Volatile anesthetics have been shown to have vasodilating or vasoconstricting actions in vitro that may contribute to their cardiovascular effects in vivo. However, the precise mechanisms of these actions in vitro have not been fully elucidated. Moreover, there are no data regarding the mechanisms of volatile anesthetic action on small resistance arteries, which play a critical role in the regulation of blood pressure and blood flow.


With the use of isometric tension recording methods, volatile anesthetic actions were studied in intact and beta-escin-membrane-permeabilized smooth muscle strips from rat small mesenteric arteries. In experiments with intact muscle, the effects of-halothane (0.25-5.0%), isoflurane (0.25-5.0%), and enflurane (0.25-5.0%) were investigated on high K(+)-induced contractions at 22 degrees C and 35 degrees C. All experiments were performed on endothelium-denuded strips in the presence of 3 microM guanethidine and 0.3 microM tetrodotoxin to minimize the influence of nerve terminal activities. In experiments with membrane-permeabilized muscle, the effects of halothane (0.5-4.0%), isoflurane (0.5-4.0%), and enflurane (0.5-4.0%) on the half-maximal and maximal Ca(2+)-activated contractions were examined at 22 degrees C in the presence of 0.3 microM ionomycin to eliminate intracellular Ca2+ stores.


In the high K(+)-stimulated intact muscle, all three anesthetics generated transient contractions, which were followed by sustained vasorelaxation. The IC50 values for this vasorelaxing action of halothane, isoflurane, and enflurane were 0.47 vol% (0.27 mM), 0.66 vol% (0.32 mM), and 0.53 vol% (0.27 mM), respectively, at 22 degrees C and were 3.36 vol% (0.99 mM), 3.07 vol% (0.69 mM), and 3.19 vol% (0.95 mM), respectively, at 35 degrees C. Ryanodine (10 microM) eliminated the anesthetic-induced contractions but had no significant effect on the anesthetic-induced vasorelaxation in the presence of high K+. In addition, no significant differences were observed in the dose dependence of the direct vasodilating action among these anesthetics with or without ryanodine at either the low or the high temperature. However, significant differences were observed in the vasoconstricting actions among the anesthetics, and the order of potency was halothane > enflurane > isoflurane. The Ca(2+)-tension relation in the membrane-permeabilized muscle yielded a half-maximal effective Ca2+ concentration (EC50) of 2.02 microM. Halothane modestly but significantly inhibited 3 microM (approximately the EC50) and 30 microM (maximal) Ca(2+)-induced contractions. Enflurane slightly but significantly inhibited 3 microM but not 30 microM Ca2+ contractions. Isoflurane did not significantly inhibit either 3 microM or 30 microM Ca2+ contractions.


Halothane, isoflurane, and enflurane have both vasoconstricting and vasodilating actions on isolated small splanchnic resistance arteries. The direct vasoconstricting action appears to result from Ca2+ release from the ryanodine-sensitive intracellular Ca2+ store. The vasodilating action of isoflurane in the presence of high K+ appears to be attributable mainly to a decrease in intracellular Ca2+ concentration, possibly resulting from inhibition of voltage-gated Ca2+ channels. In contrast, the vasodilating actions of halothane and enflurane in the presence of high K+ appears to involve inhibition of Ca2+ activation of contractile proteins as well as a decrease in intracellular Ca2+ concentration in smooth muscle.

Key words: Anesthetics, volatile: enflurane; halothane; isoflurane. Artery: resistance. Membrane, permeabilization: beta-escin. Muscle, smooth: vascular.

VOLATILE anesthetics are known to affect cardiovascular stability by causing changes in vascular reactivity, cardiac function, and autonomic reflexes that detect alterations and initiate adjustments in the cardiovascular system. [1–3] Their overall effect is a decrease in mean arterial pressure caused by peripheral vasodilation, myocardial depression, and decreased sympathetic nervous system activity. [1–3] Vasodilating or vasoconstricting actions of volatile anesthetics have been demonstrated in isolated aorta and coronary arteries, [4–10] suggesting that direct vascular action may contribute to the cardiovascular effects of volatile anesthetics in vivo. In addition, recent studies have demonstrated that volatile anesthetics have vasodilating or vasoconstricting actions on isolated small resistance arteries at clinically relevant concentrations, [11–13] more strongly suggesting that direct actions on vascular tissue do significantly contribute to changes in peripheral vascular resistance in vivo.

The precise mechanism(s) of the in vitro vasodilating or vasoconstricting actions of volatile anesthetics have not been fully clarified. The involvement of endothelium in the vascular actions of volatile anesthetics have been investigated in several vascular preparations, [4–8,11,12,14–19] and the endothelium dependence of the vasodilating or vasoconstricting actions of halothane, enflurane or isoflurane have been suggested in some preparations including canine coronary and mesenteric arteries and rat aorta. [4,6,14] In addition, several groups of investigators have demonstrated that halothane, isoflurane, enflurane, and sevoflurane can inhibit endothelium-mediated relaxations in a variety of vascular preparations. [12,16–18] However, the lack of effects of endothelial denudation or the endothelium-derived relaxing factor pathway inhibitors on vasodilating actions of isoflurane or halothane has also been documented in some vascular preparations including rat aorta, rabbit basilar and canine cerebral arteries. [7,15,19] Thus, the role of endothelium in the vascular responses to volatile anesthetics appears to be still controversial.

Vasodilating or vasoconstricting actions of volatile anesthetics have also been documented in several endothelium-denuded vascular preparations. [5–8,10,14,15,19,20] Several previous studies that have addressed the mechanisms of the direct vascular smooth muscle actions of volatile anesthetics have suggested that these agents may affect both activation of contractile proteins and intracellular Calcium sup 2+ mobilization. [5,14,20–24] Su and Zhang demonstrated that halothane slightly depressed maximal Calcium2+-activated tension in membrane-permeabilized rabbit aortic tissue, [5] suggesting that inhibition of contractile proteins may contribute to the vasodilating action of halothane. Buljubasic and coworkers, using the patch-clamp method, have demonstrated that halothane (0.75% and 1.5%) and isoflurane (2.6%) inhibit long-lasting Calcium2+ currents in canine coronary and cerebral arteries, suggesting that volatile anesthetics can decrease intracellular Calcium2+ concentration ([Calcium2+]i) by inhibiting voltage-gated Calcium2+ influx. [22,23] More recently, Tsuchida et al. have demonstrated that halothane (greater or equal to 1%) and isoflurane (greater or equal to 2%) decreased both agonist- and high Potassium sup +-induced increases in [Calcium2+] sub t in rat aortic tissues, and, by using simultaneous measurements of tension and [Calcium2+], also implied that higher concentrations of the anesthetics may inhibit Calcium2+-activation of contractile proteins. [20].

Small splanchnic resistance arteries are known to play a critical role in the regulation of systemic blood pressure and splanchnic blood flow, [25,26] and may be an important site of volatile anesthetic action. [27] In addition, maintenance of splanchnic blood flow during the perioperative period may be important in preventing splanchnic ischemia, which can significantly affect the outcome of critically ill surgical patients by allowing bacterial translocation across the gut wall. [28–30] Although a few previous studies have investigated volatile anesthetic action in isolated large mesenteric arteries, [14,31] little information is available regarding mechanisms of volatile anesthetic action in small splanchnic resistance arteries. Previous studies in vivo and in vitro have demonstrated significant differences in the vascular actions of volatile anesthetics between large conductance and smaller arteries, [9,27,32–34] indicating the need to further investigate volatile anesthetic action in the small splanchnic resistance arteries.

In this study, we investigated the mechanisms of volatile anesthetic action on small splanchnic resistance arteries. Using isometric-tension recording methods, we first investigated the vasoconstricting and vasodilating actions of halothane, isoflurane, and enflurane in endothelium-denuded tissues precontracted with high Potassium sup +. We then investigated the mechanisms of these effects in ryanodine-treated intact smooth muscle and beta-escin-permeabilized smooth muscle. This report represents the first description of volatile anesthetic action on Calcium2+-activated contractions in membrane-permeabilized resistance arteries.

Tissue Preparation for Tension Measurement

After receiving institutional approval for this animal study, Sprague-Dawley rats (200–250 g) were anesthetized with halothane in Oxygen2after preoxygenation with 100% Oxygen2for 3–5 min. After obtaining an optimal anesthetic level, the mesenteric tissue was removed and immediately placed in a dissecting chamber filled with 4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES)-buffered physiologic salt solution. The mesenteric artery was then rapidly excised. The distal portion of the third or fourth order branch (0.15–0.20 mm in diameter) of the mesenteric artery was used. Under a binocular microscope, the surrounding connective and fat tissues were removed, the vessel was cut open lengthwise, and the endothelium was removed by gently rubbing the intimal surface with the round surface of a small pin as has previously been described. [35,36] From this “vascular sheet,” a transverse strip (150–200 micro meter in width, 250–400 micro meter in length) was prepared. Both ends of the strip were then tied with two thin silk threads, and, for the isometric tension measurement, the strip was fixed between one end of a chamber (0.9 ml capacity) and a L-shaped stainless rod connected to the strain-gauge transducer (UL-2, Shinko, Tokyo) as previously reported. [12,35,36] After a 30-min equilibration, the strip was stretched to approximately 1.1 times its resting length (without tension) to obtain a maximal contractile response to high Potassium sup +. The functional removal of the endothelium was confirmed by disappearance of endothelium-dependent relaxation by acetylcholine (10 micro Meter). The solution was changed by perfusing it rapidly from one end while aspirating it simultaneously from the other end. Experiments with intact muscle were conducted both at room temperature (22 degrees Celsius) and at 35 degrees Celsius, whereas the experiments with membrane-permeabilized muscle were performed at room temperature (22 degrees Celsius) to prevent deterioration of the permeabilized strips.

Solutions and Drugs

The millimolar ionic concentrations of the HEPES-buffered physiologic salt solution (PSS) for the experiments with intact muscle were as follows: NaCl 138, KCl 5.0, MgCl21.2, CaCl21.5, HEPES 10, glucose 10. The pH was adjusted with NaOH to 7.35 at either 22 degrees Celsius or 35 degrees Celsius. The high-Potassium sup + solutions were prepared by replacing NaCl with KCl isoosmotically.

The composition of the relaxing and activating solutions used in the experiments with membrane-permeabilized muscle are listed in Table 1, which also shows calculated values of free ion concentrations and total ionic strength in each solution. The pH was adjusted with KOH to 7.00 at 22 degrees Celsius. Free ion concentrations and the ionic strengths of the relaxing and activating solutions were calculated by solving multiequilibrium equations with a hydrogen ion activity coefficient 0.75 and association constants between ions (Appendix) as previously used. [37,38] Numerical solution of a set of multiequilibrium equations was achieved by a successive approximation method by computer (Powerbook 170, Apple, Cupertino, CA).

Guanethidine, tetrodotoxin, HEPES, beta-escin, and ionomycin were obtained from Sigma Chemical (St. Louis, MO). Ethyleneglycol-bis-(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), piperazine-1,4-bis-(2-ethanesulfonic acid)(PIPES-K2), and methanesulfonic acid were obtained from Fluka Chemie AG (Buchs, Switzerland). Ryanodine was purchased from Agri Systems International (Wind Gap, PA). Isoflurane and enflurane were obtained from Abbott Laboratories (North Chicago, IL). and halothane was obtained from Ayerst Laboratories (Philadelphia, PA). All other reagents were of the highest grade commercially available.

Experimental Design

Intact Muscle.

The effects of halothane, isoflurane, and enflurane on maximum high Potassium sup +-induced contractions were studied in endothelium-denuded strips at room temperature (22 degrees Celsius) and at 35 degrees Celsius. To suppress peripheral nerve activities, 3 micro Meter guanethidine and 0.3 micro Meter tetrodotoxin were present in the solutions throughout the experiments. In the experiments at room temperature, the volatile anesthetics were cumulatively (0.25–5.0%) applied to the strip after contractile response to 143 mM Potassium sup + had reached a plateau. Because the 40 mM Potassium sup +-induced contractions were often not well maintained for a long period at 35 degrees Celsius, a single concentration (0.125–5.0%) of each anesthetic was applied to the strip after contractile response to 40 mM Potassium sup + reached a plateau. The anesthetics were not cumulatively applied in these experiments (at 35 degrees Celsius).

Because the above experiments revealed that the volatile anesthetics had a vasoconstricting action that resulted from Calcium2+ release from intracellular Calcium2+ stores as previously demonstrated, [11–13] we also examined the effects of volatile anesthetics on ryanodine-treated strips at both temperatures. In these experiments, ryanodine (10 micro Meter) was first applied to the strips with caffeine (20 mM) for approximately 1 min to deplete the Calcium2+ stores, and the strips were then treated with ryanodine (10 micro Meter)-containing PSS for 25 min. All the experiments with ryanodine were started after 20 mM caffeine-induced contractions were completely abolished.

Membrane-permeabilized Muscle.

After measuring steady contractions induced by high Potassium sup +, the strips were permeabilized with 50 micro Meter beta-escin for 22–24 min at room temperature in relaxing solution. [39–41] Low concentrations of Calcium2+ ionophores such as ionomycin or A23187 have been used to achieve nonselective and functional depletion of the intracellular Calcium2+ store. [35,40–43] In fact, in this artery, 0.3 micro Meter ionomycin was effective in abolishing caffeine (20 mM)- and phenylephrine (10 micro Meter)-induced contractions in Calcium2+-free (2mM EGTA) solutions in intact strips, and in abolishing the caffeine contractions in beta-escin-permeabilized strips. Therefore, all of the experiments with membrane-permeabilized muscle were performed in the presence of 0.3 micro Meter ionomycin to deplete the intracellular Calcium2+ stores.

In the first set of experiments with membrane-permeabilized muscle, we determined the pCa-tension relation by applying various concentrations of Calcium2+ containing activating solutions in a cumulative manner. In another set of the experiments, we examined the effects of halothane, isoflurane, and enflurane at concentrations of 0.5–4.0% on the half-maximal and maximal Calcium2+-activated contractions. The volatile anesthetics were applied to the strips in a cumulative manner after the Calcium2+-activated contractions reached a plateau (5 min after application of Calcium2+); each concentration of each anesthetic was applied for a period of 2 min.

Volatile Anesthetics: Delivery and Analysis

The volatile anesthetics were delivered by agent-specific vaporizers in line with the air gas aerating the solutions in experiments with intact and with membrane-permeabilized muscle. Each solution was equilibrated with the anesthetics for at least 10 min before introduction to the chamber, which was covered with thin glass to prevent the equilibration gas from escaping into the atmosphere. The anesthetic concentrations in the solutions were determined by gas chromatography (Figure 1).

Although the actual concentrations of volatile anesthetics in the PSS were not measured in all experiments, the relation between actual concentrations of volatile anesthetics in PSS and anesthetic concentrations (volume percentage) in the gas mixture should be theoretically linear. The anesthetic concentrations on the x-axis are displayed as volume percentage for the volatile anesthetic concentration-response relations.

Calculation and Statistical Analysis

All results were expressed as the means plus/minus standard error of the mean (SEM). The n denotes the number of preparations, and the number of animals is also noted.

The volatile anesthetic-induced relaxation in experiments with intact muscle at both temperatures was expressed as a percent change from the amplitude of high Potassium sup +-induced contraction before application of the anesthetics, and the amplitude of volatile anesthetic-induced vasoconstriction at 35 degrees Celsius was expressed relative to the amplitude of the 40 mM Potassium sup +-induced phasic contraction (100%). The concentration-dependence of the volatile anesthetic-induced vasoconstriction in experiments with intact muscle at 22 degrees Celsius was not assessed, because the amount of Calcium2+ in the intracellular Calcium2+ store was probably not constant during “cumulative” application of volatile anesthetics in these experiments.

In experiments with beta-escin-permeabilized muscle, the amplitude of the Calcium2+-activated contraction at various time points in the presence and absence of volatile anesthetics was normalized to the amplitude of Calcium2+ contraction 5 min after application of the activating solution. The effects of volatile anesthetics were then assessed by comparing the data in the presence of volatile anesthetics with the time control data.

The concentration-response data for volatile anesthetic effects or Calcium2+-activated contractions were fitted according to a four parameter logistic model described by De Lean et al. [45] The EC50(50% effective concentration) or IC50(50% inhibitory concentration) values were derived from the least-squares fit using the above equation.

Statistical analysis was made by an analysis of variance (one- or two-factor), Scheffe's F test, and Student's t test (unpaired), where appropriate. A P level of < 0.05 was considered significant.

Experiments with Intact Muscle

In intact muscle not treated with ryanodine, both halothane and enflurane generated transient contractions superimposed on the high Potassium sup +-induced contractions at 22 degrees Celsius and at 35 degrees Celsius (Figure 2and Figure 3). Isoflurane generated similar transient contractions at 22 degrees Celsius, but not at 35 degrees Celsius (Figure 2and Figure 3). The transient contractions caused by volatile anesthetics at both temperatures were followed by significant vasorelaxation (Figure 2and Figure 3). Ryanodine (10 micro Meter) abolished the volatile anesthetic-induced contractions in the presence of high Potassium sup + at both temperatures (Figure 4, Figure 5, Figure 6). Significant differences were observed in the vasoconstricting effects at 35 degrees Celsius among the anesthetics, and the order of potency was halothane > enflurane > isoflurane (Figure 6). Ryanodine treatment did not significantly affect the phasic or tonic components of high Potassium sup +-induced contractions at either temperature (P > 0.05).

The concentration-response curves for volatile anesthetic-induced relaxation under various conditions (in the absence or presence of ryanodine; at either low or high temperature) are shown in Figure 7. Two factor (concentration [volume percentage], anesthetic) analysis of variance revealed no significant difference in the vasorelaxation among the three volatile anesthetics under each condition. In addition, two-factor (concentration and treatment [with or without ryanodine]) analysis of variance showed no significant difference in the vasorelaxation produced by each anesthetic between the ryanodine-treated and untreated groups. The IC50values for vasodilating action of halothane, isoflurane, and enflurane at 22 degrees Celsius and 35 degrees Celsius in ryanodine-treated and untreated strips are summarized in Table 2.

Experiments with beta-Escin-Permeabilized Muscle

Application of various concentrations (0.1–100 micro Meter) of Calcium2+ to the beta-escin-treated strips generated the dose-dependent contractions, and the EC50value for the Calcium2+-tension relation derived from the least-squares fit with the logistic equation [4,5] was 2.02 micro Meter (Figure 8) The maximal Calcium2+ activated contraction was 1.58 plus/minus 0.25 times that of 143 mM Potassium sup +-induced maximal (phasic) contractions before permeabilization (n = 4, four animals), suggesting satisfactory membrane permeabilization.

The effects of volatile anesthetics on 3 micro Meter ([nearly equal] EC50) and 30 micro Meter (maximal) Calcium2+-activated contractions were then examined. The 3 micro Meter Calcium2+ contraction was well maintained, whereas the 30 micro Meter Calcium2+ contraction slightly decayed (time control data, Figure 9). Comparison between the volatile anesthetic data and the time control data showed that halothane modestly, but significantly inhibited both 3 micro Meter ([nearly equal] EC50) and 30 micro Meter (maximum) Calcium2+-induced contractions (Figure 9). Enflurane slightly but significantly inhibited 3 micro Meter, but not 30 micro Meter Calcium2+ contractions. Isoflurane did not significantly inhibit either 3 micro Meter or 30 micro Meter Calcium2+-induced contractions (Figure 9). All these anesthetics did not produce any significant vasoconstriction in membrane-permeabilized muscle strips that were pretreated with ionomycin to eliminate the intracellular Calcium2+ stores.

The current study demonstrates that halothane, isoflurane, and enflurane all have a relaxing action on Potassium sup +-constricted small splanchnic resistance arteries. This action observed at clinically relevant anesthetic concentrations appears to be independent of endothelial or peripheral nerve activities, although this action may be influenced by anesthetic effects on endothelium or peripheral nerves in vivo. It is well established that Potassium sup +-induced constrictions are caused by activation of voltage-gated Calcium2+ channels and that these channels play a significant role in maintaining vascular tone in vivo. Thus, the effective concentrations of volatile anesthetics that inhibited the high Potassium sup +-induced constrictions at physiologic temperature in these resistance arteries suggest that this “direct” vasodilating action of volatile anesthetics contributes to the hypotensive effects of these agents in vivo. No significant differences were observed in the direct vasodilating actions among the three anesthetics under any condition (with or without ryanodine; at either 22 degrees Celsius or 35 degrees Celsius). However, because the minimum alveolar concentration of halothane ([nearly equal] 0.90%) is known to be lower than that of isoflurane (1.38%) or enflurane (1.40%) in the rat, isoflurane or enflurane would be a more potent vasodilator than halothane in the high Potassium sup +-stimulated splanchnic resistance artery at equivalent anesthetic concentrations.

Halothane and enflurane slightly but significantly inhibited the Calcium2+-activated contractions in beta-escin-permeabilized muscle, suggesting that the observed relaxing actions of halothane and enflurane in the presence of high Potassium sup + are attributable in part to inhibition of Calcium2+ activation of contractile proteins. However, this effect was small relative to the amount of vasodilation observed in the high Potassium sup +-constricted intact vessels, suggesting that the vasodilation is caused in large part by depression of increases in [Calcium2+]i. At most, only 20% of the relaxing effect of halothane could be attributed to a direct inhibition of Calcium2+ action on the contractile protein cascade; the rest of the effect is thus presumably caused by decreases in the rise of [Calcium2+]i. The complete lack of effect of isoflurane on the Calcium2+-activated contractions implies that its relaxing action is attributable almost completely to depression of the rise in [Calcium sup 2+]i. The small amount of inhibition of Calcium2+-activated contractions by halothane in these small mesenteric arteries is consistent with the slight inhibition of maximal Calcium2+ activated tension in membrane-permeabilized rabbit aortic tissue. [5] The proposal that volatile anesthetics affect vascular tone by decreasing [Calcium2+]iis consistent with recent investigations using fura-2-loaded vascular tissue in which halothane and isoflurane were shown to decrease both agonist- and high Potassium sup +-induced increases in [Calcium2+]i. [20] Previous studies have also demonstrated that halothane and isoflurane directly inhibit depolarization-activated long-lasting Calcium2+ currents in vascular smooth muscle cells, [22,23] suggesting that the [Calcium sup 2+]i-reducing effects of volatile anesthetics result at least in part from inhibition of voltage-gated Calcium2+ influx. Finally, although still controversial, the increases in vascular smooth muscle cyclic guanosine monophosphate or cyclic adenosine monophosphate concentrations by anesthetics reported in some preparations, [21,24] may play a role in altering Calcium2+ sensitivity or affecting Calcium2+ mobilization by stimulating Calcium2+ extrusion, Calcium2+ uptake into the store or inhibiting voltage-gated Calcium sup 2+ influx.

Although we did not directly compare the data for the vasodilating actions of volatile anesthetics between 22 degrees Celsius and 35 degrees Celsius because of the differences in experimental protocols (cumulative vs. single-dose application and 143 vs. 40 mM Potassium sup +), the IC50values (volume percentage) for the volatile anesthetic-induced inhibitions of high Potassium sup +-induced contraction were approximately three to six times greater at 35 degrees Celsius than at 22 degrees Celsius. Assuming that the binding of an anesthetic to its active sites is determined by solution concentration (and not partial pressure), [46] then some of this difference can be explained by the approximately 1.5- to 2-fold increase in the solubility of these anesthetics at 22 degrees Celsius compared with 35 degrees Celsius (Figure 1). However, with or without this correction, some of the differences in potency must be related to other temperature-sensitive factors such as the voltage-gated Calcium2+ channels or enzyme activities involved in smooth muscle contraction.

In addition to the well-known Calcium2+-dependent excitation-contraction coupling pathway, recent studies also indicate the existence of either a Calcium2+-independent excitation-contraction coupling pathway or some mechanism(s) that increases Calcium2+ sensitivity of contractile proteins in smooth muscle, particularly in response to receptor agonists. [47–49] Although we did not demonstrate any substantial effect of volatile anesthetics on Calcium2+ activation of contractile proteins, the effect(s) of the volatile anesthetics on the Calcium2+-independent pathway can not be ruled out by the current study. Protein kinase C activation has been proposed to be involved in the Calcium2+-independent excitation-contraction coupling or agonist-induced increase in Calcium2+ sensitivity, [39,12,17–19] and recent studies have suggested that volatile anesthetics may inhibit the protein kinase C-mediated Calcium2+-independent pathway in canine tracheal smooth muscle. [50] However, Ozhan et al. have recently reported that halothane and isoflurane lack effects on phorbol ester-induced protein kinase C-dependent contractions in isolated coronary arterial tissue. [10] Because vascular tone in vivo is largely dependent on release of norepinephrine, [26] an agonist well known to increase protein kinase C activity by alpha1-adrenergic receptor activation, the effects of volatile anesthetics on the Calcium2+-independent contraction mechanisms should be further investigated to clarify the volatile anesthetic vascular action in vivo.

All three anesthetics also have a significant transient vasoconstricting action that does not require the presence of endothelium or peripheral nerve activity and also occurs at clinically relevant (i.e., anesthetic) concentrations. The vasoconstricting action was independent of the vasodilating action as evidenced by the finding that the relative potency of the anesthetics as vasoconstrictors was unrelated to their vasodilatory potency and the finding that ryanodine abolished the vasoconstricting action with little effect on volatile anesthetic-induced vasodilation. In the high Potassium sup +-constricted vessels, the observed vasoconstricting action was only transient, however, and was immediately followed by vasodilation as we have previously reported for halothane in the saint vessels [13] and for enflurane in rabbit small mesenteric arteries precontracted with adrenergic agonists. [11] Thus, although vasoconstriction did occur with clinically relevant concentrations of halothane and enflurane, the importance of this effect in altering splanchnic blood flow during anesthesia is uncertain. Of note, the vasoconstricting effect of isoflurane, which has not previously been reported, was observed only at 22 degrees Celsius and not at 35 degrees Celsius, probably because isoflurane is considerably less potent at producing vasoconstriction. Because the solubility of isoflurane (as well as the other anesthetics) was lower at 35 degrees Celsius than at 22 degrees Celsius, the apparent threshold for the vasoconstricting action of isoflurane observed at 22 degrees Celsius (2%[nearly equal] 1 mM) was reached only with the maximum concentration studied at 35 degrees Celsius (i.e., 5% isoflurane;Figure 1). Therefore, the apparent inability of isoflurane to produce vasoconstriction at 35 degrees Celsius may have resulted from the different concentrations achieved in the buffer at 22 degrees Celsius compared with those at 35 degrees Celsius. In addition, because the vasoconstricting action appears to involve Calcium2+ release from intracellular stores (see below), the sequential-addition protocol that was used at 22 degrees Celsius did not permit an accurate analysis of the concentration-response relation at that temperature (22 degrees Celsius). In conclusion, our data suggest that the vasoconstricting action might be a common characteristic of halogenated volatile anesthetics.

The ability of ryanodine to abolish the vasoconstricting action of volatile anesthetics in intact muscle, and the inability of volatile anesthetics to produce any significant contractions in ionomycin-treated beta-escin-permeabilized muscle strongly suggest that the volatile anesthetics cause vasoconstriction by releasing Calcium2+ from ryanodine-sensitive intracellular Calcium2+ stores, and not by inducing Calcium2+ influx or affecting Calcium2+ sensitivity. This is consistent with our previous reports in which the halothane- or enflurane-induced vasoconstricting effects were independent of extracellular Calcium2+ concentration and were blocked by ryanodine. [12,13] Similar ryanodine-sensitive vasoconstricting effects of halothane and enflurane were recently reported in canine mesenteric artery. However, the authors suggested that, in addition to Calcium2+ release from intracellular stores in smooth muscle, vasoconstriction in those vessels may also involve a change in Calcium2+ sensitivity or an endothelium effect. [14] Although our data indicate that an increase in Calcium2+ sensitivity is not involved, we did not examine their endothelium dependence. A ryanodine-sensitive intracellular Calcium2+ store has recently been documented in endothelial cells, [51,52] which may play an important role in the regulation of endothelial function. [53,54] Therefore, it might be possible that volatile anesthetics affect endothelial function by causing a change in the integrity of the endothelial ryanodine-sensitive Calcium2+ store, and this endothelial action may modify the volatile anesthetic action on vascular smooth muscle.

The observed order of potency for the Calcium2+ releasing effects of volatile anesthetics from the ryanodine-sensitive store (halothane > enflurane > isoflurane) is consistent with previous studies in which halothane and enflurane, but not isoflurane, open Calcium2+ release channels of cardiac sarcoplasmic reticulum [55,56] and stimulate [sup 3 Hydrogen] ryanodine binding to cardiac sarcoplasmic reticulum. [57] However, only isoflurane, and not halothane or enflurane, stimulated [sup 3 Hydrogen] ryanodine binding to skeletal ryanodine receptor. [57] The cardiac ryanodine receptor gene (RYR2) or skeletal ryanodine receptor gene (RYR1) is known to be different from vascular smooth muscle ryanodine receptor gene (RYR3), which is also expressed in brain or lung epithelial cells. [58] Therefore, the RYR3 presumed present in our tissue appears to resemble RYR2 more than it does RYR1.

In conclusion, halothane, isoflurane, and enflurane have direct vasoconstricting and vasodilating actions that are independent of endothelial or nerve terminal activities in rat small mesenteric arteries. Although the clinical relevance of the transient vasoconstricting action is uncertain, the vasodilating action at clinically relevant concentrations suggests that it may contribute significantly to the vascular effects of volatile anesthetics in vivo. Compared with their vasodilating actions in high Potassium sup +-constricted vessels, the relatively small effects of these anesthetics on Calcium2+ induced contractions in beta-escin-permeabilized muscle suggest that these anesthetics have little or no effect on Calcium2+ activation of contractile proteins. Thus, the direct vasodilation in high Potassium sup +-treated vascular smooth muscle must be largely explained by effects on intracellular Calcium2+ mobilization. The “direct” vasoconstricting action of volatile anesthetics depends on the Calcium2+ release from the ryanodine-sensitive Calcium2+ store and does not appear related to Calcium2+ influx or any change in the Calcium2+ sensitivity of contractile proteins. This is the first investigation to demonstrate volatile anesthetic action on Calcium2+ activation of contractile proteins in the resistance arteries that are important in the regulation of systemic blood pressure and splanchnic blood flow.

The authors are grateful to Professor Alex S. Evers (Department of Anesthesiology, Washington University) and Professor Shosuke Takahashi (Department of Anesthesiology, Kyushu University) for their continuous encouragement and helpful comments; to Dr. Masamitsu Hatakenaka (Department of Clinical Pharmacology, Kyushu University, Dr. Takeo Itoh (Department of Pharmacology, Kyushu University), and Dr. Junji Nishimura (Division of Molecular Cardiology. Research Institute of Angiocardiology, Kyushu University) for their critical advice on the preparation of solutions for experiments with membrane-permeabilized muscle, and to Sandra Leal and Gail Maher (Department of Washington University) for their kind help with the gas chromatography experiments.

The binding constants used to calculate the compositions of the solutions for the experiments with membrane-permeabilized muscle are as follows, where ATP = adenosine triphosphate and CrP = creatinine phosphate; other abbreviations are defined in main text. [HEGTA3-]/[Hydrogen sup +][EGTA4-]= 2.88 x 109, [H2EGTA 2-]/[Hydrogen sup +][H3EGTA3-]= 7.08 x 108, [H3EGTA sup -]/[Hydrogen sup +][H2EGTA2-]= 4.79 x 102, [H sub 4 EGTA]/[Hydrogen sup +][H3EGTA sup -]= 1.00 x 102;[HATP3-]/[Hydrogen sup +][ATP4-]= 8.91 x 106, [H2ATP2]/[Hydrogen sup +][HATP3-]= 1.12 x 104, [H3ATP sup -]/[Hydrogen sup +][H2ATP2-]= 1.00 x 101, [H4ATP]/[Hydrogen sup +][H3ATP sup -]= 1.00 x 101;[CaEGTA2-]/[Calcium2+][EGTA4-]= 5.01 x 1010, [CaHEGTA2-]/[Calcium2+][HEGTA3-]= 2.14 x 105;[CaATP2-]/[Calcium2+][ATP4-]= 9.55 x 103, [CaHATP sup -]/[Calcium][HATP3-]= 6.31 x 101;[CaCrP]/[Calcium2+][CrP2]= 1.41 x 101;[MgEGTA2-]/[Magnesium sup +][EGTA sup 4-]= 1.62 x 105, [MgHEGTA sup -]/[Magnesium2+][HEGTA3-]= 2.34 x 103;[MgATP2-]/[Magnesium2+][ATP4-]= 2.11 x 104, [MgHATP sup -]/[Magnesium2+][HATP3-]= 5.50 x 102;[MgCrP]/[Magnesium2+][CrP2-]= 2.0 x 101;[HCrP]/[Hydrogen sup +][CrP2]= 3.80 x 104, [H2CrP]/[Hydrogen sup +][CrP2-]= 5.01 x 102;[KATP3-]/[Potassium sup +][ATP4-]= 7.94, [K2ATP2-]/[Potassium sup +][KATP3-]= 0.6, [KHATP2-]/[Potassium sup +][HATP3-]= 0.6;[NaATP3-]/[Sodium sup +][ATP4-]= 8.8, [Na2ATP]/[Sodium sup +][NaATP3-]= 8.5, [NaHATP2-]/[Sodium sup +][HATP3-]= 5.0;[H PIPES sup -]/[Hydrogen sup +][PIPES2]= 6.31 x 106.

Pavlin EG, Su JY: Cardiopulmonary pharmacology, Anesthesia. Edited by Miller RD. New York, Churchill Livingstone. 1990, pp. 105-134.
Stevens WC, Kingston HGG: Inhalation anesthesia, Clinical Anesthesia. Edited by Barash PG, Cullen BF, Stoelting RK. Philadelphia, JB Lippincott, 1992, pp. 439-465.
Marshall BE, Longnecker DE: General anesthetics. The Pharmacological Basis of Therapeutics. Edited by Gilman AG, Rall TW, Nies AS, Taylor P, New York, Pergamon Press, 1990, pp 285-310.
Blaise G, Sill JC, Nugent M, Van Dyke AR, Vanhoutte PM: Isoflurane causes endothelium-dependent inhibition of contractile responses of canine coronary arteries. ANESTHESIOLOGY 67:513-517, 1987.
Su JY, Zhang CC: Intracellular mechanisms of halothane's effect on isolated aortic strips of the rabbit. ANESTHESIOLOGY 71:409-417, 1989.
Stone DJ, Johns AR: Endothelium-dependent effects of halothane, enflurane, and isoflurane on isolated rat aortic vascular rings. ANESTHESIOLOGY 71:420-432, 1989.
Brendel JK, Johns AR: Isoflurane does not vasolidate rat thoracic aortic rings by endothelium-derived relaxing factor or other cyclic GMP-mediated mechanisms. ANESTHESIOLOGY 77:126-131, 1992.
Witzeling TM, Sill JC, Hughes JM, Blaise GA, Nugent M, Rolie K: Isoflurane and halothane attenuate coronary artery constriction evoked by serotonin in isolated porcine vessels and in intact pigs. ANESTHESIOLOGY 73:100-108, 1990.
Hatano Y, Nakamura K, Yakushiji T, Nishiwada M, Mori K: Comparison of the direct effects of halothane and isoflurane on large and small coronary arteries isolated from dog. ANESTHESIOLOGY 73:513-517, 1990.
Ozhan M, Sill JC, Atagunduz P, Martin R, Katusic ZS: Volatile anesthetics and agonist-induced contractions in porcine coronary artery smooth muscle and Calcium sup 2+ mobilization in cultured immortalized vascular smooth muscle cells. ANESTHESIOLOGY 80:1102-1113, 1994.
Akata T, Takahasi S, Kodama K, Tominaga M, Taniyama T: Effects of volatile anesthetics on rhythmic smooth muscle activity in rabbit mesenteric arteries (abstract). ANESTHESIOLOGY 77:A676, 1992.
Akata T, Nakashima M, Kenji K, Boyle WA III, Takahasi S: Effects of volatile anesthetics on acetylcholine-induced relaxation in the rabbit mesenteric resistance artery. ANESTHESIOLOGY 82:188-204, 1995.
Boyle WA III, Maher G: Endothelium-independent vasoconstricting and vasodilating actions of halothane on rat mesenteric resistance blood vessels. ANESTHESIOLOGY 82:221-235, 1995.
Kakuyama M, Hatano Y, Nakamura K, Toda H, Terasako K, Nishiwada M, Mori K: Halothane and enflurane constrict canine mesenteric arteries by releasing Calcium sup 2+ from intracellular Calcium sup 2+ stores. ANESTHESIOLOGY 80:1120-1127, 1994.
Jensen NF, Todd MM, Kramer DJ, Leonard PA, Warner DS: A comparison of the vasodilating effects of halothane and isoflurane on the isolated rabbit basilar artery with or without intact endothelium. ANESTHESIOLOGY 76:624-634, 1992.
Muldoon SM, Hart JL, Bowen KA, Freas W: Attenuation of endothelium-mediated vasodilation by halothane. ANESTHESIOLOGY 68:31-37, 1988.
Uggeri MJ, Proctor GJ, Jones RA: Halothane: enflurane, and isoflurane attenuate both receptor-and non-receptor-mediated EDRF production in rat thoracic aorta. ANESTHESIOLOGY 76:1042-1017, 1992.
Yoshida K, Okabe E: Selective impairment of endothelium-dependent relaxation by sevoflurane: Oxygen free radicals participation. ANESTHESIOLOGY 76:440-447, 1992.
Flynn NM, Buljubasic N, Bosnjak ZJ, Kampine JP: Isoflurane produces endothelium-independent relaxation in canine middle cerebral arteries. ANESTHESIOLOGY 76:161-167, 1992.
Tsuchida H, Namba H, Yamakage M, Fujita S, Notsuki E: Namiki A: Effects of halothane and isoflurane on cytosolic calcium ion concentrations and contraction in the vascular smooth muscle of the rat aorta. ANESTHESIOLOGY 78:531-540, 1993.
Sprague DH, Yang JC, Ngai SH: Effects of isoflurane and halothane on contractility and the cyclic 3′,5′-adenosine monophosphate system in rat aorta. ANESTHESIOLOGY 40:162-167, 1974.
Buljubasic N, Rusch NJ, Marijic J, Kampine JP, Bosnjak ZJ: Effects of halothane and isoflurane on calcium currents in canine coronary arterial cells. ANESTHESIOLOGY 76:990-998, 1992.
Buljubasic N, Flynn NM, Marijic J, Rusch NJ, Kampine JP, Bosnjak ZJ: Effects of isoflurane on Potassium sup + and Calcium sup 2+ conductance in isolated smooth muscle cells of canine cerebral arteries. Anesth Analg 75:590-596, 1992.
Nakamura K, Hatano Y, Toda H, Nishiwada M, Back W-Y, Mori K: Halothane-induced relaxation of vascular smooth muscle. A possible contribution of increased cyclic GMP formation. Jpn J Pharmacol 55:165-168, 1991.
Stephenson RB: The splanchnic circulation. Textbook of Physiology. Edited by Patton HD. Fuchs AH, Hille B, Scher AM, Steiner R. Philadelphia, WH Saunders, 1989, pp 911-923.
Guytom AC: Textbook of Medical Physiology. 7th edition. Philadelphia. WB Saunders, 1986, pp 220-243.
Longnecker D, Harris P: Microcirculatory actions of general anesthetics. Fed Proc 39:1580-83, 1980.
Gianotti L, Alexander JW, Fukushima R, Childress CP: Translocation of candida albicans is related to the blood flow of individual intestinal villi. Care Shock 40:250-257, 1993.
Haglund U: Systemic mediators released from the gut in critical illness. Crit Care Med 21 (suppl):S15-S18, 1993.
Landow L: Splanchnie lactate production in cardiac surgery patients. Crit Care Med 21 (suppl):S81-S91, 1993.
Kobayashi Y, Yoshida K, Noguchi M, Wakasugi Y, Ito H, Okabe E: Effect of enflurane on contractile reactivity in isolated canine mesenteric arteries and veins. Anesth Analg 70:530-536, 1990.
Sill JC, Bove AA, Nugent M, Blaise GA, Dewey JD, Grabau C: Effects of isoflurane on coronary arteries and coronary arterioles in the intact dog. ANESTHESIOLOGY 67:513-517, 1987.
Bollen BA, McKlveen RE, Stevenson JA: Halothane relaxes preconstricted small and medium isolated porcine coronary artery segments more than isoflurane. Anesth Analg 75:9-17, 1992.
Conzen PF, Habazetti H, Vollmar B, Christ M, Baier H, Peter K: Coronary microcirculation during halothane, enflurane, isoflurane, and adenosine in dogs. ANESTHESIOLOGY 76:261-270, 1992.
Akata T, Yoshitake J, Nakashima M, Itoh T. Effects of protamine on vascular smooth muscle of rabbit mesenteric artery. ANESTHESIOLOGY 75:833-836, 1991.
Akata T, Kodama K, Takahashi S: Effects of heparin on the vasodilator action of protamine in the rabbit mesenteric artery. Br J Pharmacol 109:1247-1253, 1993.
Fabiato A, Fabiato A: Calculator programs for computing the composition of the solutions containing multiple metals and ligands used for experiments in skinned muscle cells. J Physiol Paris 75:463-505, 1979.
Lino M: Tension responses of chemically skinned fibre bundles of the guinea-pig taenia cacci under varied ionic environments. J Physiol (Lond) 320:49-467, 1981.
Kobayashi S, Kitazawa T, Somlyo AV, Somlyo AP: Cytosolic heparin inhibits muscarinic and alpha-adrenergic Calcium sup 2+ release in smooth muscle. J Biol Chem 264:17997-18004, 1989.
Satoh S, Rensland H, Plitzer G: Ras proteins increase Calcium sup 2+, responsiveness of smooth muscle contraction FEBS Lett 324:211-215, 1993.
Itoh T, Suzuki S, Kuriyama H: Effects of pinacidil on contractile proteins in high Potassium sup + -treated intact, and in beta-escin-treated skinned smooth muscle of the rabbit mesenteric artery Br J Pharmacol 103-1697-1702, 1991.
Kitazawa T, Kobayashi S, Horiuti K, Somlyo AV, Somlyo AP: Receptor-coupled, permeabilized smooth muscle. J Biol Chem 264:5339-5342, 1989.
Itoh T, Kammura Y, Kuriyama H: A23187 increases calcium permeability of store sites more than of surface membranes in the rabbit mesenteric artery. J Physiol (Lond) 359:467-484, 1985.
F. Renzi BA, Wand BE: Partition coefficients of volatile anesthetics in Krebs solution. Anesthesiology 47:62-63, 1977.
De Lean AP, Munson PJ, Rodbard D: Simultaneous analysis of families of sigmoidal curves: Application to bioassay, radioligand assay, and physiological dose-response curves. Am J Physiol 235:E97-E102, 1978.
Franks NP, Lieb WR: Molecular and cellular mechanisms of general anaesthesia. Nature 367:607-614, 1994.
Nishimura J, Moreland S, Ahn HY, Kawase T, Moreland RS, Breemen van C: Endothelin increases myofilament Calcium sup 2+ sensitivity in alpha-toxin-permeabilized rabbit mesenteric artery. Care Res 71:951-959, 1992.
Morgan KG, Khalil RA, Suematsu E, Katsuyama H: Calcium-dependent and calcium-independent pathways of signal transduction in smooth muscle. Jpn J Pharmacol 58 (suppl 2):17-53, 1992.
Nishimura J, Khalil RA, van Breemen C: Agonists-induced vascular tone. Hypertension 13:835-844, 1989.
Yamakage M, Kohno S, Kawamata T, Namiki A: Inhibitory effects of four inhaled anesthetics on canine tracheal smooth muscle contraction and intracellular Calcium sup 2+ concentration. Anesth Analg 77:67-72, 1993.
Ziegelstein RC, Spurgeon HA, Pili R, Passaniti A, Cheng L, Corda S, Lakatta EG, Capogrossi MC: A functional ryanodine-sensitive intracellular Calcium sup 2+ store is present in vascular endothelial cells. (Circ Res 74:151-156, 1994.
Kida K, Kondoh W, Horio S, Hisayama T, Moritoki H: Effects of ryanodine and cyclopiazonic acid on ACh-induced, endothelium-dependent responses in rat thoracic aorta (abstract). Jpn J Pharmacol 61:73P, 1993.
Nilius B. Ion channels and regulation of transmembrane Calcium sup 2+ influx in endothelium, Ion Channels of Vascular Smooth Muscle Cells and Endothelial Cells. Edited by Sperelakis N, Kuriyama H. New York, Elsevier Science, 1991, pp 317-325.
Kukovetz WR, Graier WF, Groschner K: Contribution of agonist-induced hyperpolarization to Calcium sup 2+ influx and formation of EDRF in vascular endothelial cells. Jpn J Pharmacol 58(suppl 2):213-219, 1992.
Connelly TJ, Coronado R: Activation of the Calcium sup 2+ release channel of cardiac sarcoplasmic reticulum by volatile anesthetics. Anesthesiology 81:459-469, 1994.
Frazer MJ, Lynch C III: Calcium release channels of cardiac sarcoplasmic reticulum are activated by halothane, but not by isoflurane (abstract). Anesthesiology 77:A630, 1992.
Connelly TJ, Hayek R-E, Rusy BF, Coronado R: Volatile anesthetics selectively alter [sup 3 Hydrogen] ryanodine binding to skeletal and cardiac ryanodine receptors. Biochem Biophys Res Commun 186:595-600, 1992.
McPherson PS, Campbell KP: The ryanodine receptor/Calcium sup 2+ release channel. J Biol Chem 268:13765-13768, 1993.