Inhibition of norepinephrine and caffeine-induced activation by ryanodine and thapsigargin in rat mesenteric arteries

Mechanism of Contraction of Rat Isolated Tail Arteries by Hyposmotic Solutions

Contraction induced by hyposmotic swelling was examined in rat tail arteries mounted on a myograph containing a modified Krebs physiological saline solution (PSS) containing 50 mM mannitol (300 mosm/l). Hyposmotic swelling was induced by removing mannitol. In arteries having basal tone or arteries precontracted with K(+) or the thromboxane mimetic U-46619, removal of mannitol caused a concentration dependent contraction of rat tail arteries. Concurrent measurement of tension and intracellular calcium [Ca(2+)](i )in arteries loaded with fura-2 showed that both tension and [Ca(2+)](i) increased on exposure to a hyposmotic solution. Removal of endothelium or inhibition of nitric oxide and cyclooxygenase together did not affect contractile responses. Removal of extracellular Ca(2+) abolished the contractile response to hyposmotic solution and NiCl(2), a nonspecific inhibitor of Ca(2+) influx pathways, blocked the rise in [Ca(2+)](i) and tension in response to a hyposmotic solution. Verapamil and nisoldipine, inhibitors of Ca(v)1.2 (L-type) calcium channels significantly reduced the contractile response to a hyposmotic solution. Addition of NiCl(2) to nisoldipine caused an additional inhibition of the response to a hyposmotic solution. Inhibition of calcium release from the sarcoplasmic reticulum by ryanodine or cyclopiazonic acid (CPA) did not cause any change in the tension response to a hyposmotic solution. CPA did not significantly inhibit the response to a hyposmotic solution in the presence of N(G)-methyl-L-arginine, oxyhaemoglobin and indomethacin. We conclude that contraction induced by a hyposmotic solution is largely due to Ca(v)1.2 calcium channels although other Ca(2+) influx pathways also contribute.

Contractile responses of isolated rat mesenteric arteries to acute episodes of severe hypoxia and subsequent reoxygenation

This study further investigates the mechanisms responsible for the effects of acute and severe hypoxia, and subsequent reoxygenation, on the contractility of isolated rat mesenteric arteries. In noradrenaline (NA)-contracted arteries, hypoxia caused a relaxation to near baseline levels. Reoxygenation resulted in an immediate transient contraction before tension returned more slowly to prehypoxia levels. Similar responses to hypoxia were observed in tissues precontracted by addition of KCl (60 mM) or U46619 (10 microM); however, the transient contraction upon reoxygenation was absent (KCl) or reduced (U46619). Responses to hypoxia were independent of changes in intracellular calcium ([Ca2+]i), while those to reoxygenation were accompanied by corresponding changes in [Ca2+]i and were completely abolished by ryanodine. In NA-contracted tissues, all responses were unaffected by endothelial removal or by inhibitors of nitric oxide synthase and cyclooxygenase. The K+ channel blockers triethylamine (TEA), glibenclamide, and 4-aminopyridine (4-AP) had no effect on the responses to hypoxia. The transient contractile response to reoxygenation was, however, significantly reduced in the presence of 4-AP. The response to reoxygenation, but not that to hypoxia, was inhibited by the antioxidant dithiothreitol (DTT) and the NAD(P)H-oxidase inhibitor diphenyliodonium (DPI). These data suggest that hypoxic vasodilation occurs independently of reductions in [Ca2+]i. Alternatively, transient contractions on reoxygenation are dependent upon the generation of reactive oxygen species and the release of stored Ca2+.

Effects of vasoactive agents on intracellular calcium and force in myometrial and subcutaneous resistance arteries isolated from preeclamptic, pregnant, and nonpregnant woman

OBJECTIVE

Preeclampsia is a common and serious complication of pregnancy, characterized by maternal hypertension and proteinuria, placental insufficiency, and fetal growth restriction. The purpose of this study was to investigate whether intracellular Ca 2+ ([Ca 2+ ] i ) and contractile responses of vascular smooth muscle to vasoactive agents are altered in preeclampsia compared with normal pregnancy and the nonpregnant state.

STUDY DESIGN

Subcutaneous and myometrial resistance arteries from women who had preeclampsia, normal pregnancy, and nonpregnant women were obtained at the time of cesarean section or hysterectomy. Arteries were mounted on an isometric myograph and loaded with the Ca 2+ indicator, fura-2AM, to permit simultaneous measurement of force and [Ca 2+ ] i . Reponses to endothelium-dependent relaxants (acetylcholine and substance P) and vasoconstrictors (depolarizing potassium solution, phenylephrine, and angiotensin II) were examined.

RESULTS

The fall in [Ca 2+ ] i and relaxation in response to acetylcholine was significantly inhibited in both myometrial and subcutaneous arteries from preeclamptic women compared with arteries from nonpregnant or normal pregnant women. However, responses to substance P did not differ between the 3 groups. There were no significant differences in [Ca 2+ ] i or force responses to high potassium, phenylephrine, or angiotensin II in myometrial and subcutaneous resistance vessels in women with preeclampsia compared with normal pregnant women. However, force, but not [Ca 2+ ] i responses to angiotensin II, in subcutaneous vessels from normal pregnant and preeclamptic women were reduced compared with subcutaneous arteries from nonpregnant women, indicating that pregnancy is associated with a reduction in Ca 2+ sensitization in this tissue. A similar effect was not seen in myometrial arteries.

CONCLUSION

Endothelial function is altered in preeclampsia, with loss of effect of acetylcholine, but not substance P. Vasoconstrictor reactivity is not increased in preeclampsia compared with uncomplicated normal pregnancy, and this is unlikely to be an explanation for the increased peripheral vascular resistance seen in preeclampsia.

Alpha1-adrenergic signaling mechanisms in contraction of resistance arteries

Our goal in this review is to provide a comprehensive, integrated view of the numerous signaling pathways that are activated by alpha(1)-adrenoceptors and control actin-myosin interactions (i.e., crossbridge cycling and force generation) in mammalian arterial smooth muscle. These signaling pathways may be categorized broadly as leading either to thick (myosin) filament regulation or to thin (actin) filament regulation. Thick filament regulation encompasses both "Ca(2+) activation" and "Ca(2+)-sensitization" as it involves both activation of myosin light chain kinase (MLCK) by Ca(2+)-calmodulin and regulation of myosin light chain phosphatase (MLCP) activity. With respect to Ca(2+) activation, adrenergically induced Ca(2+) transients in individual smooth muscle cells of intact arteries are now being shown by high resolution imaging to be sarcoplasmic reticulum-dependent asynchronous propagating Ca(2+) waves. These waves differ from the spatially uniform increases in [Ca(2+)] previously assumed. Similarly, imaging during adrenergic activation has revealed the dynamic translocation, to membranes and other subcellular sites, of protein kinases (e.g., Ca(2+)-activated protein kinases, PKCs) that are involved in regulation of MLCP and thus in "Ca(2+) sensitization" of contraction. Thin filament regulation includes the possible disinhibition of actin-myosin interactions by phosphorylation of CaD, possibly by mitogen-activated protein (MAP) kinases that are also translocated during adrenergic activation. An hypothesis for the mechanisms of adrenergic activation of small arteries is advanced. This involves asynchronous Ca(2+) waves in individual SMC, synchronous Ca(2+) oscillations (at high levels of adrenergic activation), Ca(2+) sparks, "Ca(2+)-sensitization" by PKC and Rho-associated kinase (ROK), and thin filament mechanisms.

Alterations by age of calcium handling in rat resistance arteries

This study investigated the alterations by age of Ca handling in rat small mesenteric arteries. The contractile responses to phenylephrine and caffeine in small mesenteric arteries from young and old rats were studied in Ca -containing and Ca -free medium. In Ca -containing medium the contraction to phenylephrine (100 micro M) but not to caffeine (10 mM) was greater in old than in young rats. Concentration-response curve to phenylephrine was affected to the same extent by nifedipine (1 micro M) in both age groups, whereas ryanodine (20 micro M) decreased the maximal response to phenylephrine only in young rats. These results suggest the participation of intracellular Ca handling on the observed differences by aging. In Ca -free medium, phenylephrine (10 micro M) but not caffeine (10 mM) induced a greater contraction in old than in young animals, corroborating the results obtained in Ca -containing solution. The greater response to phenylephrine observed in old rats cannot be explained by an increase in the inositol 1,4,5-trisphosphate (IP ) formation because the accumulation of inositol phosphates by phenylephrine was not affected by aging. Results obtained in Ca -free medium using caffeine after phenylephrine or vice versa suggest a common intracellular Ca pool. Pretreatment with ryanodine in Ca -free medium almost abolished contractile response to phenylephrine and caffeine in young rats but only partially decreased them in old animals, suggesting an impairment in the Ca -induced Ca release (CICR) mechanism leading to an increase in the stored Ca content. The greater amount of stored Ca could explain the higher contractile response to phenylephrine observed in aged rats. As a consequence of all these changes due to aging, an imbalance between the two Ca release mechanisms from sarcoplasmic reticulum was observed with a major role of Ca induced release by IP at the expense of an impairment of CICR mechanism. This observation will also help explain the results obtained in the presence of extracellular Ca, where phenylephrine induced a greater maximum response in old animals in spite of a decrease in the midrange sensitivity to this agonist.

Mechanism of action of angiotensin II in human isolated subcutaneous resistance arteries

1. Human isolated subcutaneous arteries were mounted in a myograph and isometric tension measured. In some experiments, intracellular calcium [Ca(2+)]i was also measured using fura-2. 2. Angiotensin II (100 pM - 1 microM) increased [Ca(2+)]i and tone in a concentration-dependent manner. The effects of angiotensin II (100 nM) were inhibited by an AT1-receptor antagonist, candesartan (100 pM). 3. Ryanodine (10 microM), had no effect on angiotensin II-induced responses, but removal of extracellular Ca(2+) abolished angiotensin II-induced rise in [Ca(2+)]i and tone. Inhibition of Ca(2+) entry by Ni(2+) (2 mM), also inhibited angiotensin II responses. The dihydropyridine, L-type calcium channel antagonist, amlodipine (10 microM), only partially attenuated angiotensin II responses. 4. Inhibition of protein kinase C (PKC) by chelerythrine (1 microM), or by overnight exposure to a phorbol ester (PDBu; 500 nM) had no effect on angiotensin II-induced contraction. 5. Genistein (10 microM), a tyrosine kinase inhibitor, inhibited angiotensin II-induced contraction, but did not inhibit the rise in [Ca(2+)]i, suggesting that at this concentration it affected the calcium sensitivity of the contractile apparatus. Genistein did not affect responses to norepinephrine (NE) or high potassium (KPSS). 6. A selective MEK inhibitor, PD98059 (30 microM), inhibited both the angiotensin II-induced contraction and rise in [Ca(2+)]i, but had no effect on responses to NE or KPSS. 7. AT1 activation causes Ca(2+) influx via L-type calcium channels and a dihydropyridine-insensitive route, but does not release Ca(2+) from intracellular sites. Activation of tyrosine kinase(s) and the ERK 1/2 pathway, but not classical or novel PKC, also play a role in angiotensin II-induced contraction in human subcutaneous resistance arteries.

Ca(2+) uptake function of sarcoplasmic reticulum during contraction of rat arterial smooth muscles

To determine the Ca(2+) uptake function of the sarcoplasmic reticulum during contraction, the effects of cyclopiazonic acid or thapsigargin, agents that inhibit sarcoplasmic reticulum Ca(2+)-ATPase, on the contractile responses to K(+) or norepinephrine were compared in endothelium-denuded strips of femoral, mesenteric and carotid arteries of the rat. The addition of K(+) (3-20 mM) to the strips caused a concentration-dependent contraction, and the sensitivity to K(+) was much higher in the carotid artery than in the other arteries. The preincubation of strips with cyclopiazonic acid (10 microM) or thapsigargin (100 nM) caused a leftward shift of the concentration-response curve for K(+), and this effect was smaller in the carotid artery than in the other arteries. Inhibition of sarcoplasmic reticulum Ca(2+) uptake caused the sensitivity to K(+) to be similar in the three arteries. Similar results were obtained when the contractile responses to norepinephrine were determined. Cyclopiazonic acid itself induced similar transient contractions in the three arteries. The addition of caffeine (20 mM) caused a transient contraction that was smaller in the carotid artery than in the other arteries. We conclude that (1) the Ca(2+) influx during stimulation with K(+) or norepinephrine is buffered by the sarcoplasmic reticulum in femoral and mesenteric arteries, (2) this function is weak in the carotid artery, probably because the sarcoplasmic reticulum of this artery is almost filled with Ca(2+) in the resting state, and (3) the Ca(2+) uptake function of the sarcoplasmic reticulum during contraction is reflected by the contractile sensitivity in these arteries.

Ca(2+) movement from leaky sarcoplasmic reticulum during contraction of rat arterial smooth muscles

To examine the Ca(2+) buffering function of the sarcoplasmic reticulum during arterial contraction, we studied Ca(2+) movement during stimulation with K(+) or norepinephrine in arteries with a leaky sarcoplasmic reticulum. Responses were compared in endothelium-denuded strips of femoral, mesenteric and carotid arteries of the rat. To make the sarcoplasmic reticulum leaky to Ca(2+), Ca(2+)-induced Ca(2+) release channels of the sarcoplasmic reticulum were locked open by treatment with ryanodine plus caffeine. After ryanodine treatment, the contractile responses to K(+) (3-20 mM) were augmented when compared with control responses in femoral and mesenteric arteries, but were inhibited in the carotid artery. Similar results were obtained when the contractile responses to norepinephrine were determined. The inhibition by ryanodine of the K(+)- or norepinephrine-contractions seen in the carotid artery was reversed by pretreatment with cyclopiazonic acid (10 microM), an inhibitor of the sarcoplasmic reticulum Ca(2+)-ATPase, but was not by charybdotoxin (100 nM), a blocker of Ca(2+)-activated K(+) channels. We conclude that (1) after ryanodine treatment, Ca(2+) entering from the extracellular space during stimulation with K(+) or norepinephrine is first taken up into the leaky sarcoplasmic reticulum and then reaches the myofilaments in femoral and mesenteric arteries, while in the carotid artery, Ca(2+) leaked from the sarcoplasmic reticulum reaches mainly the plasma membrane from where it is extruded into the extracellular space, and (2) the different movement of Ca(2+) may be due to the relative location of the sarcoplasmic reticulum in the smooth muscle cell of each artery.

Pharmacological analysis of noncholinergic action of 2-sec-butylphenyl N-methylcarbamate insecticide on the isolated rabbit aorta

We investigated noncholinergic actions of 2-sec-butylphenyl N-methylcarbamate (BPMC) on the isolated rabbit thoracic aorta to help determine the mechanisms responsible for its unique toxicological properties, which are characterized by cardiovascular collapse and low lethality compared to its anticholinesterase (anti-ChE) activity. BPMC inhibited K(+)-induced contraction more effectively than it did norepinephrine (NE)-induced contraction. The inhibitory effect on K(+)-induced contraction was not altered by changing the external K(+) concentration, but it was decreased by adding an L-type Ca(2+) channel agonist, BAY K 8644. Simultaneous measurement of tension and cytosolic Ca(2+) levels elevated by K(+) stimulation revealed that BPMC decreased the Ca(2+) levels prior to and parallel to the tension. The magnitude of the inhibitory effect on Ca(2+) levels was increased by treating BPMC before Ca(2+) application in the depolarized preparation. However, BPMC did not inhibit caffeine- or NE-induced transient contraction in Ca(2+)-free medium. On the other hand, BPMC produced tonic contraction in the resting aorta. The contraction to BPMC did not develop after removing the adventitia. The contraction was inhibited by phentolamine or guanethidine but not by atropine or tetrodotoxin. These results suggest that BPMC inhibits depolarization- or agonist-induced contraction by inhibiting Ca(2+) entry through L-type Ca(2+) channels, whereas it produces vascular contraction in the resting state by releasing NE from sympathetic nerve terminals. These apparently opposing mechanisms may contribute to the unique noncholinergic toxicological properties of BPMC.

Interactions between membrane potential and intracellular calcium concentration in vascular smooth muscle

The intracellular calcium concentration is a major determinant of vascular tone. In the steady state it is regulated mainly by membrane potential. At the same time, several mechanisms regulating the calcium concentration, including the membrane potential, are influenced by the intracellular calcium concentration itself. There are thus multiple possible positive and negative feedback loops involved in calcium regulation. This review gives a brief overview of the different mechanisms involved, including calcium-dependent ion channels, exchangers, and ATPases, and discusses their role in agonist-mediated responses, in relation primarily to studies on the portal vein and mesenteric small arteries.

The actions of some cannabinoid receptor ligands in the rat isolated mesenteric artery

1. The actions of a number of cannabinoid receptor ligands were investigated using the myograph-mounted rat isolated mesenteric artery. Anandamide, CP 55,940, HU-210, palmitoylethanolamide and WIN 55,212-2 all caused concentration-dependent relaxations of methoxamine-precontracted vessels which were not affected by removal of the endothelium. 2. Precontracting vessels with 60 mM KCl instead of methoxamine greatly reduced the vasorelaxant effects of anandamide and palmitoylethanolamide. High K+ solution caused a modest decrease in the relaxant potency of CP 55,940 and HU-210, and had no effect on relaxations induced by WIN 55,212-2. 3. Relaxations of methoxamine-induced tone by anandamide, CP 55,940 and HU-210, but not palmitoylethanolamide and WIN 55,212-2, were attenuated by the cannabinoid receptor antagonist, SR 141716A. Relaxation of vessels contracted with 60 mM KCl by CP 55,940 was also sensitive to SR 141716A. 4. Anandamide and CP 55,940 caused small but concentration-dependent contractions in resting vessels in the absence of extracellular calcium. These were not sensitive to SR 141716A. Palmitoylethanolamide and WIN 55,212-2 produced smaller contractions only at higher concentrations. 5. Anandamide and CP 55,940, but not palmitoylethanolamide and WIN 55,212-2, caused concentration-dependent inhibition of the phasic contractions induced by methoxamine in calcium-free conditions, but only anandamide caused inhibition of contractions to caffeine under such conditions. These inhibitory effects were not antagonised by SR 141716A. 6. The present study provides the first detailed investigation of the actions of cannabinoid agonists on vascular smooth muscle. Our results show that these compounds exert both receptor-dependent and -independent effects on agonist-induced calcium mobilization in the rat isolated mesenteric artery.

Action of ryanodine on neurogenic responses in rat isolated mesenteric small arteries

1. Rat mesenteric (approximately 250 microns) were set up in a single-channel isometric myograph designed to allow with 6 microM fura-2AM for 2 h and simultaneous recordings of neurogenic contraction (force) and intracellular calcium [Ca2+]i were obtained. In other experiments, arteries were loaded with 1 microCi ml-1 [3H]-noradrenaline (NA) for 30 min in order to measure release of [3H]-NA in response to field stimulation to examine whether ryanodine directly inhibited neuronal release of NA. 2. Arteries were activated by single intermittent field stimulation or continuously to excite intrinsic sympathetic nerves, or by cumulative addition of noradrenaline (1 nM-10 microM) to the bathing solution. 3. Pre-incubation with ryanodine markedly inhibited the contraction and [Ca2+]i release in response to single-pulse nerve stimulation. Ryanodine also inhibited an early phasic component of the response to continuous field stimulation and reduced the rate of rise in force in response to continuous field stimulation. However, stable maximal contraction and [Ca2+]i in response to continuous field stimulation as well as maximal responses to exogenous NA were unaffected. Release of [3H]-NA in response to single intermittent field stimulation was not affected by ryanodine when compared to vehicle. 4. Our results suggest that brief intermittent activation of intramural sympathetic nerves increases [Ca2+]i and contracts small arteries primarily by releasing Ca2+ from a ryanodine-sensitive intracellular store. In contrast, the stable rise in tone and [Ca2+]i resulting from continuous nerve stimulation may largely depend on sources of Ca2+ other than the ryanodine-sensitive intracellular store.

Nonexocytotic noradrenaline release from rat cardiac synaptosomal-mitochondrial fractions

Nonexocytotic noradrenaline (NA) release was examined in rat cardiac synaptosomal-mitochondrial fractions prelabeled with [3H]NA (300 nM; 1 h at 37 degrees C). Ischemic conditions (1 mM iodoacetate + 2 mM NaCN; 15 min at 37 degrees C) evoked a Ca(2+)-independent release of [3H]NA from isolated synaptosomes, which represented 33.4% of total content, whereas the release evoked by 56 mM K+ was Ca2+ dependent and represented 5.8% of total content. Tyramine, phencyclidine (PCP), and rimcazole also caused important Ca(2+)-independent releases of [3H]NA (from 12 to 45% of total content) with median effective concentrations (EC50s) of 6.8, 182, and 41.8 microM, respectively. The release responses evoked by ischemic conditions, tyramine, PCP, and rimcazole were mimicked by the delta-receptor ligand, 1,3-ditolyl guanidine (DTG), and blocked by the uptake 1 inhibitor, desipramine (100 microM). The delta 1-receptors ligands, (+)-3-hydroxyphenyl-N-(1-propyl)piperidine ((+)-3-PPP) and (+)N-allylnormetazocine [(+)SKF-10047], were potent blockers of the release of [3H]NA evoked by ischemic conditions but not by PCP or rimcazole. These data indicate that ischemic conditions and PCP/delta 2-receptor ligands induce carrier-mediated NA efflux from cardiac sympathetic nerve terminals, whereas delta 1-receptor ligands produce marked inhibition of the ischemic response.

Reversal of acute theophylline toxicity by calcium channel blockers in dogs and rats

Theophylline, widely used in the treatment of pulmonary diseases, has a narrow therapeutic index; the recommended plasma levels being 10-20 micrograms/ml in humans. The misuse or abuse of theophylline can cause life-threatening central nervous system and cardiovascular effects. Increased intracellular Ca2+ levels are thought to play an important role in theophylline toxicity and death. The objective of this study was to determine whether Ca2+ channel blockers, e.g. verapamil, nifedipine, or diltiazem, prevent sudden death caused by theophylline treatment in rats and dogs. Groups of Sprague-Dawley rats were treated with theophylline alone (150 mg/kg i.p.) or with theophylline pretreatment followed by administration of verapamil (0.25 to 0.5 mg/kg i.p.), nifedipine (0.25 to 1.0 mg/kg i.p.), or diltiazem (0.5 to 1.0 mg/kg i.p.), 2.5 to 15 min later. The rats were observed for toxic signs and survival over a period of 15 days. All three calcium channel blockers significantly reduced the theophylline-induced sudden death in rats. In a separate study, neither verapamil (0.5 mg/kg i.p.) nor nifedipine (1.0 mg/kg i.p.) prevented the theophylline-induced myocardial necrosis in the rat. In beagle dogs, verapamil (0.5 mg/kg i.v.) prevented theophylline (15 mg/kg/min i.v. for 10 min)-induced hypotension, arrhythmias, and sudden death. Our results support previously reported findings that calcium plays a major role in theophylline-induced toxicity and death.

The mechanism of action of alpha 2-adrenoceptors in human isolated subcutaneous resistance arteries

1. The effect of noradrenaline and the selective alpha 2-adrenoceptor agonist, azepexole, on tone and intracellular Ca2+ ([Ca2+]i) was examined in human isolated subcutaneous resistance arteries. Isolated arteries were mounted on an isometric myograph and loaded with the Ca2+ indicator, fura-2, for simultaneous measurement of force and [Ca2+]i. 2. High potassium solution (KPSS), noradrenaline and azepexole increased [Ca2+]i and contracted subcutaneous arteries in physiological saline. When extracellular Ca2+ was removed and the calcium chelator, BAPTA, added to the physiological saline (PSSo), responses to noradrenaline were transient and reduced, and responses to azepexole were markedly inhibited. 3. Ryanodine, an agent which interferes with Ca2+ release from intracellular stores, had little effect on contractile responses to KPSS, noradrenaline or azepexole in physiological saline. The response to caffeine in physiological saline was inhibited by ryanodine. In PSSo, ryanodine partially inhibited contractile responses to noradrenaline and azepexole, and completely abolished the response to caffeine. 4. Noradrenaline and azepexole both significantly increased maximum force achieved by cumulative addition of Ca2+ to a Ca(2+)-free depolarizing solution and shifted the calculated relationship between [Ca2+]i and force to the left, suggesting these agents increase the sensitivity of the contractile apparatus to [Ca2+]i. 5. (-)-202 791, a dihydropyridine antagonist of voltage-operated calcium channels partially inhibited both the contractile response and the rise in [Ca2+]i induced by azepexole. Pre-treatment of arteries with pertussis toxin inhibited responses to azepexole, but had no significant effect on tone induced by KPSS or noradrenaline. ETYA, an inhibitor of phospholipase A2, lipoxygenase and cyclo-oxygenase, had no effect on azepexole-induced contraction in the presence of N omega nitro-L-arginine methyl ester.6. Azepexole, a selective alpha2-adrenoceptor agonist, contracts human subcutaneous resistance arteries by a mechanism largely dependent on the influx of extracellular Ca2", probably through voltage-operated calcium channels. This action involves a pertussis toxin-sensitive G protein, possibly Gi.