European Journal of Pharmacology 762 (2015) 26–34

Contents lists available at ScienceDirect

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

Cardiovascular pharmacology

Nitric oxide modulation of endothelium-derived hyperpolarizing factor in agonist-induced depressor responses in anesthetized rats Shuhei Kobuchi a,b,n, Katsuyuki Miura c, Hiroshi Iwao b,d, Kazuhide Ayajiki a a

Department of Pharmacy, School of Pharmacy, Hyogo University of Health Sciences, 1-3-6 Minatojima, Chuo-ku, Kobe, Hyogo 650-8530, Japan Department of Pharmacology, Osaka City University Medical School, 1-4-3, Asahimachi, Abeno, Osaka 545-8585, Japan c Applied Pharmacology and Therapeutics, Osaka City University Medical School, 1-4-3, Asahimachi, Abeno, Osaka 545-8585, Japan d Department of Education, Shitennoji University, 3-2-1, Gakuenmae, Habikino, Osaka 583-8501, Japan b

art ic l e i nf o

a b s t r a c t

Article history: Received 11 November 2014 Received in revised form 30 April 2015 Accepted 30 April 2015 Available online 8 May 2015

Vasodilators, such as prostacyclin, nitric oxide (NO), and endothelium-derived hyperpolarizing factor (EDHF), released from the vascular endothelium are important in the maintenance of systemic blood pressure. Some studies have shown that NO affects EDHF-induced vasodilator responses in isolated perfused blood vessel segments. However, the effects of NO on EDHF-mediated dilation, and their contribution to systemic blood pressure, have not been clarified. Therefore, in the present study we investigated the mechanisms underlying acetylcholine- and bradykinin-induced depressor responses, as well as the interaction between NO and EDHF, by measuring systemic blood pressure in anesthetized rats. In the presence of indomethacin and NG-nitro-L-arginine (L-NA; an NO synthase inhibitor), apamin plus charybdotoxin significantly inhibited depressor responses to acetylcholine and bradykinin, whereas glibenclamide, iberiotoxin, quinacrine, catalase, and combination of ouabain plus BaCl2 failed to inhibit EDHF-induced depressor responses. 4-Aminopyridine significantly inhibited depressor responses to acetylcholine, but not to bradykinin. In the presence of indomethacin and L-NA, carbenoxolone, a gap junction inhibitor, significantly inhibited depressor responses to agonists. L-NA alone significantly potentiated agonist-induced depressor responses. In contrast, infusion of sodium nitroprusside, an NO donor, or 8-br-cGMP significantly inhibited depressor responses to agonist. The findings of the present study raise the possibility that agonist-induced depressor responses are elicited by propagation of endothelial hyperpolarization via apamin- plus charybdotoxin-sensitive K þ channels to smooth muscle cells through gap junctions, but not by diffusible substance(s). It is suggested that, in anesthetized rats, the EDHF-induced depressor response is attenuated in the presence of endogenous and exogenous NO via an increment in cGMP. & 2015 Elsevier B.V. All rights reserved.

Keywords: Nitric oxide (NO) Endothelium-derived hyperpolarizing factor (EDHF) Ca2 þ -activated K þ channel Gap junction cGMP

1. Introduction Endothelium-dependent relaxations induced by vasodilator substances such as acetylcholine or bradykinin are mediated by prostacyclin (PGI2), nitric oxide (NO), and endothelium-derived hyperpolarizing factor (EDHF). It has been proposed that the vasodilator action of EDHF, which hyperpolarizes cell membranes and induces smooth muscle relaxation, is mediated via initial activation of small and intermediate conductance Ca2 þ -activated potassium channels (KCa2.3, KCa3.1) present on the endothelium (Félétou and Vanhoutte, 2009). Variable responses to EDHF have n Corresponding author at: Department of Pharmacy, School of Pharmacy, Hyogo University of Health Sciences, 1-3-6 Minatojima, Chuo-ku, Kobe, Hyogo 650–8530, Japan. Tel.: þ 81 78 304 3164; fax: þ 81 78 304 2864. E-mail address: [email protected] (S. Kobuchi).

http://dx.doi.org/10.1016/j.ejphar.2015.04.053 0014-2999/& 2015 Elsevier B.V. All rights reserved.

been reported among species and vascular beds, and possible candidates for EDHF include epoxyeicosatrienoic acid (a metabolite of cytochrome P450 epoxygenase; Fisslthaler et al., 1999), endothelium-derived potassium ions (K þ ; Edwards et al., 1998), and hydrogen peroxide (H2O2; Matoba et al., 2000). It has also been suggested that the EDHF phenomenon is primarily electrical in origin and driven by endothelial hyperpolarization that spreads radially into the vessel wall via myoendothelial gap junctions (Griffith, 2004). EDHF primarily mediates vasodilation in resistance arteries, whereas NO primarily mediates vasodilation in conduit arteries (Shimokawa et al., 1996). The nature of EDHF has often been investigated using isolated arteries and arterioles; however, systematic analyses of EDHF in anesthetized rats are rarely reported because in vivo studies of EDHF measure localized blood flow, such as in the rat mesenteric, hindlimb, and sciatic nerve circulation (Parkington et al., 2002; Thomsen et al., 2000).

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

Desai et al. (2006) reported that EDHF-induced depressor responses in anesthetized rats were inhibited by combination treatment with apamin plus charybdotoxin, suggesting that EDHF-mediated systemic blood pressure responses are also mediated by KCa2.3 and KCa3.1. However, identification of EDHF in anesthetized rats remains contentious. Nitric oxide (NO) is a well-known vasodilator substance, but it may also modulate EDHF-type responses. Some investigators have demonstrated that EDHF-type responses are potentiated by inhibition of NO synthase (NOS) and reduced by exogenous NO, 8-brcGMP, or the overproduction of NO induced by proinflammatory mediators in isolated perfused arterial segments (Kessler et al., 1999; Bauersachs et al., 1996). In addition, Nishikawa et al. (2000) proposed that EDHF served as a back-up vasodilator when NO production is impaired, because a small dose of an NO donor almost completely inhibited EDHF-type responses in the canine coronary microcirculation. Together, the findings of these studies indicate that NO inhibits EDHF-induced dilation in isolated arteries, but the interaction between NO and EDHF in the maintenance of systemic blood pressure has not been clarified. Thus, the aim of the present study was to investigate the mechanisms underlying EDHF-type depressor responses induced by different agonists in anesthetized rats and to investigate the effect of endogenous and exogenous NO on EDHF-induced responses.

2. Materials and methods 2.1. Animals and experimental design Male Sprague-Dawley rats (300–400 g; Japan SLC, Shizuoka, Japan) were used in the present study. Rats were housed in a lightcontrolled room under a 12-h light–dark cycle and were allowed free access to food and water. Experimental protocols and animal care methods were approved by the Animal Experimental Committee at Hyogo University of Health Sciences (Hyogo, Japan). Rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and the femoral artery and vein were cannulated with a polyethylene catheter. The arterial catheter was connected to a pressure transducer (Deltran II; Utah Medical Products, Midvale, UT, USA) and systemic blood pressure was monitored continuously on a PowerLab (ML870; ADInstruments, Bella Vista NSW, Australia). All drugs were administered through the venous catheter. Acetylcholine and bradykinin were used as agonists in the present study; both cause endothelium-dependent vasodilation. All rats were subsequently given bolus injections of acetylcholine (0.1, 0.3, 1 and 3 μg/kg, 1 ml/kg, at 5-min intervals) and bradykinin (1, 3, 10 and 30 μg/kg, 1 ml/kg, at 5-min intervals). Responses to each dose of acetylcholine and bradykinin were recorded as the maximum fall in mean arterial blood pressure immediately after drug administration. Infusion of drug was performed through contralateral venous catheter. Five different experiments were performed, as detailed below. 2.1.1. Effects of blocking agents on agonist-induced depressor responses in the presence of indomethacin plus NG-nitro-L-arginine The effects of K þ channel inhibitors and putative EDHF inhibitors were examined to identify EDHF. Various EDHFblocking drugs have been reported by Fujioka et al. (2002). In the present study, we used carbenoxolone, a water-soluble gap junction inhibitor, rather than 18α-glycyrrhetinic acid (another gap junction inhibitor), which is not soluble in water. Rats were pretreated with indomethacin (10 mg/kg, a cyclooxygenase inhibitor), and NG-nitro-L-arginine (L-NA; 10 mg/kg, a NOS inhibitor), 30 min before construction of dose–response

27

curves to acetylcholine or bradykinin. Rats were then pretreated for 10 min with blocking agents, 4-aminopyridine (a voltagedependent K þ channel inhibitor), glibenclamide (an ATPsensitive K þ channel inhibitor), apamin (a KCa2.3 inhibitor), charybdotoxin (a KCa3.1 inhibitor), iberiotoxin (a large conductance Ca2 þ -activated K þ channel [KCa1.1] inhibitor), catalase (an H2O2-degrading enzyme), quinacrine (a phospholipase A2 inhibitor), ouabain (an Na þ /K þ -ATPase inhibitor) plus BaCl2 (an inhibitor of inward rectifier K þ channels) or carbenoxolone, before a second dose–response curve to acetylcholine or bradykinin was obtained. The doses of the blocking agents indomethacin, L-NA, 4-aminopyridine, glibenclamide, apamin, charybdotoxin, iberiotoxin, and carbenoxolone used in the present study were based on previous reports (Ayajiki et al., 2005; Berg and Koteng, 1997; Edgley et al., 2008; Desai et al., 2006, Gigout et al., 2006; Smits et al., 1997). The doses of catalase (67500–87500 U/kg), quinacrine (0.8–1.0 mg/kg), ouabain (0.39–0.51 mg/kg), and BaCl2 (0.40– 0.51 mg/kg) used in the present in vivo studies were extrapolated from concentrations of each drug used in in vitro studies (1250 U/mL, 30 μM, 10 μM, and 30 μM, respectively; Matoba et al., 2000, Fujioka et al., 2002) taking blood volume (54–70 mL/kg) into account because there are no previous reports of in vivo doses. Moreover, for some drugs we used a higher dose than the extrapolated dose to confirm effectiveness within a range that did not elicit death. 2.1.2. Effects of carbenoxolone on sodium nitroprusside- and salbutamol-induced depressor responses in the presence of indomethacin plus L-NA Non-specific effects of carbenoxolone were evaluated using sodium nitroprusside and salbutamol, which cause endotheliumindependent vasodilation. We hypothesized that carbenoxolone treatment would have no effect on sodium nitroprusside- and salbutamol-induced depressor responses. Rats were pretreated with a combination of indomethacin plus L-NA, followed 30 min later by construction of a dose–response curve to sodium nitroprusside (an exogenous source of NO; 1, 3, 10, 30 μg/kg; injected at 5-min intervals) or the selective β2adrenoceptor agonist salbutamol (0.1, 0.3, 1, 3 μg/kg; injected at 5-min intervals). Rats were then treated with carbenoxolone (150 mg/kg) followed, 10 min, by a second dose–response curve to sodium nitroprusside or salbutamol. 2.1.3. Effects of L-NA on agonist-induced depressor responses in the presence of indomethacin The effects of endogenous NO on EDHF-induced depressor responses were evaluated using L-NA, a NOS inhibitor. We hypothesized that L-NA treatment would potentiate EDHFinduced depressor responses. Rats were pretreated with indomethacin, followed by construction of dose–response curves to acetylcholine or bradykinin. Then, rats were treated with L-NA or phenylephrine (8–15 μg/kg per min, i.v. infusion to adjust basal mean arterial blood pressure) followed, 30 min later, by a second dose–response curve to acetylcholine or bradykinin. 2.1.4. Effects of sodium nitroprusside or 8-br-cGMP on agonistinduced depressor responses in the presence of indomethacin plus LNA The effects of exogenous NO or cGMP on EDHF-induced depressor responses were evaluated using the sodium nitroprusside (NO donor) and 8-br-cGMP (a cell-permeable cGMP analog). We hypothesized that treatment with either sodium nitroprusside or 8-br-cGMP would inhibit EDHF-induced depressor responses.

28

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

Table 1 Changes in mean arterial blood pressure in response to drug treatment in anesthetized rats. MAP (mmHg) Basal Indomethacin (cyclo-oxygenase inhibitor ) plus L-NA (NOS inhibitor) 4-AP (voltage-dependent K þ channel inhibitor) Glibenclamide (KATP channel inhibitor) Catalase (H2O2-degrading enzyme) Quinacrine (phospholipase A2 inhibitor) Ouabain (Na þ /K þ -ATPase inhibitor) þBaCl2 (inhibitor of inward rectifier K þ channels) Apamin (KCa2.3 inhibitor) Charybdotoxin (KCa3.1 inhibitor) Iberiotoxin (KCa1.1 inhibitor) Apamin þcharybdotoxin Carbenoxolone (gap junction inhibitor) SNP (NO donor) þ PE (α1-adrenoceptor agonist) 8-br-cGMP (cGMP analog) Indomethacin pretreatment L-NA PE

Treatment

pretreatment 130 7 4 150 77b 125 7 8 129 78 1217 5 125 75 1417 7 139 78 1137 4 137 76b 1287 7 1287 7 1327 5 1327 6 142 7 5 1527 3 152 7 6 159 7 3

130 78 138 79a 136 75 147 77a 140 76 78 78b 148 73a

104 7 4 136 72b 104 7 6 136 75b

Data are the mean 7 S.E.M. (n¼ 10). Mean arterial blood pressure (MAP) was measured 10 min after injection of drugs, except in the case of sodium nitroprusside (SNP) þphenylephrine (PE), 8-br-cGMP, NG-nitro-L-arginine (L-NA), and PE (MAP measured 30 min after drug injection). 4-AP, 4-aminopyridine; KCa, Ca2 þ -activated K þ channel; NOS, nitric oxide synthase. KATP, ATPsensitive K þ channel inhibitor a b

P o 0.05. Po 0.01 compared with basal.

Rats were pretreated with indomethacin and L-NA, followed 30 min later by the construction of dose–response curves to acetylcholine and bradykinin. Rats were then treated with sodium nitroprusside (10 μg/kg per min, i.v. infusion) plus phenylephrine (15–20 μg/kg per min, i.v. infusion) or 8-br-cGMP (0.3 mg/kg/min, i.v. infusion), followed 30 min later by construction of a second dose–response curve to acetylcholine or bradykinin. Phenylephrine was used to recover sodium nitroprusside-induced reductions in basal mean arterial blood pressure. The doses of sodium nitroprusside and 8-br-cGMP used in the present study were based on previous reports (Chintala et al., 1994). 2.1.5. Effects of L-NA on 1-ethyl-2-benzimidazolinone-induced depressor responses in the presence of indomethacin To determine whether the inhibitory effect of NO was mediated by changes in Ca2 þ influx into endothelial cells, we used 1-ethyl2-benzimidazolinone (1-EBIO; an activator of KCa2.3 and KCa3.1). We hypothesized that L-NA treatment would potentiate 1-EBIOinduced depressor responses. A dose–response curve was constructed to 1-EBIO (0.1, 0.3, 1 and 3 mg/kg, 1 mL/kg; injected at 5-min intervals) in the presence of indomethacin. Rats were then treated with L-NA, followed 30 min later by construction of a second dose–response curve to 1-EBIO. 2.2. Drugs Bradykinin acetate, indomethacin, L-NA, 4-aminopyridine, and ouabain octahydrate were obtained from Sigma Chemical (St. Louis, MO, USA). 1-EBIO was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Apamin, charybdotoxin, and iberiotoxin were purchased from the Peptide Institute (Minoh, Japan). Acetylcholine chloride and BaCl2 were obtained from Nacalai Tesque (Kyoto, Japan). Glibenclamide and quinacrine dihydrochloride

were obtained from Wako Pure Chemical Industries (Osaka, Japan). Carbenoxolone disodium was purchased from LKT Laboratories (St. Paul, MN, USA). Indomethacin was dissolved in 3% Na2CO3, glibenclamide was dissolved in 2.5% dimethylsulfoxide, and all other drugs were dissolved in saline (0.9%). 2.3. Statistical analysis All values are presented as the mean 7S.E.M. Relevant data were processed by GraphPad Prism (GraphPad Software, La Jolla, CA, USA). Unpaired Student's t-test was used for comparisons between experimental and control groups. For all comparisons, two-sided P o0.05 was considered significant.

3. Results 3.1. Effects of blocking agents on agonist-induced depressor responses in the presence of indomethacin plus L-NA Although some blocking agents caused a transient increase in mean arterial blood pressure, 4-aminopyridine, ouabain plus BaCl2, charybdotoxin, and apamin plus charybdotoxin resulted in a sustained increase in mean arterial blood pressure that did not recover to baseline levels (Table 1). Treatment with glibenclamide, catalase, quinacrine, apamin, iberiotoxin, and carbenoxolone did not increase mean arterial blood pressure (Table 1). In the presence of indomethacin plus L-NA, 4-aminopyridine (5 mg/kg) slightly but significantly inhibited acetylcholine-induced depressor response, but had no effect on bradykinin-induced depressor responses (Fig. 1A and B). In the presence of indomethacin plus L-NA, glibenclamide (10 mg/kg), catalase (80,000 U/kg), quinacrine (5 mg/kg), and ouabain (1 mg/kg) plus BaCl2 (5 mg/kg) failed to inhibit depressor responses to acetylcholine and bradykinin (Fig. 1A–D). In the presence of indomethacin plus L-NA,

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

0

control

control

-10

4-aminopyridine

-10

4-aminopyridine

-20

glibenclamide

-20

glibenclamide

*

-30

**

-40

*

MAP (%)

MAP (%)

0

-30 -40

-50

-50

-60

-60

-70

-70

0.1

1

0.3

1

3

control catalase quinacrine ouabain + BaCl2

-10

-20

30

-30 -40

control catalase quinacrine ouabain + BaCl2

0 -10

MAP (%)

0

10

3

BK (µg/kg)

ACh (µg/kg)

MAP (%)

29

-20 -30

-50

-50

-60

-60

-70

*

-40

-70

0.1

0.3

1

3

1

10

3

30

BK (µg/kg)

ACh (µg/kg)

Fig. 1. (A, B) Effects of 4-aminopyridine (5 mg/kg) and glibenclamide (10 mg/kg) on percentage change in mean arterial blood pressure (MAP) in response to (A) acetylcholine (ACh) and (B) bradykinin (BK) in the presence of indomethacin plus NG-nitro-L-arginine (L-NA). (C, D) Effects of catalase (80,000 U/kg), quinacrine (5 mg/kg), and ouabain (1 mg/kg) plus BaCl2 (5 mg/kg) on percentage change in MAP in response to ACh (C) and BK (D) in the presence of indomethacin plus L-NA. Data are the mean7 S.E.M. (n¼ 5). nP o 0.05, nnPo 0.01 compared with control.

control apamin charybdotoxin iberiotoxin apamin + charybdotoxin

** 0 -10

**

-10

**

-20

-20

* -30

**

-40

MAP (%)

MAP (%)

**

0

**

-30 -40

-50

-50

-60

-60

-70

**

*

-70 0.1

0.3

1

3

ACh (µg/kg)

1

3

10

30

BK (µg/kg)

Fig. 2. Effects of apamin (25 μg/kg), charybdotoxin (25 μg/kg), iberiotoxin (25 μg/kg), and apamin plus charybdotoxin on percentage change in mean arterial blood pressure (MAP) in response to (A) acetylcholine (ACh) and (B) bradykinin (BK) in the presence of indomethacin plus NG-nitro-L-arginine. Data are the mean 7S.E.M. (n¼ 5). nP o 0.05, nn P o 0.01 compared with control.

combined treatment with apamin (25 μg/kg) plus charybdotoxin (25 μg/kg) significantly inhibited depressor responses to acetylcholine and bradykinin, whereas apamin or iberiotoxin (25 μg/kg) alone had no effect (Fig. 2). In the presence of indomethacin plus L-NA, charybdotoxin alone significantly inhibited the depressor

response to 30 μg/kg bradykinin, but had no effect on depressor responses to acetylcholine or the lower doses of bradykinin (Fig. 2). In addition, combined treatment with apamin plus iberiotoxin failed to inhibit acetylcholine- or bradykinin-induced depressor response in the presence of indomethacin plus L-NA (data not

30

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

250

250

200

200

ABP(mmHg)

ABP(mmHg)

3

1

0.3

0.1

150 100 50

100

ACh (µg/kg)

1 min

1 min

0

0

30

10

3

1 250

250

200

200

ABP(mmHg)

ABP(mmHg)

150

50

ACh (µg/kg)

3

1

0.3

0.1

150 100

100

0

BK (µg/kg)

1 min

BK (µg/kg)

30

150

50

50

10

3

1

1 min

0

Fig. 3. (A, C) Representative traces of arterial blood pressure (ABP) recordings following injection of (A) acetylcholine (ACh) and (C) bradykinin (BK) in the presence of indomethacin plus NG-nitro-L-arginine (L-NA). (B, D) Representative traces showing the effects of carbenoxolone on ABP responses to ACh (B) and BK (D) in the presence of indomethacin plus L-NA.

control carbenoxolone 50 mg/kg carbenoxolone 150 mg/kg

0

0

**

-10

*

*

-10

**

*

-20

**

-30

** -40

MAP (%)

MAP (%)

-20

-40

-50

-50

-60

-60

-70

*

-30

-70

0.1

0.3

1

3

ACh (µg/kg)

1

3

10

30

BK (µg/kg)

Fig. 4. Effects of carbenoxolone on percentage change in mean arterial blood pressure (MAP) in response to (A) acetylcholine (ACh) and (B) bradykinin (BK) in the presence of indomethacin plus NG-nitro-L-arginine. Data are the mean 7 S.E.M. (n ¼5). nP o0.05, nnPo 0.01 compared with control.

shown). In the presence of indomethacin plus L-NA, high-dose carbenoxolone (150 mg/kg) significantly inhibited depressor responses to acetylcholine and bradykinin (Figs. 3 and 4). 3.2. Effects of carbenoxolone on sodium nitroprusside- and salbutamol-induced depressor responses in the presence of indomethacin plus L-NA In the presence of indomethacin plus L-NA, 150 mg/kg carbenoxolone significantly potentiated sodium nitroprusside-induced depressor responses, but had no effect on responses to salbutamol (Table 2). 3.3. Effects of L-NA on agonist-induced depressor responses in the presence of indomethacin In a preliminary study, we observed that indomethacin (10 mg/kg) had no effect on the depressor responses to acetylcholine and

bradykinin or on basal mean arterial blood pressure (data not shown). However, L-NA caused a significant and sustained increase in basal mean arterial blood pressure (Table 1). To adjust for this increase in basal mean arterial blood pressure, rats were infused with phenylephrine (8–15 μg/kg per min, i.v.) in the presence of indomethacin (Table 1). Under these conditions, L-NA significantly potentiated the depressor responses to acetylcholine and bradykinin compared with the responses seen during phenylephrine infusion in the presence of indomethacin (Fig. 5, Fig. 6A and B). 3.4. Effects of sodium nitroprusside and 8-br-cGMP on agonistinduced depressor responses in the presence of indomethacin plus L-NA Infusion of sodium nitroprusside (10 μg/kg per min) decreased mean arterial blood pressure (Table 1), but additional infusion of phenylephrine (15–20 μg/kg per min) recovered mean arterial blood

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

pressure to basal levels (Table 1). In the presence of indomethacin plus L-NA, sodium nitroprusside significantly inhibited depressor responses to acetylcholine and bradykinin after adjustment of mean arterial blood pressure to basal levels with phenylephrine (Fig. 6C and D). Similarly, in the presence of indomethacin plus L-NA, 8-brcGMP (0.3 mg/kg per min) significantly inhibited depressor responses to acetylcholine and bradykinin (Fig. 6C and D). 3.5. Effects of L-NA on 1-EBIO-induced depressor responses in the presence of indomethacin In the presence of indomethacin, 1-EBIO dose-dependently reduced mean arterial blood pressure, and this response was significantly potentiated by L-NA treatment (Table 3).

4. Discussion 4.1. Identification of EDHF In the present study, 4-aminopyridine (a voltage-dependent K þ channel inhibitor), ouabain (Na þ /K þ -ATPase inhibitor) plus BaCl2 Table 2 Effects of carbenoxolone (150 mg/kg) on percentage change in mean arterial blood pressure in response to sodium nitroprusside or salbutamol in the presence of indomethacin plus NG-nitro-L-arginine. ΔMAP (%)

Sodium nitroprusside (μg/kg) 1 3 10 30 Salbutamol (μg/kg) 0.1 0.3 1 3

Control

þ Carbenoxolone

–11.9 7 1.8 –23.4 7 3.3 –42.2 7 2.5 –57.7 7 1.2

–14.17 1.3 –31.2 7 2.0 –51.7 7 0.6b –62.8 7 1.3a

–14.2 7 2.3 –25.17 2.7 –29.8 7 3.9 –41.2 7 2.4

–12.9 7 1.3 –21.9 7 2.0 –30.6 7 2.8 –34.0 7 3.5

ΔMAP, change in mean arterial blood pressure. Data are the mean 7 S.E.M. (n¼ 10). b

P o0.05. Po 0.01 compared with control.

ABP(mmHg)

250

0.1

1

0.3

(inward rectifier K þ channel inhibitor) charybdotoxin (a KCa3.1 inhibitor) and apamin (a KCa2.3 inhibitor) plus charybdotoxin caused a sustained increase in baseline mean arterial blood pressure, presumably because of an increment in intracellular K þ resulting in membrane depolarization and vasoconstriction. Arachidonic acid metabolites derived from the P450 monooxygenase and lipoxygenase pathways, H2O2, and K þ are all considered candidates for EDHF. Epoxyeicosatrienoic acid, 12(S)hydroxyeicosatetraenoic (12-lipoxygenase metabolite), and H2O2 have been reported to activate KCa1.1 channels in vascular smooth muscle cells in mouse isolated mesenteric arterioles and porcine coronary arteries (Huang et al., 2005; Thengchaisri and Kuo, 2003; Zink et al., 2001). K þ activates Na þ /K þ -ATPase and inward rectifier K þ channels to cause hyperpolarization in rat isolated mesenteric or hepatic arteries (Edwards et al., 1998). However, in the present study, quinacrine (a phospholipase A2 inhibitor), catalase, iberiotoxin (a KCa1.1 inhibitor), and ouabain plus BaCl2 failed to inhibit agonist-induced depressor responses. In addition, C-type natriuretic peptide (CNP) and hydrogen sulfide (H2S) have been proposed as candidates for EDHF. It was reported that CNP and H2S cause hyperpolarization via activation of ATP-sensitive K þ channels in isolated guinea-pig carotid arteries and rat mesenteric arteries (Cheng et al., 2004; Leuranguer et al., 2008). However, in the present study, glibenclamide, an inhibitor of ATP-sensitive K þ channels, failed to inhibit agonist-induced depressor responses. These results suggest that arachidonate metabolites, H2O2, K þ , CNP, and H2S do not play an important role in the depressor actions of acetylcholine and bradykinin in anesthetized rats. It was reported that anesthetic drugs in general inhibit EDHF-induced dilator responses (Bryan et al., 2005). De Wit et al. (1999) demonstrated attenuation of EDHF-induced vasodilation by pentobarbital in hamster skin muscle arteriole, presumably mediated by a reduction in P450 metabolites. Thus, the possibility of vasodilation induced by P450 metabolites cannot be ruled out in the present study. However, De Wit and Griffith (2010) reported that epoxyeicosatrienoic acid (P450 metabolite) may not consistently contribute to the EDHF phenomenon. In rat isolated mesenteric arteries, Doughty et al. (2000) proposed that K þ did not mimic EDHF. In many studies, EDHF-induced vasodilation was not necessarily associated with activation of ATP-sensitive Kþ channels or KCa1.1 (Félétou, 2011; Ozkor and Quyyumi, 2011). Therefore, it seems unlikely that these diffusible substances are EDHF in the rat.

3

250

200

ABP(mmHg)

a

150 100 50

3

10

150 100 50 0

3

200

100 50

250

200

BK (µg/kg)

1

150

30

ABP(mmHg)

ABP(mmHg)

1

0.3

1 min

ACh (µg/kg)

0

0

250

0.1

1 min

ACh (µg/kg)

31

1

3

10

30

200 150 100 50

1 min

BK (µg/kg)

1 min

0

Fig. 5. (A, C) Representative traces of arterial blood pressure (ABP) recordings following injection of (A) acetylcholine (ACh) and (C) bradykinin (BK) in the presence of indomethacin in rats treated with phenylephrine (8–15 μg/kg per min, i.v. infusion) to adjust for NG-nitro-L-arginine (L-NA)-induced increases in blood pressure. (B, D) Representative traces showing the effects of L-NA on ABP responses to ACh (B) and BK in the presence of indomethacin.

32

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

0

PE

-20 -30 -40 -50

* 0.3

-30 -40

**

1

**

-60

*

-70 0.1

PE

-20

-50

*

-60

L-NA

-10

MAP (%)

MAP (%)

0

L-NA

-10

*

-70 1

3

control SNP + PE 8-br-cGMP

† ††

-20

††

-30

††

-40

††

††

††

-50 -60

0

††

-10

MAP (%)

MAP (%)

-10

10

30

BK (µg/kg)

ACh (µg/kg) 0

3

control SNP + PE 8-br-cGMP

††

-20

††

-30 -40

†† †† ††

-50

††

-60

-70

-70

0.1

0.3

1

3

1

3

10

30

BK (µg/kg)

ACh (µg/kg)

Fig. 6. (A, B) Effects of NG-nitro-L-arginine (L-NA) and phenylephrine (PE; 8–15 μg/kg per min, i.v. infusion) on percentage change in mean arterial blood pressure (MAP) in response to (A) acetylcholine (ACh) and (B) bradykinin (BK) in the presence of indomethacin. PE was used to adjust for L-NA-induced increases in blood pressure. (B, D) Effects of sodium nitroprusside (SNP; 10 μg/kg per min) and 8-br-cGMP (0.3 mg/kg per min) on percentage change in MAP in response to ACh (C) and BK (D) in the presence of indomethacin plus L-NA. PE (15–20 μg/kg per min, i.v. infusion) was used to adjust for SNP-induced reductions in basal MAP. Data are the mean7 S.E.M. (n¼5). nPo 0.05, nn P o0.01 compared with PE; †Po 0.05, ††P o 0.01 compared with control.

Table 3 Effects of treatment with NG-nitro-L-arginine on percentage change in mean arterial blood pressure in response to 1-ethyl-2-benzimidazolinone in the presence of indomethacin. ΔMAP (%) 1-EBIO (mg/kg)

Control þ L-NA

0.1

0.3

1

3

–1.5 7 1.5 –8.8 7 0.5b

–7.1 71.4 –21.1 73.5a

–25.2 71.6 –39.3 71.7b

–46.6 7 0.8 –57.5 7 1.1b

ΔMAP, change in mean arterial blood pressure; 1-EBIO, 1-ethyl-2-benzimidazolinone; L-NA, NG-nitro-L-arginine. Data are the mean 7 S.E.M. (n¼ 10). a b

P o0.05. Po 0.01 compared with control.

Voltage-dependent K þ channels are activated by nitroxyl (HNO) in rat small mesenteric arteries (Irvine et al., 2003). In the present study, 4-aminopyridine (a voltage-dependent K þ channel inhibitor) partially inhibited acetylcholine-induced depressor responses, but not those to bradykinin, suggesting that acetylcholine-induced depressor responses are partly mediated by the opening of voltage-dependent K þ channels, possibly via the release of HNO. Desai et al. (2006) demonstrated that agonist-induced depressor responses in anesthetized rats were inhibited by apamin plus charybdotoxin. In the present study, we also showed that combination treatment with apamin plus charybdotoxin significantly inhibited the depressor responses to acetylcholine and bradykinin, whereas apamin, charybdotoxin, or iberiotoxin alone or the combination of apamin plus iberiotoxin failed to inhibit the depressor response. These findings suggest that the depressor responses to acetylcholine and bradykinin are mediated by KCa2.3 and KCa3.1 but not KCa1.1, as reported previously (Félétou and

Vanhoutte, 2006). KCa2.3 and KCa3.1 are present in endothelial cells in various arteries (Félétou, 2011). Zygmunt et al. (1997) reported that apamin increased the density of charybdotoxin binding sites, suggesting that there may be a compensatory increase in the function of KCa2.3 or KCa3.1 in response to treatment with apamin or charybdotoxin alone and that combined blockade of both channels is necessary for inhibition of agonistinduced depressor responses. Carbenoxolone, a gap junction inhibitor, significantly inhibited the depressor responses to acetylcholine and bradykinin, suggesting that these responses are mediated by gap junctions. However, the mechanisms responsible for the depressor responses in anesthetized rats that are resistant to apamin plus charybdotoxin and carbenoxolone remain unclear. Carbenoxolone, a water-soluble derivative of 18β-glycyrrhetic acid, is known to inhibit gap junction coupling. Carbenoxolone has been shown to inhibit the transfer of dye from endothelial cells to neighboring endothelial cells through gap junctions in cultured bovine aortic endothelial cells and isolated endothelial cell tubes from mouse superior epigastric arteries (Behringer et al., 2012; Sagar and Larson, 2006), as well as the transmission of endothelial hyperpolarization to the smooth muscle in isolated guinea-pig carotid arteries and rat mesenteric and hepatic arteries (Edwards et al., 1999). It has been reported that carbenoxolone enhances the effects of endogenous corticosterone by inhibiting 11βhydroxysteroid dehydrogenase (Zhang et al., 2006) and that it blocks purinergic P2  7 (Suadicani et al., 2006) and glutamate Nmethyl-D-aspartate (NMDA) receptors (Chepkova et al., 2008). In the present study, carbenoxolone potentiated the depressor responses to sodium nitroprusside, but had no effect on responses to salbutamol. These results suggest that carbenoxolone enhances the sensitivity of the guanylate cyclase–cGMP pathway but not the adenylate cyclase–cAMP pathway. Thus, a potential non-specific, non-gap junctional effect of carbenoxolone cannot be ruled out. One of the limitations of the present study is that other gap junction inhibitors, such as 18α- or 18β-glycyrrhetic acid and oleamide, could not be used because they are not soluble in water. More specific and

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

water-soluble inhibitors are needed to investigate the physiological role of gap junction in vivo. Gap junctions are composed of two connexons that are themselves made up of six connexin molecules (Rummery and Hill, 2004). Although a large number of connexins has been identified, connexins (C  ) 37, C  40, and C  43 are primarily expressed in the mammalian vasculature (Christ et al., 1996). It has been reported that connexin-mimetic peptides, such as 43Gap 27 or 40 Gap 27, inhibit PGI2- and NO-independent renal vasodilation in rats (De Vriese et al., 2002). Figueroa and Duling (2008) demonstrated that the spread of dilation induced by local application of acetylcholine was attenuated in Cx40-deficient mice but not Cx37deficient mice. In addition, C  40-deficient mice were hypertensive (Félétou, 2011). These findings indicate that C  40 plays an important role in agonist-induced vasodilation and/or in the control of systemic blood pressure.

33

production in rats with congestive heart failure, rats fed a highsalt diet, and in mice lacking endothelial NOS (Malmsjö et al., 1999; Sofola et al., 2002; Waldron et al., 1999). This upregulation of EDHF-mediated vasodilation seems to be related to inhibition of EDHF by NO. Chadha et al. (2011) reported that EDHF-type vasodilation of human isolated mesenteric arteries was mediated by gap junctions, suggesting that understanding the role of gap junctions in endothelial dysfunction could contribute to the development of novel therapies. In conclusion, EDHF-induced depressor responses are mediated primarily by the propagation, via gap junctions, of endothelial hyperpolarization caused by the opening of small and intermediate conductance Ca2 þ -activated K þ channels on smooth muscle cells. Further, endogenous and exogenous NO exerts inhibitory effects on EDHF-mediated depressor responses via the NO–cGMP pathway.

4.2. Modification of EDHF by NO Acknowledgments In a preliminary study (data not shown), indomethacin failed to affect agonist-induced depressor responses, suggesting that the responses are not caused by cyclo-oxygenase products. In the present study, L-NA, an inhibitor of NOS, significantly potentiated the depressor responses to acetylcholine and bradykinin in anesthetized rats compared with responses seen after adjustment of mean arterial blood pressure by phenylephrine infusion. These findings suggest that the agonist-induced depressor responses, possibly mediated through EDHF, may be inhibited by NO. Furthermore, sodium nitroprusside (an NO donor) and 8-brcGMP (a cGMP analog) significantly inhibited the depressor responses in the presence of indomethacin plus L-NA. These findings suggest that both endogenous and exogenous NO inhibits depressor responses induced by EDHF via the NO–cGMP pathway. Rodenwaldt et al. (2007) demonstrated, in mice, that inhibition of NO production augmented the conduction of locally initiated constrictions along the vascular wall mediated through gap junctions and that this response was reversed by exogenous NO. Inhibition of EDHF by NO may contribute to changes in Ca2 þ concentrations in cultured bovine endothelial cells because inhibition of NOS augmented the increase in the ATP-induced Ca2 þ response (Shin et al., 1992). Bauersachs et al. (1996) demonstrated that exogenous NO inhibited bradykinin-induced increases in Ca2 þ concentrations in cultured human endothelial cells in the presence of NOS inhibitor. In the present study, we first demonstrated that administration of NOS inhibitor augmented depressor responses induced by 1-EBIO, a KCa2.3 and KCa3.1 activator. These findings suggest that NO may inhibit gap junction communication rather than changing Ca2 þ concentrations in endothelial cells. In recent study, Pogoda et al. (2014) reported that NO inhibition of Ca2 þ signal propagation was mediated by a reduction in myoendothelial gap junction coupling in endothelial and smooth muscle cell co-cultures. Thus, we speculate that NO inhibits propagation of hyperpolarization via myoendothelial gap junctions in vivo. 4.3. Implications In the present study, we demonstrated that agonist-induced depressor responses in anesthetized rats were primarily mediated by EDHF and that NO inhibited EDHF. Desai et al. (2006) reported that acetylcholine-induced depressor responses in anesthetized rats could be inhibited by acute treatment with NG-nitro-L-arginine methyl ester (L-NAME). Buxton et al. (1993) reported that LNAME antagonized muscarinic receptors, and an antimuscarinic effect could be responsible for the apparent discrepancy between the present study and that of Desai et al. (2006). EDHF-mediated vasodilation was enhanced following suppression of NO

We thank Inter-Biotech Ltd. (http://www.inter-biotech.com) for reading the manuscript. References Ayajiki, K., Fujioka, H., Shinozaki, K., Okamura, T., 2005. Effects of capsaicin and nitric oxide synthase inhibitor on increase in cerebral blood flow induced by sensory and parasympathetic nerve stimulation in the rat. J. Appl. Physiol. 98, 1792–1798. Bauersachs, J., Popp, R., Hecker, M., Sauer, E., Fleming, I., Busse, R., 1996. Nitric oxide attenuates the release of endothelium-derived hyperpolarizing factor. Circulation 94, 3341–3347. Behringer, E.J., Socha, M.J., Polo-Parada, L., Segal, S.S., 2012. Electrical conduction along endothelial cell tubes from mouse feed arteries: confounding actions of glycyrrhetinic acid derivatives. Br. J. Pharmacol. 166, 774–787. Berg, T., Koteng, O., 1997. Signalling pathways in bradykinin- and nitric oxideinduced hypotension in the normotensive rat; role of K þ -channels. Br. J. Pharmacol. 121, 1113–1120. Bryan, R.M., You, J., Golding, E.M., Marrelli, S.P., 2005. Endothelium-derived hyperpolarizing factor: a cousin to nitric oxide and prostacyclin. Anesthesiology 102, 1261–1277. Buxton, I.L., Cheek, D.J., Eckman, D., Westfall, D.P., Sanders, K.M., Keef, K.D., 1993. NG-Nitro L-arginine methyl ester and other alkyl esters of arginine are muscarinic receptor antagonists. Circ. Res. 72, 387–395. Chadha, P.S., Liu, L., Rikard-Bell, M., Senadheera, S., Howitt, L., Bertrand, R.L., Grayson, T.H., Murphy, T.V., Sandow, S.L., 2011. Endothelium-dependent vasodilation in human mesenteric artery is primarily mediated by myoendothelial gap junctions intermediate conductance calcium-activated K þ channel and nitric oxide. J. Pharmacol. Exp. Ther. 336, 701–708. Cheng, Y., Ndisang, J.F., Tang, G., Cao, K., Wang, R., 2004. Hydrogen sulfide-induced relaxation of resistance mesenteric artery beds of rats. Am. J. Physiol. Heart Circ. Physiol. 287, H2316–H2323. Chepkova, A.N., Sergeeva, O.A., Haas, H.L., 2008. Carbenoxolone impairs LTP and blocks NMDA receptors in murine hippocampus. Neuropharmacology 55, 139–147. Chintala, M.S., Bernardino, V., Chiu, P.J., 1994. Cyclic GMP but not cyclic AMP prevents renal platelet accumulation after ischemia–reperfusion in anesthetized rats. J. Pharmacol. Exp. Ther. 271, 1203–1208. Christ, G.J., Spray, D.C., El-Sabban, M., Moore, L.K., Brink, P.R., 1996. Gap junctions in vascular tissues. Evaluating the role of intercellular communication in the modulation of vasomotor tone. Circ. Res. 79, 631–646. De Vriese, A.S., Van de Voorde, J., Lameire, N.H., 2002. Effects of connexin-mimetic peptides on nitric oxide synthase- and cyclooxygenase-independent renal vasodilation. Kidney Int. 61, 177–185. Desai, K.M., Gopalakrishnan, V., Hiebert, L.M., McNeill, J.R., Wilson, T.W., 2006. EDHF-mediated rapid restoration of hypotensive response to acetylcholine after chronic, but not acute, nitric oxide synthase inhibition in rats. Eur. J. Pharmacol. 546, 120–126. De Wit, C., Esser, N., Lehr, H.a., Bolz, S.S., Pohl, U., 1999. Pentobarbital-sensitive EDHF comediates ACh-induced arteriolar dilation in the hamster microcirculation. Am. J. Physiol. 276, H1527–H1534. De Wit, C., Griffith, T.M., 2010. Connexins and gap junctions in the EDHF phenomenon and conducted vasomotor responses. Pflügers Arch. 459, 897–914. Doughty, J.M., Boyle, J.P., Langton, P.D., 2000. Potassium does not mimic EDHF in rat mesenteric arteries. Br. J. Pharmacol. 130, 1174–1182. Edgley, A.J., Tare, M., Evans, R.G., Skordilis, C., Parkington, H.C., 2008. In vivo regulation of endothelium-dependent vasodilation in the rat renal circulation

34

S. Kobuchi et al. / European Journal of Pharmacology 762 (2015) 26–34

and the effect of streptozotocin-induced diabetes. Am. J. Physiol. Regul. Integr. Comp. Physiol. 295, R829–R839. Edwards, G., Dora, K.A., Gardener, M.J., Garland, C.J., Weston, A.H., 1998. K þ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 396, 269–272. Edwards, G., Félétou, M., Gardener, M.J., Thollon, C., Vanhoutte, P.M., Weston, A.H., 1999. Role of gap junctions in the responses to EDHF in rat and guinea-pig small arteries. Br. J. Pharmacol. 128, 1788–1794. Félétou, M., Vanhoutte, P.M., 2006. Endothelium-derived hyperpolarizing factor: where are we now? Arterioscler. Thromb. Vasc. Biol. 26, 1215–1225. Félétou, M., Vanhoutte, P.M., 2009. EDHF: an update. Clin. Sci. (Lond) 117, 139–155. Félétou, M., 2011. The Endothelium: Part 2: EDHF-Mediated Responses “The Classical Pathway”. Morgan & Claypool Life Sciences Publisher, San Rafael, CA. Figueroa, X.F., Duling, B.R., 2008. Dissection of two Cx37-independent conducted vasodilator mechanisms by deletion of Cx40: electrotonic versus regenerative conduction. Am. J. Physiol. Heart. Circ. Physiol. 295, H2001–H2007. Fisslthaler, B., Popp, R., Kiss, L., Potente, M., Harder, D.R., Fleming, I., Busse, R., 1999. Cytochrome P450 2C is an EDHF synthase in coronary arteries. Nature 401, 493–497. Fujioka, H., Ayajiki, K., Shinozaki, K., Toda, N., Okamura, T., 2002. Mechanisms underlying endothelium-dependent, nitric oxide- and prostanoid-independent relaxation in monkey and dog coronary arteries. Naunyn Schmiedebergs Arch. Pharmacol. 366, 488–495. Gigout, S., Louvel, J., Pumain, R., 2006. Effects in vitro and in vivo of a gap junction blocker on epileptiform activities in a genetic model of absence epilepsy. Epilepsy Res. 69, 15–29. Griffith, T.M., 2004. Endothelium-dependent smooth muscle hyperpolarization: do gap junctions provide a unifying hypothesis? Br. J. Pharmacol. 141, 881–903. Huang, A., Sun, D., Jacobson, A., Carroll, M.A., Falck, J.R., Kaley, G., 2005. Epoxyeicosatrienoic acids are released to mediate shear stress-dependent hyperpolarization of arteriolar smooth muscle. Circ. Res. 96, 376–383. Irvine, J.C., Favaloro, J.L., Kemp-Harper, B.K., 2003. NO– activates soluble guanylate cyclase and Kv channels to vasodilate resistance arteries. Hypertension 41, 1301–1307. Kessler, P., Popp, R., Busse, R., Schini-Kerth, V.B., 1999. Proinflammatory mediators chronically downregulate the formation of the endothelium-derived hyperpolarizing factor in arteries via a nitric oxide/cyclic GMP-dependent mechanism. Circulation 99, 1878–1884. Leuranguer, V., Vanhoutte, P.M., Verbeuren, T., Félétou, M., 2008. C-Type natriuretic peptide and endothelium-dependent hyperpolarization in the guinea-pig carotid artery. Br. J. Pharmacol. 153, 57–65. Malmsjö, M., Bergdahl, A., Zhao, X.H., Sun, X.Y., Hedner, T., Edvinsson, L., Erlinge, D., 1999. Enhanced acetylcholine and P2Y-receptor stimulated vascular EDHFdilatation in congestive heart failure. Cardiovasc. Res. 43, 200–209. Matoba, T., Shimokawa, H., Nakashima, M., Hirakawa, Y., Mukai, Y., Hirano, K., Kanaide, H., Takeshita, A., 2000. Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in mice. J. Clin. Invest. 106, 1521–1530. Nishikawa, Y., Stepp, D.W., Chilian, W.M., 2000. Nitric oxide exerts feedback inhibition on EDHF-induced coronary arteriolar dilation in vivo. Am. J. Physiol. Heart Circ. Physiol. 279, H459–H465. Ozkor, M., Quyyumi, A., 2011. Endothelium-derived hyperpolarizing factor and vascular function. Cardiol. Res. Pract. 2011, 156146.

Pogoda, K., Füller, M., Pohl, U., Kameritsch, P., 2014. NO, via its target Cx37, modulates calcium signal propagation selectively at myoendothelial gap junctions. Cell Commun. Signal. 12, 33. Parkington, H.C., Chow, J.A., Evans, R.G., Coleman, H.A., Tare, M., 2002. Role for endothelium-derived hyperpolarizing factor in vascular tone in rat mesenteric and hindlimb circulations in vivo. J. Physiol. 542, 929–937. Rodenwaldt, B., Pohl, U., de Wit, C, 2007. Endogenous and exogenous NO attenuates conduction of vasoconstrictions along arterioles in the microcirculation. Am. J. Physiol. Heart Circ. Physiol. 292, H2341–H2348. Rummery, N.M., Hill, C.E., 2004. Vascular gap junctions and implications for hypertension. Clin. Exp. Pharmacol. Physiol. 31, 659–667. Sagar, G.D.V., Larson, D.M., 2006. Carbenoxolone inhibits junctional transfer and upregulates Connexin43 expression by a protein kinase A-dependent pathway. J. Cell. Biochem. 98, 1543–1551. Shimokawa, H., Yasutake, H., Fujii, K., Owada, M.K., Nakaike, R., Fukumoto, Y., Takayanagi, T., Nagao, T., Egashira, K., Fujishima, M., Takeshita, A., 1996. The importance of the hyperpolarizing mechanism increases as the vessel size decreases in endothelium-dependent relaxations in rat mesenteric circulation. J. Cardiovasc. Pharmacol. 28, 703–711. Shin, W.S., Sasaki, T., Kato, M., Hara, K., Seko, A., Yang, W.D., Shimamoto, N., Sugimoto, T., Toyo-oka, T., 1992. Autocrine and paracrine effects of endothelium-derived relaxing factor on intracellular Ca2 þ of endothelial cells and vascular smooth muscle cells. Identification by two-dimensional image analysis in coculture. J. Biol. Chem. 267, 20377–20382. Smits, G.J., Perrone, M.H., Cox, B.F., 1997. Regional hemodynamic dose-response of lemakalim and glybenclamide in anesthetized rats. J. Cardiovasc. Pharmacol. 29, 49–56. Sofola, O., Knill, A., Hainsworth, R., Drinkhill, M., 2002. Change in endothelial function in mesenteric arteries of Sprague-Dawley rats fed a high salt diet. J. Physiol. 543, 255–260. Suadicani, S.O., Brosnan, C.F., Scemes, E., 2006. P2  7 receptors mediate ATP release and amplification of astrocytic intercellular Ca2 þ signaling. J. Neurosci. 26, 1378–1385. Thengchaisri, N., Kuo, L., 2003. Hydrogen peroxide induces endothelium-dependent and -independent coronary arteriolar dilation: role of cyclooxygenase and potassium channels. Am. J. Physiol. Heart Circ. Physiol. 285, H2255–H2263. Thomsen, K., Rubin, I., Lauritzen, M., 2000. In vivo mechanisms of acetylcholineinduced vasodilation in rat sciatic nerve. Am. J. Physiol. Heart Circ. Physiol 279, H1044–H1054. Waldron, G.J., Ding, H., Lovren, F., Kubes, P., Triggle, C.R., 1999. Acetylcholineinduced relaxation of peripheral arteries isolated from mice lacking endothelial nitric oxide synthase. Br. J. Pharmacol. 128, 653–658. Zhang, Z.H., Kang, Y.M., Yu, Y., Wei, S.G., Schmidt, T.J., Johnson, A.K., Felder, R.B., 2006. 11beta-Hydroxysteroid dehydrogenase type 2 activity in hypothalamic paraventricular nucleus modulates sympathetic excitation. Hypertension 48, 127–133. Zink, M.H., Oltman, C.L., Lu, T., Katakam, P.V., Kaduce, T.L., Lee, H., Dellsperger, K.C., Spector, A.A., Myers, P.R., Weintraub, N.L., 2001. 12-Lipoxygenase in porcine coronary microcirculation: implications for coronary vasoregulation. Am. J. Physiol. Heart Circ. Physiol. 280, H693–H704. Zygmunt, P.M., Edwards, G., Weston, H., Larsson, B., Högestätt, E.D., 1997. Involvement of voltage-dependent potassium channels in the EDHF-mediated relaxation of rat hepatic artery. Br. J. Pharmacol. 121, 141–149.