Activation of adenosine A1 receptor by N6-(R)-phenylisopropyladenosine (R-PIA) inhibits forskolin-stimulated tyrosine hydroxylase activity in rat striatal synaptosomes

Life Sciences, Vol. 46, pp. 591-598 Printed in the U.S.A.

Pergamon Press

ACTIVAIION OF ADENOSINE AI RECEPIDR BY N6-(R)-PHENYLISOPROPYLADENOSINE (R-PIA) INHIBITS FORSKOLIN-STIMULATED TYROSINE HYDROXYLASE ACTIVITY IN RAT STRIATAL SYNAFIDSOMES. Maria C. Olianas

and Pierluigi Onali

Department of Neurosciences, University of Cagliari, Via Porcell 4, 09124 Cagliari, Italy. (Received in final form January 2, 1990) Summary We investigated the effect of the relatively selective A1 adenosine receptor agonist N-e(R)-phenylisopropyladenosine (R-PIA) on tyrosine hydroxylase activity ('i~) of synaptosomes obtained from rat striatum. R]q activity was assayed in supernatant obtained following sonication and centrifugation of the tissue preincu~ted with the test con~pounds. R-PIA produced a modest decrease of basal enzyme activity, but significantly reduced the activation of the enzyme by subomximal (0.1-0.5 pM) concentrations of forskolin (FSK) a stimulator of adenylate cyclase. The IC 50 value of R-PIA was 17 nM and the maximal inhibition corresponded to 3 0 - 4 ~ decrease of the enzyme activity stimulated by FSK. The S-isomer of PIA failed to affect TH activity under control and stimulated conditions. Moreover, the inhibitory effect of R-PiA was completely antagonized by 8-cyclopentyl- 1,3 -dimethylxanthine, an adenosine receptor blocker. R-PIA inhibited both basal and FSK-stimulated adenylate cyclase activity, these results indicate that in striatal dopaminergic terminals TH activity can be modulated in an inhibitory manner by activation of presynaptic AI adenosine receptors. Considerable evidence suggests that adenosine receptors modulate central dopaminergic neurotransmission (I). 'lhus, in rodents the administration of the adenosine receptor antagonist caffeine causes an increase in locomotor activity which is prevented by either inhibition of dopamine (DA) synthesis or blockade of DA receptor (2,3). The intrastriatal injection of the stable analog of adenosine 5 '-N-ethylcarboxamide adenosine can elicit homolateral turning behavior in rats treated with the dopamine receptor agonist apomorphine (4). Moreover, adenosine receptor agonists have been found to counteract the hypermotility elicited by d-amphetamine (5) and to decrease striatal dopamine metabolism (6). However, the site(s) of interaction between the central adenosine system and dopamine is not completely defined. Two main classes of adenosine receptors, termed AI and A2, have been identified on the basis of their ability to inhibit and stimulate adenylate cyclase activity, respectively (7). in rat striatum, an area highly innervated by DA, both AI and A2 0024-3205/90 $3.00 +.00 Copyright (c) 1990 Pergamon Press plc

592

R-PIA Inhibition of Synaptosomal TH

Vol. 46, No. 8, 1990

receptors have been shown to be localized on interneurons or glia rather than on dopaminergic terminals (8,9), indicating that adenosine agonists may affect the dopaminergic nigrostriatal transmission by acting postsynaptically through a neuronal feedback loop within the extrapyramidal system. Recently, we have observed that in rat striatum 2-chloroadenosine (2-CADO), an adenosine agonist which stimulates striatal adenylate cyclase via A2 receptors (10), increases synaptosomal DA synthesis and causes a change in the kinetic properties of tyrosine hydroxylase (TH). These data suggest that adenosine agonists might affect DA function by acting presynaptically on DA terminals. In the present study we have investigated the effect on striatal synaptosomal TH activity of N-S(R)-phenylisopropylader~msine (R-PIA), a relatively selective AI adenosine agonist that inhibits striatal adenylate cyclase activity (9,11). Materials and methods. Materials. L- [1!"C]t-yrosine (53.8 ~Ci/mmol),[~-2p]ATP (30-40 Ci/mmol), [2,8-3H]cyciic AMP (25 Ci/nlnol) were obtained from Dupont-New England Nuclear (Boston, MA). Forskolin was purchased from Calbiochem-Behring (San Diego, CA). N-S(R)- 2-phemylisopropyladenosine (R-PIA), N-6(S)-2-phenylisopropyladenosine (S-PIA), 8-cyclopentyl- 1,3 - d~thylxanthine (CFr) were obtained from Research Biochemicals Inc. (Natick, MA). Adenosine deaminase , catalase and the other co,~pc~mds were purchased from Sigma (St. Louis, MO). L-Aromatic amino acid decarbo×ylase was prepared according to Waymire et al. ( 12 ). Preparation of the crude mitochondrial fraction. Male Sprague-Dawley rats (170-250 g ; Charles River, Italy) were sacrified by decapitation and striata were ho,mgenized in 10 vols. of ice-cold 0.32 M sucrose using a teflon-glass h o ~ e n i z e r (clearance 0.25 mm, Kontes , Vineland, N J). A crude mitochondrial fraction (P2 fraction ) was prepared according to the method of Gray and Whittaker (13). The h o ~ e n a t e was centrifuged at 1000 g for I0 rain at 4°C . The resultant supernatant was centrifuged at 12,000 g for 20 rain at 4°C. The pellet was gently resuspended in 0.32 M sucrose and used for TH assay. ~asurement of llq activity. I~ activity was assayed in supernatant obtained following sonication and centrifugati~ of homogenate preincubated with the test co,~pounds. Routinely, one hundred microliters of crude mitochondrial fraction were incubated at 37 °C in a phosphate-buffered medium (final volume 0.5 ml) containing (in raM): NaCI 110, KH2PO41.2 , MGSO41.2, KCI 3.8, CaCIzl.2, glucose 10 and sodium phosphate 10 to give a final pH of 7.3. Adenosine deaminase was present at the concentration of 0.5 U/ ml. The medium was prepared fresh each day and gassed with ,~*dical grade O 2 prior to use. R-PIA, S-PiA and CFf were added at the beginning of the incubation. After 10 min, either vehicle or forskolin was added and the incubation was continued for 20 rain. The sables were then placed on ice and immediately centrifuged at 15,600 g for 120 s at 4°C. The supernatants were discarded and the pellets were sonicated for 10 s in an ice-cold buffer containing 20 mM sodium phosphate (pH 7.41, 50 mM NaF, I m~ dithiothreitol (DTT) and I ,,M EDTA . The sonicated san~les were centrifuged at 4°C for 10 rain at 15,600 x g and aliquots of the supernatants were immediately

Vol. 46, No. 8, 1990

R-PIA Inhibition of Synaptosomal TH

593

assayed for TH activity, lhe enzyme activity was determined following the two-steps procedure described by Haycock et al. (14). Fne reaction mixture (final volume 100 ul) contained 100 mM potassium phosphate (pH 7.0), 2000 U of catalase, 50 ~M NaF, 5 ,~M L-ascorbic acid, 5 mM EDTA, 0.2 ~M D,L6-methyl-5,6,7,8-tetrahydropterin (6-MePH 4 ) and 5 nmol of L-[I-~4C] tyrosine. Reaction was started by the addition of 50 ~i of the tissue preparation (25-40 ug of protein) and samples were incubated at 37°(: for 15 min. Tnereafter, 25 N1 of a freshly prepared solution containing 400 mM Tris-HCl (pH 9.0), 10 ~i of L-aromatic amino acid decarboxylase, 1.25 mM pyridoxal 5-phosphate and 12 .~H 3 - i o d o t y r o s i n e were added to each sample. '1~e evolved 14C02 was measured by liquid scintillation counting as p r e v i o u s l y described (10). Blank samples were incubated e i t h e r in the absence of t i s s u e or without 6-MePH4 and were in both cases less than 0.1°/o of the t o t a l r a d i o a c t i v i t y added. Assays were c a r r i e d out in triplicates. Measurement of adenylate cyclase activity. For adenylate cyclase activity the P2 fraction was resuspended in 20 vols. of hypotonic 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES)/ NaOH Duffer (pH 7.4) containing I n,M DTI, washed once by centrifugation at 28,000 g for 15 rain and resuspended in the same buffer at a protein concentration of 0.9-1.2 mg/ml. The enzyme reaction mixture (100 ~i) contained 80 mM Tris/HCl (pH 7.4), 0.5 mM MgCI2, 50 ~M [a3-2P]ATP (100 cpm/pmol), I mM DTr, I ,~M cyclic AMP, 5 mM phosphocreatine, 5 U of creatine phosphokinase, 0.2 mM papaverine, 100 ~M GI'P , 0.5 U of adenosine deaminase and 50 ~g of bovine serum albumin. [~H]cyclic AMP (about 10,000 cpm) was included to monitor cyclic AMP recovery. The reaction was started by addition of the tissue ( 25-35 pg of protein) and was carried out at 30°C for 10 min. Cyclic AMP was isolated according to the procedure of Salomon et al. (15). Assays were run in triplicates. Protein were determined by the method of Bradford (16) using bovine serum albumin as a standard. Statistical signif.icance of the results was evaluated by Student's t-test. Resul ts incubation of striatal synaptosomes with increasing concentrations of R-PIA elicited a concentration-dependent inhibition of basal enzyme activity with an IC 50 (concentration producing half-maximal inhibition) value of 20 , 3 rim (Fig I). Although this effect was consistently observed, it corresponded to only 11 % decrease of basal enzyme activity at maximally effective concentrations of R-PIA. However, when IH activity was stimulated by forskolin (FSK), an activator of adenylate cyclase (17), the inhibitory effect of R-PIA was amplified. R-PIA maximally decreased the enzyme activity stimulated by 0.5 uM FSK by 35 % with an IC 50 value of 17 ± 2 nM. In contrast, the S-isomer of PIA (S-PIA) failed to affect the enzyme activity under control add stimulated conditions. FSK stimulated synaptosomal TH with an EC 50 (concentration producing half- maximal activation) value of 0.6 ± 0. I ~M (Fig. 2). The maximal stimulation corresponded to a twofold increase of basal enzyme activity and was obtained with 50 NM FSK. R-PIA significantly inhibited the stimulation of the enzyme activity elicited by 0.1, 0.25 and 0.5 ~M FSK, which increased TH activity by about 25%, 45% and 60% respectively. At higher concentrations of

594

R-PIA Inhibition of $ynaptosomal TH

Vol. 46, No. 8, 1990

Q.

._: E \

o" ~Ta U o

f

~o'~~ -6

-i

-~

PIA

4

-g

log (M)

FIG,I Effects of R-PIA (circles) and S-PIA (triangles) on basal (open symbols) and FSK-stimulated (closed symbols) synaptosomal TH activity. Striatal tissue was incubated for ]0 rain in the presence ,

of

, . ,

the

indi¢.ated

c o n c e n t r a t i o n s o f R-PIA o r S-PIA.

V e h i c l e o r FSK

(0.5 uM) were then added and the incubation was continued for 20 rain. Data are the mean , S ~ of five experiments.Op< 0.05 vs FSK alone.

310.

270 % c "i 2 3 0 .

__ 1 9 0 .

EL t6Q

6"

-~ FSK

by

R-PIA

of

-e log

-~

-4

(M)

FIG. 2

Inhibition

activity

as a f u n c t i o n o f FSK c o n c e n t r a t i o n s .

FSK-stimulated

synaptosomal

Striatal

rH

synaptosomes

were incubated for "]0 rain with vehicle (O) or 0.5 ~M R-PiA (9#). FSK was t h e n added and t h e i n c u b a t i o n was c o n t i n u e d f o r 20 rain. Data a r e the mean t SEM of five experiments.~p
Vol. 46, No. 8, 1990

R-PIA Inhibition of Synaptosomal TH

595

TABLE I Effect of 8-Cyclopentyl-1,3-dimethylxanthine (CPT) on R-PIA Inhibition of FSK-stimulated Tyrosine Hydroxylase (TH) Activity in Rat Striatal Synaptosomes. TH activity (pmol CO2/min/ mg prot) basal R-PIA 0.5 ~M CPT O. I pM CPT + R-PIA FSK 0.5 uM FSK + R-PIA FSK + CPT FSK + CPT + R-PIA

145.0 129.1 146.4 147.3 230.7 185.6 240.7 230.8

, 8.9 , 6.5 , 10. I ± 9.8 ~ 11.5** , 8.9* , 11.9 , 7.9a

Striatal synaptosomes were incubated as in Fig. I. CPT was added together with R-PIA. TH activity was assayed in supernatants of lysed synaptosomes with 0.2 mM 6-MePH 4 . Data are the mean e SEM of three experiments. ~ p c 0.01 vs basal; * p c 0.05 vs FSK alone; a not significantly different from FSK + C.Fr. FSK, the inhibitory effect of R-PIA decreased progressively, being absent at 10 NM FSK. Thus, in the presence of R-PIA the concentration-response curve of FSK was shifted to the right yielding an EC 50 value of 1.7 , 0.3 ~M. When the synaptosomal preparation was incubated in the presence of R-PIA together with 0.1 ~M CPT, an adenosine antagonist with higher selectivity for A1 receptors (18), no significant inhibition of FSK-stimulated enzyme activity by R-PIA was observed ( Table I). When tested alone, CPT did not significantly affect TH activity. Moreover, addition of increasing concentrations of R-PIA ( from 10 nM to 10 ~M) directly to the enzyme assay mixture failed to modify the enzyme activity extracted from synaptosomes preincubated either in the presence or in the absence of 0.5 NM FSK (results not shown). FSK, tested at concentrations ranging from 10 r~M to 100 pM, maximally stimulated striatal adenylate cyclase activity by sevenfold with an EC 50 value of 0.7 ~ 0.1 pM. R-PIA inhibited basal adenylate cyclase activity with an IC 50 value of 50 * 5 nM. The maximal inhibitory effect corresponded to 2 1 % decrease and was reached with 0.5 ~M R-PIA. R-PIA also significantly inhibited the stimulation of the enzyme activity elicited by 50 nM FSK. Values of enzyme activity (expressed as pmol of cyclic AMP/sin/ mE prot , S~M, n= 4 ) were: basal 218 , 10, R-PIA 0.5 NM 173 * 8 (pc0.05 vs basal), FSK 50 nM 284 , 14 (pc0.01 vs basal), R-PIA + FSK 227 * 12 (pc 0.01 vs FSK alone). The net stimulation elicited by FSK was inhibited by R-PIA by 18 % (p(0.05). Discuss ion The present study shows that in rat striatal synaptosomes R-PIA inhibits the stimulation of *[H activity by FSK. A series of evidence indicates that this effect is mediated via stimulation of adenosine AI receptors. First, the

596

R-PIA Inhibition of Synaptosomal TH

Vol. 46, No. 8, 1990

inhibition occurs at nanomolar concentrations of R-PIA in agreement with the high affinity of this agonist for AI receptors reported in radioligand binding studies (18). Low concentrations of R-PIA are also required for eliciting other neuronal responses mediated by AI receptors, such as attenuation of cyclic A ~ formation (present study and ref. 9,11), inhibition of noradrenaline release (19) and depression of evoked potentials in brain slices (20). Second, within the range of concentrations tested, the S-isomer of PIA is inactive indicating a high degree of stereospeeificity of the inhibitory effect which is in line with the pharmacological profile of AI receptor-mediated responses (7). In contrast to R-PIA, S-PIA was also found to be ineffective in decreasing striatal DA synthesis in vivo (6). Third, the inhibition elicited by R-PIA is completely antagonized by CPT, a rather selective AI receptor antagonist (18). Basal TH activity is little sensitive to the inhibitory effect of R-PIA as, on the average, only a 117. decrease is obtained with concentrations of R-PIA ma×immlly effective in inhibiting the FSK-stimulated enzyme activity. Different factors may account for this small response. In fact , basal enzyme activity could be already inhibited by endogenous substances so that addition of R-PIA can produce only a slight increase of the inhibitory signal. One of these substances could be adenosine. However, the tissue is incubated in the presence of adenosine deaminase to decrease the concentration of endogenous adenosine and, under these conditions, the addition of the adenosine antagonist CPT fails to produce a significant increase of TH activity as expected if a consistent inhibition by endogenous adenosine is present. Alternatively, under resting condition the intracellular signal generated by A1 receptor may be too low to produce a significant decrease of TH activity. However, when llq is activated by forskolin, the AI inhibition may be amplified and can be consistently observed. Forskolin is found to stimulate striatal TH and adenylate ¢yclase activities with a similar potency, supporting the idea that this agent activates synaptosomal TH via a cyclic AMP-dependent pathway (21,22). On the other hand, R-PIA inhibits striatal adenylate cyclase activity and reduces the stimulation of the enzyme elicited by FSK. These results are in agreement with the observation that R-PIA is effective in reducing the accumulation of cyclic AMP in brain slices when adenylate cyclase activity is activated by FSK (23). ~nerefore, R-PIA may inhibit the FSK activation of TH by reducing the FSK stimulation of a presynaptic adenylate cyclase. The inhibitory effect of R-PIA on FSK-stimulated 'l~ activity is observed only at submaximal concentrations of FSK. A likely explanation is that the cyclic A~-dependent pathway regulating ~H can be modulated by A] receptors only when the intracellular level of cyclic ~ P is not saturating. This type ot response has also been observed when the inhibition of FSK-stimulated 17~ activity by DA autoreceptor activation was studied (24,25). r[he present data , however, do not rule out the involvement in the R-PIA inhibition of receptor mechanisms other than inhibition of adenylate cyclase. In fact, this adenosine agonist has been found to stimulate a particulate low Km cyclic &MP phosphodiesterase from rat brain (26), an effect that may attenuate the cyclic AMP accumulation elicited by FSK in our tissue preparation. Moreover, AI receptors have been shown to inhibit phosphoinositides hydrolysis in rat striatum (27) and therefore are expected to reduce the generation of inos[tol triphosphate and diacylglycerol which may control TH activity via the Ca+?calmodulin-dependent

Vol. 46, No. 8, ]990

R-PIA Inhibition of Synaptosomal TH

597

protein kinase and protein kinase C (28-30). Further experiments are required to ascertain the possible partecipation of these mechanisms in the inhibitory e t ~ e c t of ~-PIA. Previous studies have shown that 2-CADO , an adenosine agonist active at both AI and A2 receptors (18), causes stimulation rather than inhibition of TH activity in the strzatum (10,31). We have been unable to detect a significant inhibitory effect ot 2-CADO on synaptosomal YH activity oi rat striatum ~nen this agonist was tested at concentrations ranging from I rum to I ~M, ~nhereas a marked stin~llation was observed with higher aRonist concentrations. 2-CADO has also oeen shown to be less effective than PIA in inhibiting striatal adenylate cyclase activity (9). fhe di£ferent response elicited by R-PIA and 2-CADO might be explained by considering that 2-CADO has a lower affinity for AI receptors and higher affinity for A2 receptors than R-PIA (18). tV~reover, radioli~and binding experiments have demor~strated that 2-CADO has a lower ability to discriminate between AI and A2 sites than R-PIA (18). Fnus, if one assumes that stimulation and inhibition of striatal IH activity are n~diated via activation of A2 and AI receptors, respectively, and considers that in rat striatum the density of A2 sites is higher than that of AI sites (18), it is conceivable why the prevalent and most significant effect on TH activity of 2-CADO is stimulatory rather than inhibitory. in conclusion, the demonstration that in striatal synaptosomes activation of AI receptors can control TH activity is a further support to the idea that adenosine can affect striatal DA transmission by acting presynaptically. ~oreover, the present data together with the previous observations that 2-CADO stimulates striatal 133 activity (10,31), suggest a bim~xlal regulation of TH by adenosine agonists via activation of different types of receptors.

References I. Z. 3. 4. 5. b. 7. 8. 9. i0. 11. 12. lJ. 14. 15.

M.F. ]ARVIS and M. WILLIAMS, TIPS 8 330-332 (1987). B. WALDECK, Acta Pharmac. fox. 36 suppl. 4 1-23 (1975). B.C. WHI~E, C.C. SIMPSON, J.E. ADAMS and B. Jr. FIARKINS, Neuropharmacology i7 511-513 (197~). R.D. GREEN, H.K. PROUDFIT, S.H. YEUNG, Science 218 58-60 (1982). I".G. HEFFNER, J.N. WILEY, A.E. WILLIAMS, R.F. BRUNS, L.L. COUGHENOUR, and D.A. DOWNS, PsychopB~rmacoiogy 98 31-37 (19~9). S. MYERS and %.A. PUGSLEY, Brain Res. 375 193-197 (1986). M. WiLLIamS. 'FINS 7 164-168 (1984). J. GEIGER, Brain Res. 363 404-408 (1986). W.J. WOJCiK and N.H. NEFF, Neurosci. Lett. 41 55-60 (1983). P. ONALI, M.C. OLIANAS and B. BUNSE, Brain Res. 456 302-309 (1988) C. EBERSOLT, J. PREEN[, A. PROCHIANTZ, M. PEREZ, J. BOCKAERT, Brain Res. 267 123-129 (1983). J.C. WAYMiRE, R. BJUR and N. WEINER, Anal. Biochem. 43 588-600 (1971). E.G. GRAY and V.P. WHIFfAKER, J. Anat. 96 79-88 (1962). 3.w. HAYCOCK, J.A. MELIGENi, W.F. BENNE~, J.C. WAYMiRE, J. Biol. Chem. 257 126~I-126~ (1982). Y. SALC~ON, D. LONDOS, M. RODBELI,, Anal. Biochem. 58 541-548 (1974).

598

R-PIA Inhibition of Synaptosomal TH

Vol. 46, No. 8, 1990

16. M.M. BRADFORD, Anal. Biochem. 72 248-254 (1976). 17. K.B. SEAMON and J.W. DALY, J. Cyclic Nucleotide Res. 7 201-204 (1981). 18. R. F. BRUNS, G.H. LU, T.A. PUGSLEY, Mol. Pharmacol. 29 331-346 (1986). 19. R. JACKISCH, R. FEHR, G. HERITING, Neuropharmacology, 24 499-507 (1985). 20. M. REDDINGTON, K.S. LEE, P. SCHUBERT, Neurosci. Lett. 28 275-279 (1982). 21. S. EL MESTIKAWY, H. GOZLAN, J. GLOWlNSKY, M. HAMON, J. Neurochem. 45 173-184 (1985). 22. K.A. STRAIT and R. KL~ENSKI, Mol. Pharmacol. 29 561-569 (1986). 23. B.B. F R F ~ L M , B. JONZON, K. LINDS~ROM, Acta Physiol. Scand. 117 461-463 (1983) 24. S. EL MESTIKAWY and M. HAMON, J. Neurochem. 47 1425-1433 (1986). 25. P. ONALI and M.C. OLIANAS, Neurosci. Lett. in press. 26. P. DE MAZANOOUK~ and Y. GIUDICELLI, Brain Res. 300 211-217 (1984). 27. D. W. PETCOFF and D.M.F. COOPER, Eur. J. Pharmacol. 137 269-271 (1987). 28. T. Y A M ~ I , H. NAEATA, H. FUJISAWA, J. Biol. Chem. 256 5404-5409 (1981). 29. K.A. ALBERT, E. HEEMER-MATYjEK, A.C. NAIRN, T.H. MULLER J.W. HAYCOCK, L.A. GREENE, M. GOE/)STEIN, P. GREENGARD, Proc. Natl. Acad. Sci. USA 81 77137717 (1984). 30. P. ONALI and M.C. OLIANAS, Life Sci. 40 1219-1228 (1987). 31. Y. KURODA and K. KOBAYASHI, Proc. Jpn. Acad. 54b 640-64A (1978).