| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2003;34:1547.)
© 2003 American Heart Association, Inc.
Comments, Opinions, and Reviews |
From the Department of Pathology, Division of Autopsy and Neuropathology, Virginia Commonwealth University, Medical College of Virginia, Richmond, Va.
Correspondence to William I. Rosenblum, MD, 305 Tarrytown Dr, Richmond, VA 23229. E-mail wirosenb{at}aol.com
| Abstract |
|---|
|
|
|---|
Summary of Review Electrophysiological data show the existence of KATP in smooth muscle and endothelium of brain vessels. A much larger number of studies in virtually all experimental species have shown that classic openers of KATP dilate brain arteries and arterioles. This response can by blocked by glibenclamide, a selective inhibitor of KATP opening. Several physiological or pathophysiological responses are also blocked by glibenclamide. KATP contains a multiplicity of potential sites of interaction with drugs of diverse, sometimes unrelated, structures. Drugs with imidazole or guanidinium groups are particularly likely to have effects on KATP. This complicates interpretation of the actions of such drugs when used as supposedly selective pharmacological probes for other putative targets. A pH-sensitive site on the internal surface of cloned channels may explain the glibenclamide-inhibitable dilation produced by intracellular acidosis and perhaps by CO2. In some situations KATP appears to be involved in either the synthesis/release or action of endothelium-derived mediators of cerebrovascular tone. The importance of KATP may be dependent on the portion of the cerebrovascular tree being studied and on diverse experimental conditions, age, species, and the presence of disease.
Conclusions KATP have been shown to mediate a wide range of cerebrovascular response in physiologic or pathologic circumstances in a variety of experimental conditions. Their relevance to cerebrovascular responses in humans remains to be explored.
Key Words: adenosine triphosphate aging autoregulation biological factors carbon dioxide cerebral circulation diabetes mellitus endothelium-derived relaxing factor endothelium, vascular enzyme inhibitors hydrogen-ion concentration hypertension hypoxia nitric oxide synthase potassium channels vasodilation
| Introduction |
|---|
|
|
|---|
| Electrophysiological Evidence |
|---|
|
|
|---|
Dilation by Well-Established Openers of KATP
It has been established3,4 that cromakalim, levcromakalim, aprikalim, and pinacidil dilate brain blood vessels in rabbits,6,8,9 cats,10,11 piglets,12,13 rats,11,14,15 sheep,16 and humans.17 Where tested, these dilations are blocked by glibenclamide. However, there is heterogeneity of response, affected by species and location within the cerebrovascular tree.2,8,18,19
Channel Structure and Relation to Effects of Established Openers and Blockers
Channel structure and its relation to the effects of established openers and blockers have been well studied.3,2025 KATP channels consist of 8 subunits (Table). Four have amino acid sequences sufficiently similar to those of inward rectifying (IR) channels to warrant inclusion of KATP in the IR superfamily. These IR units are named IR6.1 or IR6.2. However, KATP have no significant inward rectifying properties. The IR directly affect pore size and contain sites with high affinity for ATP. Attachment of ATP causes reduction in open state probability of the channel. Other agents acting on the IR may also affect the open probability state of KATP. Opening small numbers can have significant effects.3
|
Four additional subunits are attached to and interact with the IR to affect channel opening through mechanisms that are not fully understood. They are labeled SUR because they contain sites with high affinity for sulfonylurea compounds, whose binding causes an interaction between SUR and IR that inhibits channel opening. There may be 4 varieties24,25 of SUR, designated SUR1, SUR2A, 2B, and 2C. The type of SUR subunits in brain blood vessels has not been determined. Examination of KATP elsewhere indicates that SUR2 bind classic openers like pinacidil, while SUR1 bind, with high affinity, the inhibiting sulfonylureas such as glibenclamide and tolbutamide. Glibenclamide may have a second, weaker binding site on SUR2. Unlike tolbutamide, the structure of glibenclamide includes a benzamido group that can bind to an additional site on SUR1. One may speculate that the absence of the additional binding sites for tolbutamide may explain the inability of tolbutamide to block dilation in situations26,27 in which glibenclamide is effective.
Other Drugs That Block KATP
The large number of sites affecting23,24,28 the open state of the channel may provide loci for binding and action of drugs with supposedly selective actions on other putative targets. Among such compounds are drugs containing imidazoline or guanidinium moieties.29,30 This may explain why tetrodotoxin, a guanidinium compound, and 7-nitroindazole, an imidazole compound, could block the dilating action of selective KATP openers on rat pial arterioles.15
Dimethyl sulfoxide and ethanol, 2 commonly used diluents of water-insoluble drugs, in commonly used final concentrations, could also prevent the dilation of pial arterioles by openers of KATP.31,32
This effect was consistent with the observation that other antioxidants also inhibit dilation produced by openers of KATP.32 Conversely, some oxidants caused dilation that was blocked by glibenclamide.33 The effects of diverse oxidizing substances on KATP and other K+ channels have recently been reviewed.34
Factors affecting the aforementioned results might include composition of irrigating fluid,35 the presence and type of paralytic drug in artificially respired animals, and the nature of the anesthetic.3638
KATP, Nitric Oxide, and Endothelium-Derived Hyperpolarizing Factor
Endothelium-derived relaxing factor (EDRF/nitric oxide [NO])39,40 and endothelium-derived hyperpolarizing factor (EDHF)41 relax vascular smooth muscle (VSM). Acetylcholine releases both from endothelium. Most studies of KATP in cerebral circulation have not distinguished between the role of KATP in release of mediator(s) from endothelium versus their role in the action of the mediator(s) on VSM.
One study42 found that blockade of KATP inhibited dilation by NO only if 2 other classes of K+ channel were also blocked, concluding that one type of K+ channel may exert effects that compensate for the loss of another type.
Dilation of piglet pial arterioles by an NO donor or by a cGMP analogue were partially inhibited by glibenclamide,13 the selective blocker of KATP. Since cGMP in VSM is the second messenger for NO, the KATP linkage for response to NO may have been in the VSM.
Rabbit cerebral arteries7,9,14 responded to EDRF/NO when KATP were blocked. Glibenclamide partly blocked dilation of rabbit middle cerebral artery while totally blocking hyperpolarization produced by acetylcholine. Dilation to a NO donor occurred without a change in membrane potential. These data suggested that response to acetylcholine had 1 component dependent on EDRF/NO but independent of KATP and 1 component dependent on EDHF and KATP. However, in guinea pig middle cerebral artery, glibenclamide failed to impair the portion of dilation thought to be caused by EDHF.43
KATP as Mediator of Other Endogenous Dilators
Dilation by calcitonin gene-related peptide (CGRP) and by vasointestinal peptide may be inhibited by glibenclamide. This may2 reflect the role of cAMP as second messenger for these dilators. When released by forskolin, cAMP dilated vessels, and the dilation could also be inhibited by glibenclamide.2 This has been contrasted2 with the failure of glibenclamide to impair the response to EDRF/NO, a cGMP-dependent mechanism. However, glibenclamide has been reported to inhibit dilation induced by cGMP in pial arterioles in piglets.13
KATP and Response to Hypercapnia and pH
Cerebral arteries and arterioles are dilated by increases in CO2 and constricted by decreases in CO2 with corresponding increases or decreases in cerebral blood flow. Because of the importance of the response to CO2,44,45 it is of interest to see whether KATP play a role in the response.
Glibenclamide inhibited all or part of the dilation of pial arterioles induced by hypercapnia11,46 in pentobarbital-anesthetized, paralyzed, artificially respired rats and cats but failed in rats under other experimental conditions.47 In another study,48 hypercapnic dilation was not affected by glibenclamide unless the rats were first treated with a nitric oxide synthase (NOS) inhibitor and then repleted with a cGMP analogue, which was presumed to compensate for the decrease of cGMP brought about by NO depletion.
In humans,49 oral glibenclamide failed to inhibit increases in cerebral blood flow produced by hypercapnia. However, the effect of glibenclamide on forearm blood flow was used as a basis for concluding that an adequate dose of glibenclamide had been given, and there was no direct evidence that the glibenclamide reached the cerebral circulation in adequate doses.
Glibenclamide pretreatment was also reported to inhibit constriction produced by hypocapnia.50 However, the inhibiting effect of glibenclamide seemed somewhat paradoxical to the investigators50 because of the following: (1) hypocapnic constriction was thought to depend on closure of KATP caused by increased pH; (2) glibenclamide was thought to prevent this constriction by closing KATP so that there was no significant proportion of KATP left open for hypocapnia to close; (3) however, this closure of channels by glibenclamide pretreatment should have constricted the vessels. The paradox resides in the absence of such constriction here and in other studies.1,8 The failure of glibenclamide to constrict had been explained by assuming1 that KATP play no role in the maintenance of resting tone. It was assumed that, unless the vessel was dilated by a stimulus that opened KATP, the open state probability of the channels was too small to affect tone. However, hypocapnic constriction then could not be explained by a further closing of the channels. To resolve the difficulties apparent in the attempt to explain all the results on the probability of KATP opening, the authors suggested that there are redundant mechanisms for maintaining tone and that these are so rapidly brought into play that they prevent closure of the KATP by glibenclamide from influencing the resting tone. If this is so, we must assume that in their experiments hypocapnia reduced the open state probability of KATP even more than glibenclamide so that only hypocapnia successfully overrode the compensatory mechanisms acting to prevent a change in tone.
The response to CO2 is actually mediated by the corresponding change of pH (reduced with hypercapnia and increased with hypocapnia) in blood and extracellular space or cerebrospinal fluid.51 Glibenclamide, the KATP blocker, had no effect on dilation of penetrating arterioles of rat cerebellum and brain stem when external pH was decreased in vitro.52 However, glibenclamide inhibited dilation of penetrating arterioles of rat cerebrum, also caused by reduction of extravascular pH.53 The difference may reflect the regional differences in KATP distribution. Dilation and constriction of penetrating arterioles of rat cerebrum were highly correlated with the resting potential, which was related linearly to the diameter.54 This is consistent with a role for KATP since the change in the open state of the channels would alter the resting potential.
However, others55 found that hyperpolarization was not required for a decrease in pH to produce dilation. The same group55,56 showed that internal pH (pHi) paralleled the external pH in cerebrovascular smooth muscle. However, their data failed to support the belief that dilation occurred when pHi was reduced by increasing external CO2. Other investigators5759 concluded that it was the external pH rather than internal pH that determined the dilation of the vessel. Despite the fact that several articles indicated that pHi did not control diameter, a diagram in a review1 of potassium channels and cerebral circulation appearing in 1998 indicated that decreases of pHi open KATP.
This conclusion foreshadowed the more recent demonstration that cloned KATP demonstrate a pH-sensitive site on the inside rather than the outside of the cell membrane.28 A decrease in pH led to an increase in the open state probability of KATP over the range of pH encountered during hypocapnia, normocapnia, and hypercapnia. The pH-sensitive site was located on the IR6.2 subunit rather than the SUR subunit.
Finding the pH-sensitive site on the inside of the cell membrane has yet to be reconciled with the studies that concluded that the response to hypercapnia is mediated by external pH. However in another setting, the weight of evidence has been judged to support control of diameter by pHi. In this setting, CO2 is elevated by acetazolamide, a carbonic anhydrase inhibitor. With the exception of 1 study,60 both hypercapnia and acetazolamide have been shown to reduce pHi in the brain. At a relatively early date, Severinghaus and Cotev61 concluded that the preponderance of evidence supported the hypothesis that reduction of pHi caused vasodilation by acetazolamide.
Some data indicate that there may be 2 mechanisms for hypercapnic dilation, one of which is KATP dependent and the other NO dependent.62 For example, a NOS inhibitor eliminated only a portion of the dilation of rat pial arterioles produced by acidosis, with the remaining portion eliminated by glibenclamide.62
The existence of dual or alternative important mechanisms not involving NO is illustrated in knockout mice lacking the neuronal isoform of NOS,63 which is the isoform thought to mediate the response to hypercapnia in normal animals.64 Whether the NO-independent mechanism is KATP dependent in the knockout mice remains to be determined. A further dissection of the relationship between NO, KATP, and the response to hypercapnia requires that drugs that inhibit NOS do not have an effect on KATP. This subject is discussed next.
KATP and Inhibitors of NOS
The basis for the belief that NO mediates the response to hypercapnia resides largely64 in studies using NOS inhibitors. However, there are now reports11,15,46 that in rats and cats both NOS inhibitors and glibenclamide can block the dilation of pial arterioles produced by KATP openers as well as by hypercapnia. The inhibiting action of the NOS inhibitors on KATP openers was not due to some (now eliminated) permissive action of NO because adding a NO donor failed to rescue the response to openers of KATP. Moreover, L-arginine, the substrate for NOS, was found necessary for the dilation produced by KATP openers. L-Lysine could substitute for L-arginine. Both amino acids may by transported by the same transport system.65 L-Lysine cannot be metabolized by NOS. Hence, its ability to substitute for L-arginine supports the interpretation that NO was not involved. These data indicate that under some circumstances KATP can mediate hypercapnic dilation in rats and cats. These data also suggest a need for caution in interpreting and designing studies using NOS inhibitors. It would appear advisable to first see whether these NOS inhibitors can block KATP under the proposed experimental conditions.
The cited studies11,15,46 employed pentobarbital anesthesia and paralyzed, artificially respired animals. No study of brain blood vessels has appeared in which either the effect of NOS inhibitors on KATP openers or the effect of glibenclamide on hypercapnic dilation has been systematically compared in the presence and absence of these other possibly pertinent experimental factors. However, there are a few studies, performed under other conditions, which, incidental to their primary purpose, describe the failure of NOS inhibitors to inhibit dilation by KATP openers.66,67
On the other hand, the data in some studies are consistent with such an effect. For example,68 NG-nitro-L-arginine (L-NAME), a NOS inhibitor, inhibited the dilation of piglet arterioles by aprikalim, an opener of KATP. The possibility that L-NAME was inhibiting KATP was not considered. Instead, it was concluded that NO was involved in the dilation produced by aprikalim. This interpretation is consistent with the data. However, if so, L-NAME must have prevented either aprikalim or KATP from releasing NO from endothelium rather than blocking its effects, since glibenclamide, the bona fide blocker of KATP, had no effect on dilation produced by a NO donor. In another study,69 CGRP dilated the basilar artery. The dilation was partially blocked by a NOS inhibitor but was completely blocked by glibenclamide, the KATP blocker. This was interpreted as indicating 2 additive mechanisms for the dilation by CGRP. The possibility was not considered that L-NAME may have also been inhibiting KATP but with less efficacy, at the dose used, than glibenclamide.
Note added in proof: Following acceptance of this review, Santa et al70 reported glibenclamide-inhibitable dilation of rat basilar artery in vivo by a drug selectively decreasing intracellular pH. Cultured basilar smooth muscle showed message for IR6.1 and for SUR-B but not for IR6.2. These data support studies cited in this review that indicate an intracellular pH-sensitive site on KATP. However, a NOS inhibitor, known to inhibit hypercapnic dilation, failed to inhibit dilation produced by the intracellular acidifier. Therefore, Santa et al restricted their conclusion to the suggestion of a role for KATP in mediating dilation produced by reduced pH but not in the context of hypercapnia.
Influence of Age, Diabetes, Hypertension, Hypotension, and Hypoxia
Maturation increased the responsiveness to a KATP opener in some branches of middle cerebral artery in sheep.16 However, there was no effect of aging on the response to KATP openers in Fischer rats.71
Diabetes66 and some forms of hypertension72 may inhibit dilation by KATP openers. In stroke-prone spontaneously hypertensive rats, superimposition of subarachnoid hemorrhage greatly enhanced the previously diminished response.72
KATP may participate in the dilation produced by hypotension73,74 and hypoxia.75 Dilation during oxygen deprivation may result in part from decrease in ATP76 below the levels at which KATP opening is suppressed and in part from the acidosis that accompanies the hypoxia76 and can increase open state probability of KATP. In rats, the autoregulatory dilation in response to hypotension may be mediated by KATP in brain stem neurons.74 However, discussion of the role of KATP in brain stem control of cerebrovascular tone is beyond the scope of this review.
Summary
Patch clamp demonstrations of KATP in cerebrovascular smooth muscle and cerebrovascular endothelium are rare. However, most pharmacological data indicate that KATP exist in cerebral arteries and arterioles in virtually all species. The importance of KATP depends on the experimental conditions or pathological state. Multiple sites on the complex KATP structure provide loci for interaction with channel inhibitors like ATP and sulfonylureas, with channel openers like pinacidil and cromakalim, and with hydrogen ions. In addition, structural complexity provides sites for interaction with a wide variety of other drugs, including imidazole or guanidinium derivatives. Therefore, many drugs, selectively interacting with other targets, may also inhibit KATP and thus have a previously unsuspected mechanism of action against a range of physiological or pathological responses. In some studies, the NOS substrate L-arginine and inhibitors of NOS interacted with KATP in pial arterioles. Therefore, under some circumstances the inhibition of responses by NOS inhibitors may not be proof of mediation by NO but instead may result from inhibition of KATP. Further studies are required to define conditions necessary for this effect. With respect to the role of KATP in endothelium-mediated dilation, further studies are necessary to distinguish between the role of KATP in the action of either EDRF/NO and/or endothelium-derived hyperpolarizing factor on VSM versus the role of KATP in the release of these dilators from endothelium. The possible role of KATP in mediating the response to pH (eg, in hypercapnia or hypocapnia) is of particular interest because pH is an important modulator of cerebrovascular tone. Therefore, it is noteworthy that a pH-sensitive site on cloned KATP has been demonstrated. This site was expressed on the inside of the cell membrane, but several investigators believe that external pH rather than pHi is the determinant of hypercapnic dilation. Resolution of this conundrum awaits further investigations.
Received September 9, 2002; revision received November 13, 2002; accepted December 13, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Adebiyi, E. M. McNally, and J. H. Jaggar Sulfonylurea Receptor-Dependent and -Independent Pathways Mediate Vasodilation Induced by ATP-Sensitive K+ Channel Openers Mol. Pharmacol., September 1, 2008; 74(3): 736 - 743. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Shi, Z. Wu, N. Cui, W. Shi, Y. Yang, X. Zhang, A. Rojas, B. T. Ha, and C. Jiang PKA phosphorylation of SUR2B subunit underscores vascular KATP channel activation by beta-adrenergic receptors Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2007; 293(3): R1205 - R1214. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. W. Leffler, H. Parfenova, J. H. Jaggar, and R. Wang Carbon monoxide and hydrogen sulfide: gaseous messengers in cerebrovascular circulation J Appl Physiol, March 1, 2006; 100(3): 1065 - 1076. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Tomura, J.-Q. Wang, M. Komachi, A. Damirin, C. Mogi, M. Tobo, J. Kon, N. Misawa, K. Sato, and F. Okajima Prostaglandin I2 Production and cAMP Accumulation in Response to Acidic Extracellular pH through OGR1 in Human Aortic Smooth Muscle Cells J. Biol. Chem., October 14, 2005; 280(41): 34458 - 34464. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2003 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |