Pharmacological characterization of adenosine A1 receptors and its functional role in brown trout (Salmo trutta) brain

Pharmacological characterization of adenosine A1 receptors and its functional role in brown trout (Salmo trutta) brain

Brain Research 837 Ž1999. 46–54 www.elsevier.comrlocaterbres Research report Pharmacological characterization of adenosine A1 receptors and its func...

241KB Sizes 0 Downloads 33 Views

Brain Research 837 Ž1999. 46–54 www.elsevier.comrlocaterbres

Research report

Pharmacological characterization of adenosine A1 receptors and its functional role in brown trout ŽSalmo trutta . brain A. Poli b

a, )

, R. Lucchi a , S. Notari a , G. Zampacavallo a , O. Gandolfi a , U. Traversa

b

a Department of Biology, UniÕersity of Bologna, Via Selmi 3, 40126 Bologna, Italy Basic Research and IntegratiÕe Neuroscience (B.R.A.I.N.), Department of Biomedical Sciences, UniÕersity of Trieste, Via Giorgieri 7, 34127 Trieste, Italy

Accepted 1 June 1999

Abstract The adenosine receptor agonist N 6-cyclohexylw3 Hxadenosine Žw3 HxCHA. was used to identify and pharmacologically characterize adenosine A1 receptors in brown trout Ž Salmo trutta. brain. In membranes prepared from trout whole brain, the A1 receptor agonist w3 HxCHA bound saturably, reversibly and with high affinity Ž K d s 0.69 " 0.04 nM; Bmax s 0.624 " 0.012 pmolrmg protein. to a single class of binding sites. In equilibrium competition experiments, the adenosine agonists and antagonists all displaced w3 HxCHA from high-affinity binding sites with the rank order of potency characteristic for an adenosine A1 receptors. A1 receptor density appeared not age-related Žfrom 3 months until 4 years., and was similar in different brain areas. The specific binding was inhibited by guanosine 5X-triphosphate ŽIC 50 s 0.778 " 0.067 mM.. GTP Ž5 mM. induced a low affinity state of A1 receptors. In superfused trout cerebral synaptosomes, 30 mM Kq stimulated the release of glutamate in a calcium dependent manner. Glutamate-evoked release was dose-dependently reduced by CHA, and the inhibition was reversed by the A1 antagonist 8-cyclopentyltheophylline ŽCPT.. In the same synaptosomal preparation, 30 mM Kq as well as 1 mM glutamate stimulated the release of adenosine in a Ca2q-independent manner and tetrodotoxin insensitive. These findings show that in trout brain adenosine A1 receptors are present which are involved in the modulation of glutamate transmitter release. Moreover, the stimulation of adenosine release by Kq depolarisation or glutamate support the hypothesis that, as in mammalian brain, a cross-talk between adenosine and glutamate systems exists also in trout brain. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Adenosine; A1 receptor; Trout brain synaptosome; Glutamate release

1. Introduction Adenosine is an inhibitory modulator of synaptic transmission and neuronal activity w14,41,43x. In the central nervous system this regulation is achieved partly by a suppression of the release of neurotransmitters, especially those with a predominantly excitatory function such as glutamate w4,15x and acetylcholine w40x. Adenosine also acts directly upon neuronal postsynaptic membranes to increase potassium conductance so as to depress excitability w16,42x. These actions seem to be related to the activa-

) Corresponding [email protected]

author.

Fax:

q0039-051-251208;

E-mail:

tion of the adenosine A1 receptor subtype which is coupled to G-proteins w7,31x. A1 adenosine receptors have been shown to have a broad phylogenetic distribution in the brains of both lower and higher vertebrates, but is not detectable in central nervous tissue of the molluscs or arthropods tested w39x, even if some findings indicated the possible involvement of adenosine receptors in the inhibition of monoamine release in the pedal ganglia of a marine bivalve w1x. Recently, in membranes from the freshwater teleost Carassius auratus Žgoldfish., we found high-affinity binding sites for w3 HxCHA which are coupled to a G-protein and show pharmacological characteristics of the A1 adenosine receptors w24x. Subsequently, in goldfish synaptosomal preparations, we have demonstrated an A1 receptor-mediated inhibition of glutamate release which seems to involve inhibition of Ca2q entry through voltage-dependent Ca2q

0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 7 0 1 - 1

A. Poli et al.r Brain Research 837 (1999) 46–54

channels in nerve terminals w34x. The A1 receptors in brain of goldfish appear pharmacologically and functionally similar to those found in mammals, but their density is much lower. In addition to the well-known role as inhibitory neuromodulator, adenosine has been reported to show a protective effect on neuronal or glial damage during anoxia or ischemia mediated by A1 receptors w6,11,37x. Mammalian brain shows a very low tolerance to anoxia since it produces a rapid fall in ATP production and a severe depletion of ATP. The ATP dependent NaqrKq pump fails as a consequence, resulting in the loss of transmembrane ionic gradient and depolarisation. Depolarisation produces an uncontrollable loss of excitatory neurotransmitters, in particular glutamate and aspartate, into the extracellular space. These neurotransmitters are thought to be a major cause of further neuronal damage since they overactivate N-methylD-aspartate ŽNMDA. receptors that when open allow more damage producing Ca2q to enter the neurons w36x. Free adenosine acts to decrease energy demand and to increase energy supply w2x. It decreases energy demand by depressing neuronal activity, by inhibition of excitatory neurotransmitter release, e.g., glutamate w17x, inhibition of calcium efflux and stimulation of potassium efflux and chloride influx w37x. The general pattern of anoxia driven degenerative changes appear to be common to the brains of all vertebrates, not only mammals, although the time scale may vary between species due to differences in body temperature and metabolic rate. Among the fishes, cyprinid of the genus Carassius Žcrucian carp and goldfish. can survive to anoxia for weeks or even months w29,32x, whereas the rainbow trout Ž Oncorhynchus mykiss . is probably a good representative of the anoxia-intolerant majority of fishes. A recent study has shown that the brain of the rainbow trout was unable to maintain Kq homeostasis during anoxia w30x. After about 30 min of anoxia at 108C the rainbow trout brain showed an abrupt rise in wKqx o indicating complete depolarisation. By contrast, the anoxia tolerant crucian carp was found to maintain a low wKqx o for at least 6 h of anoxia. If the Q10 for brain O 2 consumption is about 2.0 in fish as well as in mammals w25x, the rainbow trout time period of 30 min to anoxic depolarization at 108C is analogous to about 5 min at 378C, close to that seen in mammals. Since in mammals adenosine is considered to be an endogenous neuroprotective agent that is released into extracellular space during ischemia, the present study has been addressed to elucidate the kinetic characteristics of A1 adenosine receptors in the brain of brown trout Ž Salmo trutta. and its modulatory role on presynaptic release of excitatory amino acid neurotransmitters, as shown in different vertebrates w34,35x. In addition, as the anoxia exposure of rainbow trout brain resulted in the release of Kq and several amino acids, notably glutamate w19x, we have investigated the possible relationship between the release

47

of adenosine from trout brain synaptosomes, during depolarising conditions obtained in presence of high Kq or glutamate.

2. Materials and methods 2.1. Materials Cyclohexyladenosine, N 6 -wadenosine-2,8-3 Hx Žw3 HxCHA., 34.4 Cirmmol was obtained from Dupont-NEN ŽBoston, MA.. Unlabeled CHA, 2-chloro-N 6-cyclopentyladenosine ŽCCPA., the R- and S-diastereomers of phenylisopropyladenosine Ž R- and S-PIA., N-ethylcarboxyamidoadenosine ŽNECA., 8-cyclopentyl-1,3-dipropylxanthine ŽDPCPX., theophylline ŽTHEO. and cyclopentyltheophylline ŽCPT. were obtained from Research Biochemicals ŽWayland, MA.. Adenosine deaminase ŽADA, from calf intestine, 200 IUrmg. from Boehringer ŽMannheim, Germany.. All other chemicals were of analytical grade or the best commercially available. 2.2. Animals Brown trout Ž S. trutta. were obtained from fish-breeding of Servizio e Tutela Sviluppo Fauna, Provincia di Bologna. Except for age-related binding measurements, receptor characterization and release experiments were carried out with 3 years old trout, acclimatized in large aquaria filled with dechlorinated freshwater maintained at x w x 108C–128C and with controlled wNHq 4 and O 2 . Fish were fed daily and exposed on a 12:12 light–dark cycle for at least 10 days before the experiments were performed. 2.3. Preparation of membranes The animals were killed by decapitation; brains were excised and immediately frozen on dry ice prior to storage at y808C. On the day of the experiments the frozen brain was thawed and then homogenized in 10 vols. of 50 mM Tris–HCl buffer ŽpH 7. with an Ultra-Turrax ŽIka-Werk, Germany.. The resulting homogenate was centrifuged at 48 000 = g for 20 min at 48C and the supernatant was discarded. The pellet was washed by resuspension in an identical volume of Tris buffer using the Ultra-Turrax and recentrifuged at 48 000 = g for 20 min at 48C. The resulting pellet was resuspended in about 10 vols. of Tris buffer, gently homogenized and incubated with ADA Ž5 IUrml. for 30 min at 228C to remove endogenous adenosine. At the end of incubation the suspension was cooled on ice, then centrifuged at 48 000 = g at 48C for 10 min at 48C. The resulting pellet was resuspended in an appropriate volume of Tris–HCl buffer, pH 7, containing 5 mM MgSO4 Ž6 mg proteinrml.. Aliquots of the homogenate were used immediately in receptor binding assay, in presence of 2 UIrml of ADA.

48

A. Poli et al.r Brain Research 837 (1999) 46–54

Protein was determined by the method of Lowry et al. w22x, using bovine serum albumin as the standard. 2.4. Binding assay The specific binding of w3 HxCHA was determined using a previously described assay w3x with minor modifications. Briefly, aliquots of membrane preparations Ž0.2 mg protein. were incubated in the presence of 2 IUrml of ADA, with different concentrations of w3 HxCHA, ranging from 0.25 to 18 nM, in a total volume of 1 ml of 50 mM Tris–HCl buffer, pH 7, 5 mM MgSO4 at 228C for 120 min. Non-specific binding was determined in assays with 10 mM R-PIA. Bound and free radioligand were separated by rapid filtration of samples through Whatman GFrB glass fiber filters previously soaked in 50 mM Tris–HCl buffer. The filters were immediately washed four times with 3 ml of ice cold Tris–HCl Žless than 5 s., dried and placed in 5 ml scintillation vials containing Filter Count cocktail ŽPackard Instruments. and counted by scintillation spectrometry ŽBeckman L.S. 1801.. Membranes prepared as above from brain of different age trouts or from the different areas of 3 years old trout brain Žolfactory bulb, telencephalon, optic tectum, cerebellum and hypothalamus. were used for age-related and regional distribution of w3 HxCHA binding sites.

Displacement of the binding of w3 HxCHA Ž2 nM. to membranes from trout brain was made by increasing concentrations of unlabeled agonists and antagonists. Ten micrometers R-PIA was used to define non-specific binding. 2.5. Synaptosomal preparations The synaptosomes from 3 years old brown trout Ž S. trutta. brain were prepared according to Poli et al. w34x with minor modifications. Trout were killed by decapitation; brains were quickly excised cutting the spinal connection behind the vagal lobe, and homogenized in 10 vols. Žwrv. of ice-cold 10% sucrose pH 7.4 Ž800 rpm, eight strokes.. The homogenate was centrifuged at 1000 = g for 15 min; the supernatant was collected and centrifuged for 20 min at 17 000 = g. The pellet was washed by resuspension in 0.32 M sucrose and centrifuged as before. The final pellet ŽP2 fraction. was resuspended in an oxygenated medium composed of 120 mM NaCl, 3.5 mM KCl, 8.5 mM Na 2 HPO4 , 2.1 mM NaH 2 PO4 , 10 mM glucose, 1.2 mM MgCl 2 , 1.3 mM CaCl 2 ŽpH 7.4., and used for release experiments. 2.6. Glutamate release The resuspended P2 fraction was diluted in 10 vols. of phosphate buffer. Synaptosomes were incubated for 10

Fig. 1. Saturation of w3 HxCHA binding in membranes from whole brain of brown trout. Specific binding is defined as the difference between total and non-specific binding. Data were fitted by a computerized non-linear regression analysis, and resolved for the presence of one site with K d s 0.69 " 0.04 nM and Bma x s 0.624 " 0.012 pmolrmg protein. Values represent the mean of triplicate determinations for each dose in seven separate experiments performed as described in the text; the S.E.M. was less than 10% in all cases. Inset, the Scatchard plot from the data.

A. Poli et al.r Brain Research 837 (1999) 46–54 Table 1 K i values calculated from displacement experiments of specific binding of w3 HxCHA to whole brain membranes from S. trutta by adenosine agonist and antagonist Compound

Ki

CCo

R-PIA CCPA CHA NECA S-PIA DPCPX THEO

0.46"0.03 nM Ž4. 0.54"0.02 nM Ž6. 0.71"0.05 nM Ž4. 1.92"0.21 nM Ž3. 4.33"0.07 nM Ž4. 5.39"0.06 nM Ž5. 20,909"343 nM Ž3.

y0.99 y0.99 y0.98 y0.93 y0.98 y0.98 y0.90

Displacement of the binding of w3 HxCHA Ž2 nM. to membranes from trout brain by increasing concentrations of unlabeled agonists and antagonists. Ten micrometers R-PIA was used to define non-specific binding. Experiments were performed as in Section 2. The K i "S.E. values were calculated by Graph Pad PRISM 2.0 data analysis of triplicate determinations in different experiments Žnumber in brackets..

min at 158C in a shaking bath; then aliquots of the suspension containing 0.6–0.8 mg of protein were plated over Millipore filters Ž0.67 mm pore size. situated at the bottom of 12 parallel perfusion chambers. The synaptosomal beds were superfused for 50 min at a rate of 0.5 mlrmin with a medium of the above composition, thermostated at 158C and continuously oxygenated. When the medium containing 30 mM Kq was used, the concentration of NaCl was lowered to 93.5 mM to maintain osmolarity. When required, Ca2q-free medium containing 0.1 mM EGTA replaced the normal medium 15 min after starting. At the end of the 50-min period, a 2.5-min

49

fraction was collected so that the basal release could be measured. The medium was then changed according to the various experimental protocols, a 1-min fraction was discarded because the void volume of the perfusion system corresponded to 1.0 ml, and one further 2.5-min fraction was collected. Fractions were lyophilized and resuspended in 0.2 ml of borate buffer 50 mM ŽpH 9.5., and glutamate levels were determined by HPLC w18x. 2.7. Adenosine release The P2 fraction prepared as above, was superfused utilizing the same apparatus and the same protocol described before. At the end of the 50-min period of superfusion, three 2.5-min fractions were collected so that the basal release could be measured. The medium was then changed according to the various experimental protocols, a 1-min fraction was discarded Žvoid volume. and three further 2.5-min fraction were collected. The medium was again changed to normal medium and a final three 2.5-min fractions were collected. Fractions were lyophilized and resuspended with 0.2 ml of bidistilled water. Adenosine levels were determined by HPLC w9x. 2.8. Statistical analysis The analyses of saturation isotherm curves of binding were performed by Graph Pad PRISM 2.0 computerised program. Results from release experiments are expressed as the mean " S.E.M. The significance of the differences

Fig. 2. Ža.GTP inhibition of the specific binding of w3 HxCHA. Membranes were incubated with 2 nM w3 HxCHA and various concentrations of GTP in standard assay conditions. IC 50 value Ž0.778 " 0.067 mM. was calculated by Graph Pad PRISM data analysis of triplicate determinations in three different experiments. Žb. Saturation data of w3 HxCHA specific binding in the absence and in the presence of 5 mM GTP. The data were fitted by a computerized non-linear regression analysis and shown as Scatchard plot. The data were resolved for the presence of one site with K d s 0.707 " 0.021 nM and Bma x s 0.402 " 0.067 pmolrmg protein for controls and K d s 1.370 " 0.068 nM and Bmax s 0.294 " 0.044 pmolrmg protein for GTP. Values represent the mean of triplicate determinations for each dose in three separate experiments performed in standard assay conditions; the S.E.M. was less than 10% in all cases.

A. Poli et al.r Brain Research 837 (1999) 46–54

50

between two means was usually calculated with the Student’s two-tailed paired test. For multiple comparisons the ANOVA followed by Dunnett’s t-test was performed. The significant difference of the percentage variations was calculated on arcsin transformed percentage.

3. Results 3.1. [3 H ]CHA binding to trout brain membranes 3

The specific binding of w HxCHA appeared to be saturable when the total concentration of ligand ranged from 0.1 to 18 nM ŽFig. 1.. Non-specific binding increased linearly with the concentration of w3 HxCHA Žnot shown.. The isotherm saturation data were resolved for the presence of a single high-affinity binding site with a K d s 0.69 " 0.04 nM and Bmax s 0.624 " 0.012 pmolrmg protein. The specific binding was highly dependent on the presence of ADA Ž2 IUrmg protein.. In the absence of ADA no w3 HxCHA binding was detected. To confirm the results of the saturation experiments, K i value was determined by displacement of bound w3 HxCHA Ž2 nM. with increasing concentrations of unlabeled CHA. The displacement curve for CHA was monophasic with a K i s 1.10 " 0.13 nM ŽCCo s y0.99.. This value was in good agreement with the results of saturation experiments for one-site binding site. 3.2. Competition experiments Competition experiments with adenosine and xanthine analogs were performed in order to assay the pharmacological profile of binding sites for w3 HxCHA. In standard assay conditions several adenosine agonists and antagonists effectively displaced the binding of w3 HxCHA Ž2 nM. in a concentration-dependent manner. The displacing potency of agonists was R-PIA) CCPA) CHA ) NECA) S-PIA, the R-isomer of PIA being 10 times more potent than S-isomer. DPCPX, the most selective antagonist for A1 receptors in mammalian brain w21x, is approximately 4000 times more effective than THEO in displacing w3 HxCHA. K i values are reported in Table 1.

Table 3 Age-related K d and Bmax values of specific w3 HxCHA binding to whole brain membranes from S. trutta Age

K d ŽnM.

Bmax Žpmolrmg protein.

3 months Ž5. 6 months Ž4. 1 year Ž4. 2 years Ž5. 3 years Ž8. 4 years Ž3.

0.58"0.05 0.55"0.08 0.73"0.08 0.66"0.05 0.69"0.04 0.56"0.08

0.576"0.014 0.445"0.008 0.495"0.033 0.563"0.008 0.624"0.012 0.421"0.068

Experimental procedure was as reported in Section 2. The values are the mean"S.E.M. from different experiments Žnumber in brackets.. One-way analysis of variance ŽANOVA. did not show any significant differences of K d and Bmax values among groups.

3.3. Effect of guanosine 5X-triphosphate on [3 H ]CHA binding The results of competition experiments show that the w3 HxCHA binding sites in trout brain membranes belong to the adenosine A1 subtype receptors. It is known that these sites are G-protein coupled receptors. The influence of GTP on 2 nM w3 HxCHA binding in trout brain membranes is shown in Fig. 2a and b. The experiments were performed in the absence of Mg 2q. GTP decreased w3 HxCHA specific binding in a concentration-dependent manner up to 80% at a concentration of 10 mM ŽFig. 2a.. The calculated IC 50 value was 0.778 " 0.067 mM. The reduction in specific w3 HxCHA binding by GTP was due to a significant decrease both in affinity Ž P - 0.001. and in maximal binding capacity Ž P - 0.01. ŽFig. 2b.. The saturation experiments, performed in the absence or in the presence of 5 mM GTP gave the following results: K d s

Table 2 Regional distribution of w3 HxCHA binding in brown trout brain Brain region

K d ŽnM.

Bmax Žpmolrmg protein.

Whole brain Ž8. Olfactory bulb Ž3. Telencephalon Ž5. Optic tectum Ž6. Cerebellum Ž5. Hypothalamus Ž3.

0.691"0.038 0.503"0.045 0.493"0.030 0.644"0.023 0.541"0.012 0.737"0.068

0.624"0.012 0.512"0.049 0.642"0.028 0.584"0.022 0.601"0.019 0.487"0.055

Experimental procedure was as reported in Section 2. The values are the mean"S.E.M. from different experiments Žnumber in brackets.. Statistical comparisons were made by ANOVA followed by Dunnett’s t-test.

Fig. 3. Time course of glutamate release evoked by 30 mM Kq from trout brain synaptosomes, in presence of Ca2q ŽB. or in a Ca2q-free medium with the addition of 0.1 mM EGTA ŽI.. Experimental conditions were as reported in Section 2. The depolarizing agent was applied during the period indicated by the horizontal bar. Each point is the mean"S.E.M. of 5–7 experiments. U P - 0.005 level of significance in the comparison to Kq-evoked release of glutamate in the presence of Ca2q.

A. Poli et al.r Brain Research 837 (1999) 46–54

51

hypothalamus were not significantly different from those of whole brain ŽTable 2.. The ontogenesis of brain A1 adenosine receptors was determined in whole trout brain. As shown in Table 3, w3 HxCHA binding was highly invariant with similar K d

Fig. 4. Ža. CHA-inhibition of 30 mM Kq-evoked release of glutamate from trout brain synaptosomes. Data are the mean"S.E.M. of percentage variations with respect to the controls obtained from 5–8 separate experiments. ANOVA followed by Dunnett’s t-test was applied on arcsin transformed percentages: U P - 0.01, UU P - 0.005 vs. control. Žb. Antagonism by CTP of the CHA-inhibition of glutamate-evoked release from trout synaptosomes. Each bar represents the mean"S.E.M. of pmolrmg proteinr2.5 min obtained from four separate experiments. Statistical comparisons were made by ANOVA followed by Dunnet’s t-test: U P 0.05 vs. 30 mM Kq; UU P - 0.05 vs. 30 Kq q10y5 M CHA.

0.707 " 0.021 nM and Bmax s 0.402 " 0.067 pmolrmg protein for controls; K d s 1.370 " 0.068 nM and Bmax s 0.294 " 0.034 pmolrmg protein in the presence of GTP. 3.4. Regional distribution and ontogenetic profile of [3 H ]CHA binding The K d value of w3 HxCHA binding and the density of A1 binding sites measured in membrane preparations of optic tectum, olfactory bulb, telencephalon, cerebellum and

Fig. 5. Ža. Time course of adenosine release from brown trout synaptosomes stimulated by 30 mM Kq. Effects of Ca2q deprivation and 1 mM tetrodotoxin. Experimental conditions were as reported in Section 2. Kq was applied during the period indicated by the horizontal bar. Each point is the mean"S.E.M. of 4–6 experiments. Žb. Time course of adenosine release from trout synaptosomes stimulated by 1 mM glutamate. Effects of Ca2q deprivation and 1 mM tetrodotoxin. Experimental conditions were as reported in Section 2. Glutamate was applied during the period indicated by the horizontal bar. Each point is the mean"S.E.M. of 4–5 experiments.

52

A. Poli et al.r Brain Research 837 (1999) 46–54

values and similar levels of binding sites from the embryonic age of 3 months through adulthood Ž4 years.. 3.5. Glutamate release from trout synaptosomes In trout synaptosomes, depolarizing conditions Ž30 mM Kq. resulted in large increases of glutamate release that appeared to be largely Ca2q-dependent; the perfusion with Ca2q-free medium in the presence of 0.1 mM EGTA resulted in a strong decrease of the evoked release ŽFig. 3.. The addition of different concentrations of the A1 receptor agonist CHA to the superfusing medium significantly reduced the Kq-evoked release of glutamate ŽFig. 4a.. CHA, at the different concentrations tested, did not alter the basal release of glutamate Žnot shown.. Fig. 4b summarizes the effect of the selective A1 adenosine receptor antagonist CPT on the CHA-inhibition of Kq-evoked glutamate release. CPT 10y5 M did not affect the basal release Žnot shown., but almost completely reversed the inhibitory effect of 10y5 M CHA. 3.6. Adenosine release from trout synaptosomes Using the same apparatus described above, we measured the efflux of adenosine in the trout synaptosomal preparations. After 50 min of equilibrating perfusion, the basal rate of adenosine release was increased by 30 mM Kq ŽFig. 5a.. Release peaked during 7.5 min exposure to high Kq concentration, but remained elevated for a further 5 min following its removal. The rate of adenosine release evoked by 30 mM Kq was not significantly reduced in a Ca2q-free medium containing 0.1 mM EGTA, indicating that Ca2q is not required for Kq evoked release of adenosine. Moreover, the addition of TTX at 1 mM to the superfusion medium, did not significantly affect the evoked release of adenosine, suggesting that voltage sensitive Naq channels are not involved in 30 mM Kq-evoked adenosine release from trout synaptosomes. As shown in Fig. 5b, glutamate also, at 1 mM, stimulated the release of adenosine from trout brain synaptosomes, which appeared to be Ca2q-independent and tetrodotoxin insensitive.

4. Discussion This study demonstrated the presence of nanomolar affinity adenosine binding sites in membrane preparations from whole brain of the fresh water teleost S. trutta. The high affinity w3 HxCHA binding was saturable and reversible. Over the concentration range of w3 HxCHA tested, non-linear regression analysis of the binding saturation data showed clearly that binding was to a single population of sites. Although the mean K d of binding Ž0.69 nM. was similar to those found in membranes of rat brain w3x or in freshwater and marine teleost fish w24,27x, the total number of specific binding sites Ž624 fmolrmg protein. in the

trout brain was much higher than that found in the freshwater teleost C. auratus Ž68.68 fmolrmg protein. w24x or in marine teleost Ž120.41 fmolrmg protein. w27,33x and it appeared similar to that found in mammalian brain Žca 600 fmolrmg protein. w28x. The pharmacology of w3 HxCHA binding to trout brain membranes was consistent with the labeling of an adenosine A1 receptors. The observed rank order of displacement potency of the agonists and antagonists was characteristic for A1 receptors w10x. In the tested brain regions, the binding of w3 HxCHA showed characteristics similar to those of the binding sites detected in whole brain membranes, and no differences were observed during the development from larval phase Ž3 months. to adult Ž4 years.. Adenosine A1 receptors are coupled to GTP-regulatory proteins as shown in the brains of the rat w38,42x, goldfish w24x and marine teleost fish w27x. The affinity for the agonist of receptors coupled to a G-protein, is negatively modulated by high concentrations of GTP. In our experimental conditions, the addition of 5 mM GTP to the incubation medium promoted a low affinity state of w3 HxCHA binding sites. This is a further characteristic underlining: Ži. that these binding sites belong to the adenosine A1 receptor subtype; Žii. that these sites coupled to G-protein might be associated with neural elements where they could modulate synaptic transmission, as the A1 receptors do in mammalian brain. In fact, in synaptosomal preparations of trout brain, CHA inhibited the calcium dependent release of endogenous glutamate elicited by 30 mM Kq. The A1 antagonist CPT selectively reversed the effect of CHA without modifying the basal glutamate release. Our results suggest that inhibition of glutamate release was mediated by A1 receptors located in synaptic terminals of trout brain. Moreover, these results agree with the dose-dependent CHA inhibition of glutamate release shown in synaptosomes of different vertebrates w23,35x. Confirming previous observations in mammalian brain preparations w20,35x, Kq-induced depolarisation as well as glutamate activation of excitatory receptors released adenosine from trout brain synaptosomes in a Ca2q-independent and tetrodotoxin-insensitive manner. The released extracellular adenosine is considered to behave as real retrograde synaptic transmitter w26x mainly to inhibit the release of neutransmitters. A cross-talk between adenosine and glutamate systems has been shown in synaptosomes and slices of mammalian brain w13,20,35x as well as in astrocytes in culture w5x. Adenosine A1 receptors inhibit glutamate release and glutamate stimulates adenosine outflow. In this view, a considerable part of Kq-evoked output of adenosine from brain slices has been ascribed to the NMDA receptor-mediated activity of released glutamate even though non-NMDA sites may be involved w8,20x. Moreover, A1 receptors cooperate with mGluRs to negatively modulate release of purines and glutamate w13x and it has been suggested that the system operates as a

A. Poli et al.r Brain Research 837 (1999) 46–54

fail-safe mechanism, which avoids an excessive release of glutamate without shutting off excitatory synaptic transmission w12,37x. Also in trout brain, a cross-talk between adenosine and glutamate systems seems to exist: adenosine negatively modulated Kq-evoked glutamate release by A1 receptors and Kq depolarisation or glutamate stimulated adenosine release. While our results on the whole indicate that, consistent with other vertebrates, in brown trout brain adenosine is endowed a neuromodulatory role, the high density and the lack of area specificity of A1 receptors raises some considerations. Adenosine receptors have been shown to have a broad phylogenetic distribution w39x and they seem to be coupled to the same intracellular signaling pathways both in lower and higher vertebrates w7,23,34x. In particular, both our previous results in goldfish w34x and this study in trout, show that adenosine inhibits the presynaptic release of glutamate in these two vertebrates which are very different in their ability to tolerate anoxia. Characteristically, vertebrates are intolerant of a prolonged interruption in oxygen supply. The only exception to this rule is found among ectothermic species, notably goldfish, crucian carp and freshwater turtle. Nevertheless, ectothermy per se does not apparently provide the animal with an anoxia tolerant brain. Indeed, it has been shown that an anoxia intolerant ectothermic vertebrate such as the trout, has an anoxia intolerant brain and that this brain displays a sensitivity to anoxia comparable to that of mammalian brain w30x. These considerations suggest that phylogenetically very distant vertebrates like trout and rat can share analogous strategies to counteract intolerance to anoxia, and the high density of adenosine receptors could be related to this.

w6x

w7x

w8x

w9x

w10x

w11x

w12x

w13x

w14x

w15x

w16x

w17x

Acknowledgements This work was supported by a grant from Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica ŽMURST., Italy.

References w1x R.A. Barraco, G.B. Stefano, Pharmacological evidence for the modulation of monoamine release by adenosine in the invertebrate nervous system, J. Neurochem. 54 Ž1990. 2002–2006. w2x R.F. Bruns, Role of adenosine in energy supplyrdemand balance, Nucleosides Nucleotides 10 Ž1991. 931–943. w3x R.F. Bruns, J.W. Daly, S.H. Snyder, Adenosine receptor in brain membranes: binding of N-cyclohexylw3 Hxadenosine and 1,3-diethyl8-w3 Hxphenylxanthine, Proc. Natl. Acad. Sci. USA 77 Ž1980. 5547– 5551. w4x S.P. Burke, J.V. Nadler, Regulation of glutamate and aspartate release from slices of hippocampal CA1 area: effects of adenosine and baclofen, J. Neurochem. 51 Ž1988. 1541–1551. w5x F. Caciagli, mGluRs and purinoceptors cooperate in regulating

w18x

w19x

w20x

w21x

w22x

w23x

53

excitatory amino acids and purine release, Neuropharmacology 35– 36 Ž1996. A6. R. Ciccarelli, P. Di Iorio, P. Giuliani, P. Ballerini, F. Caciagli, M.P. Rathbone, Rat cultured astrocytes release guanine-based purine in basal conditions and after hypoxiarhypoglycemia, Glia 25 Ž1999. 93–95. M.F. Cooper, K.K. Caldwell, 1990, Signal transduction mechanisms for adenosine, in: M. Williams ŽEd.., Adenosine and Adenosine Receptors, Humana Press, Clifton, NJ, 1990, pp. 105–141. C.G. Craig, T.D. White, N-methyl-D-aspartate and non-N-methyl-Daspartate-evoked adenosine release from rat cortical slices: distinct purinergic sources and mechanisms of release, J. Neurochem. 54 Ž1993. 1073–1080. R.A. Cunha, A.N. Sebastiao, ˜ Separation of adenosine triphosphate and its degradation products in innervated muscle of the frog by reverse phase high-performance liquid chromatography, Chromatographia 28 Ž1993. 610–612. J.W. Daly, P. Butts-Lamb, W. Padgett, Subclasses of adenosine receptors in the central nervous system. Interaction with caffeine and related methylxanthines, Cell. Mol. Neurobiol. 3 Ž1983. 69–80. J. Deckert, C.H. Gleiter, Adenosine — an endogenous neuroprotective metabolite and neuromodulator, J. Neural Transm. 43 Ž1994. 23–31, Suppl. P. Di Iorio, P. Ballerini, F. Caciagli, R. Ciccarelli, Purinoceptormediated modulation of purine and neurotransmitter release from nervous tissue, Pharmacol. Res. 37 Ž1998. 169–178. P. Di Iorio, G. Battaglia, R. Ciccarelli, P. Ballerini, P. Giuliani, A. Poli, F. Nicoletti, F. Caciagli, Interaction between A1 adenosine and class II metabotropic glutamate receptors in the regulation of purine and glutamate release from rat hippocampal slices, J. Neurochem. 67 Ž1996. 302–309. Dunwiddie, T.V., The physiological role of adenosine in the central nervous system, in: International Review of Neurobiology, Vol. 27, Academic Press, San Diego, 1985, pp. 63-132. J. Fastbom, B.B. Fredhlom, Inhibition of w3 Hxglutamate release from rat hippocampal slices by L-phenylisopropyladenosine, Acta Physiol. Scand. 125 Ž1985. 121–123. H.L. Haas, R.W. Greeen, Endogenous adenosine inhibits hippocampal CA1 neurones: further evidence from extra- and intracellular recording, Naunyn-Schmiedeberg’s Arch. Pharmacol. 337 Ž1988. 561–565. H.H. Harms, G. Wardeh, A.H. Mulder, Effects of adenosine on depolarization-induced release of various radiolabeled neurotransmitters from slices of rat corpus striatum, Neuropharmacology 18 Ž1979. 577–580. D.W. Hill, F.H. Walters, T.D. Wilson, J.D. Stuart, High performance liquid chromatographic determination of amino acid in the picomole range, Anal. Chem. 51 Ž1979. 1338–1341. P. Hylland, G.E. Nilsson, D. Johansson, Anoxic brain failure in an ectothermic vertebrate: release of amino acids and Kq in rainbow trout thalamus, Am. J. Physiol. 269 Ž1995. R1077–R1084, Regulatory Integrative Comp. Physiol. 38. K. Hoehn, T.D. White, Glutamate-evoked release of endogenous adenosine from rat cortical synaptosomes is mediated by glutamate uptake and not by receptors, J. Neurochem. 54 Ž1990. 1716–1724. M.J. Lohse, K.N. Klotz, J. Lindenborn-Fottinos, M. Reddington, U. Schwabe, R.A. Olsson, 8-Cyclopentyl-1,3-dipropylxanthine ŽDPCPX.: a selective high affinity antagonist radioligand for A1 adenosine receptors, Naunyn-Schmiedeberg’s Arch. Pharmacol. 336 Ž1987. 204–210. O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193 Ž1951. 265–275. R. Lucchi, A. Poli, U. Traversa, O. Barnabei, Functional adenosine A1 receptors in goldfish brain: regional distribution and inhibition of Kq-evoked glutamate release from cerebellar slices, Neuroscience 58 Ž1994. 237–243.

54

A. Poli et al.r Brain Research 837 (1999) 46–54

w24x R. Lucchi, A. Poli, U. Traversa, O. Barnabei, Characterization of A1 adenosine receptors in membranes from whole goldfish brain, Comp. Biochem. Physiol. 102B Ž1992. 331–336. w25x J.W. Mink, R.J. Blumenschine, D.B. Adams, Ratio of central nervous system to body metabolism in vertebrates: its constancy and functional basis, Am. J. Physiol. 241 Ž1981. R203–R212, Regulatory Integrative Comp. Physiol. 10. w26x J.B. Mitchell, C. R, T.V. Lupica, Dunwiddie, Activity-dependent release of endogenous adenosine modulates synaptic responses in the rat hippocampus, J. Neurosci. 13 Ž1993. 3439–3447. w27x T.F. Murray, J.F. Siebenaller, Comparison of the binding properties of A1 adenosine receptors in brain membranes of two congeneric marine fishes living at different depths, J. Comp. Physiol. 157 Ž1987. 267–277. w28x T.F. Murray, D.L. Cheney, Neuronal location of N 6 -cyclohexylw3 Hxadenosine binding sites in rat and Guinea-pig brain, Neuropharmacology 21 Ž1982. 575–580. w29x G.E. Nilsson, Long-term anoxia in crucian carp: changes in the levels of amino acid and monoamine neurotransmitters in the brain, catecholamines in chromaffin tissue, and liver glycogen, J. Exp. Biol. 150 Ž1990. 295–320. w30x J.W. Phillis, G.A. Walther, R.E. Simpson, Brain adenosine and transmitter amino acid release from the ischemic rat cerebral cortex: effects of the adenosine deaminase inhibitor deoxycoformycin, J. Neurochem. 56 Ž1991. 644–650. w31x J.W. Phillis, R.A. Barraco, Adenosine, adenylate cyclase and transmitter release, in: D.M.F. Cooper, K.B. Seamon ŽEds.., Advances in Cycle Nucleotide and Protein Phosphorylation Research, Raven Press, New York, 1985, pp. 243–257. w32x J. Pironen, I.J. Holopainen, A note on seasonality in anoxia tolerance of crucian carp Ž Carassius carassius . in the laboratory, Ann. Zool. Fenn. 23 Ž1986. 335–338. w33x A. Poli, B. Pavan, R. Lucchi, P.G. Borasio, E. Fabbri, R. Rossi,

w34x

w35x

w36x w37x

w38x

w39x

w40x

w41x w42x

w43x

Biochemical and pharmacological characterization of adenosine A1 receptors in eel Ž Anguilla anguilla., Fish Physiol. Biochem. 16 Ž1997. 19–27. A. Poli, R. Lucchi, M. Zottini, U. Traversa, Adenosine A1 receptormediated inhibition of evoked glutamate release is coupled to calcium influx decrease in goldfish brain synaptosomes, Brain Res. 620 Ž1993. 245–250. A. Poli, R. Lucchi, M. Vibio, O. Barnabei, Adenosine and glutamate modulate each other’s release from rat hippocampal synaptosomes, J. Neurochem. 57 Ž1991. 298–306. S.M. Rothman, J.W. Olney, Glutamate and the pathophysiology of hypoxic–ischemic brain damage, Ann. Neurol. 19 Ž1986. 105–111. K.A. Rudolphi, P. Schubert, F.E. Parkinson, B.B. Fredholm, Neuroprotective role of adenosine in cerebral ischemia, Trends Pharmacol. Sci. 13 Ž1992. 439–445. R.H. Scott, A.C. Dolphin, Inhibition of neural calcium currents by adenosine: role of g-proteins, in: J.A. Ribeiro ŽEd.., Adenosine Receptors in Nervous System, Taylor & Francis, London, 1989, pp. 151–158. J.F. Siebenaller, T.F. Murray, Phylogenetic distribution of w3 Hxcyclohexyladenosine binding sites in nervous tissue, Biochem. Biophys. Res. Commun. 137 Ž1986. 182–189. G. Spignoli, F. Pedata, G. Pepeu, A 1 and A 2 adenosine receptors modulate acetylcholine release from brain slices, Eur. J. Pharmacol. 97 Ž1984. 341–342. T.W. Stone, Purine receptors and their pharmacological potential, Adv. Drug Res. 18 Ž1989. 291–429. L.O. Trussel, M.B. Jackson, Dependence of an adenosine-active potassium current on a GTP-binding protein in mammalian central neurons, J. Neurosci. 7 Ž1987. 3306–3316. M. Williams,. Adenosine receptor: an historical perspective, in: M. Williams ŽEd.., Adenosine and Adenosine Receptors, Humana, Clifton, NJ, 1990, pp. 1–15.