. The level of nonspecific binding was essentially the same (30 to 40% of total binding at the saturating concentrations of 3(H)CHA) when we used other purinergic ligands like 1-PIA or 2-chloroadenosine (Clado) as displacers. The mixtures were incubated for 1 hr and the reaction was stopped by the addition of 5 ml of SO mM Tris-HC1 buffer pH 7.5 at room temperature and filtration under reduced pressure through Watman GF/B filters. The filters were washed twice for approximately 10 sec with the same volume of buffer. The filter radioactivity was determined by liquid scintillation counting with 30-35% efficiency. Retina treatment and cyclic AMP assay procedures. Unless stated otherwise retinas were incubated for 10 min at 37°C in Basal Medium of Eagle (BME) containing 0.5 mM 4-(3-butoxy-4-methoxybenzyl-2-imidazolidinone) (R020- 1724), 0.5 unit/ml adenosine deaminase, 0.1 mM pargyline, and 0.1 mM sodium ascorbate (de Mello, 1978). The test compounds were then added at the indicated concentrations, and the retinas were further incubated for 10 min at 37°C. The reaction was stopped by the addition of TCA ( 5 % final concentration). Cyclic AMP was separated as described by Matsuzawa and Nirenberg (197.5) and measured according to the method of Gilman (1970). The protein contents were determined by the procedure of Lowry et al. (1951) by using bovine serum albumin as standard. Statistical analysis. The binding kinetics was analysed by using the computer programs Ebda and Ligand (Munson and Rodbard, 1980; McPherson, 1983).

RESULTS The concentration-dependent inhibitory effects of CHA and Clado on dopamine-induced cyclic AMP increases are demonstrated for E l 0 retinas in Figure I . IC,,'s were approximately 0.1 and 0.5 pM, respec-

238

Paes de Carvalho

;::j '1

0

-

0

140

g

120

1 8

60

40

m 0

I

2

3

4

5

6

7

8

9

10

II

12

3[H] CHA (nM)

Fig. 2. Saturation isotherm of '(H)CHA specific binding to homogenates from E l 3 retinas. The homogenates were incubated at 25°C for 1 hr in Tris-HC1 buffer pH 7.5 containing the indicated concentrations of '(H)CHA. The nonspecific binding - log nucleoside rnolariiy (1 5 to 20% of total binding at saturating concentrations) was Fig. 1 . Inhibition of dopamine-dependent cyclic AMP accu- assessed by using 10 IJ.Munlabeled CHA. The points represent mulation in the chick embryo retina by Clado and CHA. Ret- the mean of triplicate assays in which the individual values inas from E l 0 were incubated for 10 min at 37°C in BME were within 5% of the values shown. The nonspecific binding buffered with 25 mM N-2-hydroxyethylpiperazine-N'-2- was assayed in duplicate. Insert: Scatchard plot of the data ethanesulphonic acid (HEPES) (pH 7.4) containing 0.5 mM indicating the presence of one class of sites with calculated K, R020-1724, 0.5 unit/ml adenosine deaminase, 0.1 mM pargy- of 1.8 ? 0.3 nM and B,, of 160 ? 26 fmolimg protein. line, and 0.1 mM sodium ascorbate. After this time, 100 pM dopamine and the indicated concentration of CHA (open circles) or Clado (closed circles) were added to the medium. The retinas were further incubated for 10 min at 37°C and the competition experiments of ('H)CHA binding by the unreaction was stopped by the addition of TCA ( 5 % final con- labeled ligand in El2 or E l 5 retinas such as that shown centration). The cyclic AMP was extracted and assayed as in Figure 3 also produced a best fit for a one-site model indicated in Methods. The points represent the mean from two with calculated K d of 2.0 2 1.6 nM. separate experiments in which the individual values were Although the adenosine-mediated inhibition of within 10% of the values shown. dopamine-dependent cyclic AMP accumulation was no longer detected in post-hatched retinas (Paes d e Carvalho and de Mello, 1985), '(H)CHA binding to retinal hotively . Seventy percent of the dopamine-induced stimu- mogenates was still detectable at this stage. Saturation was obtained with approximately 8 nM '(H)CHA (Fig. lation was inhibited by the adenosine analogs. Binding of the selective A , receptor ligand 4) and the Scatchard plot revealed the existence of a '(H)CHA to E l 2 retinal homogenates at 25°C was rela- single class of sites (Fig. 4 insert). Accordingly, the tively slow, reaching equilibrium after 40 min of incu- computer-assisted analysis of saturation curves obtained bation. Excess concentration of unlabeled C H A or 1-PIA with homogenates from retinas at this stage indicated a competed with the labeled ligand, showing a slow rate of one-site model with a calculated K, of 2.1 ? 0 . 2 nM and of 128.6 2 13.4 fmol/mg protein. dissociation with a T,,2 of 120 min (data not shown). As a B,, The displacement of bound '(H)CHA by cold puexpected, ('H)CHA binding to embryonic retinas was saturable. Although handmade Scatchard plots of some rinergic ligands using retinas from post-hatched animals saturation isotherms appeared curvilinear, the multi-fit is shown in Figure 5 . The rank order of potency obtained computerized analysis of these curves obtained by using was CHA > I-PIA > NECA > IBMX with IC,,'s of E l 0 to E l 3 homogenates did not show statistically sig- approximately 8, 20, 110, and 3,500 nM, respectively. nificant differences between one-site or two-sites mod- This behavior is characteristic of A , adenosine receptors els. Indeed, most curves gave results showing the pres- (Van Calker et al., 1979; Bruns et al., 1980). The same ence of only a single class of sites. A representative hierarchy could be observed for embryonic retinas with experiment is shown in Figure 2 in which the computer approximately similar K, values at both stages, indicatanalyses indicated a K, of 1.8 2 0.3 n M and a B,,,, of ing that '(H)CHA was interacting with similar A,-type 160 k 26 fmol/mg protein. Moreover, the analysis of receptor sites (data not shown).

A ;A A d

100

r

A, Receptors in the Retina

l

239

z rn

60

c

o

g 8 7 -LOG [CHA](M)

6

5

Fig. 3. Displacement of 3(H)CHAbinding by the cold ligand in homogenates of retinas from E l 2 (circles) and I515 (triangles). The homogenates were incubated for 1 hr at 25OC in Tris buffer containing 5.6 nM '(H)CHA and the indicated concentrations of unlabeled ligand. The points represent the mean from triplicate assays in which the individual values were within 5% of the values shown. Maximal 3(H)CHA binding showed remarkable developmental variations (Fig. 6). Thus, there was a sharp increase from approximately 40 fmol/mg protein in E l 0 retinas to nearly 300 fmol/mg protein in El7 retinas, but the number of sites decreased thereafter to approximately 150 fmolimg protein at post-hatching day 10. Functional evidence corroborating the presence of A, receptors in post-hatched retinas is provided by the inhibition of forskolin-stimulated cyclic AMP accumulation by CHA (Table I). Thus, while CHA inhibited approximately 35% of the total cyclic AMP accumulation induced by forskolin in retinas from ElO, it inhibited as much as 52% of the stimulation observed at posthatched day 7. The inhibition of adenylate cy'clase mediated by A , receptors has been described to involve activation of GTP hydrolysis by specific G proteins, probably of the Gi type (Gavish et al., 1982; Goodman et al., 1982). Consistent with this model is the observation that the binding of 3(H)CHA to chick retinal homogenates was inhibited by about 50% in the presence of the non-hydrolyzable GTP analogue Gpp (NH)p (Fig. 7). This inhibitory effect, observed at all the developmental stages studied, indicates that the A , receptors are coupled to adenylate cyclase molecules in embryonic as well as in post-hatched retinas.

DISCUSSION The present study demonstrated the existence of A , adenosine receptors in the chick retina as revealed by the

- 0

0

Fig. 4. Saturation isotherm of '(H)CHA specific binding to homogenates from 5-day-old chick retinas. The homogenates were incubated at 25°C for 1 hr in Tris-HC1 buffer pH 7.5 containing the indicated concentrations of '(H)CHA. The nonspecific binding (30 to 40% of total binding) was assessed by using 10 p M unlabeled CHA. The points shown were pooled from two different experiments that gave similar results and represent the mean of triplicate assays in which the individual values were within 5% of the values shown. The nonspecific binding was assayed in duplicate. The use of 1-PIA to assess nonspecific binding in another experiment also gave similar results. Insert: Scatchard plot of the data indicating the presence of one class of sites with calculated B,, of 128.6 0.7 fmolimg protein and K, of 2.1 2 0.2 nM.

*

presence of ('H)CHA binding sites. These receptors could be detected in the chick retina since early embryonic stages, and showed increases during ontogenesis, leading to maximal levels at E17. The number of receptors decreased thereafter but tended to stabilize after hatching. The binding was to a single class of sites at all stages studied and showed a pharmacological profile characteristic for A , receptors. Previous studies had shown that adenosine, acting at A, receptors, inhibits adenylate cyclase activity stimulated by dopamine in the chick embryo retina (Paes de Carvalho and de Mello, 1985). This effect was already present in retinas from E10, a stage in which adenosine did not promote increase in cyclic AMP levels through interactions with A, receptors (Paes de Carvalho and de Mello, 1982). The inhibitory effects of adenosine analogs such as CHA or Clado were dramatic, attaining 70% of the total accumulation of cyclic AMP induced by dopamine. We now show that A, receptors can also be detected by binding of ('H)CHA to retina homogenates. However, developmental stagedependent increases in receptor number are not accompanied by parallel increases in adenosine inhibition of dopamine-induced adenylate cyclase activity and, in

Paes de Carvalho

240

TABLE I. Effect of CHA on Forskolin-Induced Cyclic AMP Accumulation in Embryonic and Post-Hatched Retinas*

100

Additions to the incubation medium (3

None Cyclohexy ladenosine Forskolin Forskolin cyclohex yladenosine

z n

$ 50 8a

+

-

cn 8

Lo

0 +

a

I

a ' 9

7

6

5

4

3

- LOG [DISPLP(=ER]

Fig. 5 . Displacement of '(H)CHA binding by different purinergic ligands in 5-day-old post-hatched chick retinas, The homogenates were incubated for 1 hr at 25°C in Tris buffer containing 7.6 nM '(H)CHA and the indicated concentrations of CHA (closed circles), 1-PIA (triangles), NECA (squares), and IBMX (open circles). The points represent the mean from triplicate assays in which the individual values were within 5% of the data shown.

1O-day-old

embryos

7-day-old chicks

16.5 2 0.5(2) 9.4 t 1.4(2) 1,668.3 107.4(3) 1,086.4 2 123.8(3)

28.8 -+ 3.4(2) 21.9 -+ 0.7(2) 393.7 2 41.2(3) 189.0 _t 6.3(3)

*

*The procedure was the same as described in the legend to Figure 1, except that the incubation medium did not contain pargyline and sodium ascorbate. Forskolin (50 p,M) and CHA (10 p,M) were added simultaneously to the medium. Numbers in parenthesis represent the number of experiments performed in each case. Data are expressed in pmoles cyclic AMP/mg protein + SEM (for n = 3 ) or ? deviation from the mean (for n = 2).

different degrees of coupling to G protein and adenylate cyclase at different chick retina developmental stages, as observed during rat forebrain ontogeny (Morgan and Marangos, 1987). This possibility is unlikely, in light of experiments showing that CHA inhibits cyclic AMP accumulation induced by forskolin in both embryonic and post-hatched retinas, and that the non-hydrolysable GTP analog Gpp(NH)p inhibits ('H)CHA binding both before and after hatching. The latter finding indicates receptor coupling to a G protein, as observed in other CNS areas (Gavish et al., 1982; Goodman et al., 1982). Adenylate cyclase activity measured by its stimulation by forskolin decreased approximately 10 times from embryonic to post-hatched retinas. This could be due to lower number of enzyme molecules after hatching or to the presence of an inhibitory factor in chick but not embryonic retinas. This hypothetic factor is not adenoI sine since the experiments were performed in the presence of adenosine deaminase. Furthermore, the D, reI ceptor-mediated inhibition of adenylate cyclase appears to be absent from the chick retina since our unpublished y, , I , , results showed that blocking these receptors with low 8 12 16 207 I 5 9 HATCHING concentrations of the antagonist spiroperidol did not inDEVELOPMENTAL STAGE (WE) terfere with the increase in CAMP levels mediated by the Fig. 6. Developmental profile of 3(H)CHA binding in the stimulation of D , receptors in the post-hatched chick chick retina. The incubation was performed at 25°C for 1 hr in retina. Similar findings were recently reported by Agui et Tris-HC1 buffer containing 12.4 nM '(H)CHA. The nonspe- al. (1988). cific binding was assessed by using 10 pM 1-PIA. The points It has been proposed that there are two populations represent the mean from at least three separate experiments of D, dopamine receptors in the chick retina: D , E that assayed in triplicate. Bars represent the SEM. can only be stimulated by dopamine and detected in the embryonic stages of development, and a second type (conventional D,) that can be stimulated by both dopafact, these inhibitory effects were not observed at all mine and apomorphine and can be observed throughout after hatching. Surprisingly, this lack of inhibitory ef- retinal ontogeny, including post-hatched stages (Ventura fects occurs in spite of the presence of high levels of A, et al., 1984). Recently, D, dopamine receptors were receptors in post-hatched retinas. One possible explana- studied in the chick embryo retina by binding methods tion for these findings is that the receptors could have (Agui et al.. 1988). The number of receptors increases

1 1 I

I

,

I

I

1

I

I

A, Receptors in the Retina

A

fZ-&y-dd

2

4

6

8

embryos

10

1i

12

I

0

2

4

6

8

1012

0

a

2

4

6

8

10

241

1

12

CHA ( n M ) Fig. 7. Effect of Gpp (NH)p on the binding of (3H)CHA to homogenates of embryonic and post-hatched chick retinas. The incubation procedures were the same as described in the legend to Figure 2. The concentration of Gpp (NH)p used was 100

pM. The points represent the mean from triplicate assays and the individual values were within 5% of the values shown. The nonspecific binding was assessed in duplicate by using 10 p M unlabeled CHA.

two or three times between embryonic and post-hatched retinas and no differences were detected in the binding characteristics between these developmental stages. One explanation for these facts could be that the hypothetical D I E and conventional D, receptors represent the same receptors but localized in different cell types having distinct life periods during retina embryogenesis. Interestingly, adenosine appears to discriminate between the two proposed subpopulations of dopamine receptors since it only affects cyclic AMP increases induced by stimulation of the embryonic subclass of D, receptors (named D,E), but had no effect on apomorphine-induced cyclic AMP accumulation either in embryonic or post-hatched retinas (Paes de Carvalho and de Mello, 1985). Dopamine has been shown to regulate growth cone motility and neurite outgrowth of specific developing invertebrate neurons (McCobb et al., 1988). A similar effect was recently observed to occur in specific populations of chick embryo retinal cells in culture ((Lankford et al., 1988). Moreover, a dopamine-sensitive adenylate cyclase as well as cyclic AMP-dependent protein phosphorylation (Lockerbie et al., 1988, 1989) has been demonstrated in isolated growth cones from developing forebrain. These findings indicate that dopamine receptors could have important functions during CNS development. Since adenosine regulates the stimulation of adenylate cyclase by dopamine in the developing retina, the differential availability of dopamine and adenosine in the extracellular space during development could be an important factor controlling adenylate cyclase levels in an individual class of cells and consequently their neurite

outgrowth. This possibility could have important consequences for the pattern of synapse formation in the developing CNS.

ACKNOWLEDGMENTS I wish to thank Miss M.H. de Faria for the excellent technical assistance and Drs. A.L.M. Ventura, K.M. Braas, and F.G. de Mello for the advice and criticism. I also greatly acknowledge Dr. Ruben Adler for helpful discussions and his critical review of the manuscript. This work was supported by grants from the Brazilian National Research Council (CNPq), the Financial Agency for Studies and Projects (Finep), and PROPP (Universidade Federal Fluminense).

REFERENCES Agui T, Chase TN, Kebabian JW (1988): Identification of D , dopamine receptor in chicken embryo retina with ["'IJSCH 23982. Brain Res 452:49-56. Anand-Srivastava MB, Cantin M (1983): Regulation of adenylate cyclase in cultured cardiocytes from neonatal rats by adenosine and other agonists. Arch Biochem Biophys 223:468-476. Barnes EM, Jr, Thampy KG (1982): Subclasses of adenosine receptors in brain membranes from adult tissue and from primary cultures of chick embryo. J Neurochem 39:647-652. Blazynski C, Kinscherf DA, Geary KM, Ferrendelli JA (1986): Adcnosine-mediated regulation of cyclic AMP levels in isolated incubated retinas. Brain Res 366:224-229. Blazynski C (1987): Adenosine A, receptor-mediated inhibition of adenylate cyclase in rabbit retina. J Neurosci 7:2522-2528. Braas KM, Zarbin MA, Snyder SH (1987): Endogenous adenosine

242

Paes de Carvalho

and adenosine receptors localized to ganglion cells of the retina. Proc Natl Acad Sci USA 84:3906-3910. Bruns RF, Daly JW, Snyder SH (1980): Adenosine receptors in brain membranes: Binding of Nh-cyclohexyl 3(H) adenosine and I ,3,-diethyl-8-’(H)phenylxanthine. Proc Natl Acad Sci USA 77:5547-5551. de Mello FG (1978): The ontogeny of dopamine-dependent increase of adenosine 3’-5’-cyclic monophosphate in the chick retina. J Neurochem 3 1: 1049-1 053. de Mello FG, de Mello MCF (1980): Dopamine dependent modulation of CAMP level in the chick retina. In Levi-Montalcini R (ed): Pontificiae Academiae Scientiarum, Scripta Varia, pp 343345. Pontificiae Academiae Scientiarum, Vaticano. de Me110 MCF, Ventura ALM, Paes de Carvalho R, Klein WL, de Mello FG (1982): Regulation of dopamine and adenosine-dependent adenylate cyclase systems of’ chicken embryo retina cells in culture. Proc Natl Acad Sci USA 795708-5712. dos Reis GA, Nobrega AF, Paes de Carvalho R (1986): Purinergic modulation of T-lymphocyte activation: Differential susceptibility of distinct activation steps and conelation with intracelMar 3’-5’-cyclic adenosine monophosphate accumulation. Cell Immunol 101:213-231.

possess adenylate cyclase activity which can be augmented by various receptor agonists. Dev Brain Res 38:19-25. Lockerbie RO, Edde B, Prochiantz (1989): Cyclic AMP-dependent protein phosphorylation in isolated neuronal growth cones from developing rat forebrain. J Neurochem 52:786-796. London C, Cooper DMF, Wolff J (1980): Subcl osine receptors. Proc Natl Acad Sci USA 77:255 1-2554. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951): Protein measurement with the Fohn phenol reagent. J Biol Chem 193: 265-275. Marangos PJ, Pate1 J, Stivers J (1982): Ontogeny of adenosine binding sites in rat forebrain and cerebellum. J Neurochem 39:267270. Matsuzawa H, Nirenberg MW (1975): Receptor mediated shifts in cGMP and CAMP in neuroblastoma cells. Proc Natl Acad Sci USA 72:3472-3476. McCobb DP, Haydon PG, Kater SB (1988): Dopamine and serotonin inhibition of neurite elongation of different identified neurons. J Neurosci Res 19:19-26. McPherson GA (1983): A practical computer-based approach to the analysis of radioligand binding experiments. Comput Prog 17: 107-1 14. Erfuflh A, Reddington M (1986): properties of binding sites for Morgan PF, Marangos PJ (1987): Ontogenetic appearance of the adenosine receptor precedes N-protein coupling in rat forebrain. ’(H)cyclohexyladenosine in the hippocampus and other regions Dev Brain Res 35:269-274. of rat brain: A quantitative autoradiographic study. Neurosci Munson PJ, Rodbard D (1980): Ligand: A versatile computerized Lett 64: 116-120. approach for characterization of ligand binding sites. Anal BioFastbon J, Pazos A, Probst A, Palacios JM (1986): Adenosine A, chem IO7:220-239. receptors in human brain: Characterization and autoradioPaes de Carvalho R, de Mello FG (1982): Adenosine-elicited accugraphic visualization. Neurosci Lett 65: 127-132. mulation of adenosine 3’-5’-cyclic monophosphate in the chick Gavish M, Goodman RR, Snyder SH (1982): Solubilized adenosine embryo retina. J Neurochem 38:493-500. receptors in the brain: Regulation by guanine nucleotides. SciPaes de Carvalho R, de Mello FG (1985): Expression of A, adenosine ence 21 5: 1632- 1635. receptors modulating dopamine-dependent cyclic AMP accuGeiger JD, LaBella FS, Nagy J1 (1984): Ontogenesis of adenosine mulation in the chick embryo retina. J Neurochem 44:845receptors in the central nervous system of the rat. Dev Brain 851. Res 13:97-104. Patel J , Marangos PJ, Stivers J, Goodwin FK (1982): Characterization Gilman AG (1970): Protein binding assay for adenosine 3’-5‘-cyclic of adenosine receptors in brain using N6-cyclohexyladenosine. monophosphate. Proc Natl Acad Sci USA 67:305-312. Brain Res 237:203-214. M’ Snyder SH Goodman RR’ Cooper MJ’ (hanine Perez MTR, Ehinger BE, Lindstrom K, Fredholm BB (1986): Release nucleotide and cation regulation of the binding of (3H) diethof endogenous and radioactive purines from the rabbit retina. ylphenylxanthine to adenosine A, receptors in brain memBrain Res 398:106-112. branes. Mol Pharmacol 21 :329-335. Piddington R, Moscona AA (1965): Correspondence between gluHamburger V, Hamilton HL (1951): A series of normal stages in the tamine synthetase activity and differentiation in the embryonic development of the chick embryo. J Morphol 88:49-92. retina in situ and in culture. J Cell Biol 27:247-252. Kenimer JG, Nirenberg MW (198 1): Desensitization of adendate CYSchorderet M (1982): pharmacological characterization of clase to prostaglandin E, or 2-chloroadenosine. Mol Pharmacol mediated increase in cyclic AMP in isolated rabbit retina. Fed 20585-591. Proc 41:1707. Lankford KL, de Mello FG, Klein WL (1988): DI-tYPe dopamine Snowhill EW, Williams M (1986): ‘(H) cyclohexyladenosine binding receptors inhibit growth cone motility in cultured retina neuin rat brain: A pharmacological analysis using quantivdtive aurons: Evidence that neurotransmitters act as morphogenic toradiography. Neurosci Lett 6 8 4 - 4 6 . growth regulators in the developing central nervous system. Van Calker D , Muller M, Hamprecht B (1979): Adenosine regulates Proc Natl Acad Sci USA 85:4567-457 1. via two different types of receptors, the accumulation of cyclic Lewis ME, Patel J , Edley SM, Marangos PJ (1981): Autoradiographic AMP in cultured brain cells. J Neurochem 33:999-1005. visualization of rat brain adenosine receptors using Nh- Ventura ALM, ~ l WL,~ de i ~FG (1984): Differential ontogencyclohexy13(H)adenosine. Eur J Pharmacol 73: 109-1 10. esis of D, and D, receptors in the chick embryo retina. Dev Lloyd HGE, Stone TW (1985): Cyclohexyladenosine binding in rat Brain Res I2:2 17-223, striatum. Brain Res 334:385-388. Ventura ALM, Paes de Carvalho R (1987): Development of adenoLockerbie RO, HervC D , Blanc G, Tassin J-P, Glowinski J (1988): sine-dependent cyclic AMP accumulation in the avian optic Isolated neuronal growth cones from developing rat forebrain tectum. Dev Brain Res 35:141-147.