+) mice

European Journal of Pharmacology 560 (2007) 193 – 200 www.elsevier.com/locate/ejphar

Calcium-activated potassium channel and connexin expression in small mesenteric arteries from eNOS-deficient (eNOS −/− ) and eNOS-expressing (eNOS +/+ ) mice Lisa Ceroni a,1 , Anthie Ellis b,1 , William B. Wiehler a , Yan-Fen Jiang a , Hong Ding b , Chris R. Triggle a,b,⁎ b

a Smooth Muscle Research Group, Faculty of Medicine, University of Calgary, 3330 Hospital Dr NW, Calgary, AB, Canada T2N 4N1 School of Medical Sciences and Division of Chinese Medicine, School of Health Sciences, RMIT University, Bundoora West Campus, Victoria 3083, Australia

Received 16 September 2006; received in revised form 14 December 2006; accepted 8 January 2007 Available online 19 January 2007

Abstract Endothelium-derived hyperpolarizing factor (EDHF), notably in the microcirculation, plays an important role in the regulation of vascular tone. The cellular events that mediate EDHF are critically dependent, in a vessel dependent manner, on small conductance calcium-activated potassium channels (SK) and intermediate conductance calcium-activated potassium channels (IK) as well as the presence of the gap junction connexins 37, 40, and 43. We hypothesized that the expression levels of SK, IK, as well as vascular connexins, notably 37, 40 and 43 but, potentially, connexin 45, would show correlation with the contribution of EDHF to acetylcholine-mediated vasodilatation as well as, in the absence of endothelial-derived NO, higher expression levels in eNOS−/− mice. Wire myograph studies were performed to confirm the contribution of EDHF to endothelium-dependent relaxation in 1st, 2nd and 3rd order small mesenteric arteries from C57BL/6J eNOS-expressing (eNOS+/+) and eNOS-deficient C57BL/6J (eNOS−/−) mice. Small mesenteric arteries, as well as the branch points between 1st and 2nd and 2nd and 3rd order vessels, were analysed for the expression of mRNA for SK1, SK2, SK3, IK and large conductance calcium-activated potassium channels (BK) and comparable studies were performed for connexins 37, 40, 43 and 45. Although the contribution of EDHF to endothelium-dependent relaxation was significantly greater in the 3rd order vessels from the eNOS+/+ the real-time (RT) polymerase chain reaction (PCR) data showed no differences for the expression levels of mRNA for any of the channel subtypes or the connexins within the small mesenteric arteries from either the eNOS+/+ or eNOS−/− mice, nor, based on RT PCR analysis, were there differences in expression of the potassium channels studied in the branch points versus 1st, 2nd or 3rd order vessels. These data suggest that neither the gene expression of calcium-activated potassium channels nor vascular connexins are modulated by NO; however, their functional contribution to endothelium-dependent relaxation may be modulated by other physiological parameters. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved. Keywords: eNOS; Connexin; Calcium-activated potassium channels; Nitric oxide; EDHF (endothelium-derived hyperpolarizing factor); Small mesenteric arteries

1. Introduction Endothelium-dependent vasodilatation is achieved via the generation of endothelium-derived relaxing factors, notably nitric oxide (NO) and prostacyclin, as well as the variable contribution of endothelium-derived hyperpolarizing factor, or ⁎ Corresponding author. School of Medical, School of Health Sciences, RMIT University, Bundoora West Campus, Victoria 3083, Australia. E-mail addresses: [email protected], [email protected] (C.R. Triggle). 1 Authors contributed equally in this manuscript.

EDHF (Bryan et al., 2005; Busse et al., 2002; Feletou and Vanhoutte, 2006; McGuire et al., 2001; Sandow, 2004; Vanhoutte, 2004). The contribution of EDHF to endotheliumdependent vasodilatation has been reported to increase as vessel size decreases and conversely for the expression of endothelial nitric oxide synthase, eNOS (Shimokawa et al., 1996) Thus, whereas it is well established that NO has many important functions within the vasculature, EDHF is postulated to play a key role in the regulation of blood flow in the microcirculation. In most vascular beds the EDHF-mediated response can be inhibited by a combination of the small conductance calcium-activated K-channel (SK) inhibitor, apamin, with the

0014-2999/$ - see front matter. Crown Copyright © 2007 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejphar.2007.01.018

194

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

intermediate conductance IK inhibitor, charybdotoxin or the more specific agent TRAM-34 (Eichler et al., 2003; Garland and Plane, 1996). There is also data that indicates that the target channels for apamin, charybdotoxin and TRAM-34 are SK3 for apamin and IK for charybdotoxin and TRAM-34 and both functional as well as molecular data indicates that these channels reside on endothelial cells (Edwards et al., 1998; Burnham et al., 2002; Bychkov et al., 2002). In addition, there is increasing evidence that myoendothelial gap junctions, made up in vascular tissue from connexins 37, 40 and 43, are important for EDHF-dependent vasodilatation (De Vriese et al., 2002; de Wit et al., 2003; Lang et al., 2006; Mather et al., 2005; Ujiie et al., 2003). Similarly, based on anatomical data, the association between myoendothelial gap junctions and EDHFdominant vessels has been reported in rat mesenteric arteries (Sandow and Hill, 2000). In rabbit mesenteric resistance arteries chronic administration of nitroglycerin reduces eNOS-generated NO via the generation of superoxide anions (Yamamoto et al., 2005). Similarly, acetylcholine-mediated hyperpolarization of rabbit aortic valves, assumed to be mediated by an EDHF pathway is also reduced by nitroglycerin (Kusama et al., 2005). Thus, the bioavailability of NO may determine the contribution of EDHF to the regulation of vascular tone in any given vascular bed. In the aorta from the eNOS−/− mouse, acetylcholine-mediated vasodilatation is completely absent (Huang et al., 1995; Waldron et al., 1999); however, the tissue is supersensitive, compared to aortae from eNOS+/+ mice, to the relaxant actions of the endothelium-independent vasodilator, sodium nitroprusside (Waldron et al., 1999). In contrast, in eNOS+/+ the generation of NO activates soluble guanylyl cyclase leading to the formation of cyclic GMP (cGMP) and subsequent activation of cGMP-dependent kinase with the downstream result that the activity of the cGMP-specific phosphodiesterase V is increased resulting in the consequent reduction in NO-mediated processes (Mullershausen et al., 2003). Predictably the cellular events that regulate endotheliumdependent vasodilatation are subjected to fine-tuning and, of particular interest, the contribution of both NO and EDHF to endothelium-dependent vasodilatation can be modified by cardiovascular disease (Coleman et al., 2004; De Vriese et al., 2000; Goto et al., 2004; Triggle et al., 2005). As an example in the microcirculation from the type 2 diabetic leptin receptor mutant mouse, db/db, endothelium-dependent vasodilatation is entirely mediated by EDHF whereas in the non-diabetic control mouse both NO and EDHF contribute to vasodilatation; however, the pharmacological properties of EDHF differ between the diabetic and non-diabetic controls (Pannirselvam et al., 2003, 2006). Alterations in the expression of KCa and connexin 37 and changes in the contribution of EDHF in small mesenteric arteries from a murine model of type 1 diabetes have also been reported. (Ding et al., 2005). Thus, a better understanding of the cellular mechanisms that regulate EDHF versus NO-mediated modulation of vascular tone is critical for the development of improved therapies for vascular disease. Branch points between vessels have shown a defective acetylcholine-mediated vasodilatation with constriction evident

at high concentrations of acetylcholine (McLenachan et al., 1990) thus raising the question as to whether the expression levels of KCa channels and connexins are different in such regions. Indeed, von der Weid and Coleman (2005) have argued that the plasma level of NO, and modulation thereof with a nitroglycerin patch (Yamamoto et al., 2005; Kusama et al., 2005) may affect potassium channel expression or their function (von der Weid and Coleman, 2005). Support for this latter hypothesis comes from the work of Hilgers et al. (2006) who have reported that, in the rat mesenteric arterial vasculature, semiquantitative RT-PCR data indicates that higher levels of SK and IK were expressed in 4th versus 1st order vessels, thus correlating with a greater contribution of EDHF in the smaller vessels. In the present study we tested the hypothesis that the absence of eNOS-generated NO in eNOS−/− would affect the gene expression of SK, IK, BK and connexins 37, 40 and 43, as well as, potentially, connexin 45. We also tested the hypothesis that SK, IK, BK and connexin expression would differ in branch points. Finally, we hypothesized that expression levels of calcium-activated potassium channels and endothelial-vascular connexins would demonstrate regional differences within the mesenteric vasculature. 2. Methods 2.1. Animals Male eNOS+/+ and the eNOS−/− mice were purchased from Jackson Laboratories (Bar Harbor ME, U.S.A.). Upon halothane-induced anaesthesia, mice were killed by cervical dislocation, in accordance with a research protocol complying with the standards of the Canadian Council on Animal Care and approved by the University of Calgary Animal Care Committee. 2.2. Isolation of mesenteric vessels 2.2.1. Preparation for molecular work: Superior, first (1st), second (2nd) and third (3rd) order small mesenteric arteries (100–180 μm internal diameter) and the branch points between the 1st to 2nd orders and 2nd to 3rd orders were dissected cleared of any adherent fat in a Ca2+free solution of the following composition (mM): NaCl 120; NaHCO3 25; KCl 4.2; KH2PO4 1.2; MgCl2 1.2; D-glucose 11). For real-time (RT)-polymerase chain reaction (PCR) analysis of KCa message expression, each respective vessel order and each branch point sample from one single mouse mesenteric bed were pooled individually (i.e. all 1st order vessels together; all 1st to 2nd branch points together etc.), flash frozen in liquid nitrogen and used either immediately or stored at − 80°C until use. Sufficient branch point tissue was available to pursue only RT-PCR analysis of KCa message expression. 2.2.2. Preparation for functional studies: 1st, 2nd and 3rd order branches stemming from the superior mesenteric artery were dissected, cleared of any adherent fat, cut into 2 mm segments and mounted onto a wire myograph.

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

Parkington et al. (1993) have shown that the degree of stretch to which an artery is stretched has been shown to affect the hyperpolarization response to acetylcholine and thus in the current study all vessels were subjected at a resting tension of 1 mN. Vessels were bathed in 5 ml chambers containing physiological saline solution, which had the following composition (mM): NaCl 118.0; KCl 4.7; KH2PO4 0.8; MgSO4 1.2; NaHCO3 24.9; dextrose 11.1 and CaCl2 2.5. Physiological saline solution was maintained at a temperature of 37 °C and bubbled with Carbogen (95% O2, 5% CO2). Following a 1 h equilibration period, vessels were exposed to a high concentration of KCl (120 mM) to determine vessel responsiveness. To observe vasodilatation, tissues were constricted with submaximal concentrations of cirazoline (0.03–0.3 μM) and relaxations were elicited by acetylcholine (0.01–10 μM) in the absence and presence of (i) the nitric oxide synthase inhibitor, nitro-Larginine (L-NAME 100 μM), (ii) the cyclooxygenase inhibitor, indomethacin (10 μM) and (iii) the inhibitor of guanylate cyclase, 1H-[1,2,4]oxadiazolo[4,3,-a]quinoxalin-1-one (ODQ 10 μM). EDHF-mediated relaxations are typically resistant to inhibitors of NO and prostaglandin synthesis, thus any residual relaxations observed in the presence of these inhibitors was taken as an index of the EDHF contribution to vascular tone of the vessels. 2.3. KCa channel gene expression RNA was isolated in DNase/RNase-free tubes using the RNeasy mini kit (Qiagen, Mississauga, ON, Canada). A DNase step was carried out to remove any genomic DNA using the RNase-free DNase set (Qiagen). First strand cDNA was synthesized from 12 μl of the isolated RNA using the Sensiscript Reverse Transcriptase (RT) Kit (Qiagen) and Oligo d(T) Primer (Invitrogen, Burlington, ON, Canada). For the outer PCR reaction, 2 μl of the first strand cDNA was placed in a PCR solution (50 μl) containing the following reagents: 1.5 mM MgCl2; 0.25 μM each of forward and reverse primers; 0.2 mM deoxynucleotide triphosphates; and 2.5 U Taq DNA polymerase. The detection of Rho-A was used as evidence for the successful production of cDNA. As a negative control, water was used in a standard PCR reaction at the same volume as cDNA. All PCR reactions were hot started (94 °C for 3 min) and then exposed to various PCR cycling parameters. The following primer sequences were used for qualitative RT-PCR analysis of KCa channels in small mesenteric arteries. Accession numbers for each of the KCa cDNA sequences are also provided in brackets, as well as amplicon size and a letter, which corresponds to the cycling parameters used for each primer set (given below). Rho: (BC068115) Fwd: 5′ CGGGATCCCGATGGCTGCCATC(C/A)GGAAG 3′ Rev: 5′ GAATTCCTCACAAGA(T/C)(G/A)AGGCA(A/C) 3′ Amplicon size: 597 bp Parameter: A BKCa: (NM_010610) Outer Fwd: 5′ GCCGAGGTCGGCTGGATGAC 3′ OuterRev: 5′ CAGGGACGTAGCTGGCAAAC 3′ (continued on next page)

195

Amplicon size: 697 bp Parameter: B Nested Fwd: 5′ CACATTACAGATCGACATGGC 3′ Outer Rev: 5′ TGGGTCCCCTGAATTCTCCACC 3′ Amplicon size: 343 bp Parameter: D IKCa: (AF072884) Outer Fwd: 5′ ATGCTCCTGCGTCTCTAC 3′ Outer Rev: 5′ CTCCAGCTTCCGAGCCACCA 3′ Amplicon size: 405 bp Parameter: A Nested Fwd: 5′ CCAAGTCCGCTTCCGCCAC 3′ Nested Rev: 5′ CCATGACTCCGGTGCACAGGC 3′ Amplicon size: 263 bp Parameter: A SK1: (AF357239) Outer F: 5′ CCACGCCCGAGAGATCCAGCTG 3′ Outer R: 5′ CGGGGCAGATGGTCATGAGCGT 3′ Amplicon size: 367 bp Parameter: C Nested F: 5′ CAATGGTGCCGACGACTG 3′ Nested R: 5′ CCGAGCCAGCAGGTAGAG 3′ Amplicon size: 216 bp Parameter: D SK2: (AY123778) Outer F: 5′ CAGCGGCACCAAGTCCAGCAAA 3′ Outer R: 5′ GCCGGGCTGTCCATGTGAACGT 3′ Amplicon size: 409 bp Parameter: C Nested F: 5′ ACCAGAACATCGGCTACAAG 3′ Nested R: 5′ GCACACACCAGTATTTCCAAG 3′ Amplicon size: 333 bp Parameter: D SK3: (AF357241) Outer F: 5′ GCGTTCAGGGCCCCCACTTCA 3′ Outer R: 5′ CGGCATGCTGGTGGTTGTGGG 3′ Amplicon size: 420 bp Parameter: C Nested F: 5′ CACTTCAGCCAACTCTAC 3′ Nested R: 5′ CCTGGAGGAGATGATAATC 3′ Amplicon size: 339 bp Parameter: D

The following PCR parameters were used: A. (94 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s) ×35 cycles B. (94 °C for 45 s, 55 °C for 1 min, 72 °C for 1 min) × 30 cycles (outer); 25 cycles (nested) IKCa; ×40 cycles for Rho C. (94 °C for 15 s, 61.5 °C for 30 s, 72 °C for 1 min) × 35 cycles D. (94 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s)× 30 cycles With the exception of Rho nested PCR reactions were performed using 2 μl of the first PCR reaction product as template. Products were subjected to electrophoresis on a 1% agarose gel (w/v) prepared with ethidium bromide and identity was confirmed using automated DNA sequencing. 2.4. Quantitative real-time PCR analysis for KCa and connexin mRNA expression Quantitative RT-PCR was performed using the iCycler iQ thermocycler (BioRad) and Qiagen SYBR-green kit as previously described (Ding et al., 2005). Total RNA was

196

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

extracted from 1st, 2nd and 3rd order small mesenteric arteries (and the superior mesenteric artery for the connexin mRNA expression study) (RNeasy mini kit) and used as the template to synthesize cDNA (RT Sensiscript). β-Actin mRNA levels were compared to total RNA levels for each extraction to ensure that the levels were consistent, justifying the use of B actin as a suitable housekeeping gene. Real-time PCR was carried out in a 96-well plate with each well containing a final volume of 25 μl made up of the following constituents: 12.5 μl SYBR-green reagent (Qiagen); 0.25 μM forward primer; 0.25 μM reverse primer; 0.85 μl cDNA; 10.4 μl water (Qiagen). Each sample was analysed in triplicate using the following primer sequences:

2.5. Drugs

β-actin: forward 5′ ACGGCCAGGTCATCACTATTG 3′ and reverse 5′ CCAAGAAGGAAGGCTGGAAAAGA 3′ Connexin 37: forward 5′-AGGCAGGCTTCCTCTATGGC-3′ and reverse 5′AGACATAGCAGTCCACGATGTG-3′ Connexin 40: forward 5′-GAGGCCCACGGAGAAGAATG-3′ and reverse 5′TGGTAGAGTTCAGCCAGGCT-3′ Connexin 43: forward 5′-ACAAGGTCCAAGCCTACTCCA-3′ and reverse 5′-CCCCAGGAGCAGGATTCTGA-3′ Connexin 45: forward 5′-CGGGCTGTGAGAATGTCTGC-3′ and reverse 5′CAGGTACATCACAGAGGGAGTTG-3′ BK: forward 5′ CAGCACTCCGCAGACATTG 3′ and reverse 5′ ATCACCATAACAACCACCATCC 3′ IK: forward 5′ ATGC TCCT GCG TCTC TAC 3′and reverse 5 ′ GAAGCGGACTTGGTTGAG 3′ SK1: forward 5′ GTGAAGATTGAACAAGGGAAGG 3′ and reverse 5′ TGCCTCCAACTCCTCCTG 3′ SK2: forward 5′ ACCATCAGACAGCAGCAAAGGG 3′ and reverse 5′ GACCGCCGCCTCCTGGAC 3′ SK3: forward 5′ GCCAACTCTACCGCCATC 3′ and reverse 5′ GGCTGTGGAACTTGGAGAG 3′

Experimental data were processed using Microsoft ® Excel 2000 software and expressed as mean ± standard error of the mean (SEM). One-sided P values were calculated by Mann– Whitney test using standard statistical program. Statistical significance of concentration–response curves was evaluated using two-way ANOVA, followed by a Bonferroni posthoc comparison test. Probability values of P b 0.05 were considered statistically significant. n = number of mice.

The amplification protocol used was as follows: STEP 1 — 95 °C for 15 min; STEP 2 — (94 °C for 15 s, 58.2 °C [60.1 °C used for connexin primers] for 30 s, 72 °C for 30 s.) × 45 cycles. Transcript expression was calculated relative to the reference standard, β-actin using the following equation 2-(ΔCt); where the ΔCt = Ct gene − Ct β-actin and values are expressed as 0.001 (connexin) and 0.0001 (KCa) folds relative to β-actin levels.

Acetylcholine, cirazoline, indomethacin, apamin, ODQ and were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). TRAM-34, [1-(2-chlorophenyl)diphenyl)methyl]-1 H-pyrazole, was a gift from Dr Heike Wulff — Department of Medical Pharmacology and Toxicology, University of California, Davis, CA 95616, U.S.A. Acetylcholine, cirazoline and L-NAME were dissolved in distilled water, while TRAM34 and ODQ were dissolved in 100% DMSO and indomethacin was dissolved in 1% sodium carbonate. L-NAME

2.6. Statistical analysis

3. Results 3.1. Detection and quantification of connexin mRNA expression in small mesenteric arteries Mesenteric arteries from eNOS+/+ and eNOS−/− mice were separated into superior mesenteric, 1st, 2nd and 3rd order and analyzed, for levels of connexin 37, connexin 40, connexin 43 and connexin 45 mRNA using real-time quantitative PCR analysis. Comparisons made between mRNA levels (expressed relative to 1,000 folds of β-actin expression) according to the vessel origin indicated that there were no significant differences in the mRNA levels when connexin 37, connexin 40, connexin 43, and connexin 45 were compared between the two strains of mice for any of the vessels (Fig. 1). Expression levels of connexin 43 and connexin 45 were; however, significantly (P b 0.05) lower than the levels for connexin 37 and connexin 40. Note that branch points were not used to investigate connexin mRNA levels. 3.2. Detection and quantification of KCa mRNA expression in mouse small mesenteric arteries 3.2.1. IK The mRNA expression for the IK channels was consistently present in each of the three vessel orders and their respective branch points (five replicate experiments from five different mice) for both eNOS+/+ and eNOS−/− mice. The quantitative real-time PCR data is presented in Fig. 1 relative to β-actin.

Fig. 1. Real-time PCR expression levels relative to β-actin for connexins 37, 40, 43, and 45 in superior mesenteric, 1st, 2nd, versus 3rd vessels from eNOS+/+ and eNOS−/− mice.

3.2.2. SK1, SK2 and SK3 Messenger RNA for SK1 was detected in all vessels in eNOS+/+ and −/− mice (six replicate experiments), but expression in the

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

branch point samples was quite variable. The quantitative real-time PCR data is presented in Fig. 2 relative to β-actin. The mRNA for SK2 was detected relatively equally in vessels from both eNOS+/+ and −/− mice. In the 1st and 2nd order vessels SK2 was detected in half of the 6 replicate experiments, while in the 3rd order vessels it was detected in five out of six replicate experiments. The quantitative real-time PCR data is presented in Fig. 2 relative to β-actin. SK3 mRNA was detected in all vessels from eNOS+/+ mice in each of six replicate experiments, as well as in the 1st and 3rd order vessels from eNOS−/− mice in six replicate experiments. In 2nd order vessels from eNOS−/− mice, SK3 mRNA was detected in 5 out of 6 replicate experiments in which both the positive control (RhoA) and IKCa mRNA was detected. The quantitative real-time PCR data is presented in Fig. 2 relative to β-actin. 3.2.3. BK Messenger RNA for BK was detected in all vessels in eNOS+/+ and −/− mice (5 replicate experiments) and the data is summarised in Fig. 2. Messenger RNA for BK was also detected in the branch points (four replicate experiments — data not presented). 3.2.4. Comparison of KCa mRNA expression from real-time PCR Comparison of the mRNA expression levels of IK, BK quantitative real-time PCR revealed no significant differences between the two strains of mice (n = 5 for all KCa in each vessel order and mouse with the exception of SK1 in 3rd order vessels from eNOS−/− mice (n = 4 since mRNA was unobtainable in one sample) (Fig. 2). Comparison of each KCa to one another for each vessel in each mouse (i.e. eNOS+/+ versus eNOS−/− revealed significant differences in the relative transcript levels of KCa channels within a given vessel. IK mRNA was significantly more abundant than BK which, in turn, was more abundant than any of the three SKs (P b 0.005) transcript levels in all vessel orders. The mRNA levels for SK1, SK2 and SK3 were the least abundant and were comparable to one another (Fig. 2).

Fig. 2. Real-time PCR expression levels relative to β-actin for SK1, SK2, SK3, IK and BK in 1st, 2nd, versus 3rd vessels from eNOS+/+ and eNOS−/− mice.

197

Fig. 3. Acetylcholine (ACh)-mediated relaxation of cirazoline-contracted 1st order (A), 2nd order (B), and 3rd order (C) small mesenteric arteries from eNOS+/+ mice in the absence (○), or presence ( ) of the nitric oxide synthase inhibitor L-NAME, the cyclooxygenase inhibitor, indomethacin (Indo), and the soluble guanylyl cyclase inhibitor, 1H-[1,2,4]oxadiazolo[4,3,-a]quinoxalin-1-one (ODQ). (⁎⁎P b 0.01, ⁎⁎⁎P b 0.001, two-way ANOVA; n = 4–5).⁎⁎⁎ indicates P b 0.001. ⁎⁎ indicates P b 0.01.

▪

3.3. Functional data from small mesenteric arteries Acetylcholine (0.01–10 μM) produced endothelium-dependent relaxations in 1st, 2nd and 3rd order small mesenteric arteries from eNOS+/+ mice. Subsequent treatment of these vessels with L-NAME, indomethacin and ODQ reduced relaxations to varying degrees in the different vessels, with a substantial inhibitory effect in 1st order vessels, a less pronounced (but nonetheless significant) inhibition in 2nd orders and no significant inhibition in the 3rd order vessels (Fig. 3). Further, comparison of the acetylcholine-mediated relaxations produced in the absence of the inhibitors between the different vessels indicated that responses in the 3rd order vessels were significantly smaller than those produced in 1st order vessels (P b 0.05, two-way ANOVA). 4. Discussion The relative contribution of EDHF to endothelium-dependent vasodilatation has consistently been reported to be inversely proportional to arterial diameter (Nagao et al., 1992; Shimokawa et al., 1996). Anatomic evidence has also associated the appearance of myoendothelial gap junctions in vessels where EDHF-mediated vasodilatation dominates (Sandow and Hill, 2000). The EDHF-mediated contribution to endothelium-dependent vasodilatation is also critically dependent on the activation of SK and IK channels as well as the involvement of myoendothelial gap junctions and, notably, the presence of connexins 37, 40 and 43. Our molecular studies of the presence and levels of SK, IK and connexins 37, 40 as well as 43 and 45; however, failed to reveal any difference neither when vessels from eNOS+/+ were compared to those from eNOS−/− mice nor when superior mesenteric versus 1st versus 2nd versus 3rd order vessels were compared. Furthermore, our data comparing acetylcholine-mediated vasodilatation in 1st order versus 2nd and 3rd order small mesenteric arteries from eNOS+/+ mice indicated that acetylcholine-mediated relaxations

198

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

in 1st order vessels, notably in comparison to 3rd order vessels, were the most sensitive to L-NAME, indomethacin and ODQ. Thus, in 3rd order small mesenteric arteries, the L-NAME/ indomethacin/ODQ combination had negligible effects on acetylcholine-mediated responses and was interpreted to reflect the dominance of EDHF in these small vessels. Previously we have reported that the combination of apamin and charybdotoxin essentially eliminated the L-NAME/indomethacin/ODQresistant component of relaxation in 1st order small mesenteric arteries from eNOS+/+ and eNOS−/− mice (Waldron et al., 1999; Ding et al., 2000). Further, in 2nd order vessels murine mesenteric arteries, acetylcholine-mediated relaxation was also reported to be primarily dependent on NO (McGuire et al., 2002). In comparison, we and others have previously reported that acetylcholine-mediated relaxation in small mesenteric arteries from eNOS−/− mice is entirely mediated by EDHFdependent mechanisms, which, even in the absence of inhibitors, is smaller than that seen in their wild-type counterparts (Ding et al., 2000; Matoba et al., 2000; Waldron et al., 1999). These data illustrating a graded contribution of EDHF in 1st and 2nd versus 3rd order vessels from eNOS+/+ mice together with the notable sole contribution of EDHF to acetylcholine-mediated relaxation in small mesenteric arteries from eNOS−/− mice might suggest that the expression levels of, in particular, connexins 37 and 40 and SK3 and IK may be differentially expressed. Our data, however, does not support this hypothesis. Looft-Wilson et al. (2004) reported immunohistochemical staining that indicated that whereas connexin 37 and connexin 40 expression levels were uniform in all branches of the mouse cremaster arteriole connexin 43 was only observed in the larger vessels and even in these vessels the expression level of connexin 43 was weak. Sandow et al. (2003) studied the expression of connexin 37, connexin 40, connexin 43 and connexin 45 in hamster epithelial cheek pouch arterioles versus the feed arteries for the cheek pouch arterioles retractor muscle and, despite functional studies reflecting the importance of myoendothelial gap junctions in the feed arteries, but not the cheek pouch arteries, there were no differences in the expression levels of mRNA or immunohistochemical staining of connexin proteins. Thus, our data indicate that, despite the fact that functional differences exist between the different arteries from eNOS+/+ and eNOS−/− mice, significant differences in connexin expression are not the underlying factor for this diversity in the contribution of NO versus EDHF. Differences in the contribution of EDHF may also reflect posttranslational modification of proteins, differences in the distribution and composition of gap junction hemichannels and/or potassium channels within the small mesenteric arteries and the two strains of mouse. However, protein levels of KCa or connexin protein were not determined due to the very low levels of tissue that can be obtained from the mouse mesenteric vascular bed that was further exacerbated by the analysis not only 1st, 2nd and 3rd vessels but also the branch points. Another limitation is that our approach did not allow us to distinguish between myoendothelial and homocellular (endothelial/endothelial or smooth muscle/smooth muscle) gap junctions, thus it remains to be seen whether differences in

connexin distribution would be confined to myoendothelial gap junctions. Dora et al. (2003) have previously reported, on the basis of electron microscopy data, a high incidence of myoendothelial gap junctions in 1st and 2nd order mesenteric arteries from Balb/c mice and thus electron microscopy studies of mesenteric vessels from eNOS−/− and eNOS+/+ mice is also warranted. Furthermore, analysis of the relative contribution of NO versus EDHF to vasodilatation may also be complicated by agonist-dependent differences that we have previously reported in the mouse 2nd/3rd order small mesenteric arteries for acetylcholine- versus protease-activated receptor mediated responses (McGuire et al., 2002, 2004). RT-PCR analysis revealed IK mRNA consistently in 1st, 2nd and 3rd order and the two branch point samples tested for each mouse. The transcript expression levels for the IK channel were significantly more abundant than all of the SK channels, with each of these sub-types being comparable, thus suggesting an important role for IK in mediating EDHF-type vasodilatation. There is a well-documented literature showing a functional contribution by the IKCa channel in eliciting the hyperpolarization, notably in cerebral vessels, which leads to an EDHF response (Eichler et al., 2003; Marrelli et al., 2003). In rat carotid artery endothelial cells the hyperpolarization was sensitive to apamin albeit less of the total hyperpolarization than that of component sensitive to either TRAM-34 or TRAM-39 (Eichler et al., 2003). In our study, the level of SK mRNA was far more variable than that of IK with the expression levels of the SK2 transcript being undetectable in some individual vessel and branch point samples. The variability in SK2 expression may reflect neuronal contamination since SK2 channels are associated with neuronal function (Bond et al., 2004; Villalobos et al., 2004). Despite SK1 and SK3 mRNA being detected more consistently, all three SK subtypes were comparably low in expression and significantly lower in abundance than either the BK or the IK channel. In rat mesenteric arteries acetylcholine-mediated endothelium-dependent hyperpolarization of smooth muscle has been attributed to SK channels with IK channels playing an important role during repolarization phase which was only observed following depolarization (Crane et al., 2003). It is quite conceivable that there are species differences between the rat and the mouse with respect to the relative importance of SK versus IK channels to the EDHF-mediated response. In this regard semiquantitative RTPCR data from Hilgers et al. (2006) for the rat mesenteric arterial arcade showed relatively higher SK1 and SK3 mRNA in 4th versus 1st order vessels with no difference in the expression of the mRNA for BK; protein levels were not determined. In disease states changes in the expression levels of KCa and connexins have been reported. Thus, in a study of a type 1 diabetic mouse model a decreased contribution of EDHF to endothelium-dependent relaxation was associated with a decrease in the expression of mRNA for both SK2 and SK3 as well as connexin 37 (Ding et al., 2005). In the type 2 diabetic Zucker rat an impaired SK-mediated response was reported and linked to a compensatory increase in SK3 mRNA expression; protein levels were not determined due to insufficient tissue (Burnham et al., 2006). We also determined mRNA expression

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

levels for KCa in the branch points of the small mesenteric arteries (data not presented) given by McLenachan et al. (1990) have reported that branch points in coronary vessels show a defective acetylcholine-mediated vasodilatation with constriction evident at high concentrations of acetylcholine. However, based on the mRNA data, we could not detect any such differences (data not presented). In our investigation mRNA expression levels for the SK channel could only be verified in one endothelial cell sample tested from the eNOS+/+ mouse indicating the sole presence of the SK3 channel (data not presented). The importance of endothelial cell SK3 to the regulation of vascular tone has been clearly demonstrated by the elegant study by Taylor et al. (2003) in which the expression levels of SK3 were modulated in mice and the effects on blood pressure, vascular tone and vessel structure determined. It is important to stress, however, that genetic ablation of the expression of eNOS (Huang et al., 1995) or the use of 43Gap 27 or 40Gap 27 peptide inhibitors of connexins 43 and 40 (De Vriese et al., 2002) also results in an increase in blood pressure thus suggesting a complex regulation of vascular tone that involves NO, KCa channels as well as connexins. In conclusion, despite the sole contribution of EDHF to acetylcholine-mediated vasodilatation in the small mesenteric arteries of eNOS−/− mouse quantifiable differences in IK, SK, BK or connexin transcript expression compared to the eNOS+/+ mouse could not be detected. Furthermore, the greater contribution of EDHF to acetylcholine-mediated relaxation observed in 3rd order vessels from eNOS+/+ mice was not related to differences in gene expression of IK, SK, BK, or connexins 37, 40, 43 or 45. These data suggest that the contribution of EDHF to acetylcholine-mediated endotheliumdependent vasodilatation is not directly dependent on the absolute expression levels of connexins and KCa channels but their functional contribution to endothelium-dependent relaxation may be modulated by other physiological parameters. Acknowledgements This work was supported by grants from the CIHR and AHSF to CRT and a Research Infrastructure Grant from RMIT University to CRT and HD. References Bond, C.T., Herson, P.S., Strassmaier, T., Hammond, R., Stackman, R., Maylie, J., Adelman, J.P., 2004. Small conductance Ca2+-activated K+ channel knock-out mice reveal the identity of calcium-dependent after hyperpolarization currents. J. Neurosci. 24, 5301–5306. Bryan Jr., R.M., You, J., Golding, E.M., Marrelli, S.P., 2005. Endotheliumderived hyperpolarizing factor: a cousin to nitric oxide and prostacyclin. Anesthesiology 102, 1261–1277. Burnham, M.P., Bychkov, R., Feletou, M., Richards, G.R., Vanhoutte, P.M., Weston, A.H., Edwards, G.., 2002. Characterization of an apamin-sensitive small-conductance Ca(2+)-activated K(+) channel in porcine coronary artery endothelium: relevance to EDHF. Br. J. Pharmacol. 135, 1133–1143. Burnham, M.P., Johnson, I.T., Weston, A.H., 2006. Impaired small-conductance Ca2+-activated K+ channel-dependent EDHF responses in type II diabetic ZDF rats. Br. J. Pharmacol. 148, 434–441.

199

Busse, R., Edwards, G., Feletou, M., Fleming, I., Vanhoutte, P.M., Weston, A.H., 2002. EDHF: bringing the concepts together. Trends Pharmacol. Sci. 23, 374–380. Bychkov, R., Burnham, M.P., Richards, G.R., Edwards, G., Weston, A.H., Feletou, M., Vanhoutte, P.M., 2002. Characterization of a charybdotoxinsensitive intermediate conductance Ca2+-activated K+ channel in porcine coronary endothelium: relevance to EDHF. Br. J. Pharmacol. 137, 1346–1354. Coleman, H.A., Tare, M., Parkington, H.C., 2004. Endothelial potassium channels, endothelium-dependent hyperpolarization and the regulation of vascular tone in health and disease. Clin. Exp. Pharmacol. Physiol. 31, 641–649. Crane, G.J., Gallagher, N., Dora, K.A., Garland, C.J., 2003. Small- and intermediate-conductance calcium-activated K+ channels provide different facets of endothelium-dependent hyperpolarization in rat mesenteric artery. J. Physiol. 553, 183–189. De Vriese, A.S., Verbeuren, T.J., Van de Voorde, J., Lameire, N.H., Vanhoutte, P.M., 2000. Endothelial dysfunction in diabetes. Br. J. Pharmacol. 130, 963–974. De Vriese, A.S., Van de Voorde, J., Lameire, N.H., 2002. Effects of connexinmimetic peptides on nitric oxide synthase- and cyclooxygenase-independent renal vasodilation. Kidney Int. 61, 177–185. de Wit, C., Roos, F., Bolz, S.S., Pohl, U., 2003. Lack of vascular connexin 40 is associated with hypertension and irregular arteriolar vasomotion. Physiol. Genomics 13, 169–177. Ding, H., Kubes, P., Triggle, C., 2000. Potassium- and acetylcholine-induced vasorelaxation in mice lacking endothelial nitric oxide synthase. Br. J. Pharmacol. 129, 1194–1200. Ding, H., Hashem, M., Wiehler, W.B., Lau, W., Martin, J., Reid, J., Triggle, C., 2005. Endothelial dysfunction in the streptozotocin-induced diabetic apoEdeficient mouse. Br. J. Pharmacol. 146, 1110–1118. Dora, K.A., Sandow, S.L., Gallagher, N.T., Takano, H., Rummery, N.M., Hill, C.E., Garland, C.J., 2003. Myoendothelial gap junctions may provide the pathway for EDHF in mouse mesenteric artery. J. Vasc. Res. 40, 480–490. Edwards, G., Dora, K.A., Gardener, M.J., Garland, C.J., Weston, A.H., 1998. K+ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 396, 269–272. Eichler, I., Wibawa, J., Grgic, I., Knorr, A., Brakemeier, S., Pries, A.R., Hoyer, J., Kohler, R., 2003. Selective blockade of endothelial Ca(2+)-activated small- and intermediate-conductance K(+)-channels suppresses EDHFmediated vasodilation. Br. J. Pharmacol. 138, 594–601. Feletou, M., Vanhoutte, P.M., 2006. Endothelium-derived hyperpolarizing factor: where are we now? Arterioscler. Thromb. Vasc. Biol. 26, 1215–1225. Garland, C.J., Plane, F., 1996. Endothelium-derived Hyperpolarizing Factor. Hardwood Academic Publishers, pp. 173–179. Goto, K., Fujii, K., Kansui, Y., Iida, M., 2004. Changes in endothelium-derived hyperpolarizing factor in hypertension and ageing: response to chronic treatment with renin–angiotensin system inhibitors. Clin. Exp. Pharmacol. Physiol. 31, 650–655. Hilgers, R.H., Todd Jr., J., Webb, R.C., 2006. Regional heterogeneity in acetylcholine-induced relaxation in rat vascular bed: role of calcium-activated K+ channels. Am. J. Physiol.: Heart Circ. Physiol. 291, H216–H222. Huang, P.L., Huang, Z., Mashimo, H., Bloch, K.D., Moskowitz, M.A., Bevan, J.A., Fishman, M.C., 1995. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature 377, 239–242. Kusama, N., Kajikuri, J., Yamamoto, T., Watanabe, Y., Suzuki, Y., Katsuya, H., Itoh, T., 2005. Reduced hyperpolarization in endothelial cells of rabbit aortic valve following chronic nitroglycerine administration. Br. J. Pharmacol. 146, 487–497. Lang, N.N., Luksha, L., Newby, D., Kublickiene, K., 2006. Connexin 43 mediates endothelium-derived hyperpolarising factor induced vasodilatation in subcutaneous resistance arteries from healthy pregnant women. Am. J. Physiol.: Heart Circ. Physiol. Nov 3 [Electronic publication ahead of print]. Looft-Wilson, R.C., Payne, G.W., Segal, S.S., 2004. Connexin expression and conducted vasodilation along arteriolar endothelium in mouse skeletal muscle. J. Appl. Physiol. 97, 1152–1158. Marrelli, S.P., Eckmann, M.S., Hunte, M.S., 2003. Role of endothelial intermediate conductance KCa channels in cerebral EDHF-mediated dilations. Am. J. Physiol.: Heart Circ. Physiol. 285, H1590–H1599.

200

L. Ceroni et al. / European Journal of Pharmacology 560 (2007) 193–200

Mather, S., Dora, K.A., Sandow, S.L., Winter, P., Garland, C.J., 2005. Rapid endothelial cell-selective loading of connexin 40 antibody blocks endothelium-derived hyperpolarizing factor dilation in rat small mesenteric arteries. Circ. Res. 97, 399–407. Matoba, T., Shimokawa, H., Nakashima, M., Hirakawa, Y., Mukai, Y., Hirano, K., Kanaide, H., Takeshita, A., 2000. Hydrogen peroxide is an endotheliumderived hyperpolarizing factor in mice. J. Clin. Invest. 106, 1521–1530. McGuire, J.J., Ding, H., Triggle, C.R., 2001. Endothelium-derived relaxing factors: a focus on endothelium-derived hyperpolarizing factor(s). Can. J. Physiol. Pharm. 79, 443–470. McGuire, J.J., Hollenberg, M.D., Andrade-Gordon, P., Triggle, C.R., 2002. Multiple mechanisms of vascular smooth muscle relaxation by the activation of proteinase-activated receptor 2 in mouse mesenteric arterioles. Br. J. Pharmacol. 135, 155–169. McGuire, J.J., Hollenberg, M.D., Bennett, B.M., Triggle, C.R., 2004. Hyperpolarization of murine small caliber mesenteric arteries by activation of endothelial proteinase-activated receptor 2. Can. J. Physiol. Pharm. 82, 1103–1112. McLenachan, J.M., Vita, J., Fish, D.R., Treasure, C.B., Cox, D.A., Ganz, P., Selwyn, A.P., 1990. Early evidence of endothelial vasodilator dysfunction at coronary branch points. Circulation 82, 1169–1173. Mullershausen, F., Friebe, A., Feil, R., Thompson, W.J., Hofmann, F., Koesling, D., 2003. Direct activation of PDE5 by cGMP: long-term effects within NO/ cGMP signaling. J. Cell Biol. 160, 719–727. Nagao, T., Illiano, S., Vanhoutte, P.M., 1992. Heterogeneous distribution of endothelium-dependent relaxations resistant to NG-nitro-L-arginine in rats. Am. J. Physiol. 263, H1090–H1094. Pannirselvam, M., Simon, V., Verma, S., Anderson, T., Triggle, C.R., 2003. Chronic oral supplementation with sepiapterin prevents endothelial dysfunction and oxidative stress in small mesenteric arteries from diabetic (db/db) mice. Br. J. Pharmacol. 140, 701–706. Pannirselvam, M., Ding, H., Anderson, T.J., Triggle, C.R., 2006. Pharmacological characteristics of endothelium-derived hyperpolarizing factormediated relaxation of small mesenteric arteries from db/db mice. Eur. J. Pharmacol. 551 (1–3), 98–107. Parkington, H.C., Tare, M., Tonta, M.A., Coleman, H.A., 1993. Stretch revealed three components in the hyperpolarization of guinea-pig coronary artery in response to acetylcholine. J. Physiol. 465, 459–476. Sandow, S.L., 2004. Factors, fiction and endothelium-derived hyperpolarizing factor. Clin. Exp. Pharmacol. Physiol. 31, 563–570.

Sandow, S.L., Hill, C.E., 2000. Incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat is suggestive of a role in endothelium-derived hyperpolarizing factor-mediated responses. Circ. Res. 86, 341–346. Sandow, S.L., Looft-Wilson, R., Doran, B., Grayson, T.H., Segal, S.S., Hill, C.E., 2003. Expression of homocellular and heterocellular gap junctions in hamster arterioles and feed arteries. Cardiovasc. Res. 60, 643–653. 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. Taylor, M.S., Bonev, A.D., Gross, T.P., Eckman, D.M., Brayden, J.E., Bond, C.T., Adelman, J.P., Nelson, M.T., 2003. Altered expression of small-conductance Ca2+-activated K+ (SK3) channels modulates arterial tone and blood pressure. Circ. Res. 93, 124–131. Triggle, C.R., Howarth, A., Cheng, Z.J., Ding, H., 2005. Twenty-five years since the discovery of endothelium-derived relaxing factor (EDRF): does a dysfunctional endothelium contribute to the development of type 2 diabetes? Can. J. Physiol. Pharm. 83, 681–700. Ujiie, H., Chaytor, A.T., Bakker, L.M., Griffith, T.M., 2003. Essential role of Gap junctions in NO- and prostanoid-independent relaxations evoked by acetylcholine in rabbit intracerebral arteries. Stroke 34, 544–550. Vanhoutte, P.M., 2004. Endothelium-dependent hyperpolarizations: the history. Pharmacol. Res. 49, 503–508. Villalobos, C., Shakkottai, V.G., Chandy, K.G., Michelhaugh, S.K., Andrade, R., 2004. SKCa channels mediate the medium but not the slow calciumactivated afterhyperpolarization in cortical neurons. J. Neurosci. 24, 3537–3542. von der Weid, P.Y., Coleman, H.A., 2005. Reduced vascular reactivity after chronic nitroglycerine administration: EDHF mechanism is also downregulated. Br. J. Pharmacol. 146, 479–480. Waldron, G.J., Ding, H., Lovren, F., Kubes, P., Triggle, C.R., 1999. Acetylcholine-induced relaxation of peripheral arteries isolated from mice lacking endothelial nitric oxide synthase. Br. J. Pharmacol. 128, 653–658. Yamamoto, T., Kajikuri, J., Watanabe, Y., Suzuki, Y., Suzumori, K., Itoh, T., 2005. Chronic nitroglycerine administration reduces endothelial nitric oxide production in rabbit mesenteric resistance artery. Br. J. Pharmacol. 146, 534–542.