Mechanisms of Isoflurane-mediated Hyperpolarization of Vascular Smooth Muscle in Chronically Hypertensive and Normotensive Conditions

All protocols in this study were reviewed and approved by the Animal Care and Use Committee at the Medical College of Wisconsin. A total of 218 SHR and WKY rats were studied. All animals were between 8 and 13 weeks of age and weighed between 250 and 350 g. Anesthesia was induced with 40 mg/kg intraperitoneal ketamine followed by 20 mg/kg intraperitoneal pentobarbital to facilitate surgical preparation. Subsequently, basal anesthesia was maintained throughout the course of each experimental protocol by intravenous pentobarbital administration at a constant infusion rate of 15– 30 mg · kg⁻¹· h⁻¹in physiologic salt solution (PSS) containing 2% bovine serum albumin. In each animal the femoral artery and vein were cannulated for mean arterial blood pressure (MAP) measurement and intravenous access, respectively. In addition, a tracheotomy was performed through which ventilation was controlled with a model 680 rodent respirator (Harvard Apparatus Co., South Natick, MA). Respiratory rate and tidal volumes were adjusted to maintain end-tidal carbon dioxide between 30 and 35 mmHg.

During all of the experiments, the animals breathed an inspired oxygen concentration of 30% (in an oxygen–nitrogen mixture) to reduce any possibility of hypoxia-induced effects on the VSM. 18Minute ventilation was adjusted as described above and the end-tidal carbon dioxide was measured with a POET 2 infrared capnograph and end-tidal agent monitor (Criticare Systems, Inc. Waukesha, WI).

A midline laparotomy was performed through which a loop of terminal ileum and its attached mesentery were externalized and placed on a movable, temperature-regulated microscope stage that was mounted on a micro-G vibration-free table (Technical Mfg. Co., Woburn, MA). Small (200–300 μm) mesenteric arteries and paired veins were identified. After dissection of the perivascular fat (without disturbing luminal blood flow or adventitial innervation), the surrounding connective tissue of the vessels was attached to the silastic rubber floor of the chamber with 125-μm-diameter stainless steel pins. Smaller (50 μm) pins were used to line each of the vessels on both sides to minimize pulse movement artifact. While in the chamber, the vessels were continuously superfused with PSS composed of 119 mm NaCl, 4.7 mm KCl, 1.17 mm MgSO⁴, 1.6 mm CaCl², 24.0 mm NaHCO³, 1.18 mm NaH²PO⁴, 0.026 mm EDTA. The superfusate was continuously aerated with a gas mixture of nitrogen, oxygen, and carbon dioxide to maintain pH between 7.35 and 7.45, carbon dioxide partial pressure between 35 and 45 mmHg, and oxygen partial pressure between 75 and 100 mmHg. The temperature of the PSS was maintained between 36 and 37°C.

For each protocol, isoflurane was delivered locally (not systemically) to the in situ  vessel preparation via  the PSS superfusate. The oxygen–nitrogen–carbon dioxide mixture aerating the superfusate was used as a carrier gas to deliver the isoflurane to the PSS through an Ohio Medical Products vaporizer (Airco Inc., Madison, WI). Concentrations of isoflurane were measured in PSS and blood samples with a Shymadzu model GC-8A gas chromatograph (Shimadzu Co., Kyoto, Japan). The vaporizer settings were adjusted to produce PSS isoflurane concentrations that averaged 0.6 mm. This concentration corresponded to the mean concentration that was measured in blood when 1 MAC isoflurane was administered systemically by inhalation. 19 

Local sympathetic denervation of each paired mesenteric artery and vein preparation in the present study was accomplished by a 20-min superfusion of the vessel preparations with 300 μg/ml 6-hydroxydopamine. Local application of 6-hydroxydopamine produces changes consistent with perivascular denervation, including inhibition of contractile responses to field stimulation, blockade of H³-norepinephrine uptake by sympathetic nerve endings, and histologic changes consistent with adrenergic nerve degeneration. 20,21Before denervation by this technique, the preparation was pretreated with PSS containing 10⁻⁶m phentolamine for 5 min to inhibit the vasoconstriction from catecholamines locally released by the 6-hydroxydopamine superfusion. 22 

In each preparation, single-cell VSM E^ms were measured in situ  by advancing 3 m KCl–filled glass micropipette electrodes into the VSM layer of the mesenteric arteries and veins from the adventitial side. The tip diameter of the micropipettes was approximately 0.1 μm with an impedance range of approximately 40–60 MΩ. Micropipettes were pulled from borosilicate glass with a Model P-97 Brown/Flaming micropipette puller (Sutter Instrument Company, Novato, CA). Electrodes were advanced manually using a hydraulic micro-micromanipulator (Trent Wells Inc., Coulterville, CA) and were connected to a Model 8100 biological amplifier (Dagan Corporation, Minneapolis, MN). Both arterial blood pressure and VSM E^ms were recorded with a Grass model RPS7C polygraph (Astro-Med/Grass Inc., West Warwick, RI) and a Superscope II (version 1.44) digital data acquisition system (GW Instruments Co., Somerville, MA).

Figure 1shows a schematic of the vasodilator second messenger pathways regulating VSM K⁺channel activity together with the sites of action of specific pharmacologic analogs and inhibitors used in the present study. Each analog and inhibitor concentration used was at least five times the reported ED⁵⁰or a concentration previously reported to be effective. 23–28During each step in each protocol (see below), vessel preparations were superfused with PSS containing the specified equilibrated mixture of isoflurane, analog, inhibitor, or combination for a 20–30-min period before any E^mmeasurements.

Two specific VSM membrane K⁺channel inhibitors were used to verify the participation of K⁺in the isoflurane-induced hyperpolarization of VSM in the SHR and WKY small mesenteric vessels. In situ  VSM E^mand concurrent MAP were measured during sequential step intervals with PSS superfusate containing one of the following: (1) no agent (control); (2) 0.6 mm isoflurane; (3) either a K^Cachannel inhibitor (10⁻⁷m iberiotoxin) or a K^ATPchannel inhibitor (10⁻⁶m glybenclamide) 9after washout of isoflurane; or (4) isoflurane plus one of the channel inhibitors.

The effect of inhibition of NO synthesis on isoflurane-induced VSM hyperpolarization in the small mesenteric arteries and veins in both animal types was assessed by superfusion with the NO synthase inhibitor, N  ^G-nitro-l-arginine methyl ester (l-NAME). In situ  VSM E^mand concurrent MAP were measured sequentially during superfusion with PSS containing (1) no agent (control), (2) 0.6 mm isoflurane, (3) NO synthase inhibitor, or (4) isoflurane plus NO synthase inhibitor. Both a lower (50 μm) 29and higher (1 mm) 30concentration of l-NAME was used in separate animal–vessel preparations.

The effect of an inhibition of two specific steps in the cGMP pathway on isoflurane-induced VSM hyperpolarization in the small mesenteric arteries and veins was investigated in two separate studies. In the first, guanylate cyclase-induced synthesis of cGMP was inhibited with 15 μm H-[1,2,4] oxadiazolo [4,3,-a] quinoxalin-1one (ODQ). 12In the second study, cGMP-mediated activation of protein kinase G (PKG) was inhibited with 2.5 μm (Rp)-8-(para –chlorophenylthio)guanosine-3′,5′-cyclic monophosphorothioate (Rp-8-pCPT-cGMPS), an inhibitor of PKG phosphorylation (fig. 1). 12 In situ  VSM E^mand MAP were measured sequentially during vessel superfusion with PSS containing (1) no agent (control), (2) 0.6 mm isoflurane, (3) one of the two specific cGMP pathway inhibitors (after washout of isoflurane), or (4) both isoflurane and one of the two inhibitors.

In addition to these studies, two control studies were conducted to verify the functional existence of PKG in these vessels and the efficacy of its inhibitors, respectively. In the first control study, the VSM E^mresponse to the membrane-permeable cGMP analog 8-bromoguanosine-3′,5′-cyclic monophosphate (8-Br-cGMP) was measured. The concentration used in the PSS superfusate was 100 μm.23 In situ  VSM E^mand MAP were measured sequentially during vessel superfusion with PSS containing (1) no agent (control), (2) the cGMP analog, or (3) no agent (after washout of the analog). In the second control study, the efficacy of the inhibitor of PKG activation (Rp-8-pCPT-cGMPS) was verified. In situ  VSM E^mand MAP were measured sequentially during superfusion with PSS containing (1) no agent (control), (2) the inhibitor of PKG activation, or (3) both the cGMP analog and the inhibitor of PKG activation.

Similar to the cGMP studies, the effect of an inhibition of two specific steps in the cAMP pathway on isoflurane-induced VSM hyperpolarization was investigated in two separate studies. In the first, adenyl cyclase-induced synthesis of cAMP was inhibited with 410 μm SQ22536. 24In the second study, cAMP-mediated activation of protein kinase A (PKA) was inhibited with 24.5 μm Rp-adenosine-3′,5′-cyclic monophosphorothioate (Rp-cAMPS), an inhibitor of PKA phosphorylation (fig. 1). 25 In situ  VSM E^mand MAP were measured sequentially during vessel superfusion with PSS containing (1) no agent (control), (2) 0.6 mm isoflurane, (3) one of the two inhibitors (after isoflurane washout), or (4) both isoflurane and one of the two inhibitors.

In addition to these studies, two control studies were conducted to verify the functional existence of PKA in the VSM of these vessels and the efficacy of its inhibitors, respectively. In the first, E^mresponse to a membrane-permeable activator of PKA 0.15 μm Sp-5,6-dichloro-1-b-d-ribofuranosylbenzimidazole-3′,5′-monophosphoro-thioate (Sp-5,6-DCl-cBiMPS) 26was measured. This agent activates PKA via  phosphorylation. In situ  VSM E^mand MAP were measured sequentially during vessel superfusion with PSS containing (1) no agent (control), (2) the PKA activator, or (3) no agent (after washout of the activator). In the second control study, the efficacy of the inhibitor of PKA activation was verified. In situ  VSM E^mand MAP were measured sequentially during vessel superfusion with PSS containing (1) no agent (control), (2) the inhibitor of PKA activation, or (3) both the PKA activator and its inhibitor.

Sodium chloride, potassium chloride, magnesium sulfate, calcium chloride, sodium bicarbonate, sodium phosphate, ethylenediaminetetraacetic acid, d-glucose, 6-hydroxydopamine, phentolamine, and l-NAME were purchased from Sigma Chemical Co. (St. Louis, MO). ODQ, SQ22536, Rp-cAMPS, and Sp-5,6-DCl-cBiMPS were purchased from BIOMOL Research Laboratories, Inc. (Plymouth Meeting, PA). 8-Br-cGMP was purchased from Calbiochem-Novabiochem Corporation (La Jolla, CA). Iberiotoxin and glybenclamide were purchased from Research Biochemicals International (Natick, MA). Sodium pentobarbital and isoflurane were purchased from Abbott Laboratories (North Chicago, IL). Ketamine hydrochloride was purchased from Phoenix Pharmaceutical, Inc. (St. Joseph, MO).

The VSM E^mvalue reported for each step in each of the experimental protocols is the mean ± SD obtained from 8–12 vessel–animal preparations for both artery and vein. For each step in each experimental protocol, an individual VSM E^mvalue for a vessel–animal preparation is the numerical average of at least five stable (6–10 s) individual impalements. The mean ± SD of 8–12 of these individual values was then compared using a repeated-measures multiple analysis of variance. The “between” factor was animal type (with WKY and SHR as different levels), and the levels of the “within” factor were each of the three or four conditions defining that particular protocol. The MAP recorded simultaneously with VSM E^mwas analyzed in identical fashion.

The first two groups of mean E^mmeasurements were the denervated control and the superfused isoflurane in all protocols involving isoflurane administration. Thus, the effect of superfused isoflurane on VSM E^mand MAP was determined by comparing pooled denervated control data to pooled superfused isoflurane data using repeated-measures analyses of variance. WKY versus  SHR served as the between factor and control versus  isoflurane served as the repeated factor.

Mean blood concentrations of isoflurane were determined by pooling all measurements from individual animal–vessel preparations during each of the three experimental conditions (i.e. , before, during, and after superfusion of blood vessels with isoflurane, respectively). The mean values in these three groups were compared using a one-way analysis of variance.

All t  tests in the current study were calculated using the Stat-View program for Macintosh computers (SAS Institute, Cary, NC). All analyses of variance were calculated using the Super ANOVA program for Macintosh computers (Abacus Concepts, Berkeley, CA). The significance of differences between individual means of specific groups was evaluated using the post hoc  least significance difference test. P ≤ 0.05 was used to define the significance of all differences.

The mean ± SD isoflurane concentration in the PSS superfusate was 0.6 ± 0.1 mm for the isoflurane step and 0.6 ± 0.2 mm for the isoflurane + inhibitor step (n = 66 WKY and 69 SHR vessel preparations). Combining all protocols, the mean ± SD baseline MAP during pentobarbital anesthesia for SHR (158 ± 26 mmHg) was significantly greater than for WKY (114 ± 16 mmHg). Neither SHR nor WKY MAP was significantly altered by local superfusion with isoflurane. The maximum variation (from the initial baseline MAP) within any single protocol was 15% (observed for SHR in the study of the effect of PKA activation on VSM E^m). Such MAP variations included both increases and decreases and were not correlated with any of the observed changes in VSM E^m.

The in situ  venous VSM was significantly more polarized than the arterial VSM in each animal type. In addition, the isoflurane-induced hyperpolarization response was significant and similar between SHR and WKY for each vessel type. For the SHR artery, the pooled mean VSM E^m± SD before isoflurane superfusion was −37 ± 4.3 mV versus −41 ± 4.2 mV during isoflurane superfusion. For the WKY artery, corresponding values were −37 ± 4.2 mV versus −43 ± 3.4 mV, respectively. For the SHR vein, the pooled mean VSM E^m± SD before isoflurane superfusion was −42 ± 5.7 mV versus −49 ± 5.8 mV during isoflurane superfusion. For the WKY vein, corresponding values were −40 ± 4.3 mV versus −46 ± 4.8 mV, respectively.

Table 1illustrates the effect of K⁺channel inhibitors on arterial and venous mean VSM E^min SHR and WKY and the responses to isoflurane. Superfusion of the vessel preparations with either the K^Caor K^ATPchannel inhibitor (iberiotoxin or glybenclamide, respectively) significantly depolarized the VSM in each of the two vessel types in both SHR and WKY. In addition, when superfused concurrently with isoflurane, both inhibitors abolished the isoflurane-induced VSM hyperpolarization in each of the two vessel types in both SHR and WKY.

Table 2illustrates the effect of two concentrations of the NO synthase inhibitor (l-NAME) on arterial and venous VSM E^min SHR and WKY and the VSM E^mresponse to isoflurane. When superfused alone (i.e. , after washout of isoflurane), the 50-μm concentration caused a small but significant depolarization in each vessel type relative to its respective control VSM E^m. However, at the 1-mm concentration, such depolarization was not consistent, occurring only in the SHR vein. Of particular note is that addition of either concentration of the inhibitor to the superfusate containing isoflurane did not significantly attenuate the isoflurane-induced hyperpolarization in either the SHR or WKY vessels.

The data in table 3support the functional existence of the cGMP pathway in the VSM of small mesenteric vessels in both SHR and WKY. The cGMP analog (8-Br-cGMP) significantly hyperpolarized VSM in each vessel type in both SHR and WKY. In addition, the efficacy of the inhibitor of PKG activation (2.5 μm Rp-8-pCPT-cGMPS) was verified by its ability to block the hyperpolarization produced by the cGMP analog. Superfusion with the inhibitor of PKG activation (Rp-8-pCPT-cGMPS) or with the inhibitor of guanylate cyclase–induced synthesis of cGMP (ODQ; after washout of isoflurane) produced a depolarizing response in each vessel type when compared with its respective VSM E^mcontrol value. These responses were variable in that the depolarization produced by these two inhibitors was not significant for all vessel types. However, in both vessel types of both WKY and SHR, the isoflurane-induced hyperpolarization was not significantly attenuated either by the inhibitor of PKG activation or cGMP synthesis.

The data in table 4support the functional existence of the cAMP pathway in the VSM of small mesenteric vessels in both SHR and WKY. The membrane-permeable activator of PKA (Sp-5,6-DCl-cBiMPS) significantly hyperpolarized VSM in each vessel type in both SHR and WKY. In addition, the efficacy of the inhibitor of PKA activation (24.5 μm Rp-cAMPS) was verified by its ability to inhibit the hyperpolarization induced by the PKA activator. It is particularly important to note that the isoflurane-induced VSM hyperpolarization was abolished in the presence of either the inhibitor of cAMP synthesis (SQ22536) or the inhibitor of PKA activation (Rp-cAMPS).

Potassium channel activity in the VSM membrane is the primary regulator of VSM E^mover the normal in situ  physiologic range. 7,8,10,31In the present study we observed an inhibition of isoflurane-induced in situ  VSM hyperpolarization in small mesenteric blood vessels of SHR and WKY after direct block of VSM K^Caor K^ATPchannels (with iberiotoxin or glybenclamide, respectively). Attention was focused on K^Caand K^ATPchannels only, because in a previous study with normotensive Sprague-Dawley rats, 19we observed that only these channels (and not K^Vor K^IRchannels) were activated by volatile anesthetics to produce in situ  VSM hyperpolarization. These observations suggested that neurally independent mechanisms of VSM hyperpolarization and inhibition of VSM tone by volatile anesthetics (in both hypertensive and normotensive subjects) result from enhanced K^Caand K^ATPchannel activation. However, the possible role of NO, cGMP, and cAMP second messenger vasodilator pathways in mediating such anesthetic-induced effects was not clear.

A major finding of the present study is the observed abolition of isoflurane-induced VSM hyperpolarization by inhibition of cAMP synthesis or cAMP-mediated activation of PKA (with SQ22536 or Rp-cAMPS, respectively). Other investigators have shown that phosphorylated PKA activates K^Caand K^ATPchannels. 10,32Abolition of isoflurane-induced VSM hyperpolarization by inhibition of components in the cAMP pathway suggests that volatile anesthetics can enhance activity of components in the cAMP second messenger system in addition to a possible direct action on potassium channel proteins. This is in contrast to reported isoflurane-induced inhibition of cAMP-mediated vasodilatation of in vitro  coronary arteries. 33However, a cAMP-mediated VSM hyperpolarization is in agreement with a number of other studies that have demonstrated an augmentation of cAMP concentration and activity in VSM by volatile anesthetics. 34The results of the present study suggest that the mechanisms underlying the observed isoflurane-induced VSM hyperpolarization (and presumed coupled relaxation) in the in situ  small mesenteric resistance- and capacitance-regulating blood vessels of SHR and WKY include activation of a cAMP-mediated opening of VSM K^Caand K^ATPchannels.

The small but significant in situ  VSM depolarization response to superfusion with 50 μm l-NAME (NO synthase inhibitor) observed after washout of isoflurane (table 2) provides some evidence that NO participates in the regulation of VSM E^mand tone in the SHR and WKY small mesenteric blood vessels. However, it is not clear why 1.0 mm l-NAME failed to produce a similar (or greater) depolarization response (except for the SHR vein). The data in table 3provide relatively strong evidence for the existence of the cGMP-PKG pathway in the regulation of VSM E^min these vessels (e.g. , hyperpolarizing response to 8-Br-cGMP [cGMP analog] and depolarizing response to Rp-8-pCPT-cGMPS [PKG inhibitor]). Perhaps the best recognized effect of cGMP is the phosphorylation (and activation) of cGMP-dependent PKG, 35ultimately leading to an increased extrusion of intracellular calcium ([Ca²⁺]ⁱ) by activated cell and sarcoplasmic reticulum membrane-bound Ca²⁺pumps. 36It is not clear if such reduction of [Ca²⁺]ⁱalone can alter the resting VSM E^m. However, some evidence suggests that either activated PKG or cGMP (or a cGMP derivative) hyperpolarizes VSM by activating VSM membrane potassium channels. 36Volatile anesthetics have been reported to enhance NO-mediated relaxation of in vitro  canine coronary arterial VSM 37and increase the concentration of cGMP in canine cerebral arterial VSM. 15However, other in vitro  studies concluded that isoflurane does not enhance, but rather inhibits, endothelium-derived NO release 38,39as well as cGMP-mediated VSM relaxation. 40In the present study, no differences were observed in VSM E^mresponse to isoflurane before and during inhibition of NO synthesis or block of selective steps in the cGMP-PKG pathway in respective vessels of either animal type. This suggests that the isoflurane-induced VSM hyperpolarization does not involve components in the cGMP pathway.

We recognize that our conclusions in the present study are based on VSM E^mmeasurements rather than measurements of VSM contractility or changes in [Ca²⁺]ⁱ. However, VSM tone is closely coupled to both E^mand [Ca²⁺]ⁱand critically dependent on influx of extracellular calcium through voltage-sensitive calcium channels. 9Such influx (and [Ca²⁺]ⁱ) are inversely proportional to potassium channel conductance and hence magnitude of VSM E^m. Thus, isoflurane-induced increases in the magnitude of VSM E^mshould be inversely coupled to decreases in [Ca²⁺]ⁱand VSM tone. 41 

A second major observation in the present study is the lack of a significant difference between SHR and WKY in the effects of isoflurane on K⁺channel–mediated control of VSM E^min respective small mesenteric blood vessels. The results in the present study complement those of our previous studies, in which we demonstrated a greater anesthetic-induced hyperpolarization in neurally intact (vs.  denervated) vessel preparations in SHR and WKY. 19Both studies strongly suggest that volatile anesthetics attenuate VSM tone in hypertensive subjects primarily by modulation of the elevated level of the sympathetic neural control that exists in the neurogenic SHR model. Therefore, we concluded that in situ  differences in anesthetic effect on VSM E^m(and tone) between the SHR and WKY do not involve the neurally independent vasodilator mechanisms that were the focus of the present study. Evidence exists in support of intrinsic differences in potassium channel function between WKY and SHR, 13as well as an altered NO-dependent control of VSM tone in hypertensive subjects. 14Results of the present study suggest that none of these neurally independent mechanisms account for the hemodynamic instability characteristic of anesthesia in the hypertensive condition. 3It is possible that differential effects of volatile anesthetics on vascular control in hypertensive versus  normotensive subjects may involve mechanisms not addressed in the present study (e.g. , altered sensitivity of contractile proteins to volatile anesthetics). Clearly, the SHR is only one model of human essential hypertension and does not precisely manifest all of the possible causes of this disease. 5Similar studies using other models may produce different results.

One possibility that was not addressed in our previous studies was the lack of a complete local sympathetic denervation of the vessel preparation by the 6-hydroxydopamine pretreatment. If so, subsequent administration of isoflurane (either locally or systemically) potentially may not produce neurally independent effects, but rather merely a completion of a partial denervation. To address this issue, in preliminary studies we compared the effect of local denervation on VSM E^min normotensive Sprague-Dawley animals using 300 mg/ml 6-hydroxydopamine versus  1 mg/ml tetrodotoxin. We observed no differences in the isoflurane-mediated hyperpolarization after inhibition of sympathetic neural input with either of these two agents (data not shown). Tetrodotoxin is a well-established antagonist of sodium channels in nerve and an effective agent for removal of sympathetic control of VSM tone. 42Thus, we conclude that the sympathetic neural denervation by 6-hydroxydopamine was effective and complete.

Another question in the present study is the effectiveness of the inhibitors of the potassium channels and of the analogs and inhibitors of the second messenger pathways. The concentrations of the K⁺channel inhibitors (iberiotoxin and glybenclamide) used were 10 times greater than those reported to produce effective half-block of K^ATPand K^Cachannels, respectively. Because of the selectivity of these K⁺channel inhibitors, crossover effects to other channels is unlikely. 9It is unclear why either iberiotoxin or glybenclamide alone were capable of virtually eliminating the isoflurane-mediated hyperpolarization in the present study. It might be expected that either inhibitor alone would produce only partial attenuation of the response, and complete blockade would require both agents simultaneously. However, the current data (as well as previous results) 11do not support this and instead suggest a possible interaction between K^Caand K^ATPchannels (assuming that each inhibitor is specific as reported for the concentrations used). 9The existence and possible mechanisms of such a potential interaction remain to be clarified.

Finally, in the present study we did not control for the possible isoflurane-mediated enhancement of at least two other endogenous vasodilators in the SHR and WKY vessels (prostacyclin and endothelium-derived hyperpolarizing factor [EDHF]). Evidence exists to indicate that prostacyclin vasodilates by enhancing potassium channel activity through the cAMP-PKA pathway. 43However, very little is currently known about the effects of anesthetics on this mechanism. Similarly, the identity and mechanisms of action of EDHF have not been elucidated, although a variety of substances have now been identified as having potential EDHF activity. 44The effect of anesthetics on these EDHF-related mechanisms is even less clear. Although not studied in detail, initial evidence suggests that any effect that anesthetics may have on release of EDHF appears to be inhibitory. 44,45Such an effect would not explain the isoflurane-mediated hyperpolarization observed in the present study. Further studies are needed to establish the relative importance of both of these substances (prostacyclin and EDHF) in mediating anesthetic-induced VSM hyperpolarization and vasodilatation.

In summary, the results of the present study indicate that 1.0 minimum alveolar concentration–superfused isoflurane hyperpolarizes VSM and attenuates control of VSM tone in mesenteric small resistance-regulating arteries and capacitance-regulating veins equally in SHR and WKY by enhancing the activity of membrane-bound K^Caand K^ATPchannels. The elimination of isoflurane-induced VSM hyperpolarization by inhibition of cAMP synthesis or PKA activation suggests that at least a portion of this anesthetic action results from an enhanced activity of VSM membrane-bound and intracellular components of the cAMP-mediated second messenger system. Isoflurane-induced VSM hyperpolarization appears to be independent of endothelial-derived NO and the related cGMP pathway. The lack of a difference between SHR and WKY suggests that these neurally independent effects of isoflurane (and presumably other volatile anesthetics) on control of the VSM do not account for the hemodynamic instability characteristic of anesthetic administration in hypertensive subjects. Based on previous data, it appears that such differences are the result of alterations in the level of sympathetic vascular control.