C-Type Natriuretic Peptide Hyperpolarizes and Relaxes HumanPenile Resistance Arteries

Attila Kun, PhD,*‡1 Istvan Kiraly, MD,† Janos Pataricza, MD, PhD,* Zoltan Marton, BSc,*Iren Krassoi, BSc,‡ Andras Varro, MD, PhD, DSc,*‡ Ulf Simonsen, MD, PhD,§

Julius Gy. Papp, MD, PhD, DSc,*‡ and Laszlo Pajor, MD, PhD†

*Department of Pharmacology and Pharmacotherapy, University of Szeged, Szeged, Hungary; †Department of Urology,University of Szeged, Szeged, Hungary; ‡Division of Cardiovascular Pharmacology, Hungarian Academy of Sciences,Szeged, Hungary; §Department of Pharmacology, University of Aarhus, Aarhus, Denmark

DOI: 10.1111/j.1743-6109.2008.00775.x

A B S T R A C T

Introduction. In addition to nitric oxide (NO), it is thought that an endothelium-derived hyperpolarizing factor(EDHF) plays an important role in the relaxation of penile arteries. Recently, it has been shown that C-typenatriuretic peptide (CNP) shows the characteristics of EDHF in systemic small arteries.Aim. To investigate the mechanism involved in CNP-evoked vasodilatation and to address whether CNP is anEDHF in human penile resistance arteries.Methods. Erectile tissue was obtained in connection with transsexual operations. Intracavernous penile resistancearteries were isolated and mounted in microvascular myographs for recording of isometric tension. Membranepotential was recorded by the use of a small glass electrode inserted in the smooth muscle layer.Main Outcome Measure. In vitro evidence for hyperpolarization and vasorelaxation induced by CNP.Results. Acetylcholine (ACh) and CNP hyperpolarized smooth muscle membrane potential in resting penile resis-tance arteries. In penile small arteries incubated with inhibitors of NO synthase and cyclooxygenase and contractedwith phenylephrine, ACh and CNP evoked concentration-dependent relaxations with maximum of 56 � 6% and71 � 6%, respectively. Addition of a combination of blockers of small- and intermediate-conductance calcium-activated K+ channels, apamin plus charybdotoxin, respectively, and a combination thought to block the smoothmuscle response of EDHF-type relaxation, barium plus ouabain, markedly reduced ACh- and CNP-evoked relax-ation. Iberiotoxin, a blocker of big-conductance calcium-activated K+ channels inhibited the vasorelaxant responsesevoked by ACh and CNP. A selective natriuretic peptide receptor type C (NPR-C) agonist, C-atrial natriureticfactor4-23 (cANF4-23), induced relaxations with less maximum response compared to CNP.Conclusion. The present findings suggest that CNP possesses the characteristics of an EDHF in human penileresistance arteries. By activation of natriuretic peptide receptor type B and NPR-C receptors, CNP causes relaxationby activation, respectively, of large-conductance calcium-activated K+ channels and Na+/K+-adenosine triphosphatase(ATPase), and barium-sensitive inward rectifier K+ channels. Modulation of the CNP pathway opens for newtreatment modalities of erectile dysfunction. Kun A, Kiraly I, Pataricza J, Marton Z, Krassoi I, Varro A,Simonsen U, Gy. Papp J, and Pajor L. C-type natriuretic peptide hyperpolarizes and relaxes human penileresistance arteries. J Sex Med 2008;5:1114–1125.

Key Words. CNP; Acetylcholine; Endothelium; EDHF; Human Penile Arteries

1Present address: Department of Pharmacology, University of Aarhus, Aarhus, Denmark.

1114

J Sex Med 2008;5:1114–1125 © 2008 International Society for Sexual Medicine

Introduction

T he endothelium is an important contributorto smooth muscle relaxation of the corpus

cavernosum and penile small arteries, and isthought to play an important role in erection [1,2]and erectile dysfunction [3]. Acetylcholine (ACh)-evoked vasodilation is thought to depend on nitricoxide (NO), prostanoids as well as a non-NO/nonprostanoid endothelium-derived hyperpolariz-ing factor (EDHF) [4]. The EDHF-type relaxationis present in human [5,6], horse [7], bovine [8], andrat [9] penile arteries. In systemic arteries, severalcandidates for EDHF have been suggested, includ-ing potassium ions [10], products of the cyto-chrome P450 pathway [11–14], C-type natriureticpeptide (CNP) [15], and, controversially, hydrogenperoxide [16,17]. Both K+ and CNP are believed toactivate the Na+/K+-ATPase and/or inward rectifierK+ channels in the vascular smooth muscle layer[10,15]. Other studies have attributed agonist-evoked non-NO/nonprostanoid relaxation tocommunication by myoendothelial gap junctions[18–20]. In human penile arteries, inhibition of theP450 enzyme causes marginal reduction of EDHF-type relaxation [6], while inhibition of Na+/K+-ATPase has a marked effect in horse penile arteries[7]. Therefore, several mechanisms and candidatesmay contribute to EDHF-type vasodilation inpenile arteries.

In systemic arteries, the EDHF-type relaxationis dependent on activation of endothelial small-and intermediate-conductance calcium-activatedK+ channels which can be blocked by the combi-nation of apamin and charybdotoxin [10,21,22].Another charactistics of an EDHF is that it causeshyperpolarization of the underlying smoothmuscle layer [10,23]. However, membrane poten-tial has so far not been measured to confirm thepresence of an EDHF in penile arteries.

CNP is abundant in vascular endothelial cellsand is a potent vasodilator in human blood vessels[24], and infusion of CNP increases forearm bloodflow independent of NO and cyclooxygenaseproducts [25]. CNP vasodilation can be mediatedby activation of natriuretic peptide receptor type B(NPR-B) [26] or NPR-C [27]. CNP is suggestedto be an EDHF in rat mesenteric resistancearteries causing relaxation through activation ofNPR-C [15], but in human subcutaneous resis-tance arteries, CNP relaxation involves activationof NPR-B and opening of large-conductanceCa2+-activated K (BKCa) channels [28]. Receptorsfor CNP are present in corpus cavernosum from

man [29] and in several animal species [30,31],suggesting the possibility that CNP is involved inthe regulation of penile erection.

The hypothesis of the present study is thatCNP causes hyperpolarization and is a candidateto be involved in EDHF-type relaxation in humanpenile small arteries. Several of the previousstudies have been performed in human penilearteries obtained from patients undergoingsurgery for insertion of a penile prosthesis [5,6].Therefore, in the present study, penile arterieswere obtained from patients undergoing trans-sexual operation. In isolated arteries, the mem-brane potential was measured to investigatewhether CNP causes hyperpolarization, andpenile arteries were investigated in the absenceand presence of blockers of the EDHF pathway.The findings of the present study suggest thatCNP is a potent penile vasodilator and possessesthe characteristics of an EDHF in penile arteries.

Methods

Tissue PreparationPenile erectile tissue was obtained in connectionwith transsexual operations. All patients had beencross dressing, living as women (mean age 29,range 18–49, N = 14). Before surgery, each under-went a complete psychiatric and medical evalua-tion. The procedures were in accordance withEthical Committee of the University of Szeged(No.1988/Szeged Tudományegyetem [SZTE]).

The penis was excised and submerged immedi-ately in ice-cold (4°C) physiological salt solution(PSS). The penile artery was carefully dissectedand cleaned from the adherent connective tissue.Ring segments (ca. 2-mm long) were mounted ontwo 40-mm wires of isometric double myograph(Danish Myotechnology, Aarhus, Denmark) asdescribed previously [9]. The preparations wereallowed to equilibrate in PSS solution, at 37°C andpH 7.4, for 30 minutes. The relation betweenresting wall tension and internal circumference ofthe vessels was determined, and the internal cir-cumference, L100, corresponding to a transmuralpressure of 100 mm Hg in a relaxed vessel, wascalculated [32]. Subsequently, the internal circum-ference of the vessels was set to L1, whereL1 = 0.9 ¥ L100. The effective internal lumen diam-eter was determined as l1 = L1/p.

Experimental ProcedureTo test the contractility of the preparations, theywere exposed twice to phenylephrine (PE, 10 mM).

Effect of CNP in Human Penile Arteries 1115

J Sex Med 2008;5:1114–1125

The presence of intact endothelium was evaluatedby inducing a stable contraction with phenyleph-rine (10 mM) followed by addition of ACh (10 mM).Relaxation greater than 50% was taken as evidenceof endothelial integrity. Arteries were incubatedwith NG-nitro-L-arginine (L-NOARG, 100 mM)and indomethacin (10 mM) for 20 minutes beforebeing contracted with phenylephrine (3 mM), andcumulative concentration–response curves wereconstructed for ACh (1 nM–10 mM), CNP (0.01–1 mM), and cANF4-23 (0.01–1 mM). To investigatethe role of K+ channels in CNP relaxations,concentration–response curves for CNP were con-structed in preparations contracted by 80-mMpotassium-rich PSS, where NaCl was exchangedfor KCl on an equimolar basis. When the effects ofinhibition of K+ channels or Na+-K+-ATPase weretested, the arteries were subsequently incubated for20 minutes with either the combination of charyb-dotoxin (0.1 mM) plus apamin (0.5 mM), barium(Ba2+, 30 mM) plus ouabain (5 mM), charybdotoxin(0.1 mM), or iberiotoxin (0.1 mM), and a secondconcentration–response curve was obtained for theagonist.

Membrane Potential MeasurementsArterial segments (~3 mm long) were cut longitu-dinally, and the strips were pinned to the siliconerubber floor of an organ bath. PSS solution (37°C,pH 7.4) was continuously circulated throughthe organ chamber at 3 mL/minute. Membranepotential was recorded using conventional intrac-ellular microelectrodes filled with 3 M KCl (re-sistance: 50–80 MW). The microelectrodes andreference electrode (Ag/AgCl) were connected toan Intra 767 amplifier (World Precision Instru-ments, Inc, Sarasota, FL, USA). Data wererecorded by a Cole Palmer 80807-00 paperlessrecorder (Cole-Palmer Instrument Company,Vernon Hills, IL, USA). Drugs were addeddirectly to the organ bath.

Analysis of DataThe mechanical responses of the vessels were mea-sured as force and were expressed as active walltension, DT, which is the increase in measuredforce, DF, divided by twice the segment length[32]. The effects of different blockers on basaltension were expressed as percentage of phenyle-phrine contractions. The magnitude of relaxantresponses induced by either CNP or ACh wasgiven as percentage of the contraction level justprior to the addition of the drug.

The results were expressed as means � standarderror of the mean (SEM), and the concentration–response curves were presented on a semilogarith-mic scale. N represents the number of experiments,one artery per protocol per patient. Differencesin concentration–response relationships betweentreatments were analyzed using two-way analysis ofvariance. Differences at the P < 0.05 level wereconsidered statistically significant.

DrugsThe composition of PSS was as follows (in mM):NaCl, 120.0; NaHCO3, 20.0; KCl, 4.1; KH2PO4,1.2; MgCl2, 1.2; CaCl2, 1.5; and glucose, 11.0 (pH7.4) (REANAL, Szeged, Hungary).

The following drugs were used: ACh hydro-chloride, apamin, barium chloride dihydrate,CNP, charybdotoxin, iberiotoxin, indomethacin,L-NOARG, ouabain, and phenylephrine hydro-chloride (Sigma, St. Louis, MO, USA). All drugswere dissolved in distilled water except indo-methacin (70% ethanol).

Results

Human penile small arteries with normalizedinternal lumen diameter of 509 � 17 mm werecontracted with phenylephrine (10 mM) (N = 50arteries from 14 patients). Average contraction was6.4 � 0.3N/m (N = 50).

Hyperpolarizations Induced by ACh and CNPSmooth muscle membrane potential in resting,untreated arterial preparations was -44.8 �1.3 mV (N = 12). In endothelium intact prepara-tions, infusion of ACh (10 mM) hyperpolarizedmembrane potential by 5.2 � 0.9 mV (N = 6),while infusion of an equieffective relaxant concen-tration of CNP (1.4 mM) hyperpolarized smoothmuscle membrane potential by 2.5 � 0.5 mV(N = 10; Figure 1a, b). Mechanical removal of theendothelium abolished the ACh-evoked hyperpo-larization (0.2 � 0.3 mV, N = 7), while the CNP-evoked hyperpolarization was unchanged (2.4 �0.6 mV, N = 10; Figure 1c, d).

Effect of Blockers of the EDHF Pathway onACh RelaxationIn penile arteries contracted with phenylephrine,ACh evoked concentration-dependent relaxation.Incubation with an inhibitor of NO synthase,L-NOARG (100 mM), and cyclooxygenase,indomethacin (10 mM), significantly reducedACh relaxation (Figure 2a). In the presence of

1116 Kun et al.

J Sex Med 2008;5:1114–1125

L-NOARG and indomethacin, the residual AChrelaxation was inhibited by treating the arterieswith a blocker of intermediate-conductance Ca2+-activated K+-channel, charybdotoxin (Figure 2b),and was abolished in the presence of a combina-tion of blockers of small- and intermediate-conductance Ca2+-activated K+-channels, apaminand charybdotoxin (Figure 2b), or with a combi-nation of barium chloride (30 mM) and ouabain(5 mM) which block smooth muscle inward recti-fier K+ channels and Na+-K+-ATPase, respectively(Figure 2c).

Characterization of CNP-Evoked RelaxationIn arteries with endothelium (Figure 3a) and con-tracted with phenylephrine in the presence ofL-NOARG and indomethacin, CNP (0.01–1 mM)evoked concentration-dependent relaxations(Figures 3b and 4a). The concentration–responsecurves for CNP were reproducible up to threetimes (N = 4) and were unaltered in preparationswithout endothelium (N = 2, results not shown).In the presence of L-NOARG and indomethacinand contracted with high extracellular K+

(80 mM), CNP relaxation was markedly blunted(Figures 3c and 4a).

Incubation of the arteries with charybdotoxinalone (Figure 4b), a combination of charybdotoxinplus apamin (Figure 4b), or with the combinationof barium chloride (30 mM) plus ouabain (5 mM)(Figure 4c) markedly reduced CNP relaxation.

BKCa Channels and NPR-C Are Involved in ACh- andCNP-Evoked RelaxationTo test whether BKCa channels were involved inACh- and/or CNP-evoked relaxation, the bloodvessels were incubated with iberiotoxin (0.1 mM).In the presence of iberiotoxin, the concentration–response curves for ACh and CNP were markedlyreduced (Figure 5b, d) compared to parallelcontrol experiments (Figure 5a, c).

To determine whether NPR-C was presentin penile small arteries, the selective NPR-Cagonist cANF4-23 was applied. cANF4-23 evokedconcentration-dependent relaxations of humanpenile small arteries with maximal relaxations at1 mM of 32 � 12% (N = 4), while CNP at 1 mMrelaxed the arteries 68 � 11% (N = 5) (Figure 6).

Discussion

Our findings provide, for the first time, evidencesuggesting that CNP possesses the characteristicsof an EDHF contributing to vasodilation ofhuman penile small arteries (see Figure 7). Thus,direct electrophysiological measurements revealedthat CNP hyperpolarizes penile arterial smoothmuscle. CNP, similar to the endothelium-dependent vasodilator ACh, caused potent relax-ations which were also present after inhibition ofNO synthase and cyclooxygenase. These EDHF-type relaxations were inhibited in the presence ofblockers of calcium-activated K+ channels, apamin

Figure 1 Acetylcholine (ACh) andC-type natriuretic peptide (CNP)hyperpolarizes human penile arteries.Original recordings (panels a, c) andaverages (panels b, d) of smoothmuscle cell membrane potentialchanges (DEm) induced by ACh(10 mM) and CNP (1.4 mM) in penilearterial segments with endothelium (a,b) and without endothelium (b, d). Sig-nificant differences were evaluated bypaired t-test: *P < 0.05 vs. arterialsegment without endothelium.

2 m

V

2 min

1.4 μMCNP

10 μMACh

A

2 m

V

2 min

1.4 μMCNP

10 μMACh

C

-7

-6

-5

-4

-3

-2

-1

0

1.4 μMCNP

10 μMACh

B

ΔE

m(m

V)

-7

-6

-5

-4

-3

-2

-1

0

1.4 μMCNP

10 μMACh

D

*

ΔE

m(m

V)

Effect of CNP in Human Penile Arteries 1117

J Sex Med 2008;5:1114–1125

plus charybdotoxin, and by the combinationof barium plus ouabain, respectively, blockingsmooth muscle inward rectifier K+ channels andNa+-K+-ATPase. Both NPR-B and NPR-Cappeared to be involved in CNP vasodilation inpenile arteries.

EDHF and K+ Channels in Penile ArteriesSimilar to previous studies [5,6], inhibition of NOsynthase and cyclooxygenase, in the present study,caused a rightward shift in concentration–responsecurves for ACh suggesting that NO and vasodila-tor prostanoids contribute to the endothelium-dependent vasodilation to ACh in human penilearteries. However, in the presence of inhibitors ofNO synthase and cyclooxygenase, considerableACh relaxation persisted. Our findings agree withprevious studies in penile arteries from man [5,6]and different animal species [7,9] where endo-thelium-dependent agonists such as bradykininand ACh also evoked pronounced vasodilation inthe presence of inhibitors of NO synthase andcyclooxygenase. These findings suggest that inaddition to NO and prostanoids, a non-NO non-prostanoid EDHF-type relaxation is involved inACh relaxation. The present study provides directevidence that ACh actually causes endothelium-dependent hyperpolarization in human penilearteries, a finding which, for the first time, sub-stantiates that an EDHF contributes to theresponses to ACh in human penile arteries.

EDHF-type relaxation is dependent on activa-tion of endothelial calcium-activated K+ channels[10,21], and the combination of apamin andcharybdotoxin, which blocks calcium-activated K+

channels, has been found to inhibit ACh relaxationin human penile arteries [6]. In the present study,the combination of apamin and charybdotoxin also

-9 -8 -7 -6 -5

0

25

50

75

100

ControlL-NOARG+INDO

A

*

Log ACh [M]

Rel

axat

ion

(%

of

PE

)

-9 -8 -7 -6 -5

0

25

50

75

100

L-NOARG+INDOL-NOARG+INDO+ChTX+ApaL-NOARG+INDO+ChTX

B

**

Log ACh [M]

Rel

axat

ion

(%

of

PE

)

-9 -8 -7 -6 -5

0

25

50

75

100

L-NOARG+INDO

L-NOARG+INDO+Ouabain+Ba2+

C

*

Log ACh [M]

Rel

axat

ion

(%

of

PE

)�

Figure 2 Acetylcholine (ACh) induces endothelium-derived hyperpolarizing factor-type relaxation in penilearteries contracted with phenylephrine (PE). (a)Concentration–response curves for ACh in the absence(control) and presence of inhibitors of, respectively, nitricoxide synthase and cyclooxygenase, NG-nitro-L-arginine(L-NOARG, 100 mM), and indomethacin (INDO, 10 mM)(N = 7 experiments from four patients). In the presence ofL-NOARG and indomethacin, (b) combination with charyb-dotoxin (ChTX) (N = 5 experiments from three patients) orChTX plus apamin (ChTX + Apa) (N = 5 experiments fromfour patients), or (c) combination with ouabain (5 mM) plusbarium (Ba2+, 30 mM) (N = 5 experiments from threepatients) markedly inhibited ACh relaxation. Results aremeans � standard error of the mean. Significant differ-ences were evaluated by two-way analysis of variance:*P < 0.05 vs. control curve.

1118 Kun et al.

J Sex Med 2008;5:1114–1125

inhibited the EDHF-type relaxation evoked byACh in human penile small arteries, thus confirm-ing the presence of an EDHF. This was furthersubstantiated by the use of the combination ofbarium and ouabain, which was found to block theeffect of EDHF in rat mesenteric arterial smoothmuscle [10], and also markedly inhibited EDHF-type relaxation in horse penile arteries [7]. In thepresent study, the combination of barium andouabain also inhibited ACh relaxation, suggestingthat an EDHF mediates the vasodilation.

In addition to activation of inward rectifier Kchannels and Na/K-ATPase, opening of BKCa

channels has been proposed to be involved inEDHF relaxation of small arteries. Thus, BKCa

channels have been found to be involved inEDHF-evoked vasodilation or membrane hyper-polarization in different vessels obtained fromhuman [33,34] and animal tissues [35–37]. In ratpenile arteries, we also recently found that tetra-ethylammonium, at concentrations selectivelyblocking large-conductance calcium-activated K+

channels [38], as well as charybdotoxin, which is ablocker of intermediate- and big-conductancecalcium-activated K channels also abolished AChrelaxation [9]. The latter finding is confirmed inthe present study, where charybdotoxin alone alsoinhibited ACh EDHF-type relaxation in humanpenile arteries. However, charybdotoxin does not

allow distinguishing whether the effect of charyb-dotoxin is due to inhibition of the intermediate-conductance calcium-activated K channel in theendothelium or the smooth muscle BKCa channelsthat are involved in ACh relaxation. Therefore, toaddress the role of BKCa channels, the effect ofiberiotoxin was examined and was found to mark-edly inhibit ACh-evoked EDHF-type relaxation,hence suggesting that EDHF in human arteries, inaddition to activation of inward rectifier K chan-nels and Na+/K+-ATPase, also leads to activationof smooth muscle BKCa channels.

CNP an EDHF in Penile ArteriesAs mentioned in the Introduction, several candi-dates may contribute to EDHF-type relaxation inpenile arteries [6,7]. CNP is a likely candidate foran EDHF in human penile arteries. First, CNPreceptors are present in the corpus cavernosumof different species [29–31]. Second, we haveobserved in previous studies that EDHF-typerelaxation regulates neurogenic contraction [9], anobservation which is in agreement with the recentdemonstration of both pre- and postjunctionalCNP receptors in arteries [39]. Third, in thepresent study, CNP evoked relaxations in the pres-ence of inhibitors of NO synthase and cyclooxy-genase and in arteries without endothelium.Moreover, CNP relaxations were sensitive to

L-NOARG (100 μM) + INDO (10 μM)

0.01

0.1

1

W

0.010.1

1

W

4 m

N

2 min

a b c

PE(10 μM)

KCl (80 mM) CNP (μM) CNP (μM)ACh (10 μM)

PE (3 μM)

Figure 3 Original traces showing the relaxing effect of C-type natriuretic peptide (CNP) in human penile small artery. (a)Contraction induced by phenylephrine (PE, 10 mM) followed by acetylcholine (ACh, 10 mM). (b) In the arterial segment (withan internal diameter of 410 mm) contracted by PE in the presence of NG-nitro-L-arginine (L-NOARG) (100 mM) andindomethacin (INDO, 10 mM), CNP induced concentration-dependent relaxation. (c) In the same arterial segment contractedwith 80-mM K+-rich physiological salt solution, CNP relaxation was markedly reduced. The horizontal bar indicates time, andthe vertical bar indicates increase in force (mN). W = washout.

Effect of CNP in Human Penile Arteries 1119

J Sex Med 2008;5:1114–1125

barium plus ouabain and high extracellular K+

concentration. All these findings agree with CNPbeing an EDHF in human penile arteries,although further studies are required where

release of CNP from endothelial cells is measured.However, because of the size of the penile arteries,such measurements are excluded with the cur-rently available methods.

CNP has previously been described to causehyperpolarization in rat mesenteric and porcinecoronary arteries [15,40]. To investigate whetherhyperpolarization plays a role in the effect ofCNP, membrane potential measurements wereconducted in human penile small arteries. CNPevoked hyperpolarization of human penile arterialsmooth muscle, similarly to ACh, thereby provid-ing direct evidence that it possesses an importantcharacteristic of an EDHF. However, CNPinduced less hyperpolarization compared to ACh,suggesting that in addition to CNP, other EDHFcandidates may also contribute to ACh-evokedEDHF responses in human penile arteries.

The biological activity of CNP can be mediatedboth by activation of NPR-B or NPR-C. In thepresent study, the selective NPR-C agonist,cANF4-23, induced relaxations suggesting the pres-ence of NPR-C receptors in human penile smallarteries. In rat mesenteric small arteries, theNPR-C was suggested to be linked to activation ofthe smooth muscle inward rectifier K+ channel andto mediate EDHF-type relaxation [15,27], andhence activation of NPR-C may contribute to CNPrelaxation in human penile small arteries. However,CNP relaxation is more potent and pronouncedcompared with cANF4-23, suggesting that activationof NPR-B also contributes to CNP relaxation inhuman penile arteries. NPR-B is present in thecorpus cavernosum of different species includingman [29–31], and is linked to particulate guanylyl

-8 -7 -6

0

25

50

75

100

PE

KCl

A

*

Log CNP [M]

Rel

axat

ion

-8 -7 -6

0

25

50

75

100

Control

Ouabain+ Ba 2+

C

Log CNP [M]

Rel

axat

ion

(%

of

PE

)

*

-8 -7 -6

0

25

50

75

100

ControlChTX+ ApaChTX

B

Log CNP [M]

Rel

axat

ion

(%

of

PE

)

*

�

Figure 4 C-type natriuretic peptide (CNP) relaxation inhuman penile arteries is inhibited by drugs blocking theendothelium-derived hyperpolarizing factor pathway.Average concentration–response curves for CNP obtainedin the presence of NG-nitro-L-arginine (L-NOARG) (100 mM)and indomethacin (10 mM) in (a) preparations contractedwith phenylephrine (PE) or with 80-mM K+ (N = 5 experi-ments from four patients). (b) Average concentration–response curves for CNP in the presence of L-NOARG andindomethacin alone or in combination with charybdotoxin(ChTX) (N = 4 experiments from three patients) or ChTXplus apamin (ChTX + Apa) (N = 4 experiments from threepatients) in PE-contracted penile arteries. (c) Averageconcentration–response curves for CNP in the presence ofL-NOARG and indomethacin alone or in combination withouabain (5 mM) plus barium (Ba2+, 30 mM) in PE-contractedpenile arteries (N = 5 experiments from four patients).Results are means � standard error of the mean. Signifi-cant differences were evaluated by two-way analysis ofvariance: *P < 0.05 vs. control curve.

1120 Kun et al.

J Sex Med 2008;5:1114–1125

cyclase and the formation of cyclic guanosinemonophosphate (GMP) [41]. Activation of thecyclic GMP pathway has been found to lead toactivation of both Na+/K+-ATPase and BKCa inhorse penile arteries [7,42], and of Na/K-ATPase inhuman corpus cavernosum [43]. Moreover, CNP-evoked relaxation in systemic arteries was describedto be inhibited by iberiotoxin, a blocker of BKCa

channels, [28,44–46]. In the present study, botha blocker of intermediate-conductance Ca2+-activated K channels and BKCa channels, charyb-dotoxin, and iberitoxin markedly reduced CNPrelaxation suggesting BKCa channels are involved inCNP relaxation. Moreover, addition of apamin pluscharybdotoxin did not cause further inhibitioncompared to charybdotoxin alone, hence excludinga contribution of small-conductance Ca2+-activatedK channels to CNP relaxation. Therefore, togetherwith the pronounced effect of barium and ouabain,the inhibitory effect of iberiotoxin makes it plau-sible that CNP causes relaxation in human penilearteries through activation of NPR-B and NPR-C

DC

BA0.001

0.01

0.1

1

104

mN

2 min

ACh (μM)PE (3 μM)

30 min

L-NOARG+INDO

0.001 0.01 0.1 110

ACh (μM)PE (3 μM)

L-NOARG+INDO

30 min2 min

4 m

N

IbTX (0.1 μM)

CNP (μM)

0.01

0.1

1

4 m

N

PE (3 μM)

2 min

30 min

L-NOARG+INDO

2 min

0.010.1

1

4 m

N

PE (3 μM)CNP (μM)

L-NOARG+INDO

30 min

IbTX (0.1 μM)

Figure 5 Large-conductance Ca2+-activated K (BKCa) channels involved in acetylcholine (ACh)- and C-type natriureticpeptide (CNP)-evoked relaxation. Original traces showing the relaxing effect of increasing concentrations of (a, b) ACh and(c, d) CNP in human penile small arteries in the absence and presence of a blocker of BKCa channels, iberiotoxin (IbTX). Theexperiments were performed in the presence of an inhibitor of nitric oxide synthase, NG-nitro-L-arginine (L-NOARG)(100 mM), and an inhibitor of cyclooxygenase, indomethacin (INDO, 10 mM). The traces are representative for two experi-ments obtained from two patients. The horizontal bar indicates time, and the vertical bar indicates increase in force (mN).

-8 -7 -6

0

20

40

60

80

100

CNP

cANF 4-23

*

Log [M]

Rel

axat

ion

(%

of

PE

)

Figure 6 Average relaxations evoked by C-type natriureticpeptide (CNP) and cANF4-23, a natriuretic peptide receptortype C agonist, in human penile small arteries. In the pres-ence of NG-nitro-L-arginine (10-4 M) and indomethacin(10-5 M), cANF4-23 induced less relaxation than CNP.Results are means � standard error of the mean of five andfour experiments obtained from four patients.

Effect of CNP in Human Penile Arteries 1121

J Sex Med 2008;5:1114–1125

followed, respectively, by the activation ofiberiotoxin-sensitive BKCa channels and Na+/K+-ATPase, and by the activation of barium-sensitiveinward rectifier K channels.

Perspectives and Conclusions

An apparent limitation of the samples examined inthe present study could be that they were obtainedfrom patients undergoing transsexual operations,and we cannot entirely exclude that hormonaltreatments influence the endothelial responses.However, the advantage, on the other hand, is thatnone of the patients had cardiovascular disease,which is known to alter endothelium-dependentvasodilation in both systemic [47] and penile arter-ies [48–50]. Thus, in erectile tissue and penile arter-ies from patients with diabetes, endothelium-dependent relaxation is blunted [6,51], as well as inpenile arteries and erectile tissue isolated fromhypertensive and diabetic animals [52,53]. In penilearteries mainly, the EDHF-type relaxation isimpaired in hypertensive animals [52]. Phosphodi-esterase type 5 inhibitors, e.g., sildenafil, would beexpected to enhance the effect of activation of theNPR-B/cyclic GMP pathway by CNP, but thatwould not be the case for the activation of theNPR-C/inward rectifier K+ channel pathway byCNP.

In conclusion, the present findings suggestthat CNP possesses the characteristics of anEDHF in human penile resistance arteries,because it causes hyperpolarization of the penile

smooth muscle. CNP causes potent relaxationsthat are sensitive to the blockers of calcium-activated K+ channels, inward rectifier K+ chan-nels, and Na+-K+-ATPase. Modulation of thispathway opens for new treatment modalities oferectile dysfunction.

Acknowledgment

This study was supported by grants from the HungarianSociety of Urology, Hungarian Ministry of Education,Hungarian Academy of Sciences and HungarianNational Research Foundation (OTKA NI-61902). UlfSimonsen was supported by grants from the DanishMedical Research Council.

Corresponding Author: Professor Ulf Simonsen, MD,PhD, Faculty of Health Sciences, Department of Phar-macology, University of Aarhus, 8000 Aarhus C,Denmark. Tel: +4589421713; Fax: +4586128804;E-mail: us@farm.au.dk

Conflict of Interest: None declared.

Statement of Authorship

Category 1(a) Conception and Design

Attila Kun; Janos Pataricza; Laszlo Pajor; UlfSimonsen

(b) Acquisition of DataAttila Kun; Istvan Kiraly; Zoltan Marton; IrenKrassoi

(c) Analysis and Interpretation of DataAttila Kun; Istvan Kiraly; Janos Pataricza; ZoltanMarton; Iren Krassoi

Figure 7 Model for C-type natriureticpeptide (CNP) relaxation of penilesmall arteries. Acetylcholine (ACh),through increased endothelial cellcalcium, activates small-conductanceCa2+-activated K+ channels (SKCa) andintermediate-conductance Ca2+-activated K+ channels (IKCa) followedby release of CNP. CNP binds to natri-uretic peptide receptor type B (NPR-B)and natriuretic peptide receptor type C(NPR-C), which are coupled, respec-tively, to opening of large-conductanceCa2+-activated K channels (BKCa) andNa/K-ATPase and to opening ofinward rectifier K channels (KIR).

Endothelial cell

Smooth muscle cell

SKCa IKCa

R

↑[Ca2+]i

Apamin ChTX

ACh

BKCa

Na+/K+ ATP-ase

Hyperpolarization/relaxation

IbTX

Ouabain

CNP

NPR-B

CNP

KIR

NPR-CBa2+

1122 Kun et al.

J Sex Med 2008;5:1114–1125

Category 2(a) Drafting the Article

Attila Kun; Istvan Kiraly; Janos Pataricza; ZoltanMarton

(b) Revising It for Intellectual ContentAndras Varro; Julius Gy. Papp; Ulf Simonsen;Laszlo Pajor

Category 3(a) Final Approval of the Completed Article

Andras Varro; Julius Gy. Papp; Ulf Simonsen;Laszlo Pajor

References

1 Simonsen U, Garcia-Sacristan A, Prieto D. Penilearteries and erection. J Vasc Res 2002;39:283–303.

2 Sáenz de Tejada I, Angulo J, Cellek S, González-Cadavid N, Heaton J, Pickard R, Simonsen U.Physiology of erectile function. J Sex Med2004;1:254–65.

3 Sáenz de Tejada I, Angulo J, Cellek S, González-Cadavid N, Heaton J, Pickard R, Simonsen U.Pathophysiology of erectile dysfunction. J Sex Med2005;2:26–39.

4 Busse R, Edwards G, Félétou M, Fleming I, Van-houtte PM, Weston AH. EDHF: Bringing the con-cepts together. Trends Pharmacol Sci 2002;23:374–80.

5 Simonsen U, Prieto D, Delgado JA, Hernández M,Resel L, Saenz de Tejada I, García-Sacristán A.Nitric oxide is involved in the inhibitory neu-rotransmission and endothelium-dependent relax-ations of human penile resistance arteries. Clin Sci1997;92:269–75.

6 Angulo J, Cuevas P, Fernandez A, Gabancho S,Videla S, Saenz de Tejada I. Calcium dobesilatepotentiates endothelium-derived hyperpolarizingfactor-mediated relaxation of human penile resis-tance arteries. Br J Pharmacol 2003;139:854–62.

7 Prieto D, Simonsen U, Hernandez M, Garcia-Sacristan A. Contribution of K+ channels andouabain-sensitive mechanisms to the endothelium-dependent relaxations of horse penile small arteries.Br J Pharmacol 1998;123:1609–20.

8 Simonsen U, Contreras J, García-Sacristán A, Mar-tinez AC. Effect of sildenafil on non-adrenergicnon-cholinergic neurotransmission in bovine penilesmall arteries. Eur J Pharmacol 2001;412:155–69.

9 Kun A, Martinez AC, Tanko LB, Pataricza J, PappJG, Simonsen U. Ca2+-activated K+ channels in theendothelial cell layer involved in modulation of neu-rogenic contractions in rat penile arteries. Eur JPharmacol 2003;474:103–15.

10 Edwards G, Dora KA, Gardener MJ, Garland CJ,Weston AH. K+ is an endothelium-derived hyper-polarizing factor in rat arteries. Nature 1998;396:269–72.

11 Fulton D, McGiff JC, Quilley J. Contribution ofNO and cytochrome P450 to the vasodilator effectof bradykinin in the rat kidney. Br J Pharmacol1992;107:722–5.

12 Bauersachs J, Hecker M, Busse R. Display of thecharacteristics of endothelium-derived hyperpolar-izing factor by a cytochrome P450-derived arachi-donic acid metabolite in the coronarymicrocirculation. Br J Pharmacol 1994;113:1548–53.

13 Fisslthaler B, Hinsch N, Chataigneau T, Popp R,Kiss L, Busse R, Fleming I. Nifedipine increasescytochrome P4502C expression and endothelium-derived hyperpolarizing factor-mediated responsesin coronary arteries. Hypertension 2000;36:270–5.

14 Bolz SS, Fisslthaler B, Pieperhoff S, De Wit C,Fleming I, Busse R, Pohl U. Antisense oligonucle-otides against cytochrome P450 2C8 attenuateEDHF-mediated Ca(2+) changes and dilation inisolated resistance arteries. FASEB J 2000;14:255–60.

15 Chauhan SD, Nilsson H, Ahluwalia A, Hobbs AJ.Release of C-type natriuretic peptide accounts forthe biological activity of endothelium-derivedhyperpolarizing factor. Proc Natl Acad Sci USA2003;100:1426–31.

16 Matoba T, Shimokawa H, Nakashima M, HirakawaY, Mukai Y, Hirano K, Kanaide H, Takeshita A.Hydrogen peroxide is an endothelium-derivedhyperpolarizing factor in mice. J Clin Invest2000;106:1521–30.

17 Ellis A, Triggle CR. Endothelium-derived reactiveoxygen species: Their relationship to endothelium-dependent hyperpolarization and vascular tone. CanJ Physiol Pharmacol 2003;81:1013–28.

18 Chaytor AT, Evans WH, Griffith TM. Central roleof heterocellular gap junctional communication inendothelium-dependent relaxations of rabbit arter-ies. J Physiol 1998;508:561–73.

19 Yamamoto Y, Imaeda K, Suzuki H. Endothelium-dependent hyperpolarization and intercellular elec-trical coupling in guinea-pig mesenteric arterioles. JPhysiol 1999;514:505–13.

20 Coleman HA, Tare M, Parkington HC. K+ currentsunderlying the action of endothelium-derivedhyperpolarizing factor in guinea-pig, rat and humanblood vessels. J Physiol 2001;531:359–73.

21 Zygmunt PM, Högestätt ED. Role of potassiumchannels in endothelium-dependent relaxationresistant to nitroarginine in the rat hepatic artery. BrJ Pharmacol 1996;117:1600–6.

22 Buus NH, Simonsen U, Pilegaard HK, MulvanyMJ. Nitric oxide, prostanoid and non-NO, non-prostanoid involvement in acetylcholine relaxationof isolated human small arteries. Br J Pharmacol2000;129:184–92.

23 Vanhoutte PM. Endothelium-dependent hyperpo-larizations: The history. Pharmacol Res 2004;49:503–8.

Effect of CNP in Human Penile Arteries 1123

J Sex Med 2008;5:1114–1125

24 Wiley KE, Davenport AP. Physiological antagonismof endothelin-1 in human conductance and resis-tance coronary artery. Br J Pharmacol 2001;133:568–74.

25 Honing ML, Smits P, Morrison PJ, Burnett PC Jr,Rabelink TJ. C-type natriuretic peptide-inducedvasodilation is dependent on hyperpolarization inhuman forearm resistance vessels. Hypertension2001;37:1179–83.

26 Drewett JG, Fendly BM, Garbers DL, Lowe DG.Natriuretic peptide receptor-B (guanylyl cyclase-B)mediates C-type natriuretic peptide relaxation ofprecontracted rat aorta. J Biol Chem 1995;270:4668–74.

27 Villar IC, Panayiotou CM, Sheraz A, Madhani M,Scotland RS, Nobles M, Kemp-Harper B, Ahluwa-lia A, Hobbs AJ. Definitive role for natriureticpeptide receptor-C in mediating the vasorelaxantactivity of C-type natriuretic peptide andendothelium-derived hyperpolarising factor. Car-diovasc Res 2007;74:515–25.

28 Garcha RS, Hughes AD. CNP, but not ANP orBNP, relax human isolated subcutaneous resistancearteries by an action involving cyclic GMP andBKCa channels. J Renin Angiotensin AldosteroneSyst 2006;7:87–91.

29 Küthe A, Reinecke M, Uckert S, Becker A, David I,Heitland A. Expression of guanylyl cyclase B in thehuman corpus cavernosum penis and the possibleinvolvement of its ligand C-type natriureticpolypeptide in the induction of penile erection. JUrol 2003;169:1918–22.

30 Guidone G, Muller D, Vogt K, Mukhopadhyay AK.Characterization of VIP and PACAP receptors incultured rat penis corpus cavernosum smoothmuscle cells and their interaction with guanylatecyclase-B receptors. Regul Pept 2002;108:63–72.

31 Kim SZ, Kim SH, Park JK, Koh GY, Cho KW.Presence and biological activity of C-type natri-uretic peptide-dependent guanylate cyclase-coupledreceptor in the penile corpus cavernosum. J Urol1998;159:1741–6.

32 Mulvany MJ, Halpern W. Contractile properties ofsmall arterial resistance vessels in spontaneouslyhypertensive and normotensive rats. Circ Res1977;41:19–26.

33 Archer SL, Gragasin FS, Wu X, Wang S, McMurtryS, Kim DH, Platonov M, Koshal A, Hashimoto K,Campbell WB, Falck JR, Michelakis ED.Endothelium-derived hyperpolarizing factor inhuman internal mammary artery is 11,12-epoxyeicosatrienoic acid and causes relaxation byactivating smooth muscle BK(Ca) channels. Circu-lation 2003;107:769–76.

34 Martinez-Leon JB, Segarra G, Medina P, Vila JM,Lluch P, Peiro M, Otero E, Lluch S. Ca2+-activatedK+ channels mediate relaxation of forearm veins inchronic renal failure. J Hypertens 2003;21:1927–34.

35 McNeish AJ, Garland CJ. Thromboxane A(2) inhi-bition of SK(Ca) after NO synthase block in ratmiddle cerebral artery. Br J Pharmacol2007;151:441–9.

36 Hoepfl B, Rodenwaldt B, Pohl U, De Wit C.EDHF, but not NO or prostaglandins, is critical toevoke a conducted dilation upon ACh in hamsterarterioles. Am J Physiol Heart Circ Physiol2002;283:H996–1004.

37 Huang A, Sun D, Smith CJ, Connetta JA, SheselyEG, Koller A, Kaley G. In eNOS knockout miceskeletal muscle arteriolar dilation to acetylcholine ismediated by EDHF. Am J Physiol Heart CircPhysiol 2000;278:H762–8.

38 Braun AP, Heist EK, Schulman H. Inhibition of amammalian large conductance, calcium-sensitive K+

channel by calmodulin-binding peptides. J Physiol2000;527:479–92.

39 Grant S, Lutz EM, McPhaden AR, Wadsworth RM.Location and function of VPAC1, VPAC2 andNPR-C receptors in VIP-induced vasodilation ofporcine basilar arteries. J Cereb Blood Flow Metab2006;26:58–67.

40 Barton M, Beny JL, d’Uscio LV, Wyss T, Noll G,Luscher TF. Endothelium-independent relaxationand hyperpolarization to C-type natriuretic peptidein porcine coronary arteries. J Cardiovasc Pharma-col 1998;31:377–83.

41 Potthast R, Potter LR. Phosphorylation-dependentregulation of the guanylyl cyclase-linked natriureticpeptide receptors. Peptides 2005;26:1001–8.

42 Simonsen U, Prieto D, Sánez de Tejada I, García-Sacristán A. Involvement of nitric oxide in the non-adrenergic non-cholinergic neurotransmission ofhorse deep penile arteries: Role of charybdotoxin-sensitive K(+)-channels. Br J Pharmacol 1995;116:2582–90.

43 Gupta S, Moreland RB, Munarriz R, Daley J, Gold-stein I, Saenz de Tejada I. Possible role of Na(+)-K(+)-ATPase in the regulation of human corpuscavernosum smooth muscle contractility by nitricoxide. Br J Pharmacol 1995;116:2201–6.

44 Otsuka K, Tanaka H, Horinouchi T, Koike K,Shigenobu K, Tanaka Y. Functional contribution ofvoltage-dependent and Ca2+ activated K+ (BK(Ca))channels to the relaxation of guinea-pig aorta inresponse to natriuretic peptides. J Smooth MuscleRes 2002;38:117–29.

45 Tolsa JF, Gao Y, Sander FC, Souici AC, MoessingerA, Raj JU. Differential responses of newborn pul-monary arteries and veins to atrial and C-type natri-uretic peptides. Am J Physiol Heart Circ Physiol2002;282:H273–80.

46 Banks M, Wei CM, Kim CH, Burnett JC Jr, MillerVM. Mechanism of relaxations to C-type natriureticpeptide in veins. Am J Physiol 1996;271:H1907–11.

47 Feletou M, Vanhoutte PM. EDHF new therapeutictargets? Pharmacol Res 2004;49:565–80.

1124 Kun et al.

J Sex Med 2008;5:1114–1125

48 Xie D, Odronic SI, Wu F, Pippen AM, DonatucciCF, Annex BH. A mouse model of hyper-cholesterolemia-induced erectile dysfunction. J SexMed 2007;4:898–907.

49 Hannan JL, Heaton JP, Adams MA. Recoveryof erectile function in aging hypertensive and nor-motensive rats using exercise and caloric restriction.J Sex Med 2007;4:886–97.

50 Wingard C, Fulton D, Husain S. Altered penilevascular reactivity and erection in the Zucker obese-diabetic rat. J Sex Med 2007;4:348–62.

51 Saenz de Tejada I, Goldstein I, Azadzoi K, Krane RJ,Cohen RA. Impaired neurogenic and endothelium-

mediated relaxation of penile smooth muscle fromdiabetic men with impotence. N Engl J Med1989;320:1025–30.

52 Martínez AC, Stankevicius E, Jakobsen P, SimonsenU. Blunted non-nitric oxide vasodilatory neu-rotransmission in penile arteries from renalhypertensive rats. Vasc Pharmacol 2006;44:354–62.

53 Angulo J, Cuevas P, Gabancho S, Gonzalez-Corrochano R, Videla S, Saenz de Tejada I.Enhancement of both EDHF and NO/cGMP path-ways is necessary to reverse erectile dysfunction indiabetic rats. J Sex Med 2005;2:341–6.

Effect of CNP in Human Penile Arteries 1125

J Sex Med 2008;5:1114–1125