Role of K+ and Ca2+ fluxes in the cerebroarterial vasoactive effects of sildenafil

Available online at www.sciencedirect.com

gy 581 (2008) 138–147www.elsevier.com/locate/ejphar

European Journal of Pharmacolo

Role of K+ and Ca2+ fluxes in the cerebroarterial vasoactiveeffects of sildenafil

Juan B. Salom a,b,⁎, María Castelló-Ruiz a, María C. Burguete b, Carla Guzmán a,Teresa Jover-Mengual b, Germán Torregrosa a,b, Ramiro Jover a,c,

Ignacio Lizasoain d, Enrique Alborch a,b

a Centro de Investigación, Hospital Universitario ‘La Fe’, Valencia, Spainb Departamento de Fisiología, Universidad de Valencia, Valencia, Spain

c Departamento de Bioquímica y Biología Molecular, Universidad de Valencia, Valencia, Spaind Departamento de Farmacología, Facultad de Medicina, Universidad Complutense de Madrid, Madrid, Spain

Received 17 May 2007; received in revised form 5 November 2007; accepted 15 November 2007Available online 26 December 2007

Abstract

The aim of this study was to assess the role of K+ and Ca2+ fluxes in the cerebroarterial vasoactive effects of the phosphodiesterase-5 inhibitorsildenafil. We used isolated rabbit basilar arteries to assess the effects of extracellular K+ raising on sildenafil-induced vasodilatation, and studiedthe pharmacological interaction of sildenafil with selective modulators of membrane K+ and Ca2+ channels. Expression of Kv1 subunits of K+

channels was assessed at messenger and protein levels. Parallel experiments were carried out with zaprinast for comparison. Sildenafil (10 nM–0.1 mM) induced concentration-dependent relaxation of endothelin-1 (10 nM)-precontracted arteries, which was partially inhibited bydepolarization with KCl (50 mM), 3 mM tetraethylammonium (non-selective K+ channel blocker) or 1 mM aminopyridine (inhibitor of Kv

channels), but not by 1 μM glibenclamide (inhibitor of KATP channels) or 50 nM iberiotoxin (inhibitor of KCa channels). Arterial smooth muscleexpressed messengers for Kv1.2, Kv1.3, Kv1.4, Kv1.5 and Kv1.6, and proteins of Kv1.1, Kv1.2 and Kv1.4. CaCl2 (10 μM- 10 mM) inducedconcentration-dependent contraction in Ca2+-free, depolarizing (50 mM KCl) medium. Sildenafil (0.1–100 μM) produced reversibleconcentration-dependent inhibition of the response to CaCl2, which was completely abolished by the highest sildenafil concentration. Bycontrast, only 100 μM zaprinast inhibited the response to CaCl2. The L-type Ca2+ channel activator Bay K 8644 (0.1 nM–1 μM) inducedconcentration-dependent potentiation of the response to CaCl2 inhibited by 100 μM sildenafil. Moreover, Bay K 8644 (0.1 nM–1 μM) inducedconcentration-dependent contraction in slightly depolarizing (15 mM) medium, which was inhibited to the same extent and in a concentration-dependent way by sildenafil (0.1–100 μM) and zaprinast (1 or 100 μM). These results show that sildenafil relaxes the rabbit basilar artery byincreasing K+ efflux through Kv channels, which in turn may affect Ca2+ signalling. Expression of Kv1 subunits involved in this pharmacologicaleffect occurs at the messenger and, in some cases, at the protein level. In addition to this phosphodiesterase-5-related effect, sildenafil andzaprinast inhibit cerebroarterial vasoconstriction at least in part by directly blocking L-type Ca2+ channels, although a decrease in the sensitivity ofthe contractile apparatus to Ca2+ can not be discarded.© 2007 Elsevier B.V. All rights reserved.

Keywords: Sildenafil; Vasodilatation; Phosphodiesterase-5; K+ efflux; Kv1 subunit; Ca2+ influx; Rabbit basilar artery

⁎ Corresponding author. Centro de Investigación, Hospital Universitario ‘LaFe’, Ave. Campanar 21, 46009 Valencia, Spain. Tel.: +34 963862797; fax: +34961973018.

E-mail address: [email protected] (J.B. Salom).

0014-2999/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.ejphar.2007.11.032

1. Introduction

Guanylyl cyclases are a family of enzymes that catalyze theconversion of GTP to cGMP in response to diverse extracellularand intracellular activators. There are several particulate andsoluble isoforms expressed in different cell types. NO activatessoluble guanylyl cyclase by direct binding to heme, thusinducing cGMP formation which mediates inhibition of platelet

mailto:[email protected]

http://dx.doi.org/10.1016/j.ejphar.2007.11.032

139J.B. Salom et al. / European Journal of Pharmacology 581 (2008) 138–147

aggregation, relaxation of smooth muscle, neurotransmissionand immunomodulation (Lucas et al., 2000). On the other hand,cyclic nucleotide phosphodiesterases are a superfamily ofenzymes that regulate the cellular levels of cAMP and cGMPby catalyzing their degradation to AMP and GMP respectively,thus interrupting signal transduction. There are 11 differentphosphodiesterase families showing different structures, kineticproperties, modes of regulation, intracellular localization,cellular expression, substrate specificities and inhibitor sensi-tivities. Among them, phosphodiesterase-5 is characterized by arelative specificity for cGMP hydrolysis and by its functionalrelevance as a regulator of vascular smooth muscle contraction(Bender and Beavo, 2006).

Sildenafil, the active principle of (Viagra®), is a member ofthe drug family of phosphodiesterase-5 inhibitors primarily usedin erectile dysfunction (Briganti et al., 2005) but more recentlysuggested to have therapeutic potential in the treatment ofrecurrent priapism (Burnett et al., 2006), premature ejaculation(Wang et al., 2006), chronic heart failure and pulmonaryhypertension (Patel and Katz, 2005). Moreover, preclinicalstudies with animal models have shown that phosphodiesterase-5 inhibitors may be useful in the treatment of inflammatoryairway disease (Toward et al., 2004), obesity-associatedhypertension (Beltowski et al., 2006), benign prostatic hyper-plasia and lower urinary tract symptoms (Tinel et al., 2006),myocardial infarct (Sesti et al., 2007) and stroke (Zhang et al.,2006). The rationale for these uses is always the ability of thesedrugs to interfere the NO-cGMP pathway in target tissues.

If we focus on the cerebrovascular bed, sildenafil-inducedtransient increases in cerebral blood flow have been described inrats after oral administration (Zhang et al., 2002). However,intake of sildenafil doses used to treat erectile dysfunction failedto induce changes in either middle cerebral artery diameter,blood velocity, or cerebral blood flow in both men and women(Kruuse et al., 2002, 2003; Arnavaz et al., 2003). Therapeuticalapplications in cerebrovascular disorders are being suggested assildenafil induces neurogenesis and promotes functionalrecovery after focal ischemia in rats (Zhang et al., 2002,2006), and reverses vasospasm after subarachnoid hemorrhagein dogs (Inoha et al., 2002). We and others have recently shownthe effects of sildenafil in cerebral arteries. Phosphodiesterase-5is present in human and guinea pig cerebral arteries, and isinhibited by sildenafil to induce vasodilatation which isaugmented by sodium nitroprusside (NO donor) co-adminis-tration (Kruuse et al., 2005). In the rabbit basilar artery,sildenafil also enhances the NO-cGMP signalling pathway toinduce vasodilatation, prevent vasoconstriction and potentiatethe effect of other NO-dependent vasodilators. Endothelium-borne NO is at least partially involved in these vasoactiveeffects of sildenafil (Salom et al., 2006).

The aim of this study was to assess the possible role of K+

and Ca2+ fluxes in the cerebroarterial vasoactive effects ofsildenafil. We hypothesized that sildenafil can directly orindirectly, through phosphodiesterase-5 inhibition, interact withthe cellular ion fluxes controlling membrane potential andcytoplasmic Ca2+ concentration to induce vasodilatation orprevent vasoconstriction. To prove it, we used isolated rabbit

basilar arteries to assess the effects of extracellular K+ raising onsildenafil-induced vasodilatation, and studied the pharmacolo-gical interaction of sildenafil with selective modulators ofmembrane K+ and Ca2+ channels. Parallel experiments werecarried out with zaprinast, another phosphodiesterase-5 inhi-bitor, for comparison. Finally, since expression and function ofKv1 channel subunits have been reported in cerebral vessels(Cheong et al., 2001a,b; Albarwani et al., 2003; Chen et al.,2006), we aimed to characterize this expression (mRNA and/orprotein) in the rabbit basilar artery.

2. Materials and methods

Experiments were conducted in compliance with the Spanishlegislation on “Protection of Animals used for Experimental andother Scientific Purposes”, and in accordance with theDirectives of the European Union on this subject.

2.1. Animals and isolation of tissues

Ninety-two male New Zealand White rabbits (TechnologyTransferring Centre, Polytechnic University of Valencia, Spain),weighing 2.5–3 kg, were killed by injection of 25 mg kg−1

sodium thiopental (Tiobarbital, B Braun Medical, Jaén, Spain)and 1.5 ml of 10 mM KCl solution through the ear vein. Thewhole brain, including the brainstem, was removed, the basilarartery was dissected free and in some cases a sample of cortexpyriformis was harvested. In those arteries to be used fordetermination of mRNA or protein expression the endotheliumwas removed by passing a stainless steel rod through the vessellumen. In selected animals the abdomen was open to harvestliver samples.

2.2. Vascular reactivity: isometric tension recording

The basilar artery was cut in four 3-mm long segments. Eachsegment was mounted in an organ bath by using tungsten wires89 µm in diameter. Two pins were introduced through thearterial lumen: one pin was fixed to a stationary support, whilethe other one was connected to a strain gauge (UniversalTransducing Cell UC3, Gould Statham, Oxnard, CA, USA).Isometric tension was conveniently amplified (OCTAL Bridge,ADInstruments, Castle Hill, Australia), digitized (PowerLab/8SP, ADInstruments), recorded and stored in an IBM® PCcompatible computer by means of the appropriate software(Chart 5, ADInstruments) for later analysis. Each organ bathcontained 5 ml of Ringer-Locke solution at 37 °C and bubbledwith a 95% O2 and 5% CO2 mixture to give a pH of 7.3–7.4.Previously determined optimal resting tension of 0.5 g wasapplied to the arterial segments, and they were allowed toequilibrate for 30–60 min before starting the experiments.

The contractile capacity of every arterial segment wasassessed by exposure to 50 mM KCl Ringer-Locke solution.Arteries contracting less than 0.5 g were discarded. Cumulativeconcentration–response curves to sildenafil (10 nM–0.1 mM)were obtained in arteries precontracted with 10 nM endothelin-1.In order to assess the involvement of K+ efflux in sildenafil-

140 J.B. Salom et al. / European Journal of Pharmacology 581 (2008) 138–147

induced relaxation, concentration–response curves to sildenafil(10 nM–0.1 mM) were obtained in arteries precontracted with50 mM KCl or in endothelin-1-precontracted arteries preincu-bated (30 min) with the non-selective K+ channel blockertetraethylammonium (3 mM). The involvement of specific K+

channels in the relaxant effect of sildenafil was assessed byobtaining concentration-response curves to sildenafil inendothelin-1-precontracted arteries preincubated (30 min) with1 mM 4-aminopyridine (inhibitor of voltage-dependent K+

channels), 1 μM glibenclamide (inhibitor of ATP-sensitive K+

channels) or 50 nM iberiotoxin (inhibitor of Ca2+-activated K+

channels). The ability of sildenafil to inhibit the contractioninduced by extracellular Ca2+ influx was checked by obtainingconcentration-response curves to CaCl2 (10 μM–10 mM) inCa2+-free, highly depolarizing (50 mMKCl) medium, in controlconditions or after preincubation with increasing concentrationsof sildenafil (0.1–100 μM). After exposure to the highestsildenafil concentration (100 μM), some arteries were washedout and a second concentration-response curve to CaCl2 wascarried out to check for reversal of sildenafil-induced inhibition.The involvement of L-type Ca2+ channels in the inhibitory effectof sildenafil was assessed in a double way: (1) concentration-response curves to CaCl2 were obtained in 100 μM sildenafilpreincubated arteries without or with the addition of increasingconcentrations of the L-type Ca2+ channel activator Bay K 8644(0.1 nM–1 μM), and (2) concentration–response curves to BayK 8644 (0.1 nM–10 μM) were obtained in slightly depolarizing(15 mM) medium in control conditions or after preincubationwith increasing concentrations of sildenafil (0.1–100 μM).For comparison with sildenafil, the inhibitory effects of zaprinast(1 or 100 μM) on CaCl2 and Bay K 8644 induced contractionswere assessed.

2.3. Expression of mRNA: RT–PCR

Total cellular RNA was extracted from rabbit basilar artery,brain and liver with TRIZOL Reagent (Invitrogen, Life

Table 1Oligonucleotides used for RT–PCR analysis

mRNA Primer sequences

Kv1.1 a UP: GCATCGACAACACCACGGTCDN: GGCGACTGAGGTCACTGTCAGAGGCTAA

Kv1.2 UP: GCCTTACCGGTCCCTGTCATAGDN: TTTGGACAGCTTGTCACTTGCAA

Kv1.3 UP: GCCCGTTCCTGTGATTGTTDN: GGAGTTGGGGTTATTGTTCGTG

Kv1.4 b UP: CACCGACAGAGCGGCTTTCCDN: CCTCATCCTCACGAAACTTGAGC

Kv1.5 a UP: AAGGGGCTGCAGATCCTGDN: TCACAAGTCGGTTTCCCGGCT

Kv1.6 a UP: GACCTGAAGGCAACGGACAATDN: CGATGTGGAGTTGGAAGGTAGC

β-actin UP: CGTACCACTGGCATCGTGATDN: GTGTTGGCGTACAGGTCTTTG

a Plane et al., 2005.b Sanchez et al., 2002.

Technologies, Barcelona, Spain), and contaminating genomicDNA was removed by incubation with DNase I AmplificationGrade (Invitrogen, Life Technologies, Barcelona, Spain).cDNA was synthesised from 250 ng of total RNA by usingM-MLV reverse transcriptase (Invitrogen, Barcelona, Spain)according with the manufacturer's instructions. cDNA wasdiluted at 1/10 and 3 µl were amplified with a rapid thermalcycler (LightCycler Instrument, Roche Diagnostics, Barcelona,Spain) in 15 µl of LightCycler FastStart DNA Master SYBRGreen I (Roche Molecular Biochemicals, Barcelona, Spain), 3–4 mM MgCl2 and 0.3 µM of each oligonucleotide. Primers forthe different Kv1 subunits were designed based on rabbitdatabase sequences or taken from previously published studiesas indicated (Table 1). In parallel, we analyzed the mRNAexpression of the ubiquitous β-actin as a positive control ofRT–PCR in the different rabbit tissues (Table 1). Afterdenaturing for 10 min at 95 °C, amplification was performedin 40 cycles of 15 s at 94 °C, 5 s at 58–62 °C and 10–30 s at72 °C, depending on primer Tm and amplicon length. Thepredicted sizes of the different PCR products were confirmed byagarose gel electrophoresis.

2.4. Expression of proteins: Western blot

Soluble proteins were extracted from rabbit basilar artery,brain and liver by mincing and homogenizing tissues in ice-coldlysis buffer containing HEPES (0.5 mM), MgCl2 (1 mM),EGTA (2 mM), DTT (1 mM), sucrose (10%) and Nonidet™(0.2%) in the presence of protease inhibitor cocktail (1%,Sigma-Aldrich, Madrid, Spain) and phosphatase inhibitorcocktail 1 (1%, Sigma, Madrid, Spain). Protein concentrationwas determined by BCA protein assay kit (Sigma-Aldrich,Madrid, Spain). Aliquots of proteins (30 µg) were dissolved inLDS sample buffer (Invitrogen, Barcelona, Spain), loaded onNuPAGE 4–12% Bis–Tris gels, separated by electrophoresisand transferred to nitrocellulose membranes for immunolabel-ing. Non-specific sites were blocked with 5% non-fat milk for

Size (bp) GenBank no.

Species

720 NM_000217GT Human

104 NM_001082722Rabbit

271 U38240Rabbit

535 AF493544Rabbit

591 NM_001082036Rabbit

92 XM_001067110Rat

431 NM_001101Human

Fig. 1. Effects of K+ gradient reduction and non-selective blockade of K+

channels on sildenafil-induced relaxation of the rabbit basilar artery.Concentration-dependent relaxations to sildenafil in endothelin-1-precontractedarteries (control), KCl-precontracted arteries, and endothelin-1-precontractedarteries preincubated with tetraethylamonium. Data are means±S.E.M.


60 min, and the blots were then incubated with polyclonalantibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA)against Kv1.1 (1:100), Kv1.2 (1:500) or Kv1.4 (1:100)overnight at 4 °C. Secondary antibody was horseradishperoxidase (HRP)-conjugated donkey anti-goat IgG (1:5000;Santa Cruz Biotechnology). Membranes were reprobed withanti-β-actin mouse monoclonal antibody (1:10,000; Sigma-Aldrich, Madrid, Spain) as a loading control. The proteinexpression was detected by enhanced chemiluminescence.

2.5. Data analysis

The relaxant responses elicited by sildenafil were expressedas percentage of the active tone achieved with endothelin-1 orKCl. The contractile responses to CaCl2 or Bay K 8644 wereexpressed as percentage of previous response to KCl. Maximumeffect (Emax) and half-maximal effective drug concentration(EC50) were calculated for each concentration-response curve.The pEC50 was calculated as the negative logarithm to base 10of the EC50 for statistical analysis. One-way ANOVA followedby Dunnett multiple comparison test was used to compare theeffects of sildenafil between control and KCl, tetraethylammo-nium, 4-aminopyridine, glibenclamide or iberiotoxin treatedgroups. One-way ANOVA followed by Student–Newman–Keuls multiple comparisons test was used to compare the effectsof CaCl2 or Bay K 8644 between control groups and therespective groups treated with sildenafil, sildenafil + Bay K8644, or zaprinast. Pb0.05 was considered significant.

2.6. Drugs and solutions

Sildenafil, zaprinast and glibenclamide were dissolved anddiluted in dimethyl sulfoxide (DMSO). Bay K 8644 wasdissolved and diluted in ethanol. Total DMSO or ethanol addedto the organ bath for preincubations or concentration-responsecurves was, at the most, 0.59% v/v and did not affect arterialtone. Endothelin-1 was dissolved in 5% aqueous acetic acid anddiluted in 10 mM PBS with 0.05% bovine serum albumin.CaCl2 and iberiotoxin were dissolved and diluted in salinesolution (0.9% NaCl). Tetraethylammonium chloride and 4-aminopyridine were dissolved in distilled water. All drugs werefrom Sigma-Aldrich-Fluka (Madrid, Spain), except for silde-nafil (kindly provided by Pfizer Pharmaceuticals, SandwichKent, U.K.) and Bay K 8644 (S)-(−)-enantiomer ((4S)-1,4-Dihydro-2,6-dimethyl-5-nitro-4-[2-trifluoromethyl)phenyl]-3-pyridinecarboxylic acid methyl ester; Tocris Cookson, Bristol,U.K.). The Ringer-Locke solution had the following composi-tion (mM): NaCl 120, KCl 5.4, CaCl2 2.2, MgCl2 1.0, NaHCO3

25 and glucose 5.6. In KCl-depolarizing solutions, NaCl wasreplaced by an equimolar amount of KCl. In Ca2+ free medium,CaCl2 was omitted from the composition, and 1 mM EGTAwasadded.

3. Results

Active tones achieved with 10 nM endothelin-1 (1299±60 mg;n = 54) and 50 mM KCl (1141±78 mg; n = 16) in the rabbit

basilar artery were not significantly different. Sildenafil (10 nM–0.1 mM) induced concentration-dependent relaxation of 10 nMendothelin-1-precontracted arteries. This relaxation was signifi-cantly inhibited in arteries precontracted with 50 KCl and inarteries preincubated with 3 mM tretraethylammonium (non-selective K+ channel blocker) (Fig. 1). With regard to selectiveinhibitors of K+ channels, sildenafil-induced relaxation wassignificantly inhibited in arteries preincubated with 1 mM 4-aminopyridine (inhibitor of voltage-dependent K+ channels). Incontrast, it was not modified by preincubation with 1 μMglibenclamide (inhibitor of ATP-sensitive K+ channels) or 50 nMiberiotoxin (inhibitor of Ca2+-activated K+ channels), as shown inFig. 2, except for a slight but significant reduction in the EC50 ofiberiotoxin treated arteries (Table 2).

Expression of Kv1 mRNA (encoding voltage-dependent K+

channels) by endothelium-rubbed rabbit basilar artery wasexamined by RT–PCR using subunit-specific primers. Productsof appropriate predicted size were obtained for Kv1.2, Kv1.3,Kv1.4, Kv1.5 and Kv1.6 (Fig. 3). However, a single major bandof the predicted product size for Kv1.1 was not obtained.Instead, multiple bands with no consistent reproducibilityappeared (not shown). Brain and liver tissues yielded appro-priate positive and negative results, respectively. A product ofappropriate predicted size was obtained for the ubiquitous β-actin as a positive control of RT–PCR in the three rabbit tissues(Fig. 3).

Western blot analysis was performed to determine theexpression of Kv1 proteins (forming voltage-dependent K+

channels) in endothelium-rubbed rabbit basilar artery. Basilarartery expressed proteins of Kv1.1, Kv1.2 and Kv1.4.Appropriate positive and negative results were obtained withbrain and liver tissues, respectively. Detection of β-actinexpression indicated appropriate loading of proteins from thethree tissues (Fig. 4).

Fig. 2. Effects selective blockade of K+ channels on sildenafil-inducedrelaxation of the rabbit basilar artery. Concentration-dependent relaxations tosildenafil in endothelin-1-precontracted arteries (control), and endothelin-1-precontracted arteries preincubated with 4-aminopyridine (voltage-dependentK+ channel blocker), glibenclamide (ATP-sensitive K+ channel blocker) oriberiotoxin (Ca2+-activated K+ channel blocker). Data are means±S.E.M.


On the other hand, CaCl2 (10 μM–10 mM) inducedconcentration-dependent contraction of the basilar artery inCa2+-free, highly depolarizing (50 mM KCl) medium (Fig. 5A).Preincubation with sildenafil (0.1–100 µM) produced concen-tration-dependent inhibition of the concentration-responsecurve to CaCl2, which was completely abolished by the highestsildenafil concentration (Fig. 6). Table 2 summarizes the EC50

and Emax values for the CaCl2 curve in control conditionsand after increasing sildenafil concentrations. However, asshown in Fig. 5B, sildenafil washout led to significant but notcomplete recovery of CaCl2-induced contraction (Table 2). Bycontrast, preincubation with zaprinast (1 or 100 μM) producedsignificant inhibition of the concentration-response curve toCaCl2 only at the high concentration (Fig. 7). Inhibitory effectof zaprinast was significantly lower than inhibition induced bysildenafil at the same concentrations (Table 2).

Preincubation with the L-type Ca2+ channel activator Bay K8644 (0.1 nM–1 μM) induced concentration-dependent poten-tiation of concentration–response curves to CaCl2 (10 μM–10 mM) obtained in the presence of 100 μM sildenafil (Fig. 8).As shown in Table 2, recovery of CaCl2-induced contractiontowards control values was no complete even at the highestconcentration of Bay K 8644 used.

On the other hand, Bay K 8644 (0.1 nM–1 μM) inducedconcentration-dependent contraction of the basilar artery inslightly depolarizing (15 mM) medium (Fig. 9). Slightdepolarization induced almost negligible contraction (5±1%of 50 mM KCl-induced contraction, n = 79) but it wasnecessary to start subsequent Bay K 8644-induced contraction.Higher Bay K 8644 concentrations (3 and 10 μM) produced nofurther contraction or even relaxation of previous contraction.Preincubation with sildenafil (0.1–100 µM) produced concen-tration-dependent inhibition of the concentration-responsecurve to Bay K 8644 (Fig. 9). Table 2 summarizes the EC50

and Emax values for the Bay K 8644 curve in control conditionsand after increasing sildenafil concentrations. Preincubationwith zaprinast (1 or 100 μM) produced concentration-dependentinhibition of the concentration-response curve to Bay K 8644(Fig. 10). Inhibitory effect of zaprinast was not significantlydifferent to inhibition induced by sildenafil at the sameconcentrations (Table 2).

4. Discussion

Cellular mechanisms activated by endothelin-1 and high-K+

medium to induce vasoconstriction are different. However,endothelin-1 also induces gradual membrane depolarization inbasilar artery myocytes (Salter and Kozlowski, 1998). Ourresults show that sildenafil induces relaxation of rabbit basilarartery which is inhibited by strong depolarization with high-K+

medium. Activity of membrane K+ channels in blood vesselsmooth muscle contributes to determination of membranepotential and vascular tone. The electrochemical gradient for K+

ions is such that opening of K+ channels results in diffusion ofthis cation out of the cells, membrane hyperpolarization andvascular relaxation (Jackson, 2000). Decreasing the K+ gradientacross the cell membrane by raising extracellular K+ reduces K+

efflux, which suggests that inhibition of sildenafil-inducedrelaxation takes place because sildenafil relaxes the rabbitbasilar artery by increasing K+ conductance. It has beenpreviously reported that sildenafil, by inhibiting phosphodies-terase-5 activity, enhances the NO–cGMP pathway in cerebralarteries (Kruuse et al., 2005; Salom et al., 2006), and themolecular mechanism for cGMP-mediated smooth musclerelaxation involves the activation of K+ efflux (Carvajal et al.,2000).

Vascular smooth muscle can express at least 4 types of K+

channels (Jackson, 2000). In our study, sildenafil-inducedrelaxation of the rabbit basilar artery was inhibited bytetraethylammonium, a non-selective K+ channel blocker.This result pharmacologically supports the involvement of K+

efflux activation in the response to sildenafil and confirmsindirect evidence obtained by inhibition with high-K+ medium.In order to identify the type (or types) of K+ channels involvedin the response to sildenafil we carried out experiments withselective inhibitors. Sildenafil-induced relaxation was inhibitedby 4-aminopyridine, which indicates that voltage-dependent K+

channels (Kv) are involved. By contrast, no inhibition ofsildenafil-induced relaxation was obtained after preincubationof the rabbit basilar artery with glibenclamide or iberiotoxin.This indicates that neither ATP-sensitive K+ channels (KATP)nor Ca2+-activated K+ channels (KCa) are involved in therelaxant effects of sildenafil. In line with our results, KATP

channels mediate relaxation to PGE1 and cAMP-elevatingagents, but not relaxation to cGMP-elevating agents such assildenafil in penile resistance arteries (Ruiz Rubio et al., 2004).However, the lack of role for KCa channels in the relaxation tosildenafil is quite surprising taking into account that we haveused in our experiments an iberiotoxin concentration similar tothose able to inhibit KCa channels and subsequent relaxation innon-cerebral (Prieto et al., 2006) and cerebral (Yu et al., 2002)

Table 2Pharmacological parameters for the concentration–response curves to sildenafil, CaCl2 and Bay K 8644 in the rabbit basilar artery

Precontraction Pretreatment EC50 Emax n

Sildenafil ET-1 – 0.74 (0.49–1.13) µM 101±1 54KCl – 8.51 (5.82–12.45) µMa 97±2 16ET-1 Tetraethylammonium, 3 mM 10.72 (4.79–23.99) µMa 84±8 a 17ET-1 4-Aminopyridine, 1 mM 14.45 (7.55–27.67) µMa 59±2 a 18ET-1 Glibenclamide, 1 µM 0.36 (0.13–0.98) µM 99±1 12ET-1 Iberiotoxin, 50 nM 0.22 (0.07–0.65) µMb 100±2 13

CaCl2 – – 0.20 (0.16–0.26) mM 111±3 27– Sildenafil, 0.1 μM 0.25 (0.17–0.36) mM 102±5 8– Sildenafil, 1 μM 0.30 (0.22–0.41) mM 83±6 c, d 9– Sildenafil, 10 μM 0.40 (0.24–0.65) mM 39±7 c, e 9– Sildenafil, 100 μM 0.68 (0.24–1.89) mM c 3±1 c, f 15– Sildenafil washout 1.05 (0.53–2.07) mM c 77±7 c, g 7– Sildenafil, 100 μM+Bay K 8644, 0.1 nM – 1±1 8– Sildenafil, 100 μM+Bay K 8644, 10 nM 0.78 (0.43–1.39) mM 41±3 g, h 13– Sildenafil, 100 μM+Bay K 8644, 1 µM 0.13 (0.09–0.19) mMg, i 77±6 c, g, i 13– Zaprinast, 1 µM 0.23 (0.13–0.40) mM 109±5 e 12– Zaprinast, 100 µM 0.12 (0.09–0.16) mM g 84±4 c, g 15

Bay K 8644 – – 4.17 (1.64–10.59) nM 122±4 16– Sildenafil, 0.1 μM 4.90 (1.40–17.14) nM 104±9 j 13– Sildenafil, 1 μM 12.02 (3.56–45.60) nM 74±6 k, l 10– Sildenafil, 10 μM 21.88 (11.89–40.27) nM 56±11 k 10– Sildenafil, 100 μM 0.18 (0.10–0.34) µMk 29±5 k, m 9– Zaprinast, 1 μM 5.37 (2.96–9.73) nM 84±6 k 10– Zaprinast, 100 μM 37.15 (20.00–69.02) nMk 37±8 k, n 11

Values averaged from ‘n’ arteries. EC50 (mean with 95% confidence limits) is half-maximal effective concentration. Emax (mean±S.E.M.) is maximum relaxant effectexpressed as percent of active tone in precontracted arteries, or maximum contractile effect expressed as percent of previous contraction to 50 mM KCl. ET-1,endothelin-1.a Significantly different from sildenafil in ET-1-precontracted, untreated arteries, Pb0.01.b Significantly different from sildenafil in ET-1-precontracted, untreated arteries, Pb0.05.c Significantly different from CaCl2 in untreated arteries, Pb0.01.d Significantly different from CaCl2 in sildenafil 0.1 μM treated arteries, Pb0.05.e Significantly different from CaCl2 in sildenafil 1 μM treated arteries, Pb0.01.f Significantly different from CaCl2 in sildenafil 10 μM treated arteries, Pb0.01.g Significantly different from CaCl2 in sildenafil 100 μM treated arteries, Pb0.01.h Significantly different from CaCl2 in sildenafil 100 μM+BayK8644 0.1 nM treated arteries, Pb0.01.i Significantly different from CaCl2 in sildenafil 100 μM+BayK8644 10 nM treated arteries, Pb0.01.j Significantly different from Bay K 8644 in untreated arteries, Pb0.05.k Significantly different from Bay K 8644 in untreated arteries, Pb0.01.l Significantly different from Bay K 8644 in sildenafil 0.1 μM treated arteries, Pb0.01.m Significantly different from Bay K 8644 in sildenafil 10 μM treated arteries, Pb0.05.n Significantly different from Bay K 8644 in zaprinast 1 μM treated arteries, Pb0.01.


arteries. Therefore, the pharmacological profile of K+ channelsinvolved in the relaxant response to sildenafil in the rabbitbasilar artery resembles those previously reported, for example,in histamine-induced, H2 receptor-mediated relaxation of ratcerebral arteries (Jarajapu et al., 2006) or in endothelin-1-induced, ETB receptor-mediated relaxation of the rat carotidartery (Tirapelli et al., 2005).

Aminopyridine-sensitive, voltage-dependent K+ currentshave been characterized in smooth muscle cells from rabbitbasilar artery (Robertson and Nelson, 1994) and Kv channelsregulate membrane potential and diameter in rabbit myogenicmiddle cerebral artery (Knot and Nelson, 1995). At themolecular level, K+ channels comprise primarily a tetramericarrangement of structural subunits, each one being a separatepolypeptide, and all of which are members of a diverse proteinfamily (Coetzee et al., 1999). Kv channels are composed ofpore-forming Kvα and modulatory Kvβ subunits. Kvα proteinswith six transmembrane segments are encoded by nine related

families, Kv1 to Kv9, within the superfamily of Kv channelgenes. Kv1 to Kv4 proteins assemble to form homotetramericchannels, or coassemble with members of the same family toproduce heterotetrameric channel complexes. Additional func-tional diversity and appropriate trafficking to the membrane areobtained by association of the Kvα subunits with Kvβ subunits(Thorneloe et al., 2001). Kv1 subunits are known to beexpressed in blood vessels. However, there are indications thatKv1 subunits are differentially expressed depending on the sizeand type of blood vessel and depending on the vascular bed.With regard to cerebral vessels, Kv1 expression profiles havebeen reported in rabbit arterioles (Cheong et al., 2001a), mousearterioles (Cheong et al., 2001b) and rat middle cerebral artery(Albarwani et al., 2003; Chen et al., 2006). In this study we haveshown expression of mRNA for Kv1.2, Kv1.3, Kv1.4, Kv1.5and Kv1.6 in the rabbit basilar artery. Minor discrepancies withthe predicted amplicon size can likely arise from sequencedifferences between rabbit and the species used for primer

Fig. 3. Detection of amplified products corresponding to Kv1 mRNA in rabbitbasilar artery. RT–PCR screening detected mRNA encoding Kv1.2, Kv1.3,Kv1.4, Kv1.5 and Kv1.6 in basilar artery (BA) and brain (positive control), butnot in liver (negative control). Expected product sizes are given in base pairs(bp) and can be checked against the ladder lane spanning from 100 bp to 700 bpin one hundred steps. Expression of the ubiquitous β-actin is a positive controlof RT–PCR in the different rabbit tissues. Blank lane is a control PCR withprimers but without cDNA.

Fig. 5. Typical records showing reversible inhibition by sildenafil ofvasoconstriction to CaCl2 in rabbit basilar artery. (A) Two consecutiveconcentration-dependent contractions to CaCl2 show reproducibility ofresponses; (B) abolished response to CaCl2 in the presence of 100 µMsildenafil, followed by concentration-dependent contraction to CaCl2 aftersildenafil washout. Contractions to CaCl2 were obtained in Ca2+-free, highlydepolarizing (50 mMKCl) medium and are compared with previous response to50 mM KCl in the same artery.


design (e.g. rat Kv1.6). On the other hand, we have foundprotein expression of Kv1.1, Kv1.2 and Kv1.4. Expression ofKv1.1 protein obviously implies expression of the correspond-

Fig. 4. Detection of Kv1 proteins in rabbit basilar artery. Representative Westernblots in basilar artery (BA), brain (positive control) and liver (negative control)probed with antibodies to Kv1.1, Kv1.2, Kv1.4 and β-actin (loading control).Protein expression of Kv1.1, Kv1.2 and Kv1.4 was detected in basilar artery andbrain but not in liver.

ing mRNA, which could not be shown because we tried toanalyze the expression of rabbit Kv1.1 by using previouslypublished human primers that did not properly work in this case

Fig. 6. Effects of sildenafil on CaCl2-induced contraction of the rabbit basilarartery. Concentration-dependent contractions to CaCl2 in control conditionsand during incubation with increasing sildenafil concentrations. Data aremeans±S.E.M.

Fig. 7. Effects of zaprinast on CaCl2-induced contraction of the rabbit basilarartery. Concentration-dependent contractions to CaCl2 in control conditionsand during incubation with low and high zaprinast concentrations. Data aremeans±S.E.M.

Fig. 9. Effects of sildenafil on Bay K 8644-induced contraction of the rabbitbasilar artery. Concentration-dependent contractions to Bay K 8644 in controlconditions and during incubation with increasing sildenafil concentrations. Dataare means±S.E.M.


(Plane et al., 2005). On the contrary, expression of proteins forKv1.3, Kv1.5 and Kv1.6 subunits can not be taken for granted,but we could not determine them because appropriate antibodieswere lacking. For example, rat middle cerebral artery expressmRNA encoding for the six Kv1 subunits but only Kv1.2 andKv1.5 proteins are detected (Albarwani et al., 2003). Finally,assessing the particular assembly/coassembly of these Kv1subunits to form the tetrameric Kv channels involved insildenafil induced relaxation of the rabbit basilar artery wouldneed further research.

We have also shown that sildenafil produces concentration-dependent, reversible inhibition of the concentration-responsecurve to CaCl2. This indicates that sildenafil is able tocounteract the contractile effect induced by extracellular Ca2+

entry to smooth muscle in the basilar artery. This could be due tothe ability of sildenafil to increase cGMP in smooth muscle.

Fig. 8. Effects of selective L-type Ca2+ channel activation on sildenafil-inducedinhibition of contraction to CaCl2 in the rabbit basilar artery. Concentration-dependent contractions to CaCl2 in control conditions, during incubation withthe highest sildenafil concentration, and during incubation with sildenafil plusincreasing Bay K 8644 concentrations. Data are means±S.E.M.

cGMP effects on vascular tone are mediated mainly but notalways by cGMP-dependent protein kinase (PKG) activation. Itinvolves several molecular events leading to a reduction inintracellular Ca2+ concentration and/or a decrease in thesensitivity of the contractile system to Ca2+ (Carvajal et al.,2000). For example, the relaxant effect of sildenafil in the ratpulmonary artery is mainly because of a cGMP-dependentmechanism that alters Ca2+ signalling by inhibiting the releaseof intracellular Ca2+ stores via the IP3 pathway (Pauvert et al.,2003). Alternatively, sildenafil can directly inhibit Ca2+ influxto vascular smooth muscle by binding and blocking L-type Ca2+

channels (Mochida et al., 2002), and this could be the case inour experiments with the rabbit basilar artery.

On one hand, we have shown that preincubation with thedihydropyridine activator of L-type Ca2+ channels Bay K 8644reverses in a concentration-dependent way the inhibitory effectsof sildenafil on CaCl2-induced contractions. On the other hand,

Fig. 10. Effects of zaprinast on Bay K 8644-induced contraction of the rabbitbasilar artery. Concentration-dependent contractions to Bay K 8644 in controlconditions and during incubation with low and high zaprinast concentrations.Data are means±S.E.M.


concentration-dependent contractions to Bay K 8644 areinhibited in a concentration-dependent way by sildenafil.Thus, although phosphodiesterase-5 inhibitory effects ofsildenafil can in turn affect Ca2+ signalling, these results givefurther support to a direct antagonistic effect of sildenafil on L-type Ca2+ channels. Therefore, both direct and indirect effectsof sildenafil on Ca2+ signalling contribute to the anticonstrictor/relaxant effects of sildenafil in the rabbit basilar artery. SimilarCa2+ channel antagonistic, phosphodiesterase-5-independenteffects have been reported in sildenafil-induced relaxation of theisolated rat aorta (Mochida et al., 2002). Quite interestingly,RS93522 is a dihydropyridine Ca2+ channel blocker which alsohas phosphodiesterase inhibitory effects (Marsh et al., 1988).

Zaprinast is another selective phosphodiesterase-5 inhibitorwith relaxant effects in cerebral (Kruuse et al., 2001; Salom etal., 2006) and non-cerebral (Pauvert et al., 2003) arteries. Wehave previously shown that zaprinast relaxes endothelin-1precontracted rabbit basilar artery, although with lower potencythan sildenafil (Salom et al., 2006). In the present study, thepattern for the inhibitory effects of zaprinast on CaCl2 and BayK 8644-induced contractions in the rabbit basilar artery wasdifferent from that for sildenafil. While the highest concentra-tion of sildenafil produced complete abolition of CaCl2-inducedcontraction, equimolar zaprinast only induced slight inhibitionof CaCl2-induced contraction. By contrast, both sildenafil andzaprinast inhibited Bay K 8644-induced contractions to thesame extent. Since vascular smooth muscle can expressdifferent types of voltage-operated, store-operated and stretch-activated Ca2+ channels (Jackson, 2000), may be sildenafil andzaprinast target different Ca2+ channel populations in the rabbitbasilar artery. Contraction elicited by Bay K 8644 is mediatedby selective activation of dihydropyridine-sensitive L-type Ca2+

channels (Su et al., 1984) while contraction elicited by CaCl2can be mediated by Ca2+ entry through L-type and non-L-typeCa2+ channels. Broader inhibitory effects of sildenafil on bothCaCl2- and Bay K 8644-induced contractions suggests a non-selective blocking effect on Ca2+ channels, while specificinhibition of Bay K 8644-induced contraction by zaprinastsuggests a selective blocking effect on dihydropyridine-sensitive L-type Ca2+ channels. Partial recovery of sildenafil-inhibited, CaCl2-induced contractions by Bay K 8644 alsosupports our proposal since the dihydropyridine activator of L-type Ca2+ channels could not counteract inhibition of non-L-type Ca2+ channels induced by sildenafil.

Since we have not studied binding of sildenafil to Ca2+

channels in the rabbit basilar artery and subsequent changes in[Ca2+]i of smooth muscle, other Ca2+-related mechanisms forsildenafil-induced relaxation can not be discarded. For example,endothelin-1 can induce vasoconstriction at least in part byincreasing the Ca2+ sensitivity of the contractile apparatus viathe RhoA/Rho kinase pathway. In the rabbit basilar artery,endothelin-1 activates RhoA, which leads to activation of Rhokinase, inactivation of MLC phosphatase and increase in MLCphosphorilation (Miao et al., 2002). On the other hand, RhoA-mediated Ca2+ sensitization and contraction of vascular smoothmuscle cells are inhibited by the cGMP/PKG pathway viaphosphorilation of RhoA (Sauzeau et al., 2000). Moreover,

vasodilator response to NO in rat middle cerebral artery ismediated in part by cGMP-induced decrease in the contractileresponse to elevations in [Ca2+]i (Yu et al., 2002). Therefore,sildenafil-induced relaxation of endothelin-induced tone andinhibition of CaCl2-induced vasoconstriction in our study couldbe due at least in part to cGMP-induced decrease in thesensitivity of the contractile apparatus to Ca2+.

In conclusion, sildenafil relaxes the rabbit basilar artery byincreasing K+ efflux through Kv channels, which in turn mayaffect Ca2+ signalling. Expression of Kv1 subunits involved inthis pharmacological effect occurs at the messenger and, in somecases, at the protein level. In addition to this phosphodiesterase-5-related effect, sildenafil and zaprinast inhibit cerebroarterialvasoconstriction at least in part by directly blocking L-type Ca2+

channels, although a decrease in the sensitivity of the contractileapparatus to Ca2+ can not be discarded.

Acknowledgements

This work was partially supported by grant RETICS-RD06/0026/0006 from ‘Ministerio de Salud- Instituto de Salud CarlosIII’. M. C. Burguete and M. Castelló-Ruiz are recipients of PhDfellowships from ‘Ministerio de Ciencia y Tecnología’ (grantSAF01/0398) and ‘Fondo de Investigación Sanitaria’ (grantsC03/06 and PI030323). The authors thank María C. Máñez andMaría C. Tirados for their technical assistance, and PfizerPharmaceuticals for kindly providing sildenafil.

References

Albarwani, S., Nemetz, L.T., Madden, J.A., Tobin, A.A., England, S.K., Pratt,P.F., Rusch, N.J., 2003. Voltage-gated K+ channels in rat small cerebralarteries: molecular identity of the functional channels. J. Physiol. 551,751–763.

Arnavaz, A., Aurich, A., Weissenborn, K., Hartmann, U., Emrich, H.M.,Schneider, U., 2003. Effect of sildenafil (Viagra) on cerebral blood flowvelocity: a pilot study. Psychiatry Res. 122, 207–209.

Beltowski, J., Jamroz-Wisniewska, A., Borkowska, E., Marciniak, A., 2006.Phosphodiesterase 5 inhibitor ameliorates renal resistance to atrial natriureticpeptide associated with obesity and hyperleptinemia. Arch. Med. Res. 37,307–315.

Bender, A.T., Beavo, J.A., 2006. Cyclic nucleotide phosphodiesterases:molecular regulation to clinical use. Pharmacol. Rev. 58, 488–520.

Briganti, A., Salonia, A., Gallina, A., Sacca, A., Montorsi, P., Rigatti, P.,Montorsi, F., 2005. Drug Insight: oral phosphodiesterase type 5 inhibitorsfor erectile dysfunction. Nat. Clin. Pract. Urol. 2, 239–247.

Burnett, A.L., Bivalacqua, T.J., Champion, H.C., Musicki, B., 2006. Feasibilityof the use of phosphodiesterase type 5 inhibitors in a pharmacologicprevention program for recurrent priapism. J. Sex Med. 3, 1077–1084.

Carvajal, J.A., Germain, A.M., Huidobro-Toro, J.P., Weiner, C.P., 2000.Molecular mechanism of cGMP-mediated smooth muscle relaxation. J. Cell.Physiol. 184, 409–420.

Chen, T.T., Luykenaar, K.D., Walsh, E.J., Walsh, M.P., Cole, W.C., 2006. Keyrole of Kv1 channels in vasoregulation. Circ. Res. 99, 53–60.

Cheong, A., Dedman, A.M., Beech, D.J., 2001a. Expression and function ofnative potassium channel (Kvα1) subunits in terminal arterioles of rabbit.J. Physiol. 534, 691–700.

Cheong, A., Dedman, A.M., Xu, S.Z., Beech, D.J., 2001b. Kvα1 channels inmurine arterioles: differential cellular expression and regulation of diameter.Am. J. Physiol. 281, H1057–H1065.

Coetzee, W.A., Amarillo, Y., Chiu, J., Chow, A., Lau, D., McCormack, T.,Moreno, H., Nadal, M.S., Ozaita, A., Pountney, D., Saganich, M.,


Vega-Saenz de Miera, E., Rudy, B., 1999. Molecular diversity of K+

channels. Ann. N.Y. Acad. Sci. 868, 233–285.Inoha, S., Inamura, T., Ikezaki, K., Nakamizo, A., Amano, T., Fukui, M., 2002.

Type V phosphodiesterase expression in cerebral arteries with vasospasmafter subarachnoid hemorrhage in a canine model. Neurol. Res. 24, 607–612.

Jackson, W.F., 2000. Ion channels and vascular tone. Hypertension 35,173–178.

Jarajapu, Y.P.R., Oomen, C., Uteshev, V.V., Knot, H.J., 2006. Histaminedecreases myogenic tone in rat cerebral arteries by H2-receptor-mediated KV

channel activation, independent of endothelium and cyclic AMP. Eur. J.Pharmacol. 547, 116–124.

Knot, H.J., Nelson, M.T., 1995. Regulation of membrane potential and diameterby voltage-dependent K+ channels in rabbit myogenic cerebral arteries. Am.J. Physiol. 269, H348–H355.

Kruuse, C., Rybalkin, S.D., Khurana, T.S., Jansen-Olesen, I., Olesen, J.,Edvinsson, L., 2001. The role of cGMP hydrolysing phosphodiesterases 1and 5 in cerebral artery dilatation. Eur. J. Pharmacol. 420, 55–65.

Kruuse, C., Thomsen, L.L., Jacobsen, T.B., Olesen, J., 2002. The phospho-diesterase 5 inhibitor sildenafil has no effect on cerebral blood flow or bloodvelocity, but nevertheless induces headache in healthy subjects. J. Cereb.Blood Flow Metab. 22, 1124–1131.

Kruuse, C., Thomsen, L.L., Birk, S., Olesen, J., 2003. Migraine can be inducedby sildenafil without changes in middle cerebral artery diameter. Brain 126,241–247.

Kruuse, C., Khurana, T.S., Rybalkin, S.D., Birk, S., Engel, U., Edvinsson, L.,Olesen, J., 2005. Phosphodiesterase 5 and effects of sildenafil on cerebralarteries of man and guinea pig. Eur. J. Pharmacol. 521, 105–114.

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 bycyclic GMP. Pharmacol. Rev. 52, 375–414.

Marsh, J.D., Dionne, M.A., Chiu, M., Smith, T.W., 1988. A dihydropyridinecalcium channel blocker with phosphodiesterase inhibitory activity: effectson cultured vascular smooth muscle and cultured heart cells. J. Mol. Cell.Cardiol. 20, 1141–1150.

Miao, L., Dai, Y., Zhang, J., 2002. Mechanism of RhoA/Rho kinase activation inendothelin-1-induced contraction in rabbit basilar artery. Am. J. Physiol.283, H983–H989.

Mochida, H., Inoue, H., Takagi, M., Noto, T., Yano, K., Kikkawa, K., 2002.Sildenafil and T-1032, phosphodiesterase type 5 inhibitors, showed adifferent vasorelaxant property in the isolated rat aorta. Eur. J. Pharmacol.440, 45–52.

Patel, M.D., Katz, S.D., 2005. Phosphodiesterase 5 inhibition in chronic heartfailure and pulmonary hypertension. Am. J. Cardiol. 96, 47M–51M.

Pauvert, O., Lugnier, C., Keravis, T., Marthan, R., Rousseau, E., Savineau, J.P.,2003. Effect of sildenafil on cyclic nucleotide phosphodiesterase activity,vascular tone and calcium signaling in rat pulmonary artery. Br. J.Pharmacol. 139, 513–522.

Plane, F., Johnson, R., Kerr, P., Wiehler, W., Thorneloe, K., Ishii, K., Chen, T.,Cole, W., 2005. Heteromultimeric Kv1 channels contribute to myogeniccontrol of arterial diameter. Circ. Res. 96, 216–224.

Prieto, D., Rivera, L., Benedito, S., Recio, P., Villalba, N., Hernández, M.,García-Sacristán, A., 2006. Ca2+-activated K+ (KCa) channels are involved inthe relaxations elicited by sildenafil in penile resistance arteries. Eur. J.Pharmacol. 531, 232–237.

Robertson, B.E., Nelson, M.T., 1994. Aminopyridine inhibition and voltagedependence of K+ currents in smooth muscle cells from cerebral arteries.Am. J. Physiol. 267, C1589–C1597.

Ruiz Rubio, J.L., Hernández, M., Rivera de los Arcos, L., Benedito, S.,Recio, P., García, P., García-Sacristán, A., Prieto, D., 2004. Role of ATP-sensitive K+ channels in relaxation of penile resistance arteries. Urology63, 800–805.

Salom, J.B., Burguete, M.C., Perez-Asensio, F.J., Castello-Ruiz, M., Torregrosa,G., Alborch, E., 2006. Relaxant effect of sildenafil in the rabbit basilarartery. Vasc. Pharmacol. 44, 10–16.

Salter, K.J., Kozlowski, R.Z., 1998. Differential electrophysiological actions ofendothelin-1 on Cl− and K+ currents in myocytes isolated from aorta, basilarand pulmonary artery. J. Pharmacol. Exp. Ther. 284, 1122–1131.

Sanchez, D., Lopez-Lopez, J.R., Perez-Garcia, M.T., Sanz-Alfayate, G.,Obeso, A., Ganfornina, M.D., Gonzalez, C., 2002. Molecular identificationof Kvα subunits that contribute to the oxygen-sensitive K+ current ofchemoreceptor cells of the rabbit carotid body. J. Physiol. 542, 369–382.

Sauzeau, V., Le Jeune, H., Cario-Toumaniantz, C., Smolenski, A., Lohmann,S.M., Bertoglio, J., Chardin, P., Pacaud, P., Loirand, G., 2000. CyclicGMP-dependent protein kinase signaling pathway inhibits RhoA-inducedCa2+ sensitization of contraction in vascular smooth muscle. J. Biol. Chem.275, 21722–21729.

Sesti, C., Florio, V., Johnson, E.G., Kloner, R.A., 2007. The phosphodiesterase-5 inhibitor tadalafil reduces myocardial infarct size. Int. J. Impot. Res. 19,55–61.

Su, C.M., Swamy, V.C., Triggle, D.J., 1984. Calcium channel activation invascular smooth muscle by BAY K 8644. Can. J. Physiol. Pharm. 62,1401–1410.

Thorneloe, K.S., Chen, T.T., Kerr, P.M., Grier, E.F., Horowitz, B., Cole, W.C.,Walsh, M.P., 2001. Molecular composition of 4-aminopyridine-sensitivevoltage-gated K+ channels of vascular smooth muscle. Circ. Res. 89,1030–1037.

Tinel, H., Stelte-Ludwig, B., Hutter, J., Sandner, P., 2006. Pre-clinical evidencefor the use of phosphodiesterase-5 inhibitors for treating benign prostatichyperplasia and lower urinary tract symptoms. BJU Int. 98, 1259–1263.

Tirapelli, C.R., Casolari, D.A., Yogi, A., Montezano, A.C., Tostes, R.C.,Legros, E., D'Orleans-Juste, P., de Oliveira, A.M., 2005. Functionalcharacterization and expression of endothelin receptors in rat carotidartery: involvement of nitric oxide, a vasodilator prostanoid and theopening of K+ channels in ETB-induced relaxation. Br. J. Pharmacol. 146,903–912.

Toward, T.J., Smith, N., Broadley, K.J., 2004. Effect of phosphodiesterase-5inhibitor, sildenafil (Viagra), in animal models of airways disease. Am. J.Respir. Crit. Care Med. 169, 227–234.

Wang, W.F., Minhas, S., Ralph, D.J., 2006. Phosphodiesterase 5 inhibitors in thetreatment of premature ejaculation. Int. J. Androl. 29, 503–509.

Yu, M., Sun, C.W., Maier, K.G., Harder, D.R., Roman, R.J., 2002. Mechanismof cGMP contribution to the vasodilator response to NO in rat middlecerebral arteries. Am. J. Physiol. 282, H1724–H1731.

Zhang, R., Wang, Y., Zhang, L., Zhang, Z., Tsang, W., Lu, M., Zhang, L.,Chopp, M., 2002. Sildenafil (Viagra) induces neurogenesis and promotesfunctional recovery after stroke in rats. Stroke 33, 2675–2680.

Zhang, R.L., Zhang, Z., Zhang, L., Wang, Y., Zhang, C., Chopp, M., 2006.Delayed treatment with sildenafil enhances neurogenesis and improvesfunctional recovery in aged rats after focal cerebral ischemia. J. Neurosci.Res. 83, 1213–1219.

Documents

Role of K+ and Ca2+ fluxes in the cerebroarterial vasoactive effects of sildenafil