The endothelium plays a central role in the regulation of

Transcription

1 Involvement of Myoendothelial Gap Junctions in the Actions of Endothelium-Derived Hyperpolarizing Factor Shaun L. Sandow,* Marianne Tare,* Harold A. Coleman, Caryl E. Hill, Helena C. Parkington Abstract The nature of the vasodilator endothelium-derived hyperpolarizing factor (EDHF) is controversial, putatively involving diffusible factors and/or electrotonic spread of hyperpolarization generated in the endothelium via myoendothelial gap junctions (MEGJs). In this study, we investigated the relationship between the existence of MEGJs, endothelial cell (EC) hyperpolarization, and EDHF-attributed smooth muscle cell (SMC) hyperpolarization in two different arteries: the rat mesenteric artery, where EDHF-mediated vasodilation is prominent, and the femoral artery, where there is no EDHF-dependent relaxation. In the rat mesenteric artery, stimulation of the endothelium with acetylcholine (ACh) evoked hyperpolarization of both ECs and SMCs, and characteristic pentalaminar MEGJs were found connecting the two cell layers. In contrast, in the femoral artery, ACh evoked hyperpolarization in only ECs but not in SMCs, and no MEGJs were present. Selective hyperpolarization of ECs or SMCs evoked hyperpolarization in the other cell type in the mesenteric artery but not in the femoral artery. Disruption of gap junctional coupling using the peptide Gap 27 markedly reduced the ACh-induced hyperpolarization in SMCs, but not in ECs, of the mesenteric artery. These results show that transfer of EC hyperpolarization or of a small molecule to SMCs through MEGJs is essential and sufficient to explain EDHF. (Circ Res. 2002;90: ) Key Words: endothelium-derived hyperpolarizing factor myoendothelial gap junctions endothelium smooth muscle electrical coupling The endothelium plays a central role in the regulation of vascular tone. 1 The endothelium is capable of exerting a profound relaxing influence on the underlying smooth muscle, mediated by at least three different factors, depending on the vascular bed. These include nitric oxide (NO) and prostacyclin, both diffusible factors. 2 6 In addition, after blockade of NO and prostacyclin synthesis, stimulation of the endothelium is capable of evoking vascular smooth muscle relaxation that has been attributed to a third factor, endothelium-derived hyperpolarizing factor (EDHF) The hallmark of the vasorelaxation attributed to EDHF is that it is accompanied by membrane hyperpolarization. Consensus regarding the nature of EDHF is lacking, with evidence suggesting the involvement of a diffusible factor(s) released from the endothelium in some vascular beds. 11,12,16,17 Others propose that the hyperpolarization generated in endothelial cells (ECs) is capable of spreading electrotonically to the underlying smooth muscle cells (SMCs), 18,19,21,23,25 most likely via myoendothelial gap junctions (MEGJs). 9,18,19,21,23 26 Evidence to support the transfer of a small molecule (eg, camp) via MEGJs has also been presented. 20,22 Support for the MEGJ hypothesis has come from studies in which gap junctions have been pharmacologically blocked, either with peptide mimetics of connexin 43 (Cx43) or with glycyrrhetinic acid derivatives. 13,26,31 33 Although the former have been demonstrated to indeed reduce dye coupling between Cx-transfected cells, 34 the latter seem to have some nonspecific actions ,35 An alternative and complementary approach would be to study an artery that lacked an EDHF-dependent relaxation and to test for the existence of agonist-induced hyperpolarization in the ECs and for the presence of MEGJs. The femoral artery represents such a tissue because EDHF does not contribute to endothelium-dependent SMC relaxation in this vessel. 36,37 In the present study, this artery was compared with the mesenteric artery, where vasodilation to EDHF is prominent. We tested the hypothesis that the ECs of both beds could generate hyperpolarization, but that the lack of EDHF in the SMCs of the femoral artery was due to the absence of MEGJs. Materials and Methods The experiments were carried out under the guidelines of the National Health and Medical Research Council of Australia code of practice for the care and use of animals for scientific purposes. Electrophysiological Experiments Sixteen-week-old male Wistar rats (Monash University Central Animal Services, Victoria, Australia) were anesthetized with chlo- Original received August 15, 2001; resubmission received December 27, 2001; revised resubmission received March 25, 2002; accepted April 12, From the Department of Physiology (M.T., H.A.C., H.C.P.), Monash University, Clayton, Victoria, Australia; Division of Neuroscience (S.L.S., C.E.H.), John Curtin School of Medical Research, Australian National University, Canberra, Australian Capital Territory, Australia. *Both authors contributed equally to this study. Correspondence to Marianne Tare, Department of Physiology, PO Box 13F, Monash University, Victoria 3800, Australia. marianne.tare@med.monash.edu.au 2002 American Heart Association, Inc. Circulation Research is available at DOI: /01.RES

2 Sandow et al Involvement of MEGJs in EDHF 1109 roform and killed by exsanguination. Femoral (first order) and mesenteric (first and second order) arteries were isolated. A segment of artery was then cut longitudinally and pinned to the base of a recording chamber with the endothelium uppermost. In some experiments, arterial segments were left as intact tubes. Preparations were continuously superfused with physiological saline (mmol/l: NaCl 120, KCl 5, CaCl 2 2.5, MgSO 4 1.2, KH 2 PO 4 1, NaHCO 3 25, and glucose 11) at 35 C and bubbled with O 2 95% and CO 2 5%. The NO synthase inhibitor N -nitro-l-arginine methyl ester (L-NAME; 100 mol/l) and the cyclooxygenase inhibitor indomethacin (1 mol/l) were present in all experiments to inhibit NO and prostanoid production, respectively. Membrane potentials of ECs and SMCs were recorded using intracellular glass microelectrodes with resistances of 100 to 150 M. The tips of the electrodes were filled with 2% Lucifer Yellow CH and backfilled with 1 mol/l KCl. 33,35 During the experiments, Lucifer Yellow readily diffused into the impaled cell, thus allowing unequivocal identification of every cell impaled. The ECs were stimulated for 1 minute by the addition of acetylcholine (ACh) to the superfusate to evoke the response attributed to EDHF. Selective stimulation of ECs with 1-ethyl-2- benzimidazolinone (1-EBIO) and of SMCs with levcromakalim was also for 1 minute by addition of the channel opener to the superfusate. Cell-to-cell coupling was disrupted by the connexin mimetic peptide Gap (sequence SRPTEKTIFII; Biomolecular Resource Facility, John Curtin School of Medical Research, Australian National University; purity 97%). Gap 27 was dissolved directly into the physiological saline and the solution was oxygenated, warmed, and recirculated over the preparations for 1 to 3 hours. In a pilot study, we confirmed that this recirculation procedure had no effect on the resting membrane potential or on the hyperpolarization evoked by 10 mol/l ACh in either ECs or SMCs. Responses to ACh were tested at 0.5- and/or 1-hour intervals. Electron Microscopy Animals and Tissue Preparation Sixteen-week-old male Wistar rats were anesthetized with an intraperitoneal injection of ketamine and xylazine 2% (44 and 8 mg/kg, respectively) and perfused ( 60 mm Hg) via the left ventricle with saline (0.9% NaCl) containing 0.1% NaNO 3, 0.1% BSA, and 5 U/mL heparin at 25 C. The right atria were cut to allow an outflow of perfusate. Once cleared of blood, animals were perfused with 3% glutaraldehyde and 1% paraformaldehyde in 0.1 mmol/l sodium cacodylate with 0.2 mmol/l CaCl 2 6H 2 O, 0.15 mol/l sucrose, and 10 mmol/l betaine (ph 7.35) at 25 C for 10 minutes. Twomillimeter segments of the first-order femoral and mesenteric arteries were then further immersion-fixed in the same solution at 4 C for 2 hours. Care was taken to obtain the same region of vessel from each animal. Tissues were subsequently postfixed in 2% osmium tetroxide in the same buffer for 2 hours, stained with saturated aqueous uranyl acetate for 2 hours, and embedded in Araldite 502 according to conventional procedures. Serial-Section Electron Microscopy Methods for serial-section analysis were similar to those previously described by Sandow and Hill. 25 Briefly, 50 transverse serial sections (one segment totaling 5 m of vessel, each from a different animal) were cut from both the femoral and mesenteric arteries of three animals. Low-magnification micrographs ( 2500) of the central section of each series were taken on plate film on a Hitachi 7100 transmission electron microscope, and the vessel circumference was estimated as the length of the internal elastic lamina (IEL). The number of medial SMC layers was determined by averaging the number of individual SMC profiles, 5 m in length, from the outer edge of the IEL to the inner edge of the external elastic lamina along each of four linear plots 90 apart, from each of 3 vessels each from a different animal. To detect MEGJs, the IEL in each serial section was examined at All MEGJs through serial sections were photographed on plate film at high magnification ( to ), and the numbers of MEGJs with characteristic pentalaminar membrane structure were noted. Reconstruction was performed using the MCID Imaging system (Imaging Research). Photographed MEGJ profiles and their surrounding EC and SMC regions were digitized from contact prints using a flatbed scanner (HP Scanjet 4c), at a resolution of 300 dots per inch. Selected gap junctions between adjacent SMCs and adjacent ECs were photographed at to Morphology of ECs and Determination of the Number of MEGJs per EC To determine EC dimensions, whole-mount regions of femoral and mesenteric arteries (as above) were processed for Cx40 immunohistochemistry (1:100), because this staining highlights the cell periphery. 38 Antibodies were raised in sheep against a peptide corresponding to amino acids 254 to 270 of rat Cx40, as used previously. 39 Staining was detected after incubation in Cy3-conjugated donkey anti-goat immunoglobulins (1:100, Jackson Immuno-Research) for 1 hour at 25 C. Preparations were mounted in buffered glycerol, and 1- m optical sections were taken through the EC layer using a Leica TCS 4D-confocal microscope, with images being recombined to produce a single image of the labeled ECs. Morphometric measurements of randomly selected ECs were made using the MCID system. The number of MEGJs per EC was estimated from the vessel circumference at the level of the IEL, the number of ECs in this region, and MEGJs present in the 5- m region. Statistical Analysis Data are expressed as mean SEM; n indicates the number of animals tested. Student s t test, paired or unpaired, as appropriate, was used to test significance using the software package InStat 3 (GraphPad Corp). Concentration-hyperpolarization curves were fitted using the least-squares method (GraphPad), which were used to determine the pd 2 ( log EC 50 ). For the histological data, statistical significance was tested using a one-way ANOVA followed by pairwise t tests with Bonferroni correction for multiple group comparisons and with paired or unpaired t tests for groups of two. Values of P 0.05 were considered statistically significant. Results EDHF-Attributed Responses in Identified ECs and SMCs Resting membrane potentials (RMPs) in dye-identified femoral artery ECs were only 26 1 mv(n 5) compared with 58 2 mv in the SMCs of the same tissues (Figure 1). In contrast, the RMPs in mesenteric ECs were not significantly different from those of mesenteric SMCs (Figures 1A through 1C; Table). RMPs of SMCs in femoral and mesenteric arteries were not significantly different (Figure 1C; Table). Stimulation of the endothelium of the femoral artery with ACh evoked 45 2-mV (n 5) hyperpolarization in ECs but failed to initiate any hyperpolarization in the SMCs (Figures 1B and 1C), even in SMCs that were immediately beneath the IEL. Conversely, in the mesenteric artery, ACh evoked comparable concentration-dependent hyperpolarizations in both ECs and SMCs (Figures 1B and 1C; Table). The maximum amplitude of EC hyperpolarization in the mesenteric artery was only half of that in femoral ECs, but the level of membrane potential reached during the maximum hyperpolarization was not different in the two arteries (Figure 1C). Intermediate- and small-conductance Ca 2 -activated K channels have been implicated in EDHF in many arteries. 8,14,36,40,41 Mesenteric EC and SMC hyperpolarizations in response to ACh (10 mol/l) were abolished by charybdotoxin (ChTx; 30 to 50 nmol/l) plus apamin (0.1 to 0.5

3 1110 Circulation Research May 31, 2002 Figure 1. Identified SMCs and ECs and EDHFattributed hyperpolarizations recorded in the two cell types. A, Lucifer Yellow filled SMC (left) and ECs (right). Longitudinal vessel axis runs left to right. Bar 50 m. B, Membrane potential recordings in SMCs (left) and ECs (right) from femoral and mesenteric arteries. ACh evoked hyperpolarization in ECs of both arteries. EDHF-attributed hyperpolarization was absent in SMCs of the femoral artery but was recorded in SMCs of the mesenteric artery. C, Concentration-hyperpolarization curves evoked by ACh in SMCs (left) and ECs (right) of femoral (n 5) and mesenteric (n 5) arteries. The NO synthase inhibitor L-NAME (100 mol/l) and the cyclooxygenase inhibitor indomethacin (1 mol/l) were present throughout. mol/l; n 4), which are blockers of these channels. Removal of the endothelium abolished EDHF-mediated SMC hyperpolarization in the mesenteric artery. In the femoral artery, ChTx plus apamin all but abolished (95 2%, n 4) EC hyperpolarization. That the hyperpolarization attributed to EDHF was recorded in the ECs, but not in SMCs, of the femoral artery is Properties of ECs and SMCs in Mesenteric and Femoral Arteries consistent with findings that the ChTx- and apamin-sensitive K channels underpinning this response are located on the ECs. 40,41 Identification of MEGJs MEGJs did not occur in the femoral artery: there were no MEGJs in the 5- m region examined (n 3), equivalent to an Mesenteric Artery Femoral Artery Vessel circumference, m (3) (3) No. of SMC layers (3) (3) SMCs RMP, mv 57 1 (5) 58 2 (5) Maximum EDHF-mediated hyperpolarization, mv 18 2 (5) 0 (5) pd (5) ECs RMP, mv 54 1 (5) 26 3 (5) Maximum hyperpolarization, mv 22 2 (5) 45 2 (5) pd (5) (5) No. of MEGJs in 5- m vessel length (3) 0 (3) EC dimensions, m (4; 182 cells) (4; 154 cells) Results are mean SEM, with No. of animals in parentheses. First- and second-order mesenteric arteries and first-order femoral arteries were used.

4 Sandow et al Involvement of MEGJs in EDHF 1111 Figure 3. Selective generation of hyperpolarization in SMCs and ECs. A, Levcromakalim generated hyperpolarization in SMCs of both mesenteric (n 7) and femoral (n 6) arteries. The hyperpolarization was recorded in mesenteric ECs (n 6) but not femoral ECs (n 6). B, 1-EBIO evoked hyperpolarization of ECs in both arteries. Hyperpolarization was recorded in SMCs of mesenteric (n 7) but not femoral arteries (n 6). Figure 2. Transmission electron micrographs of SMC and EC relationships in femoral and mesenteric arteries of the rat. Only nonjunctional projections from ECs and SMCs (A and B; small asterisks) were present in femoral arteries (A and B). In contrast, MEGJs were present in mesenteric arteries (C and D; pentalaminar region marked with arrows and inset, between arrows). Gap junctions were present between SMCs (smc; B, arrowhead, top inset) and ECs (ec; B, arrowhead, bottom inset) in both arteries. Gaps in the internal elastic lamina (iel) were present in both arteries (A; large asterisk), although MEGJs were present in only mesenteric arteries. Bar 0.5 m in A, C, and D; 0.25 m inb; and 25 nm in the inset of B through D. area of the endothelium occupied by 13.3 ECs (Figure 2A). In contrast, in the mesenteric artery, MEGJs occurred with a frequency of 0.79 per EC, ie, MEGJs in the 5- m region examined (n 3), equivalent to an area of the endothelium occupied by 9.2 ECs (Figures 2C and 2D). All MEGJs in the mesenteric arteries occurred on projections arising from ECs. Pentalaminar gap junctions were identified between adjacent SMCs and between adjacent ECs in both vessels (eg, Figure 2B). Vessel circumference and the number of SMC layers were significantly less, whereas EC dimensions were not significantly different in the mesenteric compared with the femoral artery (Table). Separate Generation of Hyperpolarization in ECs and SMCs We further tested the existence or otherwise of electrical crosstalk between the two cell types in each artery by generating hyperpolarizations in either SMCs or ECs. In rat and rabbit mesenteric arteries, the ATP-sensitive K channel opener levcromakalim generates hyperpolarization in SMCs. 42,43 Levcromakalim evoked SMC hyperpolarization in both femoral and mesenteric arteries, but hyperpolarization was recorded in the ECs in only the mesenteric artery (Figure 3A). Intermediate-conductance Ca 2 -sensitive K channels, present on native ECs, can be activated using 1-EBIO. 44,45 Concentrations 100 mol/l selectively evoked EC hyperpolarization in both arteries (Figure 3B). SMC hyperpolarization also occurred in the mesenteric, but not in femoral, SMCs (Figure 3B). In both arteries, EC hyperpolarization evoked by 1-EBIO was blocked by ChTx plus apamin. 1-EBIO (100 mol/l) failed to evoke SMC hyperpolarization in mesenteric arteries denuded of endothelium. Taken together, these results demonstrate that there is electrical coupling between the ECs and SMCs in the mesenteric, but not in the femoral, artery. Influence of the Gap Junction Inhibitor Gap 27 on EDHF in the Mesenteric Artery EC and SMC hyperpolarizations were recorded in response to 1-minute stimulation with 10 mol/l ACh. Incubation with 300 mol/l Gap 27 (for 1 to 3 hours) had no effect on the RMP of the SMCs, but the amplitude of EDHF-attributed hyperpolarization in SMCs was significantly attenuated (progressive reduction to 33 11% over 3 hours, n 5; Figure 4). In time controls run for 2 to 3 hours in the absence of Gap 27, EDHF-attributed hyperpolarization in SMCs was not significantly different to the initial response (97 5% of the initial response, n 7). In contrast to the effect of Gap 27 on SMCs, EDHF-attributed hyperpolarization in ECs after incubation in Gap 27 for 1 to 3 hours was not significantly different to their initial response (110 9% of the initial response, n 5, Figure 4). Discussion The present study demonstrates that a lack of MEGJs in the femoral artery accounts for the absence of electrical coupling between ECs and SMCs and a lack of EDHF in this vessel. The large difference in RMPs of ECs and SMCs in the femoral artery is consistent with this absence of functional

5 1112 Circulation Research May 31, 2002 Figure 4. Effects of Gap 27 on EDHF-attributed hyperpolarization in ECs and SMCs of mesenteric arteries. EC (n 5) and SMC (n 5) hyperpolarization recorded before and after incubation with Gap 27. Only SMC hyperpolarization was significantly reduced after exposure to Gap 27. *Significantly different (P 0.05). electrical connectivity. Endothelium-dependent relaxation in the femoral artery thus relies entirely on the release of a diffusible factor, NO. 36,37 Conversely, MEGJs allow the transmission of electrical responses between ECs and SMCs, and they function as a pathway for the spread of EC hyperpolarization to SMCs in the mesenteric artery. This also accounts for the similar RMPs in ECs and SMCs in this vessel. The role of MEGJs in conducting EDHF-attributed hyperpolarization, either electrically or via a small intermediate molecule, 12,20,22 to the SMCs in the mesenteric artery is further supported by the effectiveness of Gap 27 in reducing the hyperpolarization in only the SMCs, while leaving intact the response in the ECs, similar to previous observations in this vascular bed. 30 The prominence of EDHF generally increases as vessel size decreases. 26 A strong syncytial relationship between ECs and SMCs exists in hamster retractor muscle feed arteries 24 and mesenteric arterioles, 33 where EDHF is prominent. The role of EDHF in vessels with larger numbers of medial SMCs may be diminished, despite the existence of MEGJs, because of the dissipation of the hyperpolarizing current into the electrical sink of the increased number of layers of SMCs. 9 These large arteries thus rely on the release of a diffusible factor. In the present study, however, the absence of an EDHF response in the femoral artery seems due to the absence of MEGJs and the absence of an EDHF-mediated hyperpolarization in the SMCs. At least two EDHFs can be recognized on the basis of their susceptibility to ion channel/transport blockade. In some arteries, the hyperpolarization is abolished by blockers of large-conductance Ca 2 -activated K channels (BK Ca channels) such as iberiotoxin. Products of the cytochrome P-450 pathway, epoxyeicosatrienoic acids (EETs) produced by ECs, are likely candidates because they activate BK Ca channels on SMCs. 46,47 In other vessels, especially small arteries and arterioles, however, the EDHF hyperpolarization is resistant to iberiotoxin but is blocked by the combination of ChTx plus apamin ,31,33,41 In some of these vessels, the responses are also blocked by Ba 2 and ouabain and these have been attributed to the release of K from ECs acting on inwardly rectifying K channels and the Na /K -ATPase on the SMCs. 16 In many other vessels, however, the EDHF hyperpolarization is resistant to iberiotoxin or Ba 2 and ouabain. 13,33 The hyperpolarization generated in the ECs in both the femoral and mesenteric arteries in the present study is blocked by ChTx plus apamin. Because the ChTx- and apamin-sensitive K channels underpinning the EDHF response are located on the ECs, 33,40,41 a recurring hypothesis has been that the SMC hyperpolarization attributed to this particular EDHF relies on the flow of current generated in ECs via MEGJs. Our study has tested this hypothesis from a different perspective by using the femoral artery, which lacks EDHF-attributed SMC relaxation. 36,37 The responses in this artery were compared with those occurring in the mesenteric artery, which exhibits EDHF-attributed SMC relaxation. 16,41 We demonstrated the presence of MEGJs and the spread of a ChTx- plus apaminsensitive hyperpolarization to the SMCs in mesenteric arteries, although the involvement of a small chemical intermediary in this transfer cannot be excluded. 12,13,20,22 Block of this EC hyperpolarization with ChTx plus apamin blocks the EDHF response in the SMCs, demonstrating the essential nature of EC hyperpolarization to EDHF. In femoral arteries, we found no MEGJs linking the two cellular layers, and the ChTx- plus apamin-sensitive hyperpolarization in the ECs was not recorded in the SMCs. Thus, the ultimate model of MEGJ block, an absence of MEGJs as in the femoral artery, results in a lack of EDHF-mediated SMC hyperpolarization. The most economical explanation of these observations is that EDHF arises from the spread of EC hyperpolarization or of a small molecule in response to this hyperpolarization, via MEGJs to the SMCs. Thus, MEGJs provide the critical link for the action of the ChTx- plus apamin-sensitive EDHF. Acknowledgments This work was funded by the National Health and Medical Research Council (NH&MRC) of Australia and the National Heart Foundation of Australia. S.L.S. was supported by a NH&MRC Peter Doherty Fellowship. References 1. Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature. 1980;288: Furchgott RF. Studies on relaxation of rabbit aorta by sodium nitrite: the basis for the proposal that the acid-activatable inhibitory factor from bovine retractor penis is inorganic nitrite and the endothelium-derived relaxing factor is nitric oxide. In: Vanhoutte PM, ed. Vasodilatation: Vascular Smooth Muscle, Peptides, Autonomic Nerves, and Endothelium. New York, NY: Raven Press; 1988: Rapoport RM, Murad F. Agonist-induced endothelium-dependent relaxation in rat thoracic aorta may be mediated through cgmp. Circ Res. 1983;52: Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endotheliumderived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A. 1987;84: Palmer RM, Ashton DS, Moncada S. Vascular endothelial cells synthesize nitric oxide from L-arginine. Nature. 1988;333: Moncada S, Gryglewski R, Bunting S, Vane JR. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature. 1976;263: Chen G, Suzuki H, Weston AH. Acetylcholine releases endotheliumderived hyperpolarizing factor and EDRF from rat blood vessels. Br J Pharmacol. 1988;95: Edwards G, Weston AH. Endothelium-derived hyperpolarizing factor: a critical appraisal. Prog Drug Res. 1998;50:

6 Sandow et al Involvement of MEGJs in EDHF Beny J-L. Information networks in the arterial wall. News Physiol Sci. 1999;14: Feletou M, Vanhoutte PM. The third pathway: endothelium-dependent hyperpolarization. J Physiol Pharmacol. 1999;50: Campbell WB, Harder DR. Prologue: EDHF what is it? Am J Physiol. 2001;280:H2413 H Campbell WB, Gauthier KM. What is new in endothelium-derived hyperpolarizing factors? Curr Opin Nephrol Hypertens Res. 2002;11: McGuire JJ, Ding H, Triggle CR. Endothelium-derived relaxing factors: a focus on endothelium-derived hyperpolarizing factor(s). Can J Physiol Pharmacol. 2001;79: Plane F, Holland M, Waldron GJ, Garland CJ, Boyle JP. Evidence that anandamide and EDHF act via different mechanisms in rat isolated mesenteric arteries. Br J Pharmacol. 1997;121: Zygmunt PM, Hogestatt ED. Role of potassium channels in endothelium-dependent relaxation resistant to nitroarginine in the rat hepatic artery. Br J Pharmacol. 1996;117: Edwards G, Dora KA, Gardener MJ, Garland CJ, Weston AH. K is an endothelium-derived hyperpolarizing factor in rat arteries. Nature. 1998; 396: Fisslthaler B, Popp R, Kiss L, Potente M, Harder DR, Fleming I, Busse R. Cytochrome P450 2C is an EDHF synthase in coronary arteries. Nature. 1999;410: Little TL, Xia J, Duling BR. Dye tracers define differential endothelial and smooth muscle coupling patterns within the arteriolar wall. Circ Res. 1995;76: Beny J-L. Electrical coupling between smooth muscle cells and endothelial cells in pig coronary arteries. Pflugers Arch. 1997;433: Taylor HJ, Chaytor AT, Edwards DH, Griffith TM. Gap junctiondependent increases in smooth muscle camp underpin the EDHF phenomenon in rabbit arteries. Biochem Biophys Res Commun. 2001;283: Chaytor AT, Martin PE, Edwards DH, Griffith TM. Gap junctional communication underpins EDHF-type relaxations evoked by ACh in the rat hepatic artery. Am J Physiol. 2001;280:H2441 H Chaytor AT, Taylor HJ, Griffith TM. Gap junction-dependent and -independent EDHF-type relaxations may involve smooth muscle camp accumulation. Am J Physiol. 2002;282:H1548 H Yamamoto Y, Imaeda K, Suzuki H. Endothelium-dependent hyperpolarization and intercellular electrical coupling in guinea-pig mesenteric arterioles. J Physiol. 1999;514: Emerson GG, Segal SS. Electrical coupling between endothelial cells and smooth muscle cells in hamster feed arteries: role in vasomotor control. Circ Res. 2000;87: Sandow SL, Hill CE. The incidence of myoendothelial gap junctions in the proximal and distal mesenteric arteries of the rat is suggestive of a role in EDHF-mediated responses. Circ Res. 2000;86: Hill CE, Phillips JK, Sandow SL. Heterogeneous control of blood flow amongst different vascular beds. Med Res Rev. 2001;21: Chaytor AT, Evans WH, Griffith TM. Peptides homologous to extracellular loop motifs of connexin 43 reversibly abolish contractile activity in rabbit arteries. J Physiol. 1997;503: Chaytor AT, Evans WH, Griffith TM. Central role of heterocellular gap junctional communication in endothelium-dependent relaxations of rabbit arteries. J Physiol. 1998;508: Dora KA, Martin PEM, Chaytor AT, Evans H, Garland CJ, Griffith TM. Role of heterocellular gap junctional communication in endotheliumdependent smooth muscle hyperpolarization: inhibition by a connexinmimetic peptide. Biochem Biophys Res Comm. 1999;254: Edwards G, Feletou M, Gardener MJ, Thollon C, Vanhoutte PM, Weston AH. Role of gap junctions in the responses to EDHF in rat and guinea-pig small arteries. Br J Pharmacol. 1999;128: Imaeda K, Yamamoto Y, Fukuta H, Koshita M, Suzuki H. Hyperpolarization-induced dilatation of submucosal arterioles in the guinea-pig ileum. Br J Pharmacol. 2000;131: Santicioli P, Maggi CA. Effect of 18 -glycyrrhetinic acid on electromechanical coupling in the guinea-pig renal pelvis and ureter. Br J Pharmacol. 2000;129: Coleman HA, Tare M, Parkington HC. K currents underlying the action of endothelium-derived hyperpolarizing factor in guinea-pig and human blood vessels. J Physiol. 2001;531: Berman RS, Martin PE, Evans WH, Griffith TM. Relative contributions of NO and gap junctional communication to endothelium-dependent relaxations of rabbit resistance arteries vary with vessel size. Microvasc Res. 2002;63: Tare M, Coleman HA, Parkington HC. Glycyrrhetinic derivatives inhibit hyperpolarization in endothelial cells of guinea pig and rat arteries. Am J Physiol. 2002;282:H335 H Zygmunt PM, Ryman T, Hogestatt ED. Regional differences in endothelium-dependent relaxation in the rat: contribution of nitric oxide and nitric oxide-independent mechanisms. Acta Physiol Scand. 1995;155: Wigg SJ, Tare M, Tonta MA, O Brien RC, Meridith IT, Parkington HC. Comparison of the effects of diabetes mellitus on an EDHF-dependent and an EDHF-independent artery. Am J Physiol. 2001;281:H232 H Rummery N, McKenzie K, Whitworth JA, Hill CE. Decreased endothelial size and connexin expression in rat caudal arteries during hypertension. J Hypertens. 2002;20: Severs NJ. Cardiovascular disease. Novartis Found Symp. 1999;219: Marchenko SM, Sage SO. Calcium-activated potassium channels in the endothelium of intact rat aorta. J Physiol. 1996;492: Doughty JM, Plane F, Langton PD. Charybdotoxin and apamin block EDHF in rat mesenteric artery if selectively applied to the endothelium. Am J Physiol. 1999;276:H1107 H Murai T, Muraki K, Imaizumi Y, Watanabe M. Levcromakalim causes indirect endothelial hyperpolarization via a myo-endothelial pathway. Br J Pharmacol. 1999;128: White R, Hiley CR. Hyperpolarisation of rat mesenteric endothelial cells by ATP-sensitive K channel openers. Eur J Pharmacol. 2000;397: Edwards G, Gardener MJ, Feletou M, Brady G, Vanhoutte PM, Weston AH. Further investigation of endothelium-derived hyperpolarizing factor (EDHF) in rat hepatic artery: studies using 1-EBIO and ouabain. Br J Pharmacol. 1999;128: Walker SD, Dora KA, Ings NT, Crane GJ, Garland CJ. Activation of endothelial cell IK Ca with 1-ethyl-2-benzimidazolinone evokes smooth muscle hyperpolarization in rat isolated mesenteric artery. Br J Pharmacol. 2001;134: Campbell WB, Gebremedhin D, Pratt PF, Harder DR. Identification of epoxyeicosatrienoic acids as endothelium-derived hyperpolarizing factors. Circ Res. 1996;78: Popp R, Bauersachs J, Hecker M, Fleming I, Busse R. A transferable, -naphthoflavone-inducible, hyperpolarizing factor is synthesized by native and cultured porcine coronary endothelial cells. J Physiol. 1996; 497: