Possibility of Inhibition of Calcium-Activated Chloride Channel Rescuing Erectile Failures in Diabetes

Home , Channel blocker, Chloride channel blocker

ORIGINAL ARTICLE Possibility of inhibition of calcium-activated chloride channel rescuing erectile failures in diabetes

L-C Lau and PG Adaikan

Although calcium-activated chloride channel (CaCC) blockers, niflumic acid (NFA) and anthracene-9-carboxylic acid (A9C), have been shown as potential erectogenic agents in healthy corpus cavernosum (CC) tissues, the pharmacological characteristics of CaCC blockers in diabetic state are relatively unknown. This study compares the direct muscle relaxant property of NFA and A9C with their influence on contraction and nitrergic relaxation as elicited by electrical field stimulation in normal and 16-week-old diabetic rabbit CC (n ¼ 8). Mean blood glucose level in alloxan-treated rabbits was elevated threefold (21.9±0.5 mmol l À 1 vs 7.1±0.2 mmol l À 1 in untreated rabbits; Po0.05). There was no significant alteration in the efficacies of NFA and A9C in eliciting a concentration-dependent relaxation of noradrenaline-induced cavernosum tone and in inhibiting neurogenic contraction of CC from diabetic rabbits. The capability of NFA (100 mM) and A9C (1 mM) in augmenting nitrergic transmission was also not adversely affected by diabetes. However, in CC from diabetic rabbits, A9C markedly increased nitrergic relaxation response to 1–10 Hz by 10.6–36.6% (vs À 5.1–0.8% in nondiabetic control). CaCC sensitivity to A9C appears to be enhanced in diabetic CC tissue. Inhibiting the CaCC activity in diabetes-related ED may tip the balance between proerectile/relaxant and antierectile/contractile mechanisms in favor of cavernosum relaxation.

International Journal of Impotence Research (2014) 26, 151–155; doi:10.1038/ijir.2014.1; published online 13 February 2014 Keywords: anthracene-9-carboxylic acid; calcium-activated chloride channel; corpus cavernosum; diabetes mellitus; niﬂumic acid; rabbit

INTRODUCTION following: (i) enhanced sympathetic activity, increased expression Diabetes mellitus is a major risk factor in the development of ED. of receptors for vasoconstrictor mediators such as endothelin-1, As revealed in the Massachusetts aging study of men, the risk of increased production/release of vasoconstrictors and elevated 7–9 ED is threefold greater for diabetic men (28% as compared with responsiveness to vasoconstrictors; (ii) upregulation of the 9.6%) and occurs at an earlier age than in nondiabetic men.1 downstream signaling cascade e.g. the calcium-sensitizing path- Furthermore, men with diabetes often respond less satisfactorily way, including the protein kinase C and the RhoA/Rho-kinase, to current pharmacotherapies with PDE5 inhibitors thereby which suppresses endothelial nitric oxide synthase;10 and (iii) representing a more difficult subgroup to treat.2 modification in downstream effectors, such as function or activity At the peripheral level, penile erection is a complex hemo- of the ATP-sensitive potassium channel and calcium-sensitive dynamic process regulated by neurovascular mechanisms in the large conductance potassium channel leading to diminished corpus cavernosum (CC) and its arterioles. The prime neuro- hyperpolarising ability of cavernosal smooth muscle cell.11,12 transmitter mediating erectile responses is nitric oxide, which is Recent data showed that calcium-activated chloride current is released by the efferent (nitrergic) autonomic nerves and the an important excitatory mechanism that has a role in regulating endothelial lining of the sinusoids and cavernosal arterial system. the cavernosal tone and contractility.13–17 Moreover, there are The pathophysiology of ED related to diabetes mellitus is widely experimental findings indicating that nitric oxide-cGMP pathway recognized as multifactorial; it encompasses changes to the suppresses the intrinsic calcium-activated chloride currents in the vascular, nerve, smooth muscle and metabolic functions of the rat coronary artery and rabbit CC myocytes.13,18 It is plausible penis. Pharmacologically induced diabetes in animal models has that the diminished nitric oxide production as a consequence of been frequently used to study diabetic ED. The development nitrergic neuropathy and impaired endothelial function in of ED in animal models correlates with the clinical conditions seen diabetes may lead to increased calcium-activated chloride in humans; it is accompanied by alterations in the integrity of currents that can be damaging to normal erectile physiology. nitrergic nerves, endothelial and smooth muscle cells, by an However, to date, definitive evidence suggesting involvement of increase in advanced glycation products and oxidative stress calcium-activated chloride currents in the mechanistic behavior of and by reduced nitric oxide synthase activity,3–6 which are all diabetic CC tissue is lacking. Though calcium-activated chloride detrimental to penile hemodynamics. channel (CaCC) blockers, niflumic acid (NFA) and anthracene- In addition to impaired vasodilatory mechanisms noted in the 9-carboxylic acid (A9C) have been shown to be potential diabetic CC, hypercontractility of cavernosal smooth muscle erectogenic agents in healthy CC tissues,14 the effects of these limiting penile blood inflow may also occur as a result of the compounds on CC in diabetes have not been reported previously.

Department of Obstetrics and Gynaecology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore. Correspondence: Professor PG Adaikan, Department of Obstetrics and Gynaecology, Yong Loo Lin School of Medicine, NUS, NUHS Tower Block Level 12, 1E Kent Ridge Road, Singapore 119228, Singapore. E-mail: [email protected] Received 17 May 2013; revised 7 October 2013; accepted 20 December 2013; published online 13 February 2014 Calcium-activated chloride channel in diabetes L-C Lau and PG Adaikan 152 In this study, we attempt to determine, using NFA and A9C, Drugs whether the functional activity of CaCC in CC muscle strips excised All drugs were purchased from Sigma-Aldrich (St Louis, MO, USA) unless from diabetic rabbits is intact. otherwise indicated. Aqueous stock solutions of noradrenaline, guanethidine and atropine were prepared and stored at À 20 1C. NFA and A9C were initially prepared as a stock solution in dimethyl sulfoxide (DMSO) and subsequently diluted in deionized water before use. The final concentra- MATERIALS AND METHODS tion of dimethyl sulfoxide did not exceed 0.1%. Dimethyl sulfoxide in the Diabetic rabbit model concentrations used did not contract or relax the cavernosal muscle. New Zealand white rabbits (2.5–3.0 kg) were housed in the animal holding Working concentrations of each compound were prepared freshly on the unit of the Yong Loo Lin School of Medicine, National University of day of experiment and stored on crushed ice. Singapore. The animals were maintained under a constant 12 h light and 1 dark cycle and an environmental temperature of 21–23 C. Food and fresh Statistical analysis water were available ad libitum. Diabetes mellitus was induced in eight New Zealand white rabbits via a All data are expressed as mean±s.e.m. Statistical comparison between bolus intravenous injection of freshly prepared alloxan (100 mg kg À 1). concentration–response curves was performed with one-way analysis of Hyperglycemia was confirmed 14 days after administering alloxan injection variance, followed by Bonferroni’s post hoc correction for multiple using MediSense optium blood glucose electrodes (Abbott Laboratories, comparisons. Differences were considered significant when Po0.05. Each Doncaster, Victoria, Australia). Glycemia was assessed at 2-week intervals n refers to number of strips and N denotes number of animals. after the confirmation of diabetes mellitus. Only those rabbits that maintained blood glucose levels of 300 mg dl À 1 (16.5 mmol l À 1) or higher over the 16-week period were considered diabetic and used in this study. RESULTS Eight age-matched nondiabetic rabbits served as control. A bolus intravenous injection of alloxan in New Zealand white This study involving animals conformed to the guidelines for animal rabbits induced profound diabetes. Blood glucose levels in all experiments and the principles for the care and use of animals in research treated rabbits were significantly elevated throughout the 16- and teaching established by the National University of Singapore. week study period. The mean non-fasting blood glucose levels were increased by threefold (21.9±0.5 mmol l À 1; Po0.05) in À 1 Rabbit CC tissue comparison with untreated control animals (7.1±0.2 mmol l ). Diabetic rabbits were killed 16 weeks after induction of diabetes with a When administered to the noradrenaline-precontracted rabbit sodium pentobarbital injection (100 mg kg À 1) through the ear vein. The cavernous tissue, CaCC blockers, NFA and A9C, elicited concen- entire penis was surgically removed and placed in chilled tyrode solution tration-related relaxation of CC strips from controls (n ¼ 7) and À 1 (mmol l composition: NaCl 137, NaHCO3 11.9, CaCl2 1.8, KCl 2.7, MgSO4 16-week-old diabetic rabbits (n ¼ 5) as depicted in Figure 1. NFA 1.1, NaH2PO4 0.42 and glucose 5.6). was more potent than A9C in eliciting CC relaxation. The CC was carefully dissected and cleared of adherent adipose and The concentration–response curves for NFA and A9C in CC from muscular tissues and tunica albuginea. The cavernosal strips (1 Â 1 Â 5 diabetic animals did not deviate significantly from those esta- mm) were suspended longitudinally under 1 g resting tension in 10-ml blished in healthy nondiabetic animals. organ baths containing tyrode solution bubbled with 95% O2 and 5% CO2, and maintained at 37 1C.14 An equilibration period of 90 min was applied Both NFA and A9C inhibited in a concentration-dependent to all tissues, during which tissues were washed with fresh tyrode solution manner the neurogenic contraction elicited by stimulus frequency at regular intervals and baseline was readjusted to 1 g. of 20 Hz in CC from nondiabetic and diabetic groups. There was no statistically significant difference between CC from diabetic and healthy nondiabetic groups in terms of their sensitivity to Pharmacological assessment of chloride channel blockers on CC inhibition by NFA and A9C (Figure 2). muscle strips The relationship between nitrergic relaxation response and To compare the direct muscle relaxant activity of NFA and A9C in diabetic stimulus frequency was established in seven nondiabetic and tissues with that in nondiabetic tissues, CC strips were precontracted five diabetic CC preparations. In healthy control CC, NFA (100 mM) with 10 mM noradrenaline. After a steady state plateau contraction was augmented nitrergic relaxation response elicited by lower attained, NFA (10–100 mM) and A9C (100–1000 mM) were administered in a cumulative manner to establish the respective concentration–response stimulus frequencies (1–10 Hz) but was rather inactive against relationship of the compounds. Each relaxation response to NFA or A9C frequencies of 20 Hz and above. In diabetic CC strips, NFA at was expressed as a percentage decrease in contraction induced by 100 mM was equally competent in causing a displacement of the noradrenaline. The effect of NFA and A9C on neurogenic contraction and nitrergic transmission in normal and diabetic rabbit penile muscle was also evaluated. Briefly, electrical stimulation was delivered from a Grass S88 Concentration, Log[M] stimulator (Astro-Med Inc., West Warwick, RI, USA) via platinum electrodes. -5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 To elicit neurogenic contraction, CC tissue, which was at resting tension, 0 was stimulated with stimuli strength of 90 V and 1 ms pulses at a frequency of 20 Hz for 10 s until a series of three consistent contractile responses were established. Electrical stimulation of CC tissue was repeated in the 20 presence of increasing concentrations of NFA (10, 30 and 100 mM) or A9C (100, 300 and 1000 mM) (Figure 2a). Results were expressed as a percentage of the control contractile response, which was taken as 100%. 40 In another series of experiments, frequency–response curves for electrical stimulation of noradrenaline-precontracted CC were determined NFA (ND) in the presence of guanethidine (5 mM) and atropine (3.5 mM), which 60 blocked adrenergic neuron and muscarinic cholinergic responses, respec- % Relaxation NFA (D) tively. Under such conditions, electrical stimulation of CC strips at frequen- 80 cies of 1, 2, 5, 10, 20 and 40 Hz provoked only relaxation responses A9C (ND) mediated by nitrergic nerves. CC strips were exposed to NFA (10, 30 and A9C (D) 100 mM) or A9C (100, 300 and 1000 mM) for 20 min before the frequency– 100 response relationship was reestablished again. Changes in isometric force were recorded electronically using PowerLab Figure 1. Diabetes did not appear to adversely affect the 4sp data acquisition system (AD Instruments, Bella Vista, NSW, Australia) for concentration–response relationship of A9C and NFA in cavernosal further evaluation and statistical analysis. tissue; D, diabetic (n ¼ 5, N ¼ 5); ND, nondiabetic (n ¼ 7; N ¼ 6).

International Journal of Impotence Research (2014), 151 – 155 & 2014 Macmillan Publishers Limited Calcium-activated chloride channel in diabetes L-C Lau and PG Adaikan 153 Non-diabetic Frequency (Hz) 4 0 10203040 0 10 Control (ND) 3 20 NFA 10µM (ND) 30 NFA 30µM (ND) 2 40 NFA 100µM (ND) Tension (g) 50 1 % Relaxation 60 70

0 EFS: 20Hz 80 90

Non-diabetic Diabetic 120 Diabetic Frequency (Hz) 0 10203040 0 80 10 Control (D) 20 NFA 10µM (D) 40 NFA 30µM (D) 30 NFA 100µM (D) 40 % Control Response 0 10 30 100 50 NFA Concentration, µM

% Relaxation 60

Non-diabetic Diabetic 70 120 80 90 80 Figure 3. NFA at 100 mM augmented nitrergic relaxation in CC tissue from both the healthy, nondiabetic (a) and diabetic (b) rabbits. 40 *Pr0.05; ND, nondiabetic (n ¼ 7, N ¼ 5); D, diabetic (n ¼ 5, N ¼ 5).

% Control Response 0 100 300 1000 CC, suggesting that the capability of NFA in augmenting nitrergic A9C Concentration, µM relaxation was not adversely affected by the diabetic state. In fact, Figure 2. (a) Representative tracing demonstrating the suppressant NFA did appear to actively promote relaxation in response to a effect of A9C on contraction elicited by electrical field stimulation higher stimulus frequency, an effect not noted in nondiabetic CC. at 20 Hz frequency in CC tissue excised from diabetic rabbit. Most strikingly, A9C seems to be more effective in promoting The inhibitory effect exerted by A9C was reversible upon removal of nitrergic relaxation in CC from diabetic rabbits than in nondiabetic A9C from the organ bath. Both NFA (b) and A9C (c) were almost rabbits, alluding to an elevated sensitivity of CaCCs in association equipotent in inhibiting neurogenic contractions of CC tissue from with diabetes. diabetic (n ¼ 5, N ¼ 5) and nondiabetic (n ¼ 5–8, N ¼ 5) rabbits. The marked contrast in the results for the action of A9C between diabetic and nondiabetic CC tissues could possibly arise from an inherent weak inhibitory action of A9C at the CaCC in frequency–response curve to the left (Figure 3). Response to 20 Hz nondiabetic CC tissue. Compared with A9C, NFA has been stimulus frequency was also significantly enhanced by NFA in described as a more potent inhibitor of CaCC. This property may diabetic CC strips. By comparison, in nondiabetic CC, A9C (n ¼ 5; have contributed to the ease of demonstrating the augmentation 1000 mM) was totally inert on lower stimuli frequencies and effect of NFA on nitrergic relaxation in nondiabetic CC tissue. displayed mild potentiation of responses to stimulus frequencies There is hardly any literature with respect to the interaction of 20 Hz and above. However, in CC from diabetic rabbits, A9C between diabetes and CaCC in the contractility of smooth muscle markedly increased nitrergic relaxation response to 1, 2, 5 and preparations. However, in skeletal muscle contraction diabetes 10 Hz by 12.0%, 36.6%, 23.5% and 10.6%, respectively (Figure 4; may adversely affect the chloride channel activity giving rise to compared with À 4.8, À 5.1, 0.8 and À 3.5% in nondiabetic diabetic myopathy; such an alteration in membrane chloride ion controls). Statistical comparison of the relaxation of the control CC conductance was observed in muscle cells from the diaphragm of with the diabetic CC in the presence of A9C also showed that diabetic mice.19 In the same study, Lin-Shiau et al.19 also noted the nitrergic response to 1–10 Hz was significantly augmented that the diabetic mouse skeletal muscle was less responsive to the in diabetic CC. actions of the chloride channel blocker, A9C, suggesting that this abnormality could be attributed to a decrease in membrane chloride ion conductance. Current data on CC suggest that DISCUSSION diabetes influenced CaCCs in smooth muscles differently from The present study highlights the deferential effect of two widely those in skeletal muscles.19 used probes, NFA and A9C, in the functional role of CaCCs in CC Nonspecific effects independent of chloride channel blockade isolated from diabetic and nondiabetic rabbits. NFA enhanced have been reported for NFA and A9C. There is electrophysiological nitrergic relaxation in diabetic CC as much as it did in nondiabetic evidence suggesting that the fenamate family, which includes

& 2014 Macmillan Publishers Limited International Journal of Impotence Research (2014), 151 – 155 Calcium-activated chloride channel in diabetes L-C Lau and PG Adaikan 154 A9C 100µM 300µM 1000µM chloride current in sheep urethra and rat portal vein cells by patch clamp technique;22,23 it was also reported to inhibit the forskolin- 6 stimulated Ca2 þ current in guinea-pig heart muscle24 and to activate calcium dependent potassium current in rabbit portal vein smooth muscle cells.25 4 Although NFA and A9C are chemically dissimilar, they have been reported to act as open channel blockers of calcium-activated chloride current in vascular smooth muscle cells.25–27 Notably, the inhibitory effect of A9C is characterized to be highly voltage- 26 Tension (g) 2 dependent. This unique property may contribute to the susceptibility of CaCC to blockage by A9C if CC tissue is presumably more depolarized during diabetes than it is in the 0 1 5 20 1 5 20 1 5 20 1 5 20 nondiabetic state. Findings from a recent study in pulmonary artery 2 10 40 Hz 2 10 40 Hz 210 40Hz 210 40Hz smooth muscle cells seem to support such likelihood. Forrest et al.28 have acknowledged the existence of more depolarized pulmonary artery smooth muscle cells accompanied by an upregulation of Non-diabetic Frequency (Hz) CaCC expression/activity in a rat model of pulmonary hypertension. 010203040Whether CC myocytes from diabetic tissue are more depolarized 0 than those from healthy tissue awaits conﬁrmation from future

10 Control (ND) studies. Further study is also needed to confirm our observation of elevated CaCC sensitivity in diabetic CC, and to delineate the 20 A9C 100µM (ND) underlying mechanism in view of the report that A9C has been 30 A9C 300µM (ND) observed to have nonspecific inhibitory and stimulatory effects on Ca2 þ and calcium-dependent potassium current in heart muscles 40 A9C 1000µM (ND) andportalveinsmoothmusclecells,22,25 respectively, which could 50 account for the results obtained in the current study. Our results revealed that there was no significant alteration in % Relaxation 60 the efficacies of NFA and A9C in eliciting concentration- 70 dependent relaxation of noradrenaline-induced cavernosum tone, 80 although A9C appeared to be less potent in relaxing CC from the diabetic group than in the nondiabetic group. The efficacies of 90 both chloride channel blockers in inhibiting neurogenic contraction of CC were also not significantly influenced by the diabetic Diabetic Frequency (Hz) status of the animals, suggesting that calcium homeostasis is 010203040conserved in the diabetic animal compared with the nondiabetic 0 animal. Collectively, these results indicate that CaCC blockers are 10 Control (D) potential candidates in the development of new pharmaco- 20 A9C 100µM (D) therapies to overcome the deleterious effects of diabetes on A9C 300µM (D) penile function. Targeting the CaCC to achieve penile erection is 30 an alternative therapeutic strategy to obviate the dependence on A9C 1000µM (D) 40 the cavernosal nitric oxide synthase-cGMP pathway, which is invariably impaired in diabetes. 50 In conclusion, CaCC sensitivity to A9C appears to be enhanced

% Relaxation 60 in diabetic CC tissue. Inhibiting the CaCC activity in diabetes- 70 related ED may tip the balance between proerectile/relaxant and antierectile/contractile mechanisms in favor of cavernosum 80 relaxation. Furthermore, CaCC blockers theoretically have an 90 advantage over PDE5 inhibitors in rescuing erectile failures in diabetic condition via a mechanism independent of nitric oxide Figure 4. (a) Representative tracing showing the augmentation effect of A9C on nitrergic relaxation elicited by field stimulation at synthase system. varying frequencies, 1–40 Hz. In nondiabetic CC tissue (b), A9C did not significantly influence neurogenic relaxation, whereas in diabetic CC tissue (c), A9C was capable of potentiating neurogenic CONFLICT OF INTEREST relaxation provoked by stimulus frequencies 1–10Hz. *Pr0.05; ND, The authors declare no conflict of interest. nondiabetic (n ¼ 5, N ¼ 5); D, diabetic (n ¼ 5, N ¼ 5).

ACKNOWLEDGMENTS NFA, can activate large conductance Ca2 þ -activated K þ 20,21 This project was supported by a research grant from the Asia Pacific Society for channels; such effects could produce membrane hyper- Sexual Medicine. polarization, resulting in closure of voltage-dependent calcium channels and subsequently the relaxation of smooth muscle. In addition, NFA being a nonsteroidal anti-inflammatory drug is REFERENCES noted for inhibition of prostaglandin synthesis. However, the inability of tetraethylammonium, a K þ channel blocker, and indo- 1 Feldman HA, Goldstein I, Hatzichristou DG, Krane RJ, McKinlay JB. Impotence and its medical and psychosocial correlates: results of the Massachusetts male aging methacin, an inhibitor of cyclooxygenase, to significantly influence study. J Urol 1994; 151: 54–61. NFA-induced relaxation of noradrenaline tone in normal healthy 2 Behrend L, Vibe-Petersen J, Perrild H. Sildenafil in the treatment of erectile dys- CC excludes the likelihood of NFA acting through these function in men with diabetes: demand, efficacy and patient satisfaction. Int J 14 2 mechanisms. A9C has been shown to block Ca þ -activated Impot Res 2005; 17: 264–269.

International Journal of Impotence Research (2014), 151 – 155 & 2014 Macmillan Publishers Limited Calcium-activated chloride channel in diabetes L-C Lau and PG Adaikan 155 3 Angulo J, Cuevas P, Fernandez A, Gabancho S, Allona A, Martin-Morales A et al. 15 Chung SD, Kuo YC, Liu SP. The role of chloride channels in rat corpus cavernosum: Diabetes impairs endothelium-dependent relaxation of human penile vascular in vivo study. J Sex Med 2009; 6: 708–716. tissues mediated by NO and EDHF. Biochem Biophys Res Commun 2003; 312: 16 Lau LC, Adaikan PG, Armugam A, Jeyaseelan K. Characterising behaviour of 1202–1208. corpus cavernosum in chloride free condition. Urology 2011; 77: e17–e22. 4 Cellek S, Rodrigo J, Lobos E, Fernandez P, Serrano J, Moncada S. Selective nitrergic 17 Linton DJ, Lau LC, Adaikan PG. Calcium-activated chloride channels in the corpus neurodegeneration in diabetes mellitus—a nitric oxide-dependent phenomenon. cavernosum: recent developments and future of a key cellular component of the Br J Pharmacol 1999; 128: 1804–1812. erectile process. Int J Impot Res 2012; 24: 211–216. 5 Cellek S, Foxwell NA, Moncada S. Two phases of nitrergic neuropathy in 18 Graves JE, Greenwood IA, Large WA. Tonic regulation of vascular tone by nitric streptozotocin-induced diabetic rats. Diabetes 2003; 52: 2353–2362. oxide and chloride ions in rat isolated small coronary arteries. Am J Physiol Heart 6 Cameron NE, Cotter MA. Erectile dysfunction and diabetes mellitus: mechanistic Circ Physiol 2000; 279: H2604–H2611. considerations from studies in experimental models. Curr Diabetes Rev 2007; 3: 19 Lin-Shiau SY, Liu SH, Lin MJ. Use of ion channel blockers in the exploration of 149–158. possible mechanisms involved in the myopathy of diabetic mice. Naunyn 7 Sullivan ME, Dashwood MR, Thompson CS, Muddle JR, Mikhailidis DP, Morgan RJ. Schmiedebergs Arch Pharmacol 1993; 348: 311–318. Alterations in endothelin B receptor sites in cavernosal tissue of diabetic rabbits: 20 Ottolia M, Toro L. Potentiation of large conductance KCa channels by niflumic, potential relevance to the pathogenesis of erectile dysfunction. J Urol 1997; 158: flufenamic, and mefenamic acids. Biophys J 1994; 67: 2272–2279. 1966–1972. 21 Greenwood IA, Large WA. Comparison of the effects of fenamates on Ca-activated 8 Chang S, Hypolite JA, Changolkar A, Wein AJ, Chacko S, DiSanto ME. Increased chloride and potassium currents in rabbit portal vein smooth muscle cells. Br J contractility of diabetic rabbit corpora smooth muscle in response to endothelin Pharmacol 1995; 116: 2939–2948. is mediated via Rho-kinase beta. Int J Impot Res 2003; 15: 53–62. 22 Cotton KD, Hollywood MA, McHale NG, Thornbury KD. Ca2 þ current and 9 Zhang XH, Melman A, Disanto ME. Update on corpus cavernosum smooth muscle Ca(2 þ )-activated chloride current in isolated smooth muscle cells of the sheep contractile pathways in erectile function: a role for testosterone? J Sex Med 2011; urethra. J Physiol 1997; 505: 121–131. 8: 1865–1879. 23 Baron A, Pacaud P, Loirand G, Mironneau C, Mironneau J. Pharmacological block 10 Bivalacqua TJ, Champion HC, Usta MF, Cellek S, Chitaley K, Webb RC et al. RhoA/ of Ca(2 þ )-activated Cl- current in rat vascular smooth muscle cells in short-term Rho-kinase suppresses endothelial nitric oxide synthase in the penis: a mechan- primary culture. Pflugers Arch 1991; 419: 553–558. ism for diabetes-associated erectile dysfunction. Proc Natl Acad Sci USA 2004; 101: 24 Walsh KB, Wang C. Effect of chloride channel blockers on the cardiac CFTR 9121–9126. chloride and L-type calcium currents. Cardiovasc Res 1996; 32: 391–399. 11 Venkateswarlu K, Giraldi A, Zhao W, Wang HZ, Melman A, Spektor M et al. 25 Hogg RC, Wang Q, Large WA. Effects of Cl channel blockers on Ca activated Potassium channels and human corporeal smooth muscle cell tone: diabetes chloride and potassium currents in smooth muscle cells from rabbit portal vein. and relaxation of human corpus cavernosum smooth muscle by adenosine Br J Pharmacol 1994; 111: 1333–1341. triphosphate sensitive potassium channel openers. J Urol 2002; 168: 355–361. 26 Hogg RC, Wang Q, Large WA. Action of niflumic acid on evoked and spontaneous 12 Fan SF, Brink PR, Melman A, Christ GJ. An analysis of the Maxi-K þ (KCa) channel calcium-activated chloride and potassium currents in smooth muscle cells from in cultured human corporal smooth muscle cells. J Urol 1995; 153: 818–825. rabbit portal vein. Br J Pharmacol 1994; 112: 977–984. 13 Craven M, Sergeant GP, Hollywood MA, McHale NG, Thornbury KD. Modulation of 27 Piper AS, Greenwood IA, Large WA. Dual effect of blocking agents on Ca2 þ - spontaneous Ca2 þ -activated Cl- currents in the rabbit corpus cavernosum by the activated Cl(-) currents in rabbit pulmonary artery smooth muscle cells. J Physiol nitric oxide-cGMP pathway. J Physiol 2004; 556: 495–506. 2002; 539: 119–131. 14 Chu LL, Adaikan PG. Role of chloride channels in regulation of rabbit corpus 28 Forrest AS, Joyce TC, Huebner ML, Ayon RJ, Wiwchar M, Joyce J et al. Increased cavernosum tone: a potential therapeutic target for erectile dysfunction. J Sex TMEM16A-encoded calcium-activated chloride channel activity is associated with Med 2008; 5: 813–821. pulmonary hypertension. Am J Physiol Cell Physiol 2012; 303: C1229–C1243.

& 2014 Macmillan Publishers Limited International Journal of Impotence Research (2014), 151 – 155