European Journal of Pharmacology 744 (2014) 83–90

Contents lists available at ScienceDirect

European Journal of Pharmacology journal homepage: www.elsevier.com/locate/ejphar

Neuropharmacology and analgesia

Modulation of cGMP accumulation by adenosine A1 receptors at the hippocampus: Influence of cGMP levels and gender André Serpa a, Ana M. Sebastião c,d, José F. Cascalheira a,b,n a

CICS-UBI – Health Sciences Research Center, University of Beira Interior, Covilhã, Portugal Department of Chemistry, University of Beira Interior, Covilhã, Portugal c Institute of Pharmacology and Neurosciences, Faculty of Medicine, University of Lisbon, Lisboa, Portugal d Unit of Neurosciences, Institute of Molecular Medicine, University of Lisbon, Lisboa, Portugal b

art ic l e i nf o

a b s t r a c t

Article history: Received 5 May 2014 Received in revised form 24 September 2014 Accepted 25 September 2014 Available online 7 October 2014

Adenosine A1 receptor is highly expressed in hippocampus where it inhibits neurotransmitter release and has neuroprotective activity. Similar actions are obtained by increasing cGMP concentration, but a clear link between adenosine A1 receptor and cGMP levels remains to be established. The present work aims to investigate if cGMP formation is modulated by adenosine A1 receptors at the hippocampus and if this effect is gender dependent. cGMP accumulation, induced by phosphodiesterases inhibitors Zaprinast (100 μM) and Bay 60-7550 (10 μM), and cAMP accumulation, induced by Forskolin (20 μM) and Rolipram (50 μM), were quantified in rat hippocampal slices using specific enzymatic immunoassays. N6-cyclopentyladenosine (CPA, 100 nM) alone failed to modify basal cGMP accumulation. However, the presence of adenosine deaminase (ADA, 2 U/ml) unmasked a CPA (0.03–300 nM) stimulatory effect on basal cGMP accumulation (EC50: 4.27 1.4 nM; Emax: 17 7 0.9%). ADA influence on CPA activity was specific for cGMP, since inhibition of cAMP accumulation by CPA was not affected by the presence of ADA, though ADA inhibited cAMP accumulation in the absence of CPA. Increasing cGMP accumulation, by about four-fold, with sodium nitroprusside (SNP, 100 μM) abolished the CPA (100 nM) effect on cGMP accumulation in males but did not modify the effect of CPA in female rats. This effect was reversed by 8-Cyclopentyl-1,3-dipropylxanthine (DPCPX, 100 nM), indicating an adenosine A1 receptor mediated effect on cGMP accumulation. In conclusion, adenosine A1 receptors increase intracellular cGMP formation at hippocampus both in males and females under basal conditions, but only in females when cGMP levels are increased by SNP. & 2014 Elsevier B.V. All rights reserved.

Keywords: Adenosine A1 receptor cGMP Cyclic nucleotides Hippocampus Guanylyl cyclase

1. Introduction Extracellular adenosine exerts its regulatory actions in the brain through activation of specific G protein-coupled transmembrane receptors (see Dias et al., 2013). The Gi/o protein-coupled adenosine A1 receptor is expressed at high levels in the hippocampus where it inhibits adenylyl cyclase (AC) and therefore decreases production of the second messenger cAMP (reviewed in Fredholm et al., 2001). Besides inhibiting AC, adenosine A1 receptor also inhibits N- and P/Q-type calcium channels (Ambrósio et al., 1997), activates inwardly rectifying potassium channels (Takigawa and Alzheimer, 2002) and regulates inositol phosphates formation (Cascalheira et al., 2002).

n Corresponding author at: Departamento de Química, Universidade da Beira Interior, Rua Marquês D'Ávila e Bolama, 6200 Covilhã, Portugal. Tel.: þ 351 275242021; fax: þ 351 275319730. E-mail address: [email protected] (J.F. Cascalheira).

http://dx.doi.org/10.1016/j.ejphar.2014.09.045 0014-2999/& 2014 Elsevier B.V. All rights reserved.

cGMP, another cyclic nucleotide with second messenger action, is produced by two distinct pathways. One involves cytoplasmic soluble guanylyl cyclase (sGC), whose agonist, nitric oxide (NO), is produced by calcium-activated NO synthase (NOS), while the other involves the membrane-bound particulate guanylyl cyclase, which is stimulated by natriuretic peptides (reviewed in Lucas et al., 2000). cGMP immunostaining revealed cGMP accumulation, induced by the association of an allosteric enhancer of soluble guanylyl cyclase and an inhibitor of cGMP-degrading phosphodiesterases, in pyramidal cells and astrocytes of the hippocampus (Bartus et al., 2013). Adenosine A1 receptor activation and cGMP mediate similar actions at the central nervous system, both inhibit glutamatergic synaptic transmission (Dunwiddie and Hoffer, 1980; Feil and Kleppisch, 2008; Serpa et al., 2009), protect against excitotoxic insults (Montoliu et al., 2001; Orio et al., 2007; Ribeiro, 2005; Sebastião et al., 2001) and regulate synaptic plasticity (see Dias et al., 2013; Feil and Kleppisch, 2008) at the hippocampus. However, the relationship between adenosine A1 receptors and cGMP remains to be clarified. In the present work we investigated the adenosine A1 receptor ability to

84

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

modulate cGMP accumulation at the hippocampus. The gender dependence of the adenosine A1 receptor-mediated effect on cGMP levels was also investigated since estrogen receptors are present at the hippocampus (Petersen et al., 1998) where estradiol increases the cGMP content (Palmon et al., 1998).

2. Material and methods 2.1. cGMP and cAMP accumulation in hippocampal slices The experiments were performed using acute hippocampal slices taken from young adult Wistar rats (6–8 weeks old). Female and male rats were used and analyzed separately. Whenever not specified in figures or text, males were used. The animals were handled according to European Community guidelines and Portuguese law concerning animal care and were anesthetized with halothane before decapitation. The brain was rapidly removed and transferred to ice-cold Krebs–Henseleit buffer with the following composition (mM): NaCl 118, KCl 4.7, KH2PO4 1.2, MgSO4 1.2, CaCl2 1.3, NaHCO3 25, glucose 11.6, gassed with carbogen (95% O2 and 5% CO2), pH 7.4. The brain was cut longitudinally, the two hippocampi were dissected and cross-chopped (350 mm  350 mm) with a McIlwain tissue chopper. Sliced hippocampi were then placed in an Erlenmeyer, dispersed and washed twice with buffer. The crosschopped hippocampal slices were transferred into a conic bottom falcon tube and 50 ml-aliquots of gravity-packed slices (1–2 mg protein) were pipetted into flat-bottomed polypropylene tubes (1.65 cm  9.5 cm, 20 ml capacity) containing Krebs–Henseleit buffer and pre-incubated for 30 min at 37 1C in a shaking (1 cycle s  1) water bath. In experiments where cGMP was quantified, incubation with drugs started with addition of Zaprinast and Bay 60-7550 (100 mM and 10 mM final concentration, respectively) to build up cGMP accumulation. Zaprinast is a selective inhibitor of the cGMP-specific phosphodiesterases V and VI, but also inhibiting other cGMP-hydrolyzing phosphodiesterases, and Bay 60-7550 is mainly a phosphodiesterase II inhibitor. When used, adenosine deaminase (2 U/ml) and 8-Cyclopentyl-1,3-dipropylxanthine (DPCPX, 100 nM) were present since the beginning of incubation, together with Zaprinast and Bay 60-7550. When testing the effect of the adenosine A1 receptor agonist N6-cyclopentyladenosine (CPA, 0.03–300 nM final concentration) this was added 30 min after induction of cGMP accumulation by Zaprinast and Bay 607550, whereas sodium nitroprusside (SNP, 100 mM final concentration), when present, was added 25 min after CPA. The final volume after all drug additions was 400 μl. In experiments where cAMP was quantified, incubations with drugs started with addition of the phosphodiesterase IV inhibitor rolipram (50 μM final concentration) to produce cAMP accumulation. 45 min after rolipram addition incubation proceeded in the presence of the adenylyl cyclase activator forskolin (20 μM) for a further 35 min period. When used, adenosine deaminase (ADA, 2 U/ml) was present since the start of incubation with rolipram, while CPA (100 nM), when present, was added 30 min after rolipram. The final volume after all drugs were added was 300 μl. When testing the effect of a drug, a parallel control assay was performed, where the same volume of vehicle replaced the volume of drug solution added to the tube. Tubes were gassed with carbogen for 20 s and capped, after addition of slices or drugs. Incubations were stopped, by addition of 133 ml (in cGMP experiments) or 100 μl (in cAMP experiments) of perchloric acid (HClO4, 10% w/v) solution containing ethylenediamine tetraacetic acid (EDTA, 20 mM). Samples were sonicated for 2 min, placed on ice for 30 min, neutralized by addition (133 μl for cGMP or 100 μl for cAMP experiments) of potassium carbonate (K2CO3, 0.5 M) and vortexed for 2 min, allowing the CO2 to escape. The tubes were then placed on ice for an

additional 15 min period to precipitate potassium perchlorate. Afterwards, 400 μl of each sample were transferred to 1.5 ml eppendorfs and centrifuged (5000 g, 10 min at 4 1C) and 300 ml aliquots of the supernatants were collected and stored at  20 1C for further cGMP or cAMP analysis. The pellets were digested with NaOH (1 M) for 90 min at 37 1C, neutralized and individually assayed in duplicate for protein content by the method of Peterson (1977). The supernatants were analyzed for cGMP or cAMP concentration using cGMP- or cAMPspecific enzyme immunoassay (EIA) kits (Enzo Life Sciences). cGMP or cAMP concentration in each sample was expressed as pmol per mg of protein. 2.2. Drugs used 8-Cyclopentyl-1,3-dipropylxanthine (DPCPX), N6-cyclopentyladenosine (CPA) and zaprinast were purchased from Tocris, while rolipram, forskolin and sodium pentacyanonitrosylferrate (sodium nitroprusside, SNP) were from Sigma-Aldrich. Bay 60-7550 was purchased from Santa Cruz Biotechnology. Adenosine deaminase was from Roche Applied Science. Stock solutions of DPCPX (50 mM), rolipram (20 mM), zaprinast (100 mM) and Bay 60-7550 (2 mM) were prepared in dimethyl sulfoxide (DMSO), while CPA (2 mM) and SNP (150 mM) stock solutions were prepared in ultra-pure water. Forskolin (50 mM) stock solution was prepared in ethanol. Suitable dilutions of each stock solution with Krebs buffer were made before performing the experiments. 2.3. Data analysis The values are expressed as mean7S.E.M. from n independent experiments. The effect of a drug was calculated, for each experiment, as: 100  (D C)/C; where D is the cGMP or cAMP accumulation obtained in the presence of the drug and C is the cGMP or cAMP accumulation obtained in the corresponding control assay performed in the same conditions but in the absence of the drug. The cGMP accumulation that would be expected in the presence of both CPA and SNP if the effects of each drug were additive was calculated as: AþB C; where A, B and C are, respectively, the cGMP accumulation obtained in the presence of CPA, in the presence of SNP, and in the absence of CPA and SNP (control). The significance of the differences between the means obtained in two different conditions, or when comparing means with zero, was evaluated by Student's t-test, where the paired Student's t-test was used whenever evaluating the significance of differences between two conditions tested in a paired way in the same experiment. When more than two different conditions were simultaneously considered, the one-way or two-way (when two factor were analyzed) ANOVA were used, followed by the LSD posthoc test. When analyzing, by ANOVA, differences between the means of cGMP accumulation obtained in different conditions, adjustment for inter-experiment variability was performed. Statistical significance was considered for Po0.05. The maximal effect (Emax) and the concentration of agonist producing half-Emax (EC50) were calculated by fitting the agonist concentration-response curve data to a sigmoidal curve equation, through non-linear regression analysis using the PASW for Windows program version 18.0.

3. Results 3.1. Basal cGMP and cAMP accumulations Since basal intracellular levels of cGMP in hippocampal slices are low, and therefore hard to quantify, experiments were performed in the presence of Zaprinast, a selective inhibitor of cGMP-specific phosphodiesterase (PDE) V and VI but also inhibiting other cGMPhydrolyzing PDEs (Marte et al., 2008), and Bay 60-7550, mainly a PDE

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

II inhibitor (Bender and Beavo, 2006), in order to induce cGMP accumulation. In the presence of both Zaprinast (100 mM) and Bay 60-7550 (10 mM), the cGMP accumulation was 1577 pmol/mg protein (n¼5); when ADA (2 U/ml) was added to remove endogenous adenosine the cGMP accumulation (970.9 pmol/mg protein, n¼ 18) tended to be lower, although it was not significantly different from the accumulation in the absence of ADA (P¼0.14, Student's t-test) (Fig. 1A). In experiments where soluble guanylyl cyclase was directly stimulated by SNP (100 mM), in the presence of ADA (2 U/ml), cGMP accumulation was increased by about four-fold (Fig. 1A), this increase being identical in male and female Wistar rats. Quantification of cAMP accumulation was also performed. For this purpose rolipram, a phosphodiesterase 4 inhibitor, the main enzyme responsible for cAMP degradation in the brain, and forskolin,

# P=0.00001

30 (5)

20 (18)

15 10 5 0

-

+ -

+ +

# P=0.027 [cAMP] (pmol/mg protein)

300

(3)

250 200 150 100

Application of the adenosine A1 receptor selective agonist CPA (100 nM) alone to acute hippocampal slices failed to modify (% effect of 1.673.0%, n¼5, while comparing absence and presence of CPA in the same experiments; P¼0.61, Student´s t-test) the cGMP accumulation obtained in the presence of zaprinast (100 mM) and Bay 60-7550 (10 mM) (Fig. 2A). We therefore decided to add ADA to the incubation medium, which in some cases has been proved useful to potentiate or unmask adenosine receptor mediated effects

(3)

50

% increase in cGMP accumulation

35

ADA SNP

3.2. Adenosine deaminase is required to unmask the effect of the adenosine A1 receptor agonist CPA on cGMP accumulation

(10)

40

25

which directly stimulates adenylyl cyclase (see Bender and Beavo, 2006; Daly et al., 1982) were used. In the presence of rolipram (50 mM), the basal cAMP accumulation was 40711 pmol/mg protein (n¼3) whereas the further addition of 20 mM forskolin increased basal cAMP accumulation by about five-fold (Fig. 1B).

# P=0.028

35

30

* P=0.016

25

(7)

20 15 10

(5)

5 0

- ADA

% inhibition of cAMP accumulation

[cGMP] (pmol/mg protein)

45

Forskolin

-

+

Fig. 1. cGMP and cAMP accumulations at the rat hippocampus. (A): Hippocampal slices, from Wistar rats, were incubated for 55 min in the presence of zaprinast (100 μM), Bay 60-7550 (10 μM) and in the absence or in the presence of ADA (2 U/ml). After this period SNP (100 μM final concentration) or its vehicle was added and incubation continued for a further 25 min period. Columns represent mean7S.E.M. of cGMP accumulation obtained (from left to right) in the absence of ADA and SNP, in the presence of ADA but in the absence of SNP, and in the presence of ADA and SNP. (B): Slices were incubated for 45 min in the presence of rolipram (50 μM). After this period forskolin (20 μM final concentration) or its vehicle was added and incubation proceeded for a further 35 min period. Columns represent mean7S.E.M. of the cAMP accumulation obtained in the absence (left column) or in the presence (right column) of forskolin. The number of experiments performed, run at least in quadruplicate, is shown in brackets above the bars. (#): Significantly different from the presence of SNP (A) or forskolin (B), Student´s t-test.

+ ADA

35 30 25

* P=0.020

(3)

* P=0.00001

(9)

20 15 10 5 0

-ADA 0

85

+ADA

Fig. 2. Adenosine deaminase is required for expression of the effect of CPA on cGMP accumulation, but not on cAMP accumulation. (A): Hippocampal slices, from male Wistar rats, were incubated for 30 min in the presence of zaprinast (100 μM), Bay 60-7550 (10 μM) and in the presence of ADA (2 U/ml) or its vehicle. After this period CPA (100 nM final concentration) or vehicle (control) was added and incubation continued for a further 50 min period. Columns represent mean7S.E.M. of the percentage of CPA-induced increase in control cGMP accumulation, obtained in the absence (left column) or in the presence (right column) of ADA. (B) Slices were incubated for 30 min in the presence of rolipram (50 μM) and in the absence or in the presence of ADA (2 U/ml). After this period CPA (100 nM final concentration) or vehicle (control) was added and 15 min later forskolin (20 μM final concentration) was added and incubation proceeded for a further 35 min period. Columns represent mean7S.E.M. of the percentage of CPA-induced inhibition of control cAMP accumulation obtained in the absence (left column) or in the presence (right column) of ADA. The number of experiments performed, run at least in quadruplicate, is shown in brackets above the bars. (n): Significantly different from zero, Student´s t-test. (#): Significantly different from absence of ADA, Student´s t-test.

86

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

(see Section 4). A stimulatory effect of 100 nM-CPA (1775%, n¼ 7; Fig. 2A) on cGMP accumulation was unmasked by the presence of ADA (2 U/ml). All subsequent experiments involving cGMP quantification were conducted in the presence of ADA (2 U/ml). As a control, we assessed a well known effect of adenosine A1 receptors, the ability to inhibit adenylyl cyclase. cAMP production was stimulated by forskolin (20 mM) and degradation prevented by rolipram (50 mM); under such conditions cAMP levels were 218 750 pmol/mg protein (n ¼ 3) and as expected (see Fredholm et al., 2001) CPA (100 nM) caused a 21 73% (n ¼3; Fig. 2B) inhibition of cAMP accumulation. The further addition of ADA lead to a decrease in cAMP accumulation (98 77 pmol/mg protein, n ¼9; P ¼0.002 vs accumulation in the absence of ADA, Student's t-test) but did not influence the ability of CPA to inhibit cAMP accumulation (197 1% inhibition, n¼ 9; P ¼0.42 vs effect in the absence of ADA, Student's t-test; Fig. 2B). 3.3. Maximal effect and potency of the adenosine A1 receptor agonist CPA on cGMP accumulation As shown in Fig. 3, the adenosine A1 receptor selective agonist CPA (0.03–300 nM) dose-dependently increased cGMP accumulation at the hippocampus. The highest concentration of CPA tested (300 nM) produced a nearly maximal increase in cGMP accumulation (167 3%, n ¼4; P ¼0.019 vs zero, Student's t-test). Nonlinear curve fitting to the data shown in Fig. 3 gave an EC50 for CPA of 4.2 71.4 nM and an Emax of 17 70.9%. 3.4. Effect of the adenosine A1 receptor agonist CPA on cGMP accumulation in the presence of a nitric oxide donor: gender dependence In order to evaluate if the CPA effect on basal cGMP accumulation is still observed under increased concentrations of cGMP, accumulation of cGMP was stimulated using sodium nitroprusside (100 μM), a NO donor and activator of soluble guanylyl cyclase (see Lucas et al., 2000). The effect of CPA on cGMP accumulation, in the absence or in the presence of SNP, was evaluated in both males and females Wistar rats. While gender did not influence neither basal cGMP

% increase in cGMP accumulation

25

(8) (4)

20 P=0.016 *

15

*P=0.019

(6) 10 5

(3) 0 -5

0.01

0.1

1

10

100

1000

[CPA] (nM)

-10 Fig. 3. CPA dose-dependently increases cGMP accumulation. Hippocampal slices, from male Wistar rats, were incubated for 30 min in the presence of zaprinast (100 μM), Bay 60-7550 (10 μM) and ADA (2 U/ml). After this period CPA (0.03–300 nM final concentration) or vehicle (control) was added and incubation continued for a further 50 min period. Data points represent mean 7 S.E.M. of the percentage of CPA-induced increase of control cGMP accumulation, corresponding to 3–8 experiments (number of experiments in brackets above the data points) run at least in quadruplicate. Average cGMP accumulation in control assays was 97 0.9 pmol/mg protein (n ¼18). The solid line corresponds to the nonlinear regression curve obtained by fitting a Michaelis–Menten type equation to the experimental points. (n): Significantly different from zero, Student's t-test (P value is indicated near the symbol).

levels nor the effect of CPA upon basal cGMP accumulation (absence of SNP, Fig. 4A), in the presence of SNP (100 μM) there was a marked gender influence both upon cGMP accumulation in the absence of CPA, and upon the effect of CPA (Fig. 4B and C). In males, in the presence of SNP (100 μM), CPA (100 nM) failed to modify cGMP accumulation (% effect of 3.172.4%, n¼5; P¼ 0.27 vs zero, Student's t-test; Fig. 4C). In fact, in males the cGMP accumulation obtained in the presence of both CPA (100 nM) and SNP (100 μM) (31.273.4 pmol/mg protein, n¼5) was not significantly different from the cGMP accumulation obtained in the presence of SNP alone (30.372.9 pmol/mg protein, n¼5, see Fig. 4B), but it was also not different (P¼0.9, Student's t-test) from the cGMP accumulation that would be expected in the presence of both CPA and SNP if the effects of each drug were additive (31.772.9 pmol/mg protein). This is probably consequence of the fact that the SNP effect on cGMP accumulation (236734%, n¼ 5) is much higher than the CPA effect (16.875.1%, n¼8, in the absence of SNP), therefore the SNP effect would mask/obscure the CPA effect when measured in the presence of SNP (see Fig. 4). In contrast, in female rats CPA caused a significant increase (37711%, n¼ 4; P¼0.044 vs zero, Fig. 4C) in cGMP accumulation even in the presence of SNP. Interestingly cGMP accumulation in the presence of SNP but absence of CPA was also significantly higher in females than in males, indicating that in females cGMP production is favored (see Fig. 4B). As can be concluded from the data shown in Fig. 4, in the presence of SNP (100 μM) the effect of CPA observed in females rats was significantly different from the effect observed in males (3.172.4%, n¼ 5), indicating an interaction between gender and SNP affecting the effect of CPA (Fig. 4C). The stimulatory effect of CPA on cGMP accumulation in the presence of SNP, observed in females was completely reversed by DPCPX, an adenosine A1 receptor selective antagonist (see Fredholm et al., 2001) (Fig. 5). DPCPX alone had no significant effect on cGMP accumulation (Fig. 5), which could be expected since endogenous adenosine had been removed by addition of ADA.

4. Discussion 4.1. Effect of the adenosine A1 receptor on cGMP levels The main finding in the present work is that adenosine A1 receptor activation increases cGMP levels at the hippocampus in a gender dependent way. While adenosine A1 receptor-induced increase in basal cGMP production is evident both in males and in females (providing that ADA is added to the incubation medium), the increase in cGMP production under NOS activation conditions is evident in females but not in males. Previous studies had already addressed the influence of adenosine A1 receptors on cGMP levels but in peripheral tissues, an effect which we now confirm to be also present at the hippocampus. Increase in cGMP concentration by adenosine A1 receptor activation was observed in cardiac atrium, where cGMP mediates the adenosine A1 receptor-induced decrease in contractibility (Sterin-Borda et al., 2002), in vascular smooth muscle cells of rat aorta (Kurtz, 1987) and in rat kidney cells (Kurtz et al., 1988). At the central nervous system, previous studies reported an adenosine-induced increase in cGMP levels in slices of cerebellum and cerebral cortex of guinea pig (Ohga and Daly, 1977; Saito, 1977), but no selective agonists or antagonists were used to identify the receptor involved. Subsequent work showed that this effect of adenosine was mediated by A2B receptor, at least in the cerebellum (Hernández et al., 1993). Reports of interaction between effects mediated by adenosine A1 receptor and cGMP, has been described at the hippocampus. The activation of adenosine A1 receptor together with the simultaneous increase in cGMP concentration elicited by zaprinast, was enough to

§ P=0.002

14 12 10

-SNP

(3)

(8) (8)

(3)

8

-CPA

6

+CPA

4 2 0

a

b

a'

87

+SNP

§ P=0.003

[cGMP] (pmol/mg protein)

[cGMP] (pmol/mg protein)

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

80

(4)

β P=0.020

70 60

(4)

NS P=0.74

50 40

α P=0.002 § P=0.001

(5)

(5)

c

d

-CPA +CPA

30 20 10 0

b'

c'

d'

# P=0.001

% increase in cGMP accumulation

60

* P=0.044

(4)

50

# P=0.035

40

30

δ

*

* P=0.012

(3) δ

P=0.014

(8) 20

(5)

10

0

- SNP

+ SNP

- SNP

+ SNP

Fig. 4. Influence of SNP and gender on the CPA-induced increase in cGMP accumulation. Hippocampal slices were incubated for 30 min in the presence of zaprinast (100 μM), Bay 60-7550 (10 μM) and ADA (2 U/ml). After this period CPA (100 nM final concentration) or vehicle (control) was added and 25 min later SNP (100 μM final concentration) or its vehicle was added and incubation proceeded for a further 25 min period. (A, B): cGMP accumulation obtained in the absence (A) or in the presence (B) of SNP (100 μM). Columns represent mean 7 S.E.M. of cGMP accumulation obtained in the absence (controls, open columns: a, a0 , c and c0 ) or in the presence (solid columns: b, b0 , d and d0 ) of CPA (100 nM) in males (♂) and females (♀) rats. Note that in the presence of SNP (B), but not in the absence (A), the cGMP accumulation was higher in females than in males (either in the absence or in the presence of CPA). (C): CPA effect on cGMP accumulation. Columns represent mean7 S.E.M. of the percentage of CPA-induced increase in cGMP accumulation, when compared with the corresponding control, obtained (from left to right) in males in the absence (b vs a, in A) or in the presence of SNP (d vs c, in B), and in females in the absence (b0 vs a0 , in A) or in the presence of SNP (d0 vs c0 , in B). Note that after applying Two-way ANOVA to the data presented in (C), an overall significant difference between the CPA effect obtained in males and females was observed (P¼ 0.002), but not between the CPA effect obtained in the presence and in the absence of SNP (P¼ 0.23); however a significant interaction between gender and presence of SNP was observed (P ¼0.025). The number of experiments performed, run at least in quadruplicate, is shown in brackets above the bars. (§): Significantly different from control; (α): Significantly different from d; (β): Significantly different from c; NS: Non significantly different; One-way ANOVA followed by LSD test (P value is indicated near the symbol). (n): Significantly different from zero; Student's t-test (P value is indicated near the symbol). (#): Significantly different from males in the presence of SNP; One-way ANOVA, followed by LSD test (P value is indicated near the symbol). (δ): Significantly different from males (P ¼0.002); two-way ANOVA.

induce chemical LTD (Santschi et al., 2006). On the other hand, the potentiation, by a NO donor, of the adenosine A1 receptor inhibition of neurotransmission, was blocked by a sGC inhibitor, suggesting a facilitatory effect of cGMP on the adenosine A1 receptor effect at the hippocampus (Fragata et al., 2006). In addition, the adenosine A1 receptor selective antagonist DPCPX, blocked the inhibitory effect of a NO donor and zaprinast on hippocampal synaptic transmission (Broome et al., 1994), suggesting an involvement of adenosine A1 receptor. Later, it was shown that both NO donor and zaprinast increase the release of adenosine – an effect not mediated by cGMP – a finding explaining the adenosine A1 receptor-dependent inhibitory

action of NO donors and zaprinast on neurotransmission (Arrigoni and Rosenberg, 2006). An indirect evidence of adenosine A1 receptor action on cGMP levels at the nervous system came from a recent study, where the inhibitory effect of peripheral adenosine A1 receptor on inflammatory hypernociception was blocked by sGC and PKG inhibitors, suggesting a cGMP-mediated effect of the adenosine A1 receptor (Lima et al., 2010); however a direct effect of adenosine A1 receptor on cGMP levels was not evaluated. Direct observation of adenosine A1 receptormediated effect on cGMP levels might be difficult to observe due to low intracellular levels and high compartmentalization of cGMP

88

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

[cGMP] (pmol/mg protein)

80

NS P=0.42

(4)

70 60

§

§

(4) (4)

(4)

50 40 30 20 10 0

CPA (100 nM)

% increase in cGMP accumulation

DPCPX (100 nM)

a

b

c

d

-

+ -

+

+ +

# 58 48

* (4)

38

4.2. Adenosine deaminase dependence of the CPA effect on cGMP accumulation

28

(4)

18

(4)

8 -2 CPA

DPCPX

concentrations diffuse and reach the bulk of the slice; this is particularly true for lipophilic drugs such as zaprinast, Bay 60-7550 and rolipram (Boess et al., 2004; van Staveren et al., 2001). Using these conditions, we were able to observe a stimulatory effect of the adenosine A1 receptor on cGMP formation at the rat hippocampal slice; a maximal increase of 17% and an EC50 of 4.2 nM being obtained for the effect of the selective adenosine A1 receptor agonist CPA on basal cGMP accumulation. The EC50 obtained in the present study was very similar to those reported for the CPA-induced increase of cGMP accumulation at the atrium (between 1 and 10 nM; SterinBorda et al., 2002), and for CPA inhibition of neurotransmission (EC50E12 nM; de Mendonça and Ribeiro, 1997) and inositol phosphates accumulation (EC50¼ 10 nM; Cascalheira and Sebastião, 1998) at the hippocampus. Given the similar actions of adenosine A1 receptor activation and cGMP at the nervous system, the present findings suggest that the adenosine A1 receptor-mediated adenosine effects on neurotransmission and neuroprotection, might be mediated, at least in part, by cGMP. It also opens the possibility that modulation of cGMP levels – either pharmacologically or by activation of other receptors – might modify adenosine A1 receptor-mediated effects in the nervous system.

CPA/DPCPX

Fig. 5. Specificity of the effect of CPA, in the presence of SNP. Hippocampal slices from female Wistar rats were incubated for 30 min in the presence of zaprinast (100 μM), Bay 60-7550 (10 μM), ADA (2 U/ml) and in the absence or in the presence of DPCPX (100 nM). After this period CPA (100 nM final concentration) or vehicle (control) was added and 25 min later SNP (100 μM final concentration) was added and incubation proceeded for a further 25 min period. (A): Columns represent mean 7S.E.M. o cGMP accumulation obtained (from left to right) in the absence of CPA and DPCPX (a), in the presence of CPA (b), in the presence of DPCPX (c) and in the presence of CPA and DPCPX (d). (B): CPA effect on cGMP accumulation. Columns represent mean 7 S.E.M. of the percentage of increase in cGMP accumulation, when compared with the corresponding control, produced by (from left to right) CPA (b vs a, in A), DPCPX (c vs a, in A), and CPA in the presence of DPCPX (d vs c, in A). The number of experiments performed, run at least in quadruplicate, is shown in brackets above the bars. (§): Significantly different from CPA in the absence of DPCPX (b); NS: non significantly different; One-way ANOVA, followed by LSD test (P value is indicated near the symbol). (*): Significantly different from zero; Student's t-test (P value is indicated near the symbol). (#): Significantly different from CPA in the presence of DPCPX; One-way ANOVA, followed by LSD test (P value is indicated near the symbol).

(see Arora et al., 2013). Therefore, in the present study, we used experimental conditions that would prevent cGMP degradation and induce maximal cGMP accumulation. For that purpose, we used a selective inhibitor of PDE II (Bay 60-7550), the main enzyme responsible for cGMP degradation at the hippocampus (Bartus et al., 2013), together with a more general inhibitor of cGMP-degrading PDEs (zaprinast) to induce cGMP accumulation. A 100 μM concentration of zaprinast was chosen because it was necessary to inhibit most cGMP-degrading PDEs present at the hippocampus, namely PDEs I, III, VI and IX (see Marte et al., 2008; van Staveren et al., 2001), though at this concentration zaprinast does not inhibit efficiently PDE II, so that we have to use, in addition, a more potent PDE II inhibitor, Bay 60-7550. A previous study also showed that at least a 100 μM concentration of zaprinast was necessary to produce significant increases of cGMP accumulation at the hippocampal slice (van Staveren et al., 2001). On the other hand, usually higher concentrations of drugs, particularly lipophilic ones, are necessary when using brains slices (specially in batch experiments) than when using cultured cells or cells homogenates, so that effective drug

Since basal levels of endogenous adenosine might activate adenosine A1 and A2A receptors, which could interfere with the adenosine A1 receptor-mediated effect of CPA on cGMP formation, the effect of CPA was assessed both in the absence and in the presence of adenosine deaminase to remove endogenous adenosine. ADA was required to unmask the effect of CPA as modulator of intracellular cGMP levels. The no observation of CPA effect on cGMP accumulation in the absence and its rescue in the presence of extracellular ADA, could be due to: i) adenosine A1 receptor desensitization by endogenous adenosine; ii) inhibition of adenosine A1 receptor by tonic activation of the adenosine A2A receptor by endogenous adenosine, which has the ability to directly attenuate the activity of adenosine A1 receptors (Lopes et al., 1999); iii) action of ADA on adenosine A1 receptor, independent of adenosine removal; iv) receptor occupation with endogenous adenosine, preventing further activation with the exogenous agonist. The first and second hypothesis seem unlikely since ADA did not affect the adenosine A1 receptor-mediated inhibition of cAMP accumulation – although ADA alone did decrease cAMP accumulation, reflecting a tonic activation of adenylyl cyclase by A2A receptors. The third hypothesis seems more likely since previous studies showed that extracellular ADA (including commercial calf intestine ADA) binds adenosine A1 receptor, increasing its affinity towards agonists, and acts as a co-stimulatory molecule facilitating specific signalling events, independently from its enzymatic activity (reviewed in Franco et al., 1997, 2005). In fact, extracellular ADA was absolutely required for adenosine A1 receptor agonist-induced Ca2 þ mobilization and inositol phosphate formation, independently of ADA catalytic activity, in a smooth muscle cell line (Ciruela et al., 1996). It could be possible that in situations or cell locations where extracellular ADA is increased (e.g. Correia-de-Sá et al., 2006; Pimentel et al., 2011), modulation of cGMP formation by adenosine A1 receptor could be facilitated. The fourth possibility is also likely, and not mutually exclusive with the third, considering that ADA tend to cause an inhibition of cGMP accumulation, therefore opposite to the change caused by CPA; the adenosine A1 receptor efficacy to enhance cGMP accumulation is probably lower than the efficacy to inhibit adenylyl cyclase, therefore rendering the cGMP transducing system more sensitive to receptor occupancy by the endogenous ligand.

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

4.3. Nitric oxide and gender dependence of the adenosine A1 receptor-mediated effect To determine if the CPA effect on cGMP accumulation depended on intracellular levels of cGMP, we increased cGMP levels using the NO donor SNP, which activates the soluble form of guanylyl cyclase. A larger response to SNP was observed in females than in males. Interestingly, in females the facilitatory effect of CPA on cGMP accumulation was not modified in the presence of SNP, therefore under high intracellular cGMP levels. In contrast, males failed to respond to CPA when the intracellular cGMP levels were raised by SNP, probably because the large SNP effect on cGMP accumulation masked the CPA effect in the presence of SNP. Therefore, it was not possible to conclude if the CPA response in males is occluded by SNP – suggesting a sGC-mediated effect – or if SNP and CPA responses were additive – suggesting in this case a particulate guanylyl cyclase (pGC)-mediated effect. In cortical astrocytes adenosine A1 receptors were found to increase NO production (Janigro et al., 1996). Also in cultured organotypic hippocampal slices, adenosine A1 receptors activation was able to increase NO formation (Barth et al., 1997); in this study 2-chloro-N6-cyclopentyladenosine (CCPA), an adenosine A1 receptor selective agonist with a potency similar to that of CPA, at a concentration of 100 nM was able to produce a full response on NO formation. Similarly, in the present work, a 100 nM CPA concentration was also able to produce a nearly maximal effect on cGMP accumulation. Taken together, these results suggest that adenosine A1 receptor-mediated activation of NOS might contribute, at least in part, to the adenosine A1 receptor action on cGMP basal levels at the hippocampus. The CPA stimulation of cGMP accumulation in the presence of SNP, observed in females, was reversed by DPCPX, stressing that adenosine A1 receptors kept modulating cGMP accumulation despite the increase in soluble guanylyl cyclase activity induced by SNP. The persistence of an adenosine A1 receptor-mediated effect on SNPstimulated cGMP accumulation in females suggests that adenosine A1 receptor stimulation of sGC not mediated by NO increase, could also be operating in females – in fact, non nitric oxide-mediated mechanisms for sGC activation have been described (Agulló et al., 2005; Jones et al., 2008). The possibility that this effect of adenosine A1 receptor in females results from activation of pGC seems unlikely, since increasing total cGMP levels through NO-mediated activation of sGC by SNP, would reduce the adenosine A1 receptor-induced relative increase of cGMP accumulation if this increase was mediated by pGC activation. The results therefore suggest that the mechanism of adenosine A1 receptor-mediated modulation of cGMP levels at the hippocampus depends on gender, which deserves future investigation. 4.4. Conclusions The results obtained in the present work indicate that adenosine A1 receptors are upstream modulators of cGMP levels at the hippocampus and suggest that both NO-mediated and nonmediated mechanisms might be involved in the adenosine A1 receptor action, depending on gender. Since both adenosine A1 receptor and cGMP have similar regulatory actions at the hippocampus, the possibility that cGMP mediates some of the adenosine A1 receptors effects on neuronal activity must be considered and deserves future investigation.

Acknowledgements This work was partially supported by a project grant from the Portuguese Foundation for Science and Technology (FCT, POCI/

89

SAU-FCF/57973/2004). A. Serpa received a scholarship from FCT (SFRH/BD/65112/2009).

References Agulló, L., Garcia-Dorado, D., Escalona, N., Ruiz-Meana, M., Mirabet, M., Inserte, J., Soler-Soler, J., 2005. Membrane association of nitric oxide-sensitive guanylyl cyclase in cardiomyocytes. Cardiovasc. Res. 68, 65–74. Ambrósio, A.F., Malva, J.O., Carvalho, A.P., Carvalho, A.M., 1997. Inhibition of N-, P/Qand other types of Ca2 þ channels in rat hippocampal nerve terminals by adenosine A1 receptor. Eur. J. Pharmacol. 340, 301–310. Arora, K., Sinha, C., Zhang, W., Ren, A., Moon, C.S., Yarlagadda, S., Naren, A.P., 2013. Compartmentalization of cyclic nucleotide signaling: a question of when, where, and why? Pflug. Arch. 465, 1397–1407. Arrigoni, E., Rosenberg, P.A., 2006. Nitric oxide-induced adenosine inhibition of hippocampal synaptic transmission depends on adenosine kinase inhibition and is cyclic GMP independent. Eur. J. Neurosci. 24, 2471–2480. Barth, A., Newell, D.W., Nguyen, L.B., Winn, H.R., Wender, R., Meno, J.R., Janigro, D., 1997. Neurotoxicity in organotypic hippocampal slices mediated by adenosine analogues and nitric oxide. Brain Res. 762, 79–88. Bartus, K., Pigott, B., Garthwaite, J., 2013. Cellular targets of nitric oxide in the hippocampus. PLoS One 8, e57292. Bender, A.T., Beavo, J.A., 2006. Cyclic nucleotide phosphodiesterases: molecular regulation to clinical use. Pharmacol. Rev. 58, 488–520. Boess, F.G., Hendrix, M., van der Staay, F.J., Erb, C., Schreiber, R., van Staveren, W., de Vente, J., Prickaerts, J., Blokland, A., Koenig, G., 2004. Inhibition of phosphodiesterase 2 increases neuronal cGMP, synaptic plasticity and memory performance. Neuropharmacology 47, 1081–1092. Broome, M.R., Collingridge, G.L., Irving, A.J., 1994. Activation of the NO-cGMP signalling pathway depresses hippocampal synaptic transmission through an adenosine receptor-dependent mechanism. Neurophurmacology 33, 1511–1513. Cascalheira, J.F., Sebastião, A.M., 1998. Adenosine A1 receptor activation inhibits basal accumulation of inositol phosphates in rat hippocampus. Pharmacol. Toxicol. 82, 189–192. Cascalheira, J.F., Sebastião, A.M., Ribeiro, J.A., 2002. Pertussis toxin-sensitive G proteins mediate the inhibition of basal phosphoinositide metabolism caused by adenosine A1 receptors in rat hippocampal slices. Neurochem. Res. 27, 1707–1711. Ciruela, F., Saura, C., Canela, E.I., Mallol, J., Lluis, C., Franco, R., 1996. Adenosine deaminase affects ligand-induced signalling by interacting with cell surface adenosine receptors. FEBS Lett. 380, 219–223. Correia-de-Sá, P., Adães, S., Timóteo, M.A., Vieira, C., Magalhães-Cardoso, T., Nascimento, C., Duarte-Araújo, M., 2006. Fine-tuning modulation of myenteric motoneurons by endogenous adenosine: on the role of secreted adenosine deaminase. Auton. Neurosci 126–127, 211–224. Daly, J.W., Padgett, W., Seamon, K.B., 1982. Activation of cyclic AMP-generating systems in brain membranes and slices by the diterpene forskolin: augmentation of receptor-mediated responses. J. Neurochem. 38, 532–544. de Mendonça, A., Ribeiro, J.A., 1997. Influence of metabotropic glutamate receptor agonists on the inhibitory effects of adenosine A1 receptor activation in the rat hippocampus. Br. J. Pharmacol. 121, 1541–1548. Dias, R.B., Rombo, D.M., Ribeiro, J.A., Henley, J.M., Sebastião, A.M., 2013. Adenosine: setting the stage for plasticity. Trends Neurosci. 36, 248–257. Dunwiddie, T.V., Hoffer, B.J., 1980. Adenine nucleotides and synaptic transmission in the in vitro rat hippocampus. Br. J. Pharmacol. 69, 59–68. Feil, R., Kleppisch, T., 2008. NO/cGMP-dependent modulation of synaptic transmission. In: Sϋdhof, T.C., Starke, K. (Eds.), Handbook of Experimental Pharmacology (Pharmacology of Neurotransmitter Release), 184; 2008, pp. 529–560. Fragata, I.R., Ribeiro, J.A., Sebastião, A.M., 2006. Nitric oxide mediates interactions between GABAA receptors and adenosine A1 receptors in the rat hippocampus. Eur. J. Pharmacol. 543, 32–39. Franco, R., Casadó, V., Ciruela, F., Saura, C., Mallol, J., Canela, E.I., Lluis, C., 1997. Cell surface adenosine deaminase: much more than an ectoenzyme. Prog. Neurobiol. 52, 283–294. Franco, R., Ciruela, F., Casadó, V., Cortes, A., Canela, E.I., Mallol, J., Agnati, L.F., Ferré, S., Fuxe, K., Lluis, C., 2005. Partners for adenosine A1 receptors. J. Mol. Neurosci. 26, 221–232. Fredholm, B.B., Ijzerman, A.P., Jacobson, K.A., Klotz, K.N., Linden, J., 2001. International Union of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol Rev. 53, 527–552. Hernández, F., Kendall, D.A., Alexander, S.P., 1993. Adenosine receptor-induced second messenger production in adult guinea-pig cerebellum. Br. J. Pharmacol. 110, 1085–1090. Janigro, D., Wender, R., Ransom, G., Tinklepaugh, D.L., Winn, H.R., 1996. Adenosineinduced release of nitric oxide from cortical astrocytes. Neuroreport 7, 1640–1644. Jones, J.D., Carney, S.T., Vrana, K.E., Norford, D.C., Howlett, A.C., 2008. Cannabinoid receptor-mediated translocation of NO-sensitive guanylyl cyclase and production of cyclic GMP in neuronal cells. Neuropharmacology 54, 23–30. Kurtz, A., 1987. Adenosine stimulates guanylate cyclase activity in vascular smooth muscle cells. J. Biol. Chem. 262, 6296–6300.

90

A. Serpa et al. / European Journal of Pharmacology 744 (2014) 83–90

Kurtz, A., Della Bruna, R., Pfeilschifter, J., Bauer, C., 1988. Role of cGMP as second messenger of adenosine in the inhibition of renin release. Kidney Int. 33, 798–803. Lima, F.O., Souza, G.R., Verri Jr, W.A., Parada, C.A., Ferreira, S.H., Cunha, F.Q., Cunha, T.M., 2010. Direct blockade of inflammatory hypernociception by peripheral A1 adenosine receptors: involvement of the NO/cGMP/PKG/KATP signaling pathway. Pain 151, 506–515. Lopes, L.V., Cunha, R.A., Ribeiro, J.A., 1999. Cross talk between A1 and A2A adenosine receptors in the hippocampus and cortex of young adult and old rats. J. Neurophysiol. 82, 3196–3203. Lucas, K.A., Pitari, G.M., Kazerounian, S., Ruiz-Stewart, I., Park, J., Schulz, S., Chepenik, K.P., Waldman, S.A., 2000. Guanylyl cyclases and signaling by cyclic GMP. Pharmacol. Rev. 52, 375–414. Marte, A., Pepicelli, O., Cavallero, A., Raiteri, M., Fedele, E., 2008. In vivo effects of phosphodiesterase inhibition on basal cyclic guanosine monophosphate levels in the prefrontal cortex, hippocampus and cerebellum of freely moving rats. J. Neurosci. Res. 86, 3338–3347. Montoliu, C., Llansola, M., Monfort, P., Corbalan, R., Fernandez-Marticorena, I., Hernandez-Viadel, M.L., Felipo, V., 2001. Role of nitric oxide and cyclic GMP in glutamate-induced neuronal death. Neurotox. Res. 3, 179–188. Ohga, Y., Daly, J.W., 1977. The accumulation of cyclic AMP and cyclic GMP in guinea pig brain slices. Effect of calcium ions, norepinephrine and adenosine. Biochim. Biophys. Acta. 498, 46–60. Orio, M., Kunz, A., Kawano, T., Anrather, J., Zhou, P., Iadecola, C., 2007. Lipopolysaccharide induces early tolerance to excitotoxicity via nitric oxide and cGMP. Stroke 38, 2812–2817. Palmon, S.C., Williams, M.J., Littleton-Kearney, M.T., Traystman, R.J., Kosk-Kosicka, D., Hurn, P.D., 1998. Estrogen increases cGMP in selected brain regions and in cerebral microvessels. J. Cereb. Blood Flow Metab. 18, 1248–1252. Petersen, D.N., Tkalcevic, G.T., Koza-Taylor, P.H., Turi, T.G., Brown, T.A., 1998. Identification of estrogen receptor beta2, a functional variant of estrogen receptor beta expressed in normal rat tissues. Endocrinology 139, 1082–1092.

Peterson, G.L., 1977. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem. 83, 346–356. Pimentel, V.C., Pinheiro, F.V., De Bona, K.S., Maldonado, P.A., da Silva, C.R., de Oliveira, S.M., Ferreira, J., Bertoncheli, C.M., Schetinger, M.R., Da Luz, S.C., Moretto, M.B., 2011. Hypoxic-ischemic brain injury stimulates inflammatory response and enzymatic activities in the hippocampus of neonatal rats. Brain Res. 1388, 134–140. Ribeiro, J.A., 2005. What can adenosine neuromodulation do for neuroprotection? Curr. Drug Targets CNS Neurol. Disord. 4, 325–329. Saito, M., 1977. Elevation of guanosine 30 ,50 -monophosphate level by adenosine in cerebellar slices of guinea pig. Biochim. Biophys. Acta 498, 316–324. Santschi, L.A., Zhang, X.L., Stanton, P.K., 2006. Activation of receptors negatively coupled to adenylate cyclase is required for induction of long-term synaptic depression at Schaffer collateral-CA1 synapses. J. Neurobiol. 66, 205–219. Sebastião, A.M., de Mendonca, A., Moreira, T., Ribeiro, J.A., 2001. Activation of synaptic NMDA receptors by action potential-dependent release of transmitter during hypoxia impairs recovery of synaptic transmission on reoxygenation. J. Neurosci. 21, 8564–8571. Serpa, A., Ribeiro, J.A., Sebastião, A.M., 2009. Cannabinoid CB(1) and adenosine A (1) receptors independently inhibit hippocampal synaptic transmission. Eur. J. Pharmacol. 623, 41–46. Sterin-Borda, L., Gómez, R.M., Borda, E., 2002. Role of nitric oxide/cyclic GMP in myocardial adenosine A1 receptor-inotropic response. Br. J. Pharmacol. 135, 444–450. Takigawa, T., Alzheimer, C., 2002. Phasic and tonic attenuation of EPSPs by inward rectifier K þ channels in rat hippocampal pyramidal cells. J. Physiol. 539, 67–75. van Staveren, W.C., Markerink-van Ittersum, M., Steinbusch, H.W., de Vente, J., 2001. The effects of phosphodiesterase inhibition on cyclic GMP and cyclic AMP accumulation in the hippocampus of the rat. Brain Res. 888, 275–286.