Functional crosstalk of prejunctional receptors on the modulation of noradrenaline release in mesenteric vessels: A differential study of artery and vein

European Journal of Pharmacology 652 (2011) 33–39

Contents lists available at ScienceDirect

European Journal of Pharmacology

j ourna l homepage: www.e lsev ie r.com/ locate /e jphar

Neuropharmacology and Analgesia

Functional crosstalk of prejunctional receptors on the modulation of noradrenalinerelease in mesenteric vessels: A differential study of artery and vein

Carlos Talaia, Manuela Morato, Clara Quintas, Jorge Gonçalves, Glória Queiroz ⁎Laboratory of Pharmacology, REQUIMTE, Faculty of Pharmacy, University of Porto, Rua Aníbal Cunha N° 164, 4050-047 Porto, Portugal

⁎ Corresponding author. Laboratory of Pharmacology, Fof Porto, Rua Aníbal Cunha N° 164, 4050-047 Porto, Portfax: +351 22 207 89 69.

E-mail address: [email protected] (G. Queiroz).

0014-2999/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.ejphar.2010.10.075

a b s t r a c t

Article history:Received 1 April 2010Received in revised form 15 October 2010Accepted 31 October 2010Available online 27 November 2010

Keywords:Mesenteric vesselsReceptor interactionα2-autoreceptorsβ-adrenoceptorsAngiotensin IIPurinergic receptors

The role of angiotensin II receptors, bradykinin receptors and β-adrenoceptors in the modulation ofnoradrenaline release and the influence of α2-autoinhibition in these effects was investigated in themesenteric artery and vein. Rings of mesenteric vessels of male Wistar rats were labelled with [3H]-noradrenaline and the effects of modulators on tritium overflow evoked by 100 pulses at 2 Hz (marked α2-autoinhibition) and by 20 pulses at 50 Hz or 100 pulses at 2 Hz plus yohimbine (1 μM; reduced α2-autoinhibition) were evaluated. Angiotensin II and bradykinin enhanced noradrenaline release evoked by 100pulses at 2 Hz, in a concentration-dependent manner, in both vessels. These effects were attenuated underconditions of reduced α2-autoinhibition. The attenuation was partially reversed by activation of adenosine A1

receptors in both vessels and by activation of P2Y receptors in the vein. Isoprenaline and the selective β2-adrenoceptor agonist formoterol enhanced tritium overflow independently of α2-autoinhibition in the vein.In the artery, the enhancement by formoterol was only observed under reduced α2-autoinhibition.Pharmacological characterization of the β-adrenoceptors indicated that in the artery the effect of isoprenalinewas mediated by the β1-subtype under marked α2-autoinhibition and by the β2-subtype under reduced α2-autoinhibition whereas in the vein the effect was independent of α2-autoinhibition. The results indicate thatα2-autoinhibition is a key determinant of the magnitude of facilitation caused by angiotensin II andbradykinin in both types of mesenteric vessels and regulates the effects mediated by β1-and β2-adrenoceptorswhich co-exist in the artery.

aculty of Pharmacy, Universityugal. Tel.: +351 22 207 89 70;

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Alterations in sympathetic neurotransmission have been impli-cated in cardiovascular diseases. This is true for central noradrenergicneurons associated with cardiovascular control (Kasparov andTeschemacher, 2008) and also for the peripheral sympathetic driveto the vasculature (Demel and Galligan, 2008). Arteries and veins playdistinct roles in the maintenance of a proper circulatory dynamics(Kreulen, 2003) and differences in the sympathetic control of bothtypes of vessels may contribute to blood flow regulation. In the veinssympathetic vasoconstriction is mainly mediated by noradrenaline,whereas in the arteries the co-transmitters ATP and neuropeptide Yalso play an important role (Donoso et al., 1997; Smyth et al., 2000).Differences in the regulation of sympathetic neuronal activity mayalso contribute to the distinct roles of arteries and veins on the controlof blood flow and volume distribution (Smyth et al., 2000; Bobalovaand Mutafova-Yambolieva, 2001; Park et al., 2007). Therefore, studies

on the characterization of vascular sympathetic neurotransmissionare of major relevance, since they may help to identify newtherapeutic strategies for cardiovascular diseases that course withincreased sympathetic nerve activity.

Postganglionic sympathetic neurons are endowed with prejunc-tional receptors that facilitate or inhibit transmitter release and oftenregulate the activity of one another, contributing to a fine-regulationof sympathetic transmission (Langer, 2008). Facilitation of noradren-aline release can be mediated by angiotensin II AT1 and bradykinin B2

receptors (Cox et al., 2000;Mota and Guimarães, 2003; Trendelenburget al., 2003), adenosine A2 (Fresco et al., 2002; Queiroz et al., 2003),and by β2-adrenoceptors in several tissues (Guimarães and Moura,2001). The quantitative effects of these receptors may depend on theongoing activation of receptors that mediate opposite effects (Kubistaand Boehm, 2006). Neuronal mechanisms are especially important inthe regulation of blood flow in vascular territories where this kind ofcontrol is more relevant (such as the large mesenteric arteries andveins whose blockade may lead to infarction) than in territories suchas the skeletal muscle vessels during exercise where the local controlmechanisms predominate. The mesenteric vessels participate inphysiological responses such as the postprandial reactive hyperaemiaor the regulation of mean arterial pressure (Jeays et al., 2007), whichare under control of neuronal mechanisms.

34 C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

Therefore, in the present studywe investigated, in the ratmesentericartery and vein; i) the relative contribution of angiotensin II, bradykinin,purines and β-adrenoceptor agonists to the regulation of noradrenalinerelease and ii) the occurrence of a functional interaction betweenprejunctional receptors activated by these modulators and the maininhibitory receptors present, the α2-autoreceptors. Few studies haveaddressed this issue in arteries and nothing is known about this type ofsignal integration events in veins considering their relevance to thesympathetic regulation of the blood circulation (Dzimiri, 2002).

2. Materials and methods

2.1. Chemicals

The followingdrugswere used: levo-[ring-2,5,6-3H]-noradrenaline([3H]-NA, specific activity 57.8 Ci/mmol)was fromDuPontNEN (Garal,Lisboa, Portugal); desipramine hydrochloride, yohimbine hydrochlo-ride, 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), reactive blue 2(RB2), angiotensin II human acetate, bradykinin acetate, N6-cyclo-pentyladenosine (CPA), (−)-isoproterenol(+)-bitartrate, atenolol,bisindoylmaleimide XI (RO 32,0432), 2-p-(2-carboxyethyl)phenethy-lamino-5′-N-ethylcarboxamidoadenosine hydrochloride (CGS21680), (±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol hydrochloride (ICI 115,558); NG-nitro-L-arginine methyl ester hydrochloride (L-NAME) and L-argininewere from Sigma-Aldrich (Sintra, Portugal); formoterol hemifuma-rate, 2-methylthioadenosine triphosphate tetrasodium (2-MeSATP)and (9S,10S,12R)-2,3,9,10,11,12-Hexahydro-10-hydroxy-9-methyl-1-oxo-9,12-epoxy-1H-diindolo[1,2,3-fg:3′,2′,1′-kl]pyrrolo[3,4-i][1,6]benzodiazocine-10-carboxylic acid hexyl ester (KT 5720) were fromTocris (Bristol, UK). Stock solutions of drugs were prepared withdistilled water or dimethylsulphoxide and diluted in buffer immedi-ately before use. Solvent was added to the superfusion medium inparallel control experiments.

2.2. Animal and tissue preparation

Animal handling and experiments were conducted according tothe guidelines of the European Communities Council Directive (86/609/EEC) in agreement with the NIH guidelines and the Portugueselaw (Portarias n. 1005/92 and n. 1131/97). Adult male Wistar rats(250–300 g; Charles River, Barcelona, Spain) were killed by cervicaldislocation and exsanguination. The superior mesenteric artery andvein were immediately excised and placed in cold Krebs solution ofthe following composition (mM): NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4

1.2, NaH2PO4 1.2, NaHCO3 25, glucose 11, ascorbic acid 0.3 anddisodium EDTA 0.03. Vessels were dissected out, labelled with [3H]-noradrenaline and rings of 3–4 mg were used in release experiments.

2.3. [3H]-Noradrenaline release experiments

The procedures used to label vessels with [3H]-noradrenaline andto estimate changes in electrically evoked tritium overflow as anindicator of neuronal noradrenaline release have been previouslydescribed (Fresco et al., 2002). Briefly, preparations were pre-incubated in Krebs solution containing 0.1 μM [3H]-noradrenalinefor 40 min, transferred to superfusion chambers and superfused, at aconstant flow rate of 1 ml/min, with the buffer solution continuouslyaerated (95% O2 and 5% CO2) and maintained at 37 °C. Electrical fieldstimulation (Hugo Sach Elektronik, Type 215, Hugstetten, Germany)consisted of trains of 100 pulses at 2 Hz or 20 pulses at 50 Hz (pulsewidth of 1 ms; 50 mA current strength; voltage drop of 20 V/cm).These electric field stimulation parameters where chosen becauseprevious studies (Cox et al., 2000) demonstrated they set upconditions of high (100 pulses, 2 Hz) and low (20 pulses, 50 Hz)levels of α2-autoinhibition.

A primer stimulation period, applied at t=45 min (S0), t=0 minbeing the start of superfusion was not used for determination oftritium overflow. Following S0, there were either two more periods ofstimulation (S1 at t=90 min and S2 at t=120 min) or six morestimulation periods (S1 at t=90 min, S2 at t=110 min, S3 at t=130 min, S4 at t=150 min, S5 at t=170 min and S6 at=190 min).Superfusate samples were collected at 5-min intervals, from t=85minonwards. At the end of the superfusion, tritium contentwas determinedin the collected samples and in the correspondent tissue preparationsbyscintillation spectrometry (Beckman LS 6500, Beckman Instruments,Fullerton, USA).

Desipramine (0.4 μM; a noradrenaline neuronal uptake inhibitor)was present throughout superfusion in all experiments. Yohimbine(1 μM; α2-adrenoceptor antagonist) was also present in severalexperiments where it was added throughout the superfusion. In asmall group of experiments yohimbine was added to the superfusionmedium 20 min before S2 and remained until the end of theexperiment. In some experiments, in addition to yohimbine, DPCPX(0.1 μM; adenosine A1 receptor antagonist) and RB2 (30 μM; P2receptor antagonist) were also present throughout superfusion.Angiotensin II and bradykinin were added to the superfusion medium10 min, isoprenaline and formoterol 7 min, CGS 21680 6 min beforeS2–S6, all at increasing concentrations and kept until the end of therespective stimulation period. CPA and 2-MeSATP were added 10 minbefore S2 and were kept up to the end of the stimulation period.Antagonists, enzyme inhibitors and L-arginine were added 20 minbefore S2 and were kept until the end of the experiment.

2.4. Data evaluation and statistical analysis

Tritium outflow was estimated as a fraction of the tissue tritiumcontent at the onset of the respective collection period (fractional rate ofputflow/min); b1 was the fractional rate of outflow in the 5-min periodbefore S1; bn (b2–b6) was the fractional rate of outflow in the 5-minperiod before Sn (S2–S6). Drug effects on basal outflowwere evaluated asratios bn/b1. Tritium overflow evoked by electrical stimulation wasestimated by subtracting basal outflow from total outflow observedduring and in the 10-min period subsequent to each stimulation periodbeing expressed as a percentage of the total tritium present in the tissueat the onset of stimulation. Effects of drugs added after S1 on tritiumoverflow were evaluated by the ratios Sn/S1 and were expressed as apercentage of change (increase or inhibition) from the respective meanratio obtained with the appropriate control. The Sn/S1 values obtained inindividual experiments in which a test compound A was added after S1were calculated as a percentage of the respective mean ratio obtained inthe appropriate control group (solvent instead of A). When theinteraction of A, added after S1, and a drug B added either after S1 or atthe beginning of superfusion, was studied, the “appropriate control”wasa group in which only B was used.

2.5. Statistical analysis

Data are expressed as means±S.E.M. from n number of tissuepreparations. Statistical analysis was performed using the unpairedStudent's t-test or one-way analysis of variance (ANOVA) followed byDunnett's multiple comparison tests. P values less than 0.05 wereconsidered to indicate significant differences.

3. Results

3.1. Influence of stimulation parameters in tritium overflow

The fractional rate of basal tritium outflow and electrically evokedoverflow of tritium from mesenteric vessels stimulated with trains of100 pulses at 2 Hz (100 p/2 Hz) or 20 pulses at 50 Hz (20 p/50 Hz),are shown in Table 1. Basal outflow remained constant throughout

Table 1Basal outflow and electrically evoked tritium overflow from preparations of the ratmesenteric artery and vein.

Drugs present throughoutsuperfusion (in additionto desipramine)

Basal tritium outflow (b1)(% of tissue tritium/min)

Tritium overflow (S1)(% of tissue tritium)

Mesenteric artery100 pulses/2 Hz

Solvent 0.131±0.006 (n=150) 0.574±0.025 (n=150)Yohimbine (1 μM) 0.131±0.003 (n=160) 2.911±0.082 (n=160) a

20 pulses/50 Hz 0.119±0.004 (n=40) 0.494±0.016 (n=40) a

Mesenteric vein100 pulses/2 Hz

Solvent 0.123±0.003 (n=90) 0.380±0.014 (n=90)Yohimbine (1 μM) 0.132±0.004 (n=130) 1.956±0.078 (n=130) a

20 pulses/50 Hz 0.107±0.004 (n=35) 0.256±0.022 (n=35) a

Tissues were incubated with [3H]-NA and superfused with medium containingdesipramine (0.4 μM) and in addition yohimbine or solvent. Up to 7 trains of 100 p/2 Hzor 20 p/50 Hz were applied (S0–S6); b1 is the tritium outflow during the 5 min periodbefore the first stimulation period (S1). Evoked tritium overflow was calculated bysubtracting basal outflow from total outflowmeasured during and in the 10 min after to S1and is expressed as percentage of the total tissue tritium at the onset of the stimulation.Values aremeans±S.E.M. for (n) tissue preparations. aPb0.05, significant differences fromtritium overflow in the presence of solvent, at 100 p/2 Hz.

Fig. 1. Influenceofα2-autoinhibitionon themodulationof tritiumoverflowbyangiotensin II(A,B) and bradykinin (C,D) in mesenteric vessels. Tritium overflow was evoked underdifferent levels ofα2-autoinhibition: stimulation by 100 p/2 Hz (markedα2-autoinhibition)and 100 p/2 Hz plus yohimbine or 20 p/50 Hz (reduced α2-autoinhibition). Tissues wereelectrically stimulated with 7 trains of 100 p/2 Hz or 20 p/50 Hz (S0–S6). Angiotensin II andbradykinin were added 10 min before Sn (S2–S6), at increasing concentrations. Yohimbine,when present, was added at the beginning of superfusion and kept throughout. Forevaluation of the angiotensin II and bradykinin effects on the electrically evoked tritiumoverflow, Sn/S1 ratios obtained in the presence of agonists were expressed as percentages ofthe corresponding average control Sn/S1 value. Ordinates, tritium overflow expressed aspercentage of increase from respective control. Values aremeans±S.E.M. from 5 to 8 tissuepreparations. *Pb0.05, significant differences from respective control (solvent); +Pb0.05,from the effect observed when tissues were stimulated with 100 p/2 Hz.

35C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

superfusion with bn/b1 close to unity (not shown). When yohimbine(1 μM) was added 20 min before S2 tritium overflow elicited by 100p/2 Hz increased by 468±28% (n=4, Pb0.05) in the artery and by489±31% (n=4, Pb0.05) in the vein, indicating a marked ongoingα2-autoinhibition under these stimulation conditions. Stimulationwith 20 p/50 Hz lead to a reduced level ofα2-autoinhibition: yohimbine(1 μM) only increased tritium overflow by 28±2% (n=6, Pb0.05) and30±6% (n=4, Pb0.05) in the artery and vein, respectively.

In the absence of drugs added after S1 the evoked tritium overflowremained constant throughout experiments with Sn/S1 close to unity(not shown), in both types of vessels, and none of the drugs used ortheir solvents changed the basal tritium outflow.

In the mesenteric vessels, ATP is co-released with noradrenaline(Donoso et al., 1997; Smyth et al., 2000), beingmetabolised into ADP andadenosine which modulate noradrenaline release in vessels, throughactivation of P2Y and adenosine A1 receptors (Ralevic, 2009). Tonicinfluence of these modulators was investigated with the selectiveadenosine A1 antagonist DPCPX (0.1 μM) and the non-selective P2receptor antagonist RB2 (30 μM) under conditions of reduced α2-autoinhibition obtained by stimulation with 100 p/2 Hz in the presenceof yohimbine (1 μM). DPCPX (0.1 μM) increased tritiumoutflowby 24±5% (n=4; Pb0.05) and 32±4% (n=5; Pb0.05) in the artery and vein,respectively, whereas RΒ2 (30 μM) had no effect in the artery (notshown) but increased tritium overflow by 21±2% (n=4; Pb0.05) inthe vein, indicating that under these conditions, there is a tonic influenceof A1 receptors in both vessels and of P2 receptors in the vein.

3.2. Effects of angiotensin II and bradykinin in tritium overflow: influence ofα2-autoinhibition

Interaction between angiotensin II and bradykinin in the modula-tion of noradrenaline release from sympathetic nerve terminals in-nervating the mesenteric artery and vein was investigated byevaluating the effects of these modulators on tritium overflow evokedunder different levels of α2-autoinhibition. Angiotensin II (0.001–0.1 μM)enhanced tritiumoverflowevoked by 100 p/2 Hz (markedα2-autoinhibition), in a concentration-dependentmanner, in both vessels(Fig. 1A,B). Bradykinin (0.001–0.1 μM) also markedly enhancedtritium overflow, with a similar profile in both vessels (Fig. 1C,D).When yohimbine (1 μM) was present, the enhancement causedby angiotensin II was attenuated in both vessels (Fig. 1A,B) whereasthat of bradykinin (Fig. 1C,D) was abolished in the artery and

attenuated in the vein. In tissues stimulated by 20 p/50 Hz (reducedα2-autoinhibition), the effects of angiotensin II and bradykinin wereabolished in both vessels (Fig. 1). These results suggest that facilita-tion of tritium overflow mediated by angiotensin II, in both vessels,and by bradykinin in the vein, are partially dependent of ongoing α2-autoinhibition whereas the effect of bradykinin in the artery iscompletely dependent on ongoing α2-autoinhibition.

3.3. Effects of angiotensin II and bradykinin in tritium overflow: influence ofpurinergic tonus

As described in Section 3.1, when α2-adrenoceptors were blockedwith yohimbine (1 μM) tritium overflowwas under an inhibitory tonusmediatedbyA1 receptors, in bothvessels, andof P2 receptors in the vein.We investigatedwhether these purinergic receptors could be favouringthe remaining facilitation causedbyangiotensin II andbradykinin.Usingthe conditions referred in Section 3.1, the remaining facilitation oftritium overflow caused by angiotensin II (0.01 μM) and bradykinin(0.01 μM) was not changed by DPCPX (0.1 μM) and/or RB2 (30 μM),either in the artery or in the vein (not shown), indicating that it is

36 C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

independent of an ongoing tonus mediated by adenosine A1 and/or P2receptors. However, when α2-adrenoceptors were blocked and thepurinergic influence was enhanced by activation of A1 receptors withthe selective agonist CPA (1 μM), the enhancement of tritium overflowcaused by angiotensin II was of similar magnitude to that observed inthe absence of yohimbine, in both vessels (Fig. 2). The recovery of theangiotensin II mediated facilitation of tritium overflow was alsoobserved when it was tested in the presence of the P2 agonists 2-MeSATP (100 μM), but only in the vein (Fig. 2B). CPA (1 μM) per seinhibited tritium overflow by 84±2% (n=4, Pb0.05) in the artery and89±2% (n=6, Pb0.05) in the vein whereas 2-MeSATP (100 μM)inhibited tritium overflow by 33±3% (n=6, Pb0.05) in the artery andby 47±8% (n=6, Pb0.05) in the vein.

3.4. Effect of adenosine A2A receptors on tritium overflow: influence ofα2-autoinhibition

The prejunctional effect of adenosine in the vasculature consists ofan inhibition of transmitter release mediated by A1 receptors, but insome vessels adenosine A2A receptors may also be present andmediatethe opposite effect (Maynard and Burnstock, 1994; Gonçalves andQueiroz, 1996). In themesenteric vessels, the role of A2A receptors in themodulation of transmitter release is unknown; therefore we investi-gated its influence. The selective adenosine A2A receptor agonist CGS21680 (0.001–0.1 μM) did not change tritium overflow evoked by100 p/2 Hz neither with norwithout blockade ofα2-autoreceptorswith1 μM yohimbine, in both vessels (not shown).

3.5. Effects of β-adrenoceptors on tritium overflow: influence ofα2-autoinhibition

The influence of α2-autoinhibition on the β-adrenergic mediatedfacilitation of sympathetic transmitter release has been studied inseveral tissues leading to the conclusions that it may have noinfluence (Cox et al., 2000; Mota et al., 2000) or it may attenuatethe effects mediated by these receptors (Majewski and Rand, 1981;Johnston and Majewski 1986; Queiroz et al., 2003).

Fig. 2. Influence of the adenosine A1 receptor agonist, CPA and the P2 receptor agonist,2-MeSATP, on angiotensin II mediated increase of tritium overflow from themesentericartery (A) and vein (B). Tissues were electrically stimulated with 3 trains of 100 p/2 Hz(S0–S2). Angiotensin II, CPA and 2-MeSATP were added 10 min before S2 up to the endof the stimulation. Yohimbine, when present, was added at the beginning ofsuperfusion and kept throughout. For evaluation of effects of angiotensin II on theelectrically evoked tritium overflow, S2/S1 ratios obtained in the presence ofangiotensin II were expressed as a percentage of the corresponding average controlS2/S1 value, which was for each vessel solvent (left column) or yohimbine plus eachpurinergic agonist or solvent (in the following columns). Ordinates, tritium overflowexpressed as a percentage of increase from respective control (see above). Values aremeans±S.E.M. from 6 tissue preparations. *Pb0.05, significant differences fromrespective control (solvent); +Pb0.05, from the effect of angiotensin II alone;#Pb0.05, from the effect of angiotensin II in the presence of yohimbine.

In the mesenteric vessels, the non-selective β-adrenoceptoragonist isoprenaline (0.001–0.1 μM) enhanced tritium overflow, in aconcentration-dependent manner, when tissues were stimulated by100 p/2 Hz (Fig. 3A,B). Blockade of α2-adrenoceptors with yohimbine(1 μM) or stimulating the vessels with 20 p/50 Hz did not modify theeffect of isoprenaline in the artery but attenuated the effect caused by0.1 μM in the vein (Fig. 3B). When the selective β2-agonist formoterol(0.001–0.1 μM)was tested, an enhancement of tritium overflow in theartery was only observed when α2-autoreceptors were blocked byyohimbine (Fig. 3C), whereas in the vein it enhanced tritium overflowevoked by 100 p/2 Hz independently of α2-autoinhibition (Fig. 3D).

In order to understand the discrepancy between the effects of thetwo agonists in the artery, the β-adrenoceptors were furthercharacterized in both types of vessels stimulated by 100 p/2 Hz inthe absence and presence of 1 μM yohimbine.

In the artery, the effect of isoprenaline was attenuated by the β1-adrenoceptor antagonist atenolol (0.3 μM) in the absence of yohimbineand by the β2-adrenoceptor antagonist ICI 118,551 (0.01 μM) whenyohimbine was present (Fig. 4A). In the vein, the effect of isoprenalinewas attenuated by ICI 118,551 (0.01 μM; Fig. 4B) in both conditions.Neither ICI 118,551 (0.01 μM) nor atenolol (0.3 μM) changed tritiumoverflow, in any of the conditions tested (not shown).

Fig. 3. Influence of α2-autoinhibition on the modulation of tritium overflow by the non-selective β-adrenoceptor agonist isoprenaline and the β2 selective agonist formoterol inmesenteric artery (A,C) and vein (B,D). Tissueswere electrically stimulatedwith 7 trains of100 p/2 Hzor20 p/50 Hz (S0–S6). Isoprenaline and formoterolwere added 7 minbefore Sn(S2–S6), at increasing concentrations. Yohimbine, when present, was added at thebeginning of superfusion and kept throughout. For evaluation of effects of isoprenaline andformoterol on the electrically evoked tritium overflow, Sn/S1 ratios obtained in thepresence of the agonists were expressed as percentages of the corresponding averagecontrol Sn/S1 value. Ordinates, tritiumoverflowexpressed as a percentage of increase fromrespective control. Values are means±S.E.M. from 5 to 8 tissue preparations. *Pb0.05,significant differences from respective control (solvent); +Pb0.05, from the effectobserved when tissues were stimulated with 100 p/2 Hz.

Fig. 4. Pharmacological characterization of β-adrenoceptors in the mesenteric artery(A) and vein (B): interaction with the selective β1 antagonist atenolol or the selective β2

antagonist, ICI 118,551. Tissues were electrically stimulated with 3 trains of 100 p/2 Hz.Isoprenaline and formoterol were added 7 min before S2 and kept up to the end ofstimulation. Atenolol and ICI 118,551 were added 20 min before S2 and keptthroughout. Yohimbine, when present, was added at the beginning of superfusionand kept throughout. For evaluation of effects of drugs on the electrically evokedtritium overflow, Sn/S1 ratios obtained in the presence of agonists were expressed aspercentages of the corresponding average control Sn/S1 value. Ordinates, tritiumoverflow expressed as a percentage of increase from respective control. Values aremeans±S.E.M. from 6 to 8 tissue preparations. *Pb0.05, significant differences fromrespective control (solvent); +Pb0.05, from the effect of isoprenaline alone.

37C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

3.6. Intracellular signalling mechanism of angiotensin II, bradykinin andβ-adrenoceptors

The explanation for the different interactions observed betweenthe α2-autoreceptors and the angiotensin II and bradykinin receptors,or the different subtypes of β-adrenoceptors, may reside on theintracellular mechanisms they activate. In order to clarify this issue,we investigated the intracellular signalling pathway activated bythese receptors. The signalling mechanism activated by angiotensin IIand bradykinin was investigated only in the artery, whereas that

Fig. 5. Signalling mechanisms of β1-adrenoceptors in the artery (A), β2-adrenoceptors informoterol with the PKC inhibitor RO 32,0432, the PKA inhibitor KT 5720, the NO synthase inwere electrically stimulated by 3 trains of 100 p/2 Hz. Isoprenaline or formoterol were addedor L-NAME plus L-arginine were added 20 min before S2 and kept throughout. Yohimbine (1evaluation of effects of drugs on the electrically evoked tritium overflow, Sn/S1 ratios obtaaverage control Sn/S1 value. Ordinates, tritium overflow expressed as a percentage of increa*Pb0.05, significant differences from respective control (solvent); +Pb0.05, from the effect

activated by β-adrenoceptors was investigated in the artery and in thevein since the effects mediated by β2-adrenoceptors in the two vesselsdiffered in their dependence of α2-autoinhibition.

The increase in tritium overflow caused by angiotensin II (0.01 μM;89±7%, n=5; Pb0.05) or bradykinin (0.01 μM; 62±6%, n=5;Pb0.05) was abolished by the protein kinase C (PKC) inhibitor RO32,0432 (1 μM) to −8±7% (n=5) and −9±6% (n=5), respectively,indicating that facilitation of tritium overflow by these two modulatorsis mediated by PKC.

The signalling pathways activated by β1-adrenoceptors in the arteryand β2-adrenoceptors in the vein were investigated using isoprenaline(0.1 μM; Fig. 5). The effect of isoprenaline was not changed by the PKCinhibitor RO 32,432 (1 μM) or by the nitric oxide (NO) synthaseinhibitor L-NAME(0.3 mM) inboth types of vessels, being attenuatedbythe protein kinase A (PKA) inhibitor KT 5720 (1 μM; Fig. 5A,B). Thesignalling pathway of β2-adrenoceptor in the artery was also investi-gated in the presence of 1 μM yohimbine by testing the influence of thePKC, PKA and NOS inhibitors in the enhancement of tritium overflowcaused by formoterol (0.1 μM; Fig. 5C). The PKC and PKA inhibitorsdid not change the effect of formoterol, which was only attenuated byL-NAME (0.3 mM), an effect reversed by L-arginine (1 mM; Fig. 5C).Neither the protein kinase inhibitors, L-NAME nor the combinationof L-NAME plus L-arginine, when tested alone, change tritium overflow(not shown).

Results indicate that effects of β1-adrenoceptors in the artery andβ2-adrenoceptors in the vein are mediated by PKA and suggest thatthe effect mediated by β2-adrenoceptors in the artery is independentof PKA or PKC activation being partially mediated by NO.

4. Discussion

In the present study, the effect of angiotensin II, bradykinin, purinesand β-adrenoceptor agonists on noradrenaline release in the ratmesenteric artery and veinwas characterized, and putative interactionsbetween the prejunctional facilitatory receptors activated by thesemodulators and the inhibitoryα2-autoreceptors were also investigated.

As described in other cardiovascular tissues (Onaka et al., 1991; Coxet al., 2000; Guimarães and Pinheiro, 2005; Morato et al., 2006),angiotensin II and bradykinin facilitate noradrenaline release.Our results show that in the mesenteric vessels the magnitude offacilitation depends on the ongoing activation of α2-adrenoceptors, inagreement with previous observations made in the rabbit pulmonary

the vein (B) and β2-adrenoceptors in the artery (C). Interaction of isoprenaline andhibitor L-NAME or the combination of L-NAME plus the NO precursor L-arginine. Tissues7 min before S2 being kept up to the end of stimulation. RO 32,0432, KT 5720, L-NAME

μM), when present, was added at the beginning of superfusion and kept throughout. Forined in the presence of agonists were expressed as percentages of the correspondingse from respective control. Values are means±S.E.M. from 4 to 12 tissue preparations.of agonist alone; #Pb0.05, from the effect of formoterol plus L-NAME.

Table 2Summary of the influence of α2-autoreceptors on the modulation of noradrenalinerelease mediated by angiotensin II, bradykinin, A2A and β-adrenoceptors in themesenteric vessels.

Modulators/receptors

Mesenteric artery Mesenteric vein

Marked α2-autoinhibition

Reduced α2-autoinhibition

Marked α2-autoinhibition

Reduced α2-autoinhibition

AngiotensinII (AT1)

↑↑↑↑ ↑a ↑↑↑ ↑a

Bradykinin ↑↑↑↑ → ↑↑↑↑ ↑Isoprenaline (β1) ↑↑ → → →Isoprenaline (β2) → ↑↑ ↑↑↑ ↑↑↑b

Formoterol (β2) → ↑↑ ↑↑ ↑↑CGS 21680 (A2A) → → → →

↑ and → indicate facilitation and no effect, respectively, on noradrenaline release frommesenteric vessels.

a The facilitatory effect was partially restored to the levels observed under markedα2-autoinhibition by activation of inhibitory adenosine A1 receptors in both types ofvessels and P2Y receptors in the vein.

b Indicates an attenuation of the effect caused by the highest concentration tested,suggesting receptor desensitization.

38 C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

arteries (Costa and Majewski, 1988), mouse atria (Cox et al., 2000;Trendelenburg et al., 2003) and rat tail arteries (Mota and Guimarães,2003). Additionally, these results demonstrate, for the first time that,such dependence is extensive to veins. Moreover, differences betweenthe two modulators in both types of vessels were identified. A residualfacilitation mediated by angiotensin II could still be observed, in bothvessels, when α2-autoreceptors were blocked, whereas the effect ofbradykinin was attenuated in the vein but was abolished in the artery.The possibility that noradrenaline release was already maximal whenα2-adrenoceptors were blocked by yohimbine, preventing furtherincreases from being observed is unlikely, since under the sameexperimental conditions the β-adrenoceptor agonist isoprenalinefacilitated noradrenaline. Furthermore, two different conditions thatlead to reduced α2-autoinhibition also lead to an attenuation of theangiotensin II and bradykinin effects indicating that an endogenousinhibitory tonus mediated by α2-autoreceptors is required for bothmodulators to cause a maximum effect in mesenteric vessels. In theartery, the effect of bradykinin was completely dependent on thatinhibitory tonus. However, the residual facilitation caused by angioten-sin II in both vessels and by bradykinin in the vein observed in thepresence of yohimbine could depend on the tonic activation of otherinhibitory receptors, namely thepurinergic receptors. ATP is co-releasedwith noradrenaline from mesenteric vessels (Kügelgen and Starke,1985; Donoso et al., 1997; Smyth et al., 2000) and may be metabolisedinto ADP or adenosine, which activate prejuctional inhibitory P2Y andadenosine A1 receptors, respectively (Tabrizchi and Bedi, 2001; Ralevic,2009). Our results indicate that adenosine A1 receptors mediate aninhibitory tonus in both vessels whereas the P2Y receptors are onlyactive in the vein. The absence of P2Y inhibitory tonus in the artery is inagreement with a higher ectonucleotidase activity in these vessels(Bobalova and Mutafova-Yambolieva, 2003), resulting in a fasternucleotide metabolism. The adenosine A1 and/or P2Y inhibitory tonuswere not responsible for the remaining facilitation of noradrenalinerelease observedwhenα2-autoreceptorswere blockedasdemonstratedby the lack of influence of the respective antagonists DPCPX andRB2. However, when the influence of adenosine A1 and P2Y receptorswas increased using exogenous agonists of these receptors, thefacilitation caused by angiotensin II was partially restored to thelevels observed when α2-autoreceptors were operating. Like the α2-autoreceptors (Delmas et al., 1999), adenosine A1 receptors are coupledto Gi/o-proteins (Munshi et al., 1991), suggesting that besides α2-autoreceptors other prejunctional Gi/o-protein coupled receptors mayinteract with angiotensin II receptors.

Several mechanisms have been proposed to explain the functionalinteraction between Gq-phospholipase C-PKC coupled receptors, suchas the angiotensin II or bradykinin receptors, and Gi/o-protein coupledreceptors such as the α2-autoreceptors (Hamid et al., 1999; Cooperet al., 2000; Kubista and Boehm, 2006). Such mechanisms mayoperate at mesenteric vessels since angiotensin II and bradykininmediated facilitation of noradrenaline release in these vessels wasprevented by inhibition of PKC.

In addition to angiotensin II and bradykinin receptors, adenosineA2A receptors have been shown to enhance noradrenaline release inthe guinea pig pulmonary (Wiklund et al., 1989), rabbit ear (Maynardand Burnstock, 1994) and rat tail arteries (Gonçalves and Queiroz,1996). However, the selective agonist of adenosine A2A receptors, CGS21680, failed to change noradrenaline release, independently of theinhibitory tonus, indicating that these receptors do not play a role inthe modulation of noradrenaline release in the mesenteric vessels, atleast under the experimental conditions used.

The β2-adrenoceptors are also present in sympathetic nerveterminals and mediate an increase in noradrenaline release in severalcardiovascular tissues (Guimarães andMoura, 2001). Our results revealinteresting differences between the mesenteric artery and veinconcerning the effects mediated by β-adrenoceptors and their depen-dence on α2-autoinhibition. In the artery, noradrenaline release is

modulated by β1- and β2-adrenoceptors with each subtype interactingdifferentlywith theα2-autoreceptors: theβ2-adrenoceptorsmediated afacilitation of noradrenaline release only under conditions of reducedα2-autoinhibition, while the β1-adrenoceptor-mediated facilitationrequired activation of α2-autoreceptors. Although β-adrenoceptorsare usually coupled to the Gs-adenylyl cyclase-PKA pathway, one maysuggest that β1 and β2-subtypes may be coupled to different signallingmechanisms, a hypothesis that could help to explain the differencesobserved. The results obtained indicate that seems to be the case in themesenteric artery where the β1-adrenoceptors are coupled to PKAactivation whereas the β2-adrenoceptors seem to be coupled to amechanism that is independent of PKA or PKC activation but involvesthe production of NO.

The involvement of NO on the effectsmediated by β2-adrenoceptorshas been described in other vessels (Lee et al., 2000; Blanco-Rivero et al.,2006). Also, in rat atria (Yamamoto et al., 1993), mesenteric artery(Gironacci et al., 1997; Yamamoto et al., 1997) and in the sheepmiddlecerebral artery (Mbaku et al., 2000) inhibition of endogenous NOproduction has been shown to attenuate noradrenaline release inducedbyelectrical stimulation,which is in favourof a prejunctional facilitatoryaction of NO.

In the vein, the β2-adrenoceptor is the only β-adrenoceptormediating the facilitation of noradrenaline release, since the effects ofisoprenaline and of the β2-selective agonist formoterol caused anincrease in noradrenaline release of similar magnitude and the effectof isoprenaline was not changed by the β1-adrenoceptor antagonistatenolol, being attenuated by the selective β2-antagonist ICI 118,551.The β2-adrenoceptors of the vein, like the β1-receptors in the artery,are coupled to a signalling mechanism that leads to PKA activation.Interestingly, in the vein, the effect of the highest concentration ofisoprenaline tested was attenuated by preventing α2-autoinhibition.Activation of β2-adrenoceptors is known to induce receptor endocy-tosis, followed by receptor desensitization by a mechanism involvingthe participation of β-arrestins (Claing et al., 2002). The maintenanceof the isoprenaline response at the highest concentration tested whenα2-autoreceptors were operating suggests that they may help toregulate the rate of β2-adrenoceptor-desensitization.

This work describes several mechanisms (summarized in Table 2)operating in sympathetic nerve terminals from the rat mesentericartery and vein, which might contribute to the understanding ofthe regulation mechanisms operating in the two vascular territories.Theα2-adrenoceptors are key receptors in the regulation of angiotensinII and bradykinin mediated facilitation of sympathetic transmission.The α2-autoreceptors may also influence the rate of desensitization ofβ2-adrenoceptors in the vein and in the artery they are responsible for a

39C. Talaia et al. / European Journal of Pharmacology 652 (2011) 33–39

differential activation of β1-adrenoceptors and β2-adrenoceptors. Thefunctional crosstalk between these prejunctional facilitatory andinhibitory receptors may represent an important mechanism bywhich vessels maintain a proper regulation of the sympathetic tonus.

Acknowledgments

This study was supported by the Fundação para a Ciência eTecnologia (POCI/SAU-FCF/60714/2004) and Grant SFRH/BD/13459/2003. The authors would like to thank the helpful technical assistanceof Maria do Céu Pereira during the ongoing experiments.

References

Blanco-Rivero, J., Aras-López, R., Ferrer, M., 2006. Orchidectomy increases β-adrenoceptor activation-mediated neuronal nitric oxide and noradrenaline releasein rat mesenteric artery. Neuroendocrinology 84, 378–385.

Bobalova, J., Mutafova-Yambolieva, V.N., 2001. Presynaptic α2-adrenoceptor-mediatedmodulation of adenosine 5′ triphosphate and noradrenaline corelease: differencesin canine mesenteric artery and vein. J. Auton. Pharmacol. 21, 47–55.

Bobalova, J., Mutafova-Yambolieva, V.N., 2003. Membrane-bound and releasablenucleotidase activities: differences in canine mesenteric artery and vein. Clin.Exp. Pharmacol. Physiol. 30, 194–202.

Claing, A., Laporte, S.A., Caron, M.G., Lefkowitz, R.J., 2002. Endocytosis of G protein-coupled receptors: roles of G protein-coupled receptor kinases and β-arrestinproteins. Prog. Neurobiol. 66, 61–79.

Cooper, C.B., Arnot, M.I., Feng, Z.P., Jarvis, S.E., Hamid, J., Zamponi, G.W., 2000. Cross-talkbetween G-protein and protein kinase C modulation of N-type calcium channels isdependent on the G-protein beta subunit isoform. J. Biol. Chem. 275, 40777–40781.

Costa, M., Majewski, H., 1988. Facilitation of noradrenaline release from sympatheticnerves through activation of ACTH receptors, beta-adrenoceptors and angiotensinII receptors. Br. J. Pharmacol. 95, 993–1001.

Cox, S.L., Schelb, V., Trendelenburg, A.U., Starke, K., 2000. Enhancement of noradren-aline release by angiotensin II and bradykinin in mouse atria: evidence for cross-talk between Gq/11 protein- and Gi/o protein-coupled receptors. Br. J. Pharmacol.129, 1095–1102.

Delmas, P., Abogadie, C., Milligan, G., Buckley, N.J., Brown, D., 1999. βγ Dimers derivedfrom Go and Gi proteins contribute to different components of adrenergicinhibition of Ca2+ channels in rat sympathetic neurons. J. Physiol. Paris 518, 23–36.

Demel, S.L., Galligan, J.J., 2008. Impaired purinergic neurotransmission to mesentericarteries in deoxycorticosterone acetate-salt hypertensive rats. Hypertension 52,322–329.

Donoso, M.V., Steiner, M., Huidobro-Toro, J.P., 1997. BIBP 3226, suramin and prazosinidentify neuropeptide Y, adenosine 5′-triphosphate and noradrenaline as sympa-thetic cotransmitters in the rat arterial mesenteric bed. J. Pharmacol. Exp. Ther. 282,691–698.

Dzimiri, N., 2002. Receptor crosstalk: implications for cardiovascular function, diseaseand therapy. Eur. J. Biochem. 269, 4713–4730.

Fresco, P., Diniz, C., Queiroz, G., Gonçalves, J., 2002. Release inhibitory receptorsactivation favours the A2A-adenosine receptor-mediated facilitation of noradren-aline release in isolated rat tail artery. Br. J. Pharmacol. 136, 230–236.

Gironacci, M.M., Lorenzo, P.S., Adler-Graschinsky, E.A., 1997. Possible participation ofnitric oxide in the increase of norepinephrine release caused by angiotensinpeptides in rat atria. Hypertension 29, 1344–1350.

Gonçalves, J., Queiroz, G., 1996. Purinoceptor modulation of noradrenaline release in rattail artery: tonic modulation mediated by inhibitory P2Y- and facilitatory A2A-purinoceptors. Br. J. Pharmacol. 117, 156–160.

Guimarães, S., Moura, D., 2001. Vascular adrenoceptors: an update. Pharmacol. Rev. 53,319–356.

Guimarães, S., Pinheiro, H., 2005. Functional evidence that in the cardiovascular systemAT1 angiotensin II receptors are AT1B prejunctionally and AT1A postjunctionally.Cardiovasc. Res. 67, 208–215.

Hamid, J., Nelson, D., Spaetgens, R., Dubel, S.J., Snutch, T.P., Zamponi, G.W., 1999.Identification of an integration center for cross-talk between protein kinase C and Gprotein modulation of N-type calcium channels. J. Biol. Chem. 274, 6195–6202.

Jeays, A.D., Lawford, P.V., Gillott, R., Spencer, P.A., Bardhan, K.D., Hose, D.R., 2007. Aframework for the modeling of gut blood flow regulation and postprandialhyperaemia. World J. Gastroenterol. 13, 1393–1398.

Johnston, H., Majewski, H., 1986. Prejunctional β-adrenoceptors in rabbit pulmonaryartery and mouse atria: effect of α-adrenoceptor blockade and phosphodiesteraseinhibition. Br. J. Pharmacol. 87, 553–562.

Kasparov, S., Teschemacher, A.G., 2008. Altered central catecholaminergic transmissionand cardiovascular disease. Exp. Physiol. 93, 725–740.

Kreulen, D.L., 2003. Properties of the venous and arterial innervation in the mesentery.J. Smooth Muscle Res. 39, 269–279.

Kubista, H., Boehm, S., 2006. Molecular mechanisms underlying the modulation ofexocytotic noradrenaline release via presynaptic receptors. Pharmacol. Ther. 112,213–242.

Kügelgen, I., Starke, K., 1985. Noradrenaline and adenosine triphosphate as co-transmitters of neurogenic vasoconstriction in rabbit mesenteric artery. J. Physiol.367, 435–455.

Langer, S.Z., 2008. Presynaptic autoreceptors regulating transmitter release. Neuro-chem. Int. 52, 26–30.

Lee, T.J., Zhang,W., Sarwinski, S., 2000. Presynaptic β2-adrenoceptors mediate nicotine-induced NOergic neurogenic dilation in porcine basilar arteries. Am. J. Physiol. 279,H808–H816.

Majewski, H., Rand, M.J., 1981. An interaction between prejunctional alpha-adreno-ceptors and prejunctional beta-adrenoceptors. Eur. J. Pharmacol. 69, 493–4988.

Maynard, K.I., Burnstock, G., 1994. Evoked noradrenaline release in the rabbit earartery: enhancement by purines, attenuation by neuropeptide Y and lack of effectof calcitonin gene-related peptide. Br. J. Pharmacol. 112, 123–126.

Mbaku, E.N., Zhang, L., Duckles, S.P., Buchholz, J., 2000. Nitric oxide synthase-containingnerves facilitate adrenergic transmitter release in sheep middle cerebral arteries.J. Pharmacol. Exp. Ther. 293, 397–402.

Morato, M., Pinho, D., Sousa, T., Guimarães, S., Moura, D., Albino-Teixeira, A., 2006. Pre-and postjunctional effects of angiotensin II in hypertension due to adenosinereceptor blockade. Eur. J. Pharmacol. 531, 209–216.

Mota, A., Guimarães, S., 2003. Influence of α2-autoreceptor-stimulation on thefacilitation by angiotensin II and bradykinin of noradrenaline release. Naunyn-Schmiedeberg's Arch. Pharmacol. 368, 443–447.

Mota, A., Paiva, M.Q., Moura, D., Guimarães, S., 2000. Lack of interaction between α2-autoreceptors and prejunctional receptors mediating a facilitatory effect onnoradrenaline release. Pharmacol. Res. 42, 383–387.

Munshi, R., Pang, I.H., Sternweis, P.C., Linden, J., 1991. A1 adenosine receptors of bovinebrain couple toguanine nucleotide-binding proteins Gi1, Gi2, and Go. J. Biol. Chem.266, 22285–22289.

Onaka, U., Fujii, K., Abe, I., Fujishima, M., 1997. Enhancement by exogenous and locallygenerated angiotensin II of purinergic neurotransmission via angiotensin type 1receptor in guinea-pig isolated mesenteric artery. Br. J. Pharmacol. 122, 942–948.

Park, J., Galligan, J.J., Fink, G.D., Swain, G.M., 2007. Differences in sympatheticneuroeffector transmission to rat mesenteric arteries and veins as probed by invitro continuous amperometry and video imaging. J. Physiol. 584, 819–834.

Queiroz, G., Talaia, C., Gonçalves, J., 2003. Adenosine A2A receptor-mediated facilitation ofnoradrenaline release involves protein kinase C activation and attenuation ofpresynaptic inhibitory receptor-mediated effects in the rat vas deferens. J.Neurochem.85, 740–748.

Ralevic, V., 2009. Purines as neurotransmitters and neuromodulators in blood vessels.Curr. Vasc. Pharmacol. 7, 3–14.

Smyth, L., Bobalova, J., Ward, S.M., Keef, K.D., Mutafova-Yambolieva, V.N., 2000.Cotransmission from sympathetic vasoconstrictor neurons: differences in guinea-pig mesenteric artery and vein. Auton. Neurosci. 86, 18–29.

Tabrizchi, R., Bedi, S., 2001. Pharmacology of adenosine receptors in the vasculature.Pharmacol. Ther. 91, 133–147.

Trendelenburg, A.U., Meyer, A., Klebroff, W., Guimarães, S., Starke, K., 2003. Cross talkbetween presynaptic angiotensin receptors, bradykinin receptors and α2-auto-receptors in sympathetic neurons: a study in α2-adrenoceptor-deficient mice. Br. J.Pharmacol. 138, 1389–1402.

Wiklund, N.P., Cederqvist, B., Gustafsson, L.E., 1989. Adenosine enhancement of adrenergicneuroeffector transmission in guinea-pig pulmonary artery. Br. J. Pharmacol. 96,425–433.

Yamamoto, R., Wada, A., Asada, Y., Niina, H., Sumiyoshi, A., 1993. N-ω-nitro-L-arginine,an inhibitor of nitric oxide synthesis, decreases noradrenaline outflow in ratisolated perfused mesenteric vasculature. Naunyn-Schmiedeberg's Arch. Pharma-col. 347, 238–240.

Yamamoto, R., Wada, A., Asada, Y., Yanagita, T., Yuhi, T., Niina, H., Sumiyoshi, A.,Kobayashi, H., Lee, T.J., 1997. Nitric oxide dependent and independent norepi-nephrine release in rat mesenteric arteries. Am. J. Physiol. 272, H207–H210.