NO-induced activation of cGMP-dependent protein kinase (PKG) increases the open probability of large conductance Ca2؉ -activated K ؉ channels and results in smooth muscle relaxation. However, the molecular mechanism of channel regulation by the NO-PKG pathway has not been determined on cloned channels. The present study was designed to clarify PKG-mediated modulation of channels at the molecular level. The cDNA encoding the ␣-subunit of the large conductance Large conductance Ca 2ϩ -activated K ϩ (BK Ca ) 1 channels are ubiquitously distributed among tissues and are particularly abundant in smooth muscle (1, 2). The activity of BK Ca channels is regulated by membrane potential, intracellular Ca 2ϩ , and phosphorylation (3, 4). Although BK Ca channels are usually not involved in setting resting potential, they play a key role as a negative feedback mechanism to limit depolarization and contraction (5-7). Activation of BK Ca channels is increased by nitric oxide (NO) and atrial natriuretic peptide, which hyperpolarize the membrane and increase the sensitivity of BK Ca channels to Ca 2ϩ (8 -11). Membrane hyperpolarization closes voltage-dependent Ca 2ϩ channels, reduces Ca 2ϩ influx, and leads to a reduction in intracellular Ca 2ϩ concentration and relaxation (1). NO has been reported to stimulate BK Ca channels directly as well as through stimulation of guanylate cyclase and the subsequent increase in cGMP (12-15). In addition, activation of BK Ca channels plays an important role in NO-induced relaxation of smooth muscle (16 -20). cGMP activates cGMP-dependent protein kinase (PKG), which phosphorylates various cytosolic and membrane proteins that regulate smooth muscle tone either directly or indirectly (21,22). Recent studies in native cells suggest that PKG activates BK Ca channels through phosphorylation of the channel (23). These results are supported by biochemical studies of cloned BK Ca channels, which demonstrate PKG-induced phosphorylation of the channel (24).The primary sequence of BK Ca has been determined using molecular cloning techniques in Drosophila (25) and mammals (26 -28). These studies indicate that BK Ca isoforms belong to the voltage-gated K ϩ (K V ) channel superfamily. The primary sequence of the S1-S6 segment of BK Ca channels is homologous to the corresponding regions in K V channels. The long carboxyl terminus is the region of Ca 2ϩ -sensing (29, 30), and cslo-␣ contains a single high affinity phosphorylation site for PKG at Ser-1072 (3). However additional putative PKG phosphorylation sites have been identified in other splice variants (31). Expression of the slo channel in Xenopus oocytes or mammalian cells gives rise to voltage-gated, Ca 2ϩ -sensitive currents with electrophysiological and pharmacological features similar to those of native BK Ca (32-34). However, although many studies of native cells suggest that BK Ca channel activity is also modulated by various protein kinases (35-38), this property has been difficult to reproduce in cloned channels. Two studies in which slo chann...

1 The aim of this study was to determine whether endothelium-dependent hyperpolarization and relaxation are altered during experimental diabetes mellitus. Membrane potentials were recorded in mesenteric arteries from rats with streptozotocin-induced diabetes and age-matched controls. The resting membrane potentials were not signi®cantly di erent between control and diabetic mesenteric arteries (755.3+0.5 vs 755.6+0.4 mV). However, endothelium-dependent hyperpolarization produced by acetylcholine (ACh; 10 78 ± 10 75 M) was signi®cantly diminished in amplitude in diabetic arteries compared with that in controls (maximum 710.4+1.1 vs 717.2+0.8 mV). Furthermore, the hyperpolarizing responses of diabetic arteries were more transient. 2 ACh-induced hyperpolarization observed in control and diabetic arteries remained unaltered even after treatment with 3610 74 M N G -nitro-L-arginine (L-NOARG), 10 75 M indomethacin or 60 u ml 71 superoxide dismutase. 3 Endothelium-dependent hyperpolarization with 10 76 M A23187, a calcium ionophore, was also decreased in diabetic arteries compared to controls (78.3+1.4 vs 718.0+1.9 mV). However, endothelium-independent hyperpolarizing responses to 10 76 M pinacidil, a potassium channel opener, were similar in control and diabetic arteries (720.0+1.4 vs 719.2+1.1 mV). 4 The altered endothelium-dependent hyperpolarizations in diabetic arteries were almost completely prevented by insulin therapy. Endothelium-dependent relaxations by ACh in the presence of 10 74 M L-NOARG and 10 75 M indomethacin in diabetic arteries were also reduced and more transient compared to controls. 5 These data indicate that endothelium-dependent hyperpolarization is reduced by diabetes, and this would, in part, account for the impaired endothelium-dependent relaxations in mesenteric arteries from diabetic rats.

Epoxyeicosatrienoic acids (EETs) are arachidonic acid metabolites of cytochrome P450 monooxygenase, which are released from endothelial cells and dilate arteries. Dilation seems to be caused by activation of large-conductance Ca2+ activated K+ channels (BK(Ca)) leading to membrane hyperpolarization. Previous studies suggest that EETs activate BK(Ca) channels via ADP-ribosylation of the G protein Galphas with a subsequent membrane-delimited action on the channel [Circ Res 78:415-423, 1996; 80:877-884, 1997; 85:349-356, 1999]. The present study examined whether this pathway is present in human embryonic kidney (HEK) 293 cells when the BK(Ca) alpha-subunit (cslo-alpha) is expressed without the beta-subunit. 11,12-EET increased outward K+ current in whole-cell recordings of HEK293 cells. In cell-attached patches, 11,12-EET also increased the activity of cslo-alpha channels without affecting unitary conductance. This action was mimicked by cholera toxin. The ADP-ribosyltransferase inhibitors 3-aminobenzamide and m-iodobenxylguanidine blocked the stimulatory effect of 11,12-EET. In inside-out patches 11,12-EET was without effect on channel activity unless GTP was included in the bathing solution. GTP and GTPgammaS alone also activated cslo-alpha channels. Dialysis of cells with anti-Galphas antibody completely blocked the activation of cslo-alpha channels by 11,12-EET, whereas anti-Galphai/o and anti-Gbetagamma antibodies were without effect. The protein kinase A inhibitor KT5720 and the adenylate cyclase inhibitor SQ22536 did not reduce the stimulatory effect of 11,12-EET on cslo-alpha channels in cell-attached patches. These data suggest that EET leads to Galphas-dependent activation of the cslo-alpha subunits expressed in HEK293 cells and that the cslo-beta subunit is not required.

1. The aim of the present study was to identify the sources of Ca2+ contributing to acetylcholine (ACh)-induced release of endothelium-derived hyperpolarizing factor (EDHF) from endothelial cells of rat mesenteric artery and to assess the pathway involved. The changes in membrane potentials of smooth muscles by ACh measured with the microelectrode technique were evaluated as a marker for EDHF release. 2. ACh elicited membrane hyperpolarization of smooth muscle cells in an endothelium-dependent manner. The hyperpolarizing response was not affected by treatment with 10 microM indomethacin, 300 microM NG-nitro-L-arginine or 10 microM oxyhaemoglobin, thereby indicating that the hyperpolarization is not mediated by prostanoids or nitric oxide but is presumably by EDHF. 3. In the presence of extracellular Ca2+, 1 microM ACh generated a hyperpolarization composed of the transient and sustained components. By contrast, in Ca(2+)-free medium, ACh produced only transient hyperpolarization. 4. Pretreatment with 100 nM thapsigargin and 3 microM cyclopiazonic acid, endoplasmic reticulum Ca(2+)-ATPase inhibitors, completely abolished ACh-induced hyperpolarization. Pretreatment with 20 mM caffeine also markedly attenuated ACh-induced hyperpolarization. However, the overall pattern and peak amplitude of hyperpolarization were unaffected by pretreatment with 1 microM ryanodine. 5. In the presence of 5 mM Ni2+ or 3 mM Mn2+, the hyperpolarizing response to ACh was transient, and the sustained component of hyperpolarization was not observed. On the other hand, 1 microM nifedipine had no effect on ACh-induced hyperpolarization. 6. ACh-induced hyperpolarization was nearly completely eliminated by 500 nM U-73122 or 200 microM 2-nitro-4-carboxyphenyl-N, N-diphenylcarbamate, inhibitors of phospholipase C, but was unchanged by 500 nM U-73343, an inactive form of U-73122. Pretreatment with 20 nM staurosporine, an inhibitor of protein kinase C, did not modify ACh-induced hyperpolarization. 7. These results indicate that the ACh-induced release of EDHF from endothelial cells of rat mesenteric artery is possibly initiated by Ca2+ release from inositol 1,4,5-trisphosphate (IP3)-sensitive Ca2+ pool as a consequence of stimulation of phospholipid hydrolysis due to phospholipase C activation, and maintained by Ca2+ influx via a Ni(2+)- and Mn(2+)-sensitive pathway distinct from L-type Ca2+ channels. The Ca(2+)-influx mechanism seems to be activated following IP3-induced depletion of the pool.

1 In rat mesenteric artery, acetylcholine (ACh) causes endothelium-dependent hyperpolarization by releasing endothelium-derived hyerpolarizing factor (EDHF). Recent evidence suggests that EDHF may be a cytochrome P450-derived arachidonic acid metabolite. The aim of the present study was to investigate whether such a metabolite is indeed contributing to ACh-induced hyperpolarization observed in rat mesenteric artery. 2 The phospholipase A 2 inhibitor quinacrine (30 mM) nearly completely eliminated ACh-induced hyperpolarization. However, the hyperpolarizing eect of pinacidil was also abolished in the presence of quinacrine. 3 The imidazole antimycotic agents ketoconazole (50 mM), clotrimazole (30 mM) and miconazole (10 mM), which bind to the heme moiety of cytochrome P450, eliminated not only ACh-induced hyperpolarizations but also those induced by pinacidil. SKF525A (30 mM), a prototype inhibitor of the enzyme, also abolished the hyperpolarizing responses to both agents. In contrast, neither 17-octadecynoic acid (10 mM), a mechanism-based inhibitor of cytochrome P450 metabolism of fatty acids, nor eicosatetraynoic acid (20 mM), an inhibitor of all arachidonic acid metabolic pathways, altered AChinduced hyperpolarization. Furthermore, the hyperpolarization was unaected by the preferential inhibitors of speci®c cytochrome P450 isozymes, a-naphto¯avone (1 mM), diedthyldithiocarbamate (50 mM), metyrapone (20 mM) and troleandomycin (10 mM). 4 Pretreatment of rats with lipopolysaccharide (2 mg kg 71 ) and exposure to nitroprusside (10 mM), both of which are expected to inhibit cytochrome P450 activity due to nitric oxide overproduction, were without eect on ACh-induced hyperpolarization. Pretreatment of rats for 3 days with pentobarbitone (80 mg kg 71 day 71), a cytochrome P450 inducer, also did not aect the hyperpolarizing response to ACh. 5 Arachidonic acid in concentrations up to 100 mM had no detectable eect on smooth muscle membrane potential. 11,12-Epoxyeicosatrienoic acid (EET, 10 mM), one of cytochrome P450-derived epoxygenase metabolites of arachidonic acid, elicited a small endothelium-independent membrane hyperpolarization. The hyperpolarizing response to EET was blocked by glibenclamide (30 mM), in contrast to the response to ACh. 6 These results suggest that the contribution of a cytochrome P450-derived metabolite of arachidonic acid to ACh-induced hyperpolarization via EDHF release is minimal or absent in rat mesenteric artery.