Cerebral ATP-Sensitive Potassium Channels During Acute Reduction of Carotid Blood Flow
Abstract The ATP-sensitive potassium channels (KATP) are activated either by a decrease in intracellular ATP content or by a lowering of the ATP-ADP ratio such as during stroke. We studied the role of cerebral KATP on arterial pressure during acute reduction of cerebral blood flow in 12-week-old male Wistar rats anesthetized with urethane by recording arterial pressure and heart rate continuously. After bilateral ligation of the common carotid arteries, glibenclamide, a specific blocker of KATP, was injected intracerebroventricularly into the cerebral lateral ventricle. Glibenclamide elicited a sustained vasopressor response in a dose-dependent manner in rats with bilateral carotid artery ligation (10 nmol, +15±2 mm Hg; 1 nmol, +5±1 mm Hg, P<.01 versus vehicle), but hemodynamic alterations were barely recorded with glibenclamide in sham-operated control rats. The abdominal sympathetic discharge was not increased significantly enough to explain the pressor mechanism. Similarly, pretreatments with intravenous injections of bunazosin, an α1-adrenoceptor antagonist, did not affect the pressor response of intracerebroventricular glibenclamide. To investigate the vasopressor mechanism further, we measured plasma and pituitary concentrations of arginine vasopressin and determined the effects of vasopressin receptor antagonists. The intracerebroventricular injections of glibenclamide significantly increased the plasma concentration of vasopressin (P<.05) and significantly decreased the pituitary concentration of vasopressin (P<.05) in rats with bilateral carotid artery ligation. Intravenous pretreatment with the vasopressin V1 receptor antagonist OPC-21268 abolished the vasopressor response to intracerebroventricular glibenclamide (+16±2 versus +1±1 mm Hg, P<.01). These findings indicate that KATP in the brain may inhibit an excess rise in arterial pressure in part by decreasing the release of vasopressin from the pituitary during bilateral carotid artery ligation.
The ATP-sensitive potassium channels (KATP), 45-kD proteins with ATP-binding domains,1 are thought to play a key role in the glucose-induced release of insulin from pancreatic β-cells.2 Such activity is inhibited more than 95% by 1 mmol/L ATP3 but is activated by a decrease in the intracellular concentration of ATP or by a lowering of the ratio of ATP to ADP,4 5 which may occur during brain ischemia. KATP are widely distributed in the heart, peripheral vasculature, and the brain as well as in the pancreas3 and are considered to play a specific physiological role in each organ. The activation of KATP shortens the action potential, inhibits the increase in intracellular concentration of Ca2+, and reduces the oxygen consumption in heart muscle cells.6 The activation of KATP also induces vasodilatation by relaxing vascular smooth muscle cells.7 Autoradiographic studies of KATP using the high-affinity ligand revealed that KATP are abundant in the globus pallidus, hippocampus, and anterior pituitary but also broadly in other cerebral regions in the telencephalon, diencephalon, and cerebellum of rats.8 Northern blot analysis in rats showed that signal is detected in the thalamus, hypothalamus, and pituitary of the brain, but no signal is detected in the other brain regions,1 meaning that cerebral KATP may be related to cerebral regulation of cardiovascular and endocrine functions.
Patients with stroke, such as a cerebral thrombosis or hemorrhage, show a fluctuation in arterial pressure, with an elevation seen in the acute phase.9 Bilateral carotid artery ligation of normotensive Wistar rats raises arterial pressure,10 11 although it does not produce pronounced cerebral ischemia. Not only the baroreceptor and chemoreceptor reflexes via the carotid sinus but also hyperactivation of the catecholaminergic system may be the main causes of this rise in arterial pressure during cerebrovascular accidents in humans as well as in rats after ligation of the carotid artery.12 13 Since bilateral ligation of the carotid arteries reduces regional cerebral blood flow in the diencephalon of normotensive Wistar rats,10 11 KATP in the brain may be activated during the ligation. The opening of cerebral KATP may affect arterial pressure during acute cerebral ischemia.
Centrally induced cardiovascular responses are usually mediated by both sympathetic nervous system activity and release of pituitary hormones such as vasopressin. Arginine vasopressin (AVP), which is supposed to be released from pituitary glands during cerebral ischemia, is reported to block KATP from outside the cell membrane.14 Therefore, AVP may be involved in the regulatory mechanism of the brain KATP and blood pressure.
The aim of the present study was to elucidate the role of cerebral KATP in the possible regulation of arterial pressure and the underlying mechanism during acute reduction of cerebral blood flow by measuring not only blood pressure and heart rate but also peripheral sympathetic discharge and plasma AVP levels in rats with bilateral common carotid arterial ligation.
Twelve-week-old male Wistar rats (n=109, 257±3 g) were purchased from Oriental Bio-Service Laboratory (Kyoto, Japan). The experimental procedure was authorized by the Committee for Animal Research of Kyoto Prefectural University of Medicine. Rats were anesthetized with urethane (120 mg/100 g IP, Nakarai Tesque). Catheters (PE-50, Clay Adams, Becton-Dickinson) were inserted into the femoral artery for recording of arterial pressure and into the femoral vein for injection of drugs. Arterial pressure was recorded continuously by connecting the catheter to a small-volume displacement pressure transducer (TP-200T, Nihon Kohden). Heart rate (pulse rate, in beats per minute [bpm]) was automatically calculated by triggering the femoral arterial pulse pressure with a tachometer (AT-601G, Nihon Kohden). The trachea was intubated with a cannula (PE-240, Clay Adams, Becton-Dickinson) for artificial ventilation after skeletal muscle was paralyzed by injection of decamethonium bromide (0.2 mg/100 g IV, Sigma Chemical Co).
Bilateral Ligation of Common Carotid Arteries
With rats under urethane anesthesia, bilateral carotid arteries were dissected free from the vagosympathetic trunks in the neck and ligated with double sutures (3-0 silk, Nippon Shoji Co). A sham operation revealed the carotid arteries without ligation. To prevent respiratory arrest, all experimental rats, including sham-operated rats, received artificial ventilation with room air during the experiment. When both arterial pressure and heart rate had become stable approximately 60 minutes after bilateral ligation of the carotid arteries, we started the experiments described below.
A guide cannula (23-gauge stainless steel tube, 30 mm long with a 30-gauge stylet) was inserted stereotaxically into the right lateral cerebral ventricle. The stereotaxic coordinates were 1.6 mm right laterally and 0.8 mm dorsoventrally from the bregma, with the incisor bar set 3.3 mm below the interaural line. Agents were injected into the cerebral ventricle (ICV) by insertion of an injection cannula (30-gauge stainless steel tube) connected to a 25-μL syringe into the guide cannula. In each injection, 10 μL was delivered manually into the ventricle over 10 seconds. At the end of each experiment, methylene blue solution was injected through the injection cannula to verify its correct placement in the right lateral ventricle.
Recording of Abdominal Sympathetic Nerve Activity
After an extensive transverse dissection of the left lateral abdominal wall was made, the sympathetic nerve bundle emerging from the celiac ganglion and accompanied by a superior mesenteric artery was placed over a bipolar electrode. These methods have been described in detail elsewhere.15
Radioimmunoassay for AVP
The plasma concentration of AVP and the amount of AVP in the pituitary glands were measured with a radioimmunoassay kit (Mitsubishi Medical Science Co). Blood was collected into plastic tubes containing Na2EDTA (final concentration, 1 mg/mL) at 5 minutes after ICV and intravenous injections of glibenclamide or vehicle, when the peak response was recorded. Rats were killed by decapitation, and the pituitary glands were immediately removed and stored at −20°C. After determination of the wet weight of the pituitary gland, the tissue was homogenized (Ultra Disperser, model LK-22, Yamato) with 2 mL of 0.1 mol/L acetate buffer. The homogenate was then boiled for 10 minutes and centrifuged at 40 000g for 30 minutes. The supernatant was evaporated with an air stream. The dried residue was redissolved in the assay buffer for assay. Plasma samples were applied to Sep-Pak C18 columns (Pharmacia Fine Chemicals) and eluted with 1.5 mL methanol. The extract was evaporated with an air stream. The intra-assay coefficient of variability was 9.9% (n=10); the interassay coefficient of variability was 15.3% (n=8).
Measurement of Plasma Glucose Concentration
To exclude the possibility that glibenclamide administration may induce hypoglycemia, we measured the plasma glucose level of test animals 30 minutes after injections of glibenclamide or vehicle using an automatic analyzer (Ektachem 700 analyzer, Eastman Kodak).
Glibenclamide (Sigma) and bunazosin hydrochloride (Eisai Pharmaceutical Co) were dissolved in saline (0.9% NaCl), and pH was adjusted to 7.5 with sodium hydroxide. Physiological saline (0.9% NaCl, pH 7.5) was used as the control vehicle for glibenclamide and bunazosin. OPC-21268 (Otsuka Pharmaceutical Co), a nonpeptide blocker of AVP V1 receptors,16 and OPC-31260 (Otsuka), an AVP V2 receptor blocker,17 were dissolved in 0.1 mol/L dimethyl sulfoxide (Sigma), which was used as the control vehicle for OPC-21268 and OPC-31260 solutions.
We recorded arterial pressure, heart rate, and abdominal sympathetic nerve activity for no less than 30 minutes after ICV injections of glibenclamide (1 and 10 nmol) or vehicle into the rats with or without bilateral ligation of carotid arteries and intravenous injections of glibenclamide (10 nmol) or vehicle into the rats with bilateral carotid artery ligation. For measurements of plasma and pituitary AVP, we killed rats 5 minutes after glibenclamide administration, when the peak vasopressor response was recorded with the ICV injection of glibenclamide, and collected 1 mL of blood and pituitary glands. Intravenous pretreatments with OPC-21268 (5 mg/kg) and OPC-31260 (10 mg/kg) were performed 5 minutes before the ICV injections of glibenclamide, and intravenous pretreatments with bunazosin hydrochloride (50 μg) were performed 10 minutes before ICV injections of glibenclamide, when arterial pressure was stable.
Data are expressed as mean±SEM. Differences between experimental and control groups were evaluated by one-way ANOVA followed by Duncan’s multiple range test. A level of P<.05 was accepted as statistically significant.
Effects of Bilateral Ligation of the Carotid Arteries
Arterial pressure was markedly elevated immediately after bilateral ligation of the carotid arteries. Approximately 60 minutes later, mean arterial pressure was still higher in rats with bilateral ligation of the carotid arteries (n=17) than in sham-operated rats (n=10) (134±4 versus 92±10 mm Hg, P<.01). Heart rate did not differ in the two groups (366±24 versus 358±18 bpm).
Cardiovascular Responses to ICV Injections of Glibenclamide
ICV injections of glibenclamide produced a significant and dose-dependent increase in mean arterial pressure in rats with bilateral ligation of the carotid arteries compared with ICV injections of vehicle. The peak pressor response was observed 3 to 5 minutes after injection of either 1 or 10 nmol glibenclamide. ICV injections of 10 nmol glibenclamide increased arterial pressure for more than 30 minutes. Heart rate was decreased at 25 and 30 minutes after ICV injections of 10 nmol glibenclamide compared with vehicle; ICV injections of 1 nmol glibenclamide produced no appreciable change in heart rate (Figs 1⇓ and 2⇓). ICV injections of glibenclamide had no influence on abdominal sympathetic nerve activity in rats with bilateral ligation of the carotid arteries (changes in abdominal sympathetic nerve activity from baseline, 3 minutes after injection: vehicle, +0.05±0.03 [n=5]; 1 nmol glibenclamide, +0.05±0.12 [n=5]; 10 nmol glibenclamide, −0.03±0.18 [n=5]; 30 minutes after injection: vehicle, +0.03±0.05 [n=5]; 1 nmol glibenclamide, +0.06±0.13 [n=5]; 10 nmol glibenclamide, +0.08±0.06 [n=5]).
ICV injections of 10 nmol glibenclamide in sham-operated rats did not elicit a significant response in mean arterial pressure or heart rate (Figs 1⇑ and 3⇓). Intravenous injections of 10 nmol glibenclamide had no significant effect on mean arterial pressure or heart rate of rats with bilateral ligation of the carotid arteries compared with injection of vehicle (change in mean arterial pressure: vehicle, −2±4 mm Hg [n=5]; 10 nmol glibenclamide, +1±3 mm Hg [n=5]; change in heart rate: vehicle, −4±6 bpm [n=5]; 10 nmol glibenclamide, −2±6 bpm [n=5]).
ICV and intravenous injections of glibenclamide had no effect on plasma glucose levels in rats with bilateral ligation of the carotid arteries (ICV injection: vehicle, 8.7±0.9 mmol/L [n=7]; 1 nmol glibenclamide, 10.4±2.3 mmol/L [n=6]; 10 nmol glibenclamide, 10.5±2.1 mmol/L [n=7]; intravenous injection: vehicle, 9.4±1.9 mmol/L [n=5]; 10 nmol glibenclamide, 9.7±1.4 mmol/L [n=5]).
Plasma Concentrations and Pituitary Contents of AVP
The Table⇓ shows plasma and pituitary concentrations of AVP after injection of glibenclamide. ICV injections of 10 nmol glibenclamide significantly increased the plasma AVP concentration in rats with bilateral ligation of the carotid arteries and produced a significant decrease in the pituitary AVP content compared with injection of vehicle. Plasma and pituitary concentrations of AVP were not significantly affected in sham-operated rats administered 10 nmol glibenclamide ICV. Intravenous injections of 10 nmol glibenclamide did not influence AVP levels in rats with bilateral ligation of the carotid arteries.
Effects of Intravenous Pretreatment With OPC-21268 on the Pressor Response to ICV Injections of Glibenclamide
Intravenous administration of OPC-21268 reduced arterial pressure (control, +2±3 mm Hg [n=6]; OPC-21268, −22±4 mm Hg [n=6], P<.01) and abolished the vasopressor response to ICV injections of 10 nmol glibenclamide in rats with bilateral ligation of the carotid arteries (Figs 4⇓ and 5⇓). Heart rate was not influenced by intravenous pretreatment with OPC-21268 (30 minutes after ICV injection of 10 nmol glibenclamide: control, −56±12 bpm [n=6]; OPC-21268, −64±14 bpm [n=6]). Intravenous pretreatment with OPC-31260 (10 mg/kg) affected neither basal arterial pressure (control, +2±3 mm Hg [n=5]; OPC-31260, −5±5 mm Hg [n=5]) nor vasopressor response to ICV injection of 10 nmol glibenclamide in rats with bilateral ligation of the carotid arteries (vehicle, +15±3 mm Hg [n=5]; OPC-31260, +16±4 mm Hg [n=5]).
Effects of α1-Adrenoceptor Blockade on the Pressor Responses to Glibenclamide
Intravenous injections of bunazosin hydrochloride (50 μg), an α1-adrenergic blocker, produced a prolonged lowering of arterial pressure in rats with bilateral ligation of the carotid arteries (−42±3 mm Hg) (Fig 6⇓). Injections of 10 nmol glibenclamide ICV produced a marked vasopressor response compared with vehicle (vehicle, +3±2 mm Hg [n=5]; glibenclamide, +16±3 mm Hg [n=5], P<.01).
We found in the present study that ICV injections of glibenclamide produced a dose-dependent pressor effect in rats with bilateral ligation of the carotid arteries but not in sham-operated rats. Because glibenclamide is a specific and potent blocker of KATP18 and because intravenous administration of glibenclamide did not elicit any cardiovascular response in rats with bilateral ligation of the carotid arteries, this pressor effect must be closely related to inhibition of KATP in the ischemic brain. Intravenous glibenclamide can reach the brain after bilateral ligation of the carotid artery because the residual blood flow still exists, although the amount of glibenclamide must be lower than the ICV injection because of the dilution with blood. A lack of vasopressor effects with intravenous injections of glibenclamide suggests that a large amount of glibenclamide is needed to inhibit KATP in rat brain. Otherwise, glibenclamide may not penetrate the blood-brain barrier.
Although KATP in the peripheral vasculature are the targets of antihypertensive agents,7 the relationship between cerebral KATP and the regulation of arterial pressure has not been previously investigated. Inhibition of KATP in the substantia nigra reportedly elicits the release of γ-aminobutyric acid.19 The bradycardia observed after ICV injections of 10 nmol glibenclamide in rats with bilateral ligation of the carotid arteries may be explained by the action of γ-aminobutyric acid released by glibenclamide.
Both plasma and pituitary AVP concentrations were significantly increased in rats with bilateral ligation of the carotid arteries compared with sham-operated rats. ICV injections of 10 nmol glibenclamide significantly increased the plasma concentrations of AVP and decreased the pituitary concentrations of AVP. ICV injections of glibenclamide increased AVP release and turnover rate in the pituitary glands of rats with bilateral ligation of the carotid arteries. This finding means that cerebral KATP inhibit AVP release from the pituitary glands in the hypoperfused brain. This mechanism may resemble that for KATP in the pancreatic β-cell, which inhibits insulin release.2 Since AVP exhibits a marked vasoconstrictive effect in vascular smooth muscle cells20 and potentiates the vasoconstrictive activity of such vasoactive peptides as angiotensin II21 and endothelin,22 the increased synthesis and release of AVP into the peripheral circulation may be involved in causing the hypertension that follows ligation of the carotid arteries. In support of this, the pressor effect produced by injection of glibenclamide ICV into rats with bilateral ligation of the carotid arteries was abolished by intravenous pretreatment with OPC-21268 and was unaffected by pretreatment with either a V2 receptor antagonist or an α1-adrenoceptor blocker. ICV injections of glibenclamide did not influence abdominal sympathetic nerve activity. Thus, the pressor response was apparently not mediated by peripheral sympathetic nerve activity or by the stimulation of AVP V2 receptors but by activation of AVP V1 receptors in the peripheral vascular bed. AVP V1 receptors are distributed widely in the vascular smooth muscle cells and produce vasoconstriction via a phosphatidylinositol pathway.23 Both plasma and pituitary concentrations of AVP were increased in rats with bilateral ligation even by ICV injections of vehicle. Thus, bilateral ligation of the carotid arteries per se elicited AVP synthesis and release. Intravenous injections of OPC-21268 reduced arterial pressure in rats with bilateral ligation of the carotid arteries. These findings suggest that an elevated level of circulating AVP may participate in the vasopressor mechanism during bilateral carotid artery ligation via the AVP V1 receptors. In addition, AVP is reported to block KATP in peripheral smooth muscle cells.22 Both the increase in plasma levels of AVP and the blocking of KATP in peripheral smooth muscle cells may raise arterial pressure. During the acute reduction of cerebral blood flow, activated KATP in the brain may inhibit the rise in arterial pressure by inhibiting the hypersecretion of AVP from the pituitary.
The carotid arteries supply blood exclusively to the cerebral cortex, midbrain, thalamus, and hypothalamus but not to the lower brain stem, as measured by injection of radioactive microspheres into the carotid arteries.24 Bilateral ligation of the carotid arteries in spontaneously hypertensive rats (SHR) causes acute ischemia of the brain areas in which the carotid artery is responsible for the blood supply.10 11 After carotid ligation, cortical blood flow was reduced to less than 10% of the resting level, and thalamic blood flow was decreased to less than 20% in SHR.11 In normotensive Wistar rats, bilateral carotid artery ligation also caused the reduction of regional cerebral blood flow, although its extent was less than in SHR: carotid ligation decreased cortical blood flow to 36% to 38% of the resting level and thalamic blood flow to 40% in Wistar rats.11 Therefore, present bilateral carotid artery ligation is the model of acute reduction of carotid blood flow with the minimum amount of ischemia, which does not always produce cerebral infarction.25
The cardiovascular regulatory roles of cerebral KATP have not been clarified, despite their abundance in the brain. The present report is the first to show that cerebral KATP are involved in cardiovascular regulation during bilateral common carotid artery ligation by inhibiting the secretion of AVP from the pituitary.
- Received August 8, 1994.
- Revision received September 8, 1994.
- Accepted December 23, 1994.
Petersen OH, Dunne MJ. Regulation of K+ channels plays a crucial role in the control of insulin secretion. Pflugers Arch. 1989;414:S115-S120.
Ribalet B, Ciani S. Regulation by cell metabolism and adenine nucleotides of a K channel in insulin-secreting B cells (RIN m5F). Proc Natl Acad Sci U S A. 1987;84:1721-1725.
Nichols CG, Lederer WJ. Adenosine triphosphate-sensitive potassium channels in the cardiovascular system. Am J Physiol. 1991;261:H1675-H1686.
Standen NB, Quayle JM, Davis NW, Brayden JE, Huang Y, Nelson MT. Hyperpolarizing vasodilators activate ATP-sensitive K+ channels in arterial smooth muscle. Science. 1989;245:177-180.
Britton M, Carlsson A, De Faire U. Blood pressure course in patients with acute stroke and matched controls. Stroke. 1986;17:861-864.
Choki J, Yamaguchi T, Takeya Y, Morimoto Y, Omae T. Effect of carotid artery ligation on regional cerebral blood flow in normotensive and spontaneously hypertensive rats. Stroke. 1977;8:374-379.
Fujishima M, Ishitsuka T, Nakatomi Y, Tamaki K, Omae T. Changes in local cerebral blood flow following bilateral carotid occlusion in spontaneously hypertensive and normotensive rats. Stroke. 1981;12:874-876.
Meyer JS, Stoica E, Pascu I, Shimazu K, Hartmann A. Catecholamine concentrations in CSF and plasma of patients with cerebral infarction and haemorrhage. Brain. 1973;96:277-288.
Wakatsuki T, Nakaya Y, Miyoshi Y, Nomura M, Saito K, Xiao-Rong Z, Ito S, Inoue I. Effects of vasopressin on K+ channels of vascular smooth muscle cells. Therapeutic Research. 1992;13:311-317.
Yamamura Y, Ogawa H, Yamashita H, Chihara T, Miyamoto H, Nakamura S, Onogawa T, Yamashita T, Hosokawa T, Mori T, Tominaga M, Yabuuchi Y. Characterization of a novel aquaretic agent, OPC-31260, as an orally effective, nonpeptide vasopressin V2 receptor antagonist. Br J Pharmacol. 1992;105:787-791.
Amoroso S, Schmid-Antomarchi H, Fosset M, Lazdunski M. Glucose, sulfonylureas, and neurotransmitter release: role of ATP-sensitive K+ channels. Science. 1990;247:852-854.
Mitchell RH, Kirk CJ, Billah MM. Hormonal stimulation of phosphatidylinositol breakdown with particular reference to the hepatic effects of vasopressin. Biochem Soc Trans. 1979;7:861-865.
Takahashi H, Bunag RD. Augmentation of centrally induced alpha-adrenergic vasodepression in spontaneously hypertensive rats. Hypertension. 1980;2:198-206.
Ogata J, Fujishima M, Morimoto Y, Omae T. Cerebral infarction following bilateral carotid artery ligation in normotensive and spontaneously hypertensive rats: a pathological study. Stroke. 1976;7:54-60.