Mechanisms of Direct Inhibitory Action of Isoflurane on Vascular Smooth

Home , Nifedipine

Anesthesiology 2003; 99:666–77 © 2003 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Mechanisms of Direct Inhibitory Action of Isoﬂurane on Vascular Smooth Muscle of Mesenteric Resistance Arteries Takashi Akata, M.D., Ph.D.,* Tomoo Kanna, M.D.,† Jun Yoshino, M.D.,‡ Shosuke Takahashi, M.D., Ph.D.§

Background: Isoflurane has been shown to directly inhibit tractile response to norepinephrine (i.e., a neurotrans- vascular reactivity. However, less information is available re- mitter that plays a central role in sympathetic garding its underlying mechanisms in systemic resistance arteries. maintenance of vascular tone in vivo) was not inhibited Methods: Endothelium-denuded smooth muscle strips were during exposure to isoflurane. In addition, contractile prepared from rat mesenteric resistance arteries. Isometric response to KCl (i.e., contractile response mediated by force and intracellular Ca2؉ concentration ([Ca2؉] ) were mea- 2ϩ i voltage-gated Ca channels [VGCCs] that play a crucial Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 sured simultaneously in the fura-2–loaded strips, whereas only ␤ role in the regulation of vascular tone in vivo) was not the force was measured in the -escin membrane–permeabi- 12 lized strips. inhibited during exposure to isoflurane. Therefore, we Results: Isoflurane (3–5%) inhibited the increases in both proposed that in subjects with intact endothelial func- 2؉ ␮ [Ca ]i and force induced by either norepinephrine (0.5 M)or tion, the direct action of isoflurane on mesenteric resis- KCl (40 mM). These inhibitions were similarly observed after tance arteries may not contribute to systemic hypoten- depletion of intracellular Ca2؉ stores by ryanodine. Regardless of the presence of ryanodine, after washout of isoflurane, its sion during isoflurane anesthesia. However, in the -2؉ absence of endothelium, contractile responses to norepi inhibition of the norepinephrine response (both [Ca ]i and force) was significantly prolonged, whereas that of the KCl nephrine and KCl were both inhibited during exposure response was quickly restored. In the ryanodine-treated strips, to isoflurane. Thus, we speculated that in subjects with 2؉ the norepinephrine- and KCl-induced increases in [Ca ]i were -؉ impaired endothelial function, the direct (i.e., endothe both eliminated by nifedipine, a voltage-gated Ca2 channel blocker, whereas only the former was inhibited by niflumic lium-independent) inhibitory action of isoflurane on acid, a Ca2؉-activated Cl؊ channel blocker. Isoflurane caused a mesenteric arterial VSM cells (VSMCs) may contribute to -rightward shift of the Ca2؉–force relation only in the fura-2– systemic hypotension during isoflurane anesthesia. Inter loaded strips but not in the ␤-escin–permeabilized strips. estingly, only the norepinephrine response, not the KCl Conclusions: In mesenteric resistance arteries, isoflurane de- response, was significantly inhibited for a while (15 min presses vascular smooth muscle reactivity by directly inhibiting -both Ca2؉ mobilization and myofilament Ca2؉ sensitivity. or more) after washout of isoflurane in either the pres Isoflurane inhibits both norepinephrine- and KCl-induced volt- ence or absence of endothelium, indicating that its en- ؉ age-gated Ca2 influx. During stimulation with norepinephrine, dothelium-independent inhibitory action on contractile 2؉ ؊ isoflurane may prevent activation of Ca -activated Cl chan- response to norepinephrine is prolonged.12 We thus -nels and thereby inhibit voltage-gated Ca2؉ influx in a pro longed manner. The presence of the plasma membrane appears proposed that the direct inhibitory action of isoflurane essential for its inhibition of the myofilament Ca2؉ sensitivity. on norepinephrine response might contribute to the previously observed prolonged systemic hypotension af- 13 ISOFLURANE produces systemic hypotension1 and sig- ter isoflurane anesthesia. nificantly alters distribution of blood flow to various Previous studies using isolated aorta and cultured aor- 4,5 organs.2 Previous studies3–11 performed in a variety of tic VSMCs have suggested that the direct inhibitory vascular beds have suggested that isoflurane causes action of isoflurane on VSM is a result of both reduction 2ϩ 2ϩ changes in vascular tone through its direct action on of the intracellular Ca concentration ([Ca ]i) and 2ϩ vascular smooth muscle (VSM) and/or endothelial cells. inhibition of the myofilament Ca sensitivity. However, In our previous study with isolated mesenteric resis- less information is available regarding its underlying tance arteries,12 in the presence of endothelium, con- mechanisms in systemic resistance arteries, which are different from conduit arteries in many of their properties, including Ca2ϩ-mobilization processes and respon- * Lecturer, † Research Associate, ‡ Postgraduate Student, § Professor and siveness to pharmacological agents.14–16 We previously Chair. proposed that the direct inhibitory action of isoflurane Received from the Department of Anesthesiology and Critical Care Medicine, 2ϩ Faculty of Medicine, Kyushu University, Fukuoka, Japan. Submitted for publica- on VSM is primarily a result of reduction of the [Ca ]i in tion February 17, 2003. Accepted for publication May 28, 2003. Supported in mesenteric resistance arteries.8 However, our proposal part by a Grant-in-Aid (B-09470330) from Ministry of Education, Science, Sports, and Culture, Japan (1997-1999), and Grants-in-Aid for Scientific Research (B- was based on the results obtained in experiments per- 09470330, C-13671590) from Japan Society of the Promotion of Science (1999- formed at approximately 22°C in the cell membrane– 2004, Tokyo, Japan). Presented in part at the 49th annual meeting of the Japanese 8 Society of Anesthesiologists, Fukuoka, Japan, April 18–20, 2002; to be presented permeabilized condition, in which we might have failed at the Annual Meeting of the American Society of Anesthesiologists, San Fran- to detect significant effects of isoflurane on myofilament cisco, California, October 11–15, 2003. Ca2ϩ sensitivity that require the intact cell membrane. In Address reprint requests to Dr. Akata: Department of Anesthesiology and Critical Care Medicine, Faculty of Medicine, Kyushu University, Fukuoka 812- addition, little information is available regarding the 8582, Japan. Address electronic mail to: [email protected]. In- mechanisms behind the aforementioned prolonged individual article reprints may be purchased through the Journal Web site, www.anesthesiology.org. hibitory action on norepinephrine response in mesen-

Anesthesiology, V 99, No 3, Sep 2003 666 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 667 teric resistance arteries.12 Furthermore, although there is Our method on the fura-2 fluorometry was also detailed direct evidence to indicate that isoflurane influences Ca2ϩ previously.22,23 Briefly, to allow loading of the fura-2 into mobilization from the intracellular stores in VSMCs of iso- the VSMCs, the strips were incubated in normal physio- 17 lated resistance arteries, no direct evidence is currently logic salt solution (PSS) containing 10 ␮M acetoxymethyl available to indicate that isoflurane inhibits plasmalemmal ester of fura-2 (fura-2/AM) and 2% albumin for approxi- Ca2ϩ influx in VSMCs of systemic resistance arteries. mately2hatapproximately 35°C. After this period, the In this study, using the fura-2 fluorometry and thereby solution containing fura-2/AM was washed out with nor- 2ϩ measuring force and [Ca ]i simultaneously, we further mal PSS for approximately1htoensure sufficient ester- investigated the mechanisms behind the direct inhibi- ification of fura-2/AM in the cells and to equilibrate the tory action of isoflurane on VSM in membrane-intact strips before the measurements.22,23 Changes in the flu- mesenteric resistance arteries. Specifically, we tested orescence intensity of the fura-2–Ca2ϩ complex were Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 two major hypotheses, i.e., (1) the hypothesis that isoflu- measured by a fluorometer equipped with a dual-wave- rane inhibits myofilament Ca2ϩ sensitivity of VSMCs in length excitation device (CAM-230; Japan Spectroscopic, the presence of intact cell membrane in mesenteric Tokyo, Japan) connected to the microscope with optical resistance arteries and (2) the hypothesis that isoflurane fibers. The VSM tissue was illuminated with ultraviolet directly inhibits plasmalemmal Ca2ϩ influx in VSMCs of lights at wavelengths of 340 and 380 nm alternatively mesenteric resistance arteries. Our experiments to test limited to a frequency of 1000 Hz. The fura-2 fluores- the latter hypothesis were performed in the presence of cence signals induced by excitation at 340 and 380 nm ryanodine, which has been shown to deplete the intra- were collected through the 10ϫ objective lens (Plan cellular Ca2ϩ stores of mesenteric arterial VSMCs used in Fluor; Nikon, Tokyo, Japan) and measured through a this study.17 We also tested the hypothesis that in mes- 500-nm filter with a photomultiplier. The background enteric resistance arteries, isoflurane causes the pro- fluorescence as excited by 340- and 380-nm ultraviolet longed inhibition of contractile response to norepineph- light was obtained after completion of each experiment rine by inhibiting Ca2ϩ mobilization in VSMCs. In this by breaking the cell membranes with Triton X-100 (1%) study, we also demonstrate a novel possibility that in and subsequently quenching the fura-2 fluorescence sig- 2ϩ VSMCs, isoflurane prevents activation of Ca -activated nals with MnCl2 (20 mM). The ratio (R340/380) of fura-2 Ϫ Cl channels and thereby inhibits norepinephrine-in- fluorescence intensities excited by 340 nm (F340)to 2ϩ duced Ca influx in a prolonged manner. those excited by 380 nm (F380) was calculated after the background fluorescence had been subtracted. None of the agents used during Ca2ϩ measurements influenced 2ϩ Materials and Methods the fluorescence signals. In all the Ca measurements, 23 as we showed previously, changes in F340 and F380 Tissue Preparation were constantly in opposite directions. All experiments With approval from the Kyushu University Animal Care with the fura-2–loaded strips were performed during the and Use Committee (Fukuoka, Japan), by use of the 18 period in which constant vascular responses were ob- method previously detailed, endothelium-denuded strips tained, i.e., for approximately 3 h.23 were prepared from the third- or fourth-order branches of In the next series of experiments, only isometric force male Sprague-Dawley rat (250–350 g, 7–10 W) mesenteric was measured in the non–fura-2–loaded endothelium- arteries. These branches are known to contribute to the 19,20 denuded strips, the smooth muscle membrane of which systemic vascular resistance. One strip was prepared was permeabilized with ␤-escin. To achieve the mem- from one animal. brane permeabilization, the strips were incubated with ␤-escin (50 ␮M for 25 min) at room temperature (approx- Force and Ca2ϩ Measurements imately 22°C) in relaxing solution after steady contrac- ϩ 8,18 Isometric force was measured by attaching the strip to a tions induced by 40 mM K had been measured. 8,18 strain-gauge transducer as previously detailed. Briefly, Ionomycin (0.3 ␮M) was present throughout the ␤-escin– the strip was mounted horizontally in a chamber attached permeabilized muscle experiments to eliminate the in- to the stage of a microscope, and the resting tension was fluence of intracellular Ca2ϩ stores. adjusted to obtain a maximal response to KCl. The solution To prevent early deterioration of the thin vascular was changed by infusing it into one end while aspirating strips, the aforementioned experiments with membrane- simultaneously from the other end. Removal of endothe- intact and ␤-escin membrane–permeabilized strips were lium was verified by the inability of acetylcholine (10 ␮M) performed at 35°C and room temperature (approximate- to cause significant (10% or more) relaxation during con- ly 22°C), respectively, as done previously.8,18 tractions induced by norepinephrine (10 ␮M). In the first series of experiments, changes in the Solutions and Drugs 2ϩ [Ca ]i were measured simultaneously with those in The ionic concentrations of the normal PSS were as 2ϩ 21 force by use of fura-2, a fluorescent Ca -indicator dye. follows (mM): NaCl 138, KCl 5.0, MgCl2 1.2, CaCl2 1.5,

Anesthesiology, V 99, No 3, Sep 2003 668 AKATA ET AL.

HEPES 10, and glucose 10. The pH was adjusted with norepinephrine) before and during the subsequent appli- NaOH to 7.35 at 22°C. The high-Kϩ solutions were cations of either stimulant until the steady-state effects prepared by replacing NaCl with KCl isoosmotically and were observed (for 15 min for KCl; for 59 min for norepi- 24 adding guanethidine (3 ␮M) to prevent norepinephrine nephrine). Our rationale to use these protocols was de- outflow from the sympathetic nerve terminals. The tailed previously.12,27 2ϩ Ca -free solution was prepared by removing CaCl2 with The underlying mechanisms of contractile response to or without adding EGTA. either KCl or norepinephrine could be different between The compositions of relaxing or activating solutions its initial phase (during development of force) and its sus- used in the ␤-escin–permeabilized muscle experiments tained phase (during maintenance of force). Therefore, in were determined by solving multiequilibrium equations some experiments, isoflurane was applied to the strip pre- using a hydrogen ion activity coefficient of 0.75 and contracted with KCl (40 mM) after the vascular response 2ϩ Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 association constants for the various ions as detailed ([Ca ]i and force) to KCl had reached a plateau. However, previously.8 The composition of the relaxing solution isoflurane was not applied to the strips precontracted with ␮ 2ϩ was 80 mM potassium methanesulfonate (KMS), 20 mM 0.5 M norepinephrine, because vascular response ([Ca ]i ␮ PIPES, 7 mM Mg(MS)2,5mM adenosine 5'-triphosphate, and force) to 0.5 M norepinephrine continued, gradually 10 mM creatinine phosphate, and 4 mM EGTA. The 4 mM diminishing progressively after the maximum was reached, EGTA–containing activating solution was prepared by and a plateau was not shaped for some time.

adding a specific amount of Ca(MS)2 to obtain the de- In the above-described experiments, we investigated the sired concentration of free Ca2ϩ ions based on the cal- vascular effects of 3 and 5% isoflurane, because these con- culations previously reported.8 The pH was adjusted centrations of isoflurane distinctly inhibited both norepi- with KOH to 7.00 at 22°C, and the ionic strength was nephrine- and KCl-induced contractions in our previous kept constant at 0.2 mM by adjusting the concentration experiments with the non–fura-2–loaded, endothelium-de- 12 of KMS. Guanosine 5'-triphosphate (50 ␮M) was present nuded strips prepared from this artery. As detailed below, throughout the experimental periods to minimize run- the aqueous concentrations produced by 3–5% isoflurane down of contractile responses in the ␤-escin–permeabil- in our experiments would be considered as anesthetic (i.e., ized strips.25 clinically relevant) concentrations. 2ϩ Nifedipine stock solution (1 mM) was prepared in 70% To investigate mechanisms of the observed [Ca ]i-re- ethanol under conditions of reduced illumination, ducing effects of isoflurane, we next attempted to charac- whereas niflumic acid was prepared as a stock solution terize the vascular response to either norepinephrine or 26 (10 mM) in dimethyl sulfoxide. KCl by evaluating the effects of ryanodine (10 ␮M), nifedi- Adenosine 5'-triphosphate, creatinine phosphate, pine (0.01–10 ␮M), SKF-96365 (0.3–10 ␮M), and niflumic guanosine 5'-triphosphate, HEPES, ionomycin, ␤-escin, nor- acid (3–100 ␮M) on the norepinephrine- or KCl-induced 2ϩ ␮ epinephrine, acetylcholine, nifedipine, and niflumic acid increases in [Ca ]i and force. Ryanodine (10 M,20min) 2ϩ were obtained from Sigma Chemical Co. EGTA, PIPES-K2, was previously shown to deplete the intracellular Ca and methanesulfonic acid were obtained from Fluka Che- stores (presumably sarcoplasmic reticulum, SR) in this mie AG. SKF-96365 was purchased from Calbiochem. Ry- mesenteric artery17 and thus was used to eliminate the anodine was purchased from Agri Systems International. influence of SR on the vascular responses in this study. Fura-2/AM was purchased from Dojindo Laboratories. Nifedipine is a selective blocker of the L-type VGCCs,28 Isoflurane was obtained from Dainabot Co. All other re- and SKF-96365 was reported previously to block recep- agents were of the highest grade commercially available. tor-operated Ca2ϩ channels (ROCCs).29,30 Niflumic acid has been reported to selectively inhibit Ca2ϩ-activated Ϫ Ϫ Experimental Design Cl (Cl Ca) currents without influencing the VGCC ac- In experiments with the fura-2–loaded strips, we first tivity in VSMCs.26,31 Preliminary experiments indicated 2ϩ examined the effects of isoflurane on increases in [Ca ]i that 5 min is sufficient for all of these channel blockers and force caused by norepinephrine or KCl, using proto- to exert their maximal effects on the response to nor- cols identical to those we used previously to examine the epinephrine or KCl. Thus, in the experiments with ni- direct action of isoflurane on this artery.12 Because the fedipine, SKF-96365, or niflumic acid, the strips were sympathetic nervous system plays a central role in the incubated with each blocker for 5 min before and during maintenance of resting vascular tone in vivo, norepineph- subsequent application of either norepinephrine or KCl. ␮ rine (0.5 M [EC40]) was chosen as a test stimulant as in our To investigate the effects of isoflurane on norepineph- 12 2ϩ previous study. Conversely, KCl (40 mM) was used as a rine-induced plasmalemmal Ca influx, we examined tool to activate VGCCs. Each stimulant was applied for 3 or the effects of isoflurane (3–5%) on the norepinephrine ␮ 2ϩ 5 min (3 min for KCl; 5 min for norepinephrine) at 7- or (0.5 M)–induced increases in [Ca ]i and force after 17 17-min intervals (7 min for KCl; 17 min for norepinephrine) treatment with ryanodine (10 ␮M, 20 min). so as to obtain reproducible responses, and then isoflurane To investigate the effects of isoflurane on myofilament was applied for 5 or 15 min (5 min for KCl; 15 min for Ca2ϩ sensitivity, we examined its effects on increases in

Anesthesiology, V 99, No 3, Sep 2003 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 669

2ϩ 35 [Ca ]i and force evoked by stepwise incremental in- [MAC] in this rat ) isoflurane. Accordingly, the aqueous 2ϩ 2ϩ creases in the extracellular Ca concentrations ([Ca ]e) concentration produced by 5% isoflurane would corre- ϩ from0to5mM during 40 mM K depolarization or those spond to its blood concentration in this rat under steady- 2ϩ 35 evoked by stepwise incremental increases in the [Ca ]e state anesthesia with 2.5% (1.67 MAC in this rat ) isoflu- from 0 to 1.5 mM during stimulation with norepinephrine rane. Because its partition coefficients in blood and Krebs (0.5 ␮M) in the fura-2–loaded strips. Our rationale to use solution at 37°C are 1.43 and 0.55, respectively, isoflurane these protocols has been detailed previously.23,32 In our is much more (approximately 2.6 times more) soluble in previous experiments12 as well as in the above-described blood than in the buffer solution. On the basis of calcula- experiments, 5 and 37 min were considered sufficient for tion using the blood/gas partition coefficient of isoflurane isoflurane to exert its maximal (i.e., steady-state) effects on (i.e., 1.43), its blood concentration would reach to 1.0 mM M ␮M the responses to KCl (40 m ) and norepinephrine (0.5 ), during steady-state anesthesia with 1.8% (approximately Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 respectively. Therefore, the strips were incubated for 1.5 MAC in human) isoflurane. Usually, 1.5 to 2.0 times 10 and 40 min before and during subsequent applications MAC is required to maintain anesthesia with only a single 2ϩ ϩ of Ca in these experiments with 40 mM K depolariza- inhalational agent and to suppress cardiovascular re- tion and norepinephrine stimulation, respectively. In the sponse to incision.36 We thus believe that the aqueous above-described experiments, both 3 and 5% isoflurane concentrations produced by 3–5% isoflurane in our ex- distinctly inhibited the KCl response, whereas only 5% periments can be considered anesthetic (i.e., clinically isoflurane consistently produced distinct inhibition of the relevant) concentrations. norepinephrine response. Thus, the effects of both 3 and 5% isoflurane were examined in the experiments with ϩ Calculation and Data Analysis 40 mM K depolarization, whereas only the effects of 5% ϩ Although absolute values of [Ca2 ] could be calcu- isoflurane were examined in the experiments with i lated on the basis of the fura-2 fluorescence ratio and the norepinephrine. ϩ dissociation constant of fura-2 for Ca2 binding obtained Finally, to further investigate the effects of isoflurane on 21 2ϩ ϩ in vitro, the dissociation constant of fura-2 for Ca myofilament Ca2 sensitivity, we also examined the effects ϩ binding in cytoplasm is significantly different (threefold of isoflurane on the Ca2 –force relation in the ␤-escin– to fourfold increase) from that measured in the absence permeabilized, non–fura-2–loaded strips in the absence of of protein, because more than half of the fura-2 mole- norepinephrine; ␣-adrenergic receptor coupling was not cules in cytoplasm are protein-bound.37 Therefore, we retained in the ␤-escin–permeabilized strips prepared from used the ratio of F to F (R / ), which was this artery, as reported previously.25,33 340 380 340 380 calculated after the background fluorescence had been 2ϩ subtracted, as an indicator of [Ca ]i. Isoflurane Delivery and Analysis Changes in the R340/380 and force were expressed as Isoflurane was delivered via a calibrated isoflurane vapor- the percentage value of the reference (i.e., values before izer (Forawick; Muraco Medical Co., Tokyo, Japan) in line application of isoflurane). The basal values in normal PSS with the air gas aerating the HEPES-buffered solutions. Each were assumed to be 0% in all experiments. In experiments solution was equilibrated with isoflurane for at least 15 min in which the strips were pretreated with isoflurane, its before introduction to the chamber, which was covered effects on the response to norepinephrine or KCl were with thin glass plates to prevent the equilibration gas from evaluated 3 or 5 min (3 min for KCl; 5 min for norepineph- escaping into the atmosphere. Using gas chromatography, rine) after application of each stimulant. In experiments in we previously reported concentrations of isoflurane in the which isoflurane was applied to the strips precontracted PSS produced by 0.5, 1.0, 1.5, and 3.0% isoflurane under with KCl, the effects of isoflurane were evaluated 20, 60, exactly the same experimental conditions,8,34 and the val- 120, and 180 s after its application. ues obtained were within 92% (91.8–98.5%) of theoretical The concentration–response data for the effects of values predicted by the partition coefficient of isoflurane in nifedipine, SKF-96365, and niflumic acid and the Ca2ϩ– Krebs solution at 37°C (0.55). Excellent linear relationship force relation in the ␤-escin–permeabilized strips were was obtained between the aqueous concentrations of fitted according to a four-parameter logistic model de- isoflurane (y) and its concentrations (vol%) in the gas mix- scribed by De Lean et al.38 The 50% inhibitory concen- ϭϪ ϩ ϭ 34 ture (x); y 0.0068 0.21x, r 0.998). Therefore, tration (IC50) and the 50% effective concentration (EC50) the concentrations produced by 1, 2, 3, and 5% isoflurane were derived from the least-squares fit by use of the in the PSS can be predicted as 0.21, 0.42, 0.63, and above-described model. Because the relationship be-

1.05 mM, respectively. The aqueous concentration pro- tween the R340/380 value (nonphysiologic value) and the 2ϩ duced by 3% isoflurane (i.e., 0.63 mM) is almost equal to a [Ca ]i is not theoretically linear, the R340/380 values previously reported concentration of isoflurane (0.65 were not transformed to a logarithmic scale on the x axis 13 mM) in blood sampled from this rat under steady-state in the representation of the R340/380–force relation- anesthesia with 1.5% (1 minimum alveolar concentration ship, and attempts were not made to fit the data for

Anesthesiology, V 99, No 3, Sep 2003 670 AKATA ET AL.

Fig. 1. Effects of isoﬂurane (ISO) on norepinephrine (NE)–induced increases in

force and R340/380 in either the absence (؊ryanodine) or presence (؉ryanodine)

of ryanodine. (A) An example of time-de- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 pendent effects of isoflurane on the norepinephrine response. The norepinephrine responses before application of isoflurane (control), 59 min after application of isoflurane, 15 min after washout of isoflurane, and 59 min after washout of isoflurane. (B) Examples of steady-state (i.e., maximal) effects of isoflurane on the norepinephrine response in either the absence (left) or presence (right) of ryanodine. (C) Analyzed data (left, in the absence of ryanodine; right, in the presence of ry- .P < 0.05 versus control* .(5 ؍ anodine) (n **P < 0.01 versus control.

the R340/380–force relation according to the above- cluding comparisons between the present and our previ- described logistic model. Because the relationship between ous data23 in the Discussion, were made by either the actual concentrations of isoflurane in the solutions and Student t test or Welch’s t test. As previously detailed,23 anesthetic concentrations (vol%) in the gas mixture is the- statistical analysis was not made in the overall comparison oretically linear, the anesthetic concentrations on the x axis of the R340/380–force relations. Alternatively, we at- are displayed as vol% for the isoflurane concentration– tempted to find data points in the isoflurane-treated 8,27 response relationships, as done previously. group where the increases in R340/380 are not significantly different from those of certain data points in the control group and then compare the force levels at

Statistics the certain R340/380 levels between the control and All results are expressed as mean Ϯ SD. The term n isoﬂurane-treated groups. A value of P Ͻ 0.05 was denotes the number of strips. Data were analyzed by one- considered signiﬁcant. or two-factor ANOVA, Scheffé F test, contrast, Student t test, and Welch’s t test. Comparisons among groups were Results performed by two-factor ANOVA for repeated measures.

When overall differences were detected, individual com- Effects of Isoflurane on Increases in R340/380 and parisons among groups at each time or concentration were Force Induced by Norepinephrine or KCl performed by either Scheffé F test (for multiple compari- Norepinephrine (0.5 ␮M) produced increases in both sons), two-tailed, unpaired Student t test (for comparison R340/380 and force in the normal PSS (fig. 1). The ryano- between two groups with homogeneous population vari- dine treatment slightly but significantly inhibited the ances), or Welch’s t test (for comparison between two norepinephrine-induced increases in both R340/380 and Ͻ ϭ Ϯ groups with heterogeneous population variances). Com- force (P 0.05; n 10; R340/380, 79.3 6.9% of control; parisons within each group were made by one-factor force, 96.7 Ϯ 4.4% of control). The norepinephrine-

ANOVA for repeated measures, and post hoc comparisons induced increases in both R340/380 and force were were made by use of the contrast for multiple comparisons. inhibited during application of isoﬂurane (3–5%) in All other necessary comparisons between two groups, in- either the presence or absence of ryanodine (ﬁg. 1).

Anesthesiology, V 99, No 3, Sep 2003 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 671

␮ Table 1. Norepinephrine (0.5 M)-induced Increases in R340/380 and Force 15 Minutes after Washout of Isoﬂurane (3–5%) from the Chamber in Either the Presence or Absence of Ryanodine

ϪRyanodine (n ϭ 4) ϩRyanodine (n ϭ 5) Concentration of

Isoﬂurane R340/380 Force R340/380 Force

3% 55.2 Ϯ 35.8 42.9 Ϯ 40.4* 56.7 Ϯ 33.3 39.4 Ϯ 38.9* 5% 46.7 Ϯ 20.7** 36.8 Ϯ 30.7* 39.8 Ϯ 46.6** 18.8 Ϯ 39.2*

Values are expressed as the percent value of the control (before application of isoﬂurane). * P Ͻ 0.05 vs. control (100%), ** P Ͻ 0.01 vs. control (100%). ϪRyanodine ϭ in the absence of ryanodine; ϩRyanodine ϭ in the presence of ryanodine. Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021

These inhibitions were still observed 15 min after decreases in R340/380 and force, respectively (fig. 3). Such washout of isoflurane from the chamber (table 1), and transient increases in R340/380 and force caused by 5% it took more than 15 min for VSMCs to recover from isoflurane were totally eliminated by the ryanodine treat- the inhibitions (fig. 1). ment (n ϭ 5). KCl (40 mM) also produced increases in both R340/380 and force in the normal PSS (fig. 2). The KCl-induced increase in force was inhibited by both 3 and 5% isoflu- Effects of Nifedipine, SKF-96365, and Niflumic Acid rane; however, the KCl-induced increase in R was 340/380 on Increases in [Ca2ϩ] (R ) Induced by inhibited only by 5% isoflurane but not by 3% isoflurane i 340/380 Norepinephrine (figs. 2 and 3). The ryanodine treatment did not significantly influence either the KCl-induced increases in both In the ryanodine-treated strips, both nifedipine and SKF-96365 significantly inhibited both the norepineph- R340/380 and force or the isoflurane-induced inhibitions of the KCl-induced increases in both R and force rine- and KCl-induced increases in R340/380 (figs. 4 and 5), 340/380 ␮ (n ϭ 5) (fig. 2). whereas niflumic acid (30–100 M) inhibited only the Isoflurane (3–5%) did not influence the basal levels of norepinephrine-induced, not KCl-induced, increase in R (fig. 6). There was no concentration range in either R340/380 or force; however, in the strips precon- 340/380 tracted with KCl, it produced small transient increases in which SKF-96365 selectively inhibited the norepineph-

R340/380 and force, which were followed by sustained rine-induced increase in R340/380 (ﬁg. 5).

Fig. 2. Effects of isoﬂurane (ISO) on KCl-

induced increases in force and R340/380 in either the absence or presence of ryanodine. (A) Examples of the effects of isoflurane on the KCl response. The KCl responses before application of isoflurane (precontrol), 5 min after application of isoflurane, and 5 min after washout of isoflurane (postcontrol) were shown. (B) Analyzed data (left, in the absence of ryanodine; right, in the presence of ryano- ;P < 0.01 versus control** .(5 ؍ dine) (n not significantly different from ؍ NS control.

Anesthesiology, V 99, No 3, Sep 2003 672 AKATA ET AL.

force were inhibited by 5% isoflurane, and the R340/380– force relation was shifted modestly to the right during exposure to isoflurane (fig. 8). Despite the identical (P Ͼ

0.05) increase in R340/380, the increases in force caused 2ϩ by 1.5 mM extracellular Ca after exposure to isoflurane (5%) were smaller (P Ͻ 0.05) than that caused by 0.5 mM extracellular Ca2ϩ before exposure to isoflurane (fig. 8).

Effects of Isoflurane on Ca2ϩ–Force Relation in ␤-Escin-Membrane–permeabilized Muscle In the ␤-escin–treated strips, the stepwise increment Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 of [Ca2ϩ] in the bath solution produced concentration- dependent increases in force (fig. 9). Isoflurane (5%) did not significantly influence the Ca2ϩ–force relation (fig. 9).

Discussion Using fura-2 fluorometry, this study, for the first time, provides direct evidence to indicate that isoflurane inhibits myofilament Ca2ϩ sensitivity and plasmalemmal Ca2ϩ influx during stimulation with either norepineph-

Fig. 3. Effects of isoﬂurane (ISO) on sustained increases in force and R340/380 caused by KCl. (A) Examples: Arrows indicate the transient increases in force and R340/380 observed immediately after application of isoﬂurane. (B, C) Analyzed data (left, force; ؍ right,R340/380)(n 6). *P < 0.05 versus control (100%) within each group. **P < 0.01 versus control (100%) within each group.

Effects of Isoflurane on R340/380–Force Relation in Fura-2–Loaded Muscle Stimulated with Either KCl or Norepinephrine In the fura-2–loaded strips, the stepwise increment of 2ϩ [Ca ]e during KCl depolarization produced concentration-dependent increases in both R340/380 and force (fig. 7). These increases in R340/380 were inhibited only by 5% isoflurane, not by 3% isoflurane; however, the increases in force were inhibited by both 3 and 5% isoflurane (fig.

7). As a result, the R340/380–force relation was shifted to the right after exposure to isoflurane (fig. 7). Despite the Ͼ identical (P 0.05) increases in R340/380, the increases in 2ϩ force caused by 3 and 5 mM extracellular Ca during exposure to 3% isoflurane were smaller (P Ͻ 0.05) than 2ϩ those caused by 3 and 5 mM extracellular Ca before exposure to isoflurane, respectively (fig. 7). Similarly, Ͼ regardless of the similar (P 0.05) increases in R340/380, 2ϩ Fig. 4. Effects of nifedipine on norepinephrine (NE)– and KCl- the increase in force caused by 3 mM extracellular Ca induced increases in force and R340/380 in the presence of ryan- -during exposure to 5% isoflurane was smaller (P Ͻ 0.05) odine (؉ryanodine). (A) Examples of steady-state (i.e., maxi 2ϩ than that produced by 1.5 mM extracellular Ca before mal) effects of nifedipine on the responses to either norepinephrine (left) or KCl (right). (B) Analyzed data (left, ؍ .(exposure to isoflurane (fig. 7 force; right,R340/380)(n 4). *P < 0.05 versus control (100%) In the fura-2–loaded strips, the stepwise increment of within each group. The IC50 values for the inhibitions of nore- 2ϩ pinephrine-induced increases in force and R were 3.4 and [Ca ]e during stimulation with norepinephrine also 340/380 6.8 nM, respectively, whereas those for the inhibitions of KCl- produced concentration-dependent increases in R ␮ 340/380 induced increases in force and R340/380 were 0.016 and 0.12 M, and force (fig. 8). These increases in both R340/380 and respectively.

Anesthesiology, V 99, No 3, Sep 2003 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 673

isoflurane on the R340/380–force relation suggest that the depressed contractile response to either KCl or norepinephrine during exposure to 5% isoflurane is a result of 2ϩ both reduction of the [Ca ]i and inhibition of the myofilament Ca2ϩ sensitivity. In this artery, treatment with ryanodine (10 ␮M) de- pletes not only the caffeine/ryanodine–sensitive SR but also the norepinephrine/inositol 1,4,5-triphosphate–sensitive SR.17,32 Thus, the norepinephrine-induced in-

crease in R340/380 observed after ryanodine treatment is 2ϩ

presumably because of plasmalemmal Ca inﬂux. The Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 effects of nifedipine on either norepinephrine- or KCl-

induced increases in R340/380 observed after ryanodine treatment suggest that not only the KCl-induced but also 2ϩ the norepinephrine (0.5 ␮M)–induced Ca inﬂuxes are attributable exclusively to activation of the VGCCs. This is consistent with the previously reported threshold concentrations of norepinephrine for membrane depolariza- 39–41 tion in small mesenteric arteries (i.e., 0.3–1 ␮M) and the recent recognition that VGCCs play a major role, whereas ROCCs play only a minor role, in the norepi-

Fig. 5. Effects of SKF-96365 on norepinephrine (NE)– and KCl-

induced increases in force and R340/380 in the presence of ryan- -odine (؉ryanodine). (A) Examples of steady-state (i.e., maxi mal) effects of SKF-96365 on the responses to either norepinephrine (left) or KCl (right). (B) Analyzed data (left, ؍ force; right,R340/380)(n 4). *P < 0.05 versus control (100%) within each group. **P < 0.01 versus control (100%) within each group. The IC50 values for the inhibitions of norepineph- ␮ rine-induced increases in force and R340/380 were 1.8 and 2.5 M, respectively, whereas those for the inhibitions of KCl-induced ␮ increases in force and R340/380 were 1.3 and 5.3 M, respectively.

rine or KCl in VSMCs of systemic resistance arteries. In addition, this study demonstrates for the first time the novel possibility that isoflurane prevents activation of Ϫ the Cl Ca channels and thereby causes a prolonged inhibition of norepinephrine-activated signaling pathway in VSMCs. Because this study was designed to gain ac- cess to the mechanisms behind the direct (i.e., endothelium-independent) inhibitory action of isoflurane on mesenteric arterial VSMCs, the following discussion fo- cuses on its underlying mechanisms but not its relevance to the overall circulatory effects of isoflurane in vivo, which was already discussed in our previous article.12 The results obtained indicate that in rat mesenteric resistance arteries, isoflurane at anesthetic concentra- Fig. 6. Effects of niflumic acid on norepinephrine (NE)– and 2ϩ KCl-induced increases in force and R340/380 in the presence of tions directly inhibits VSM reactivity by inhibiting Ca ؉ ϩ ryanodine ( ryanodine). (A) Examples of steady-state (i.e., mobilization and/or myofilament Ca2 sensitivity. Spe- maximal) effects of niflumic acid on the responses to either norepinephrine (left) or KCl (right). (B) Analyzed data (left, ؍ -cifically, the low concentration (3%) of isoflurane pre force; right,R340/380)(n 4). *P < 0.05 versus control (100%) sumably depresses contractile response to KCl exclu- within each group. **P < 0.01 versus control (100%) within 2ϩ sively by inhibiting the myofilament Ca sensitivity, each group. The IC50 values for the inhibitions of norepinephrine-induced increases in force and R340/380 were 15.7 and whereas it appears to depress contractile response to ␮ 2ϩ 22.5 M, respectively, whereas the IC50 value for the inhibitions of norepinephrine at least in part by reducing the [Ca ]i. KCl-induced increases in force was 19.5 ␮M. Note that the KCl-

The observed effects of the high concentration (5%) of induced increases in R340/380 were not signiﬁcantly inhibited.

Anesthesiology, V 99, No 3, Sep 2003 674 AKATA ET AL.

Fig. 7. Effects of isoﬂurane on increases

in R340/380 (A) and force (B) caused by incremental increases in the extracellu- 2؉ 2؉ lar Ca concentration ([Ca ]e) during KCl (40 mM) depolarization in the fura-2– loaded strips. After the control responses to KCl (40 mM) had been recorded, the strips were treated with isoﬂurane for 10 min before and during subsequent application of various concentrations of Ca2؉ Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 during the depolarization. In these anal-

yses, the maximal increase in either R340/ 380 or force induced by KCl before exposure to isoﬂurane was assumed to be 100% and their basal values in normal physiologic salt solution 0%. *P < 0.05 -versus control at each Ca2؉ concentra C) Effects of isoﬂurane on) .4 ؍ tion. n

the R340/380–force relation. The R340/380– force relation either in the absence or in the presence of isoﬂurane was constructed from the data shown in A and B.#P < 0.05 versus a in the control group. †P < 0.05 versus b in the control group. ‡P < 0.05 versus c in the control group.

Fig. 8. Effects of isoﬂurane on increases in

R340/380 (A) and force (B) caused by incremental increases in the extracellular Ca2؉ 2؉ concentration ([Ca ]e) during stimulation with norepinephrine (0.5 ␮M) in the fura- 2–loaded strips. After the control responses to KCl (40 mM) had been recorded, the strips were treated with isoﬂurane for 40 min before and during subsequent application of various concentrations of Ca2؉ in the presence of norepinephrine. In these analyses, the maximal increase in either R340/380 or force induced by KCl before exposure to isoﬂurane was assumed to be 100% and their basal values in normal physiologic salt solution 0%. P < 0.05 versus control at each Ca2؉* -C) Effects of iso) .4 ؍ concentration. n

ﬂurane on the R340/380–force relation in the presence of norepinephrine. The

R340/380–force relation either in the absence or in the presence of isoﬂurane was constructed from the data shown in A and B. ‡P < 0.05 between the control and isoﬂurane groups.

Anesthesiology, V 99, No 3, Sep 2003 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 675

Fig. 9. Effects of isoflurane on Ca2؉–force relation in the ␤-escin membrane–permeabilized strips. (A) An example of the effect of 5% isoflurane on the Ca2؉–force relation. Note that isoflurane influences -the Ca2؉–force relation very little. (B)An alyzed data. The Ca2؉–force relation was assessed after the normalization of the Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 Ca2؉-activated contraction to 100%. The 2؉ EC50 values for the Ca –force relations before (control) and after exposure to isoflurane were 1.9 and 2.4 ␮M, respec- .4 ؍ tively. n

nephrine-induced Ca2ϩ influx in rat mesenteric arterial cells.46,47 Thus, in the mesenteric arterial cells, arteries.42 isoflurane also may inhibit the activation of VGCCs and Ϫ The ability of niflumic acid, a Cl Ca channel block- thereby depress both the norepinephrine- and KCl-induced er,26,31 to selectively inhibit the norepinephrine-induced Ca2ϩ influx. However, the observed difference in recovery increase in R340/380 in the presence of ryanodine indi- time course after washout of isoflurane between the nor- cates that activation of the ClϪ channels is also in- Ca epinephrine- and KCl-induced increases in R340/380 (after 2ϩ volved in the norepinephrine-induced Ca influx. In ryanodine treatment) indicates that the mechanisms be- Ϫ VSM, activation of the Cl channels would result in hind the isoflurane-induced inhibition of the norepineph- membrane depolarization and hence activation of 2ϩ Ϫ rine-induced Ca influx are at least in part different from VGCCs, because the Cl equilibrium potential (i.e.,ap- those behind its inhibition of the KCl-induced Ca2ϩ influx. Ϫ Ϫ Ϫ proximately 20 to 30 mV) is more positive than both Because activation of the Cl channels appears to be Ϫ Ca the resting membrane potential (i.e., approximately 50 involved in the norepinephrine-induced but not the KCl- to Ϫ70 mV) and the threshold potential for opening 2ϩ 43 induced Ca influx, isoflurane may prevent the activation VGCCs (i.e., approximately Ϫ60 to Ϫ45 mV). Thus, as Ϫ 26,44,45 of Cl Ca channels and hence the norepinephrine-induced has been proposed in some vascular preparations, ϩ Ca2 influx through VGCCs in a prolonged manner. In norepinephrine presumably activates the ClϪ channels Ca other words, isoflurane may act at some steps between and thereby depolarizes the cell membrane, leading to Ϫ receptor binding and activation of the Cl channels to the opening of VGCCs and hence contraction. In con- Ca cause a prolonged inhibition of the norepinephrine-acti- trast, KCl directly depolarizes the cell membrane and vated signaling pathway. Because of its high lipophilicity, thereby activates VGCCs. Previous studies have suggested that during stimulation with norepinephrine, isoflurane may remain present in the VSMCs for some time 2ϩ Ϫ 44 after its removal from the extracellular space to exert such Ca released from SR activates the Cl Ca channels. Ϫ a prolonged effect. However, we speculate that the Cl Ca channels were presumably activated by Ca2ϩ entering the cells through Ryanodine strongly inhibited the initial phase of the nore- the nonselective cation channels during stimulation with pinephrine-induced increase in R340/380 but only modestly norepinephrine in our experiments with the ryanodine- inhibited its tonic phase, as shown previously in this ar- 23,32 treated strips. tery. This suggests that the norepinephrine-induced 2ϩ 2ϩ The ability of isoflurane to inhibit both the norepineph- initial increase in [Ca ]i is attributable primarily to Ca 2ϩ rine- and KCl-induced increases in R in the presence release from SR, whereas its tonic increase in [Ca ]i is 340/380 ϩ of ryanodine indicates that isoflurane inhibits both the primarily a result of the plasmalemmal Ca2 influx. Because norepinephrine- and KCl-induced plasmalemmal Ca2ϩ in- isoflurane does not seem to influence the norepinephrine- ϩ fluxes, which are presumably through VGCCs, as discussed induced Ca2 release from SR in this artery,17 the isoflu- above. Isoflurane was previously shown to inhibit the rane-induced inhibition of the norepinephrine-induced in- 2ϩ 2ϩ whole cell L-type Ca currents in cerebral or coronary crease in [Ca ]i observed in the absence of ryanodine was

Anesthesiology, V 99, No 3, Sep 2003 676 AKATA ET AL. probably exclusively a result of its inhibition of the norepi- tracellular regulatory mechanisms of contractile proteins nephrine-induced Ca2ϩ influx. that are impaired in the ␤-escin–permeabilized strips. It The inability of ryanodine to influence the KCl-induced is currently unclear whether isoflurane inhibits only the 2ϩ 2ϩ increase in R340/380 suggests that the KCl-induced in- Ca activation of contractile proteins (i.e., basal Ca 2ϩ crease in [Ca ]i is exclusively a result of activation of sensitivity of contractile proteins) or inhibits both the the plasmalemmal Ca2ϩ influx, consistent with our pre- basal Ca2ϩ sensitivity and the norepinephrine-induced vious studies in this artery.8,23,32 Because isoflurane Ca2ϩ sensitizing mechanism(s). Further investigations seems to enhance the caffeine-induced Ca2ϩ release are necessary to clarify this issue. (presumably Ca2ϩ-induced Ca2ϩ release) from SR in this More than half of the fura-2 molecules in myoplasma are artery,17 if the Ca2ϩ-induced Ca2ϩ release mechanism in a protein-bound form,37 possibly influencing responsive- operates significantly during the KCl response, the ryan- ness or sensitivity of the VSMCs to isoflurane. However, no Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 odine treatment would enhance the isoflurane-induced significant differences were found in its steady-state effects inhibition of KCl-induced increase in R340/380. However, on contractile responses to either norepinephrine or KCl it was not influenced by the ryanodine treatment, sug- between the fura-2–loaded and nonloaded strips (accord- gesting that the isoflurane-induced inhibition of the KCl- ing to comparison between our previous data12 and the 2ϩ induced increase in [Ca ]i is exclusively a result of its present data). Thus, the fura-2 loading, a nonphysiologic effect on the plasmalemmal Ca2ϩ influx. The Ca2ϩ-in- intervention as well, does not seem to significantly influ- duced Ca2ϩ release mechanism may play only a minor, if ence the action of isoflurane on VSMCs. any, role in the KCl response in this artery. The direct vascular action of isoflurane observed in this Isoflurane has been suggested to increase vascular tone study was substantially identical to that of sevoflurane pre- through its direct effects on SR,8,9 plasmalemmal Ca2ϩ viously observed in this artery.23 In our experiments, the influx,48 or the protein kinase C system.11,49 In this study, aqueous concentrations produced by 3% isoflurane (i.e., isoflurane caused small transient increases in R340/380 and 0.63 mM) and 5% sevoflurane (i.e., 0.67 mM) are almost force during stimulation with KCl (but not in the resting equal to the respective anesthetic concentrations in blood state). The observed increases in R340/380 and force were sampled from this rat under steady-state anesthesia at 1 eliminated by the ryanodine treatment, indicating that the MAC (1.5% for isoflurane,35 2.8% for sevoflurane51), i.e., 2ϩ 13 isoflurane-induced contraction results from the Ca re- 0.65 and 0.66 mM, respectively. We thus compared the lease from SR, consistent with the previous proposal in rat effect of 3% isoflurane on contractile response to either aorta.9 Because the amount of Ca2ϩ stored in SR would be norepinephrine or KCl with that of 5% sevoflurane ob- increased during stimulation with KCl, which activates served in our previous study.23 No significant difference (P voltage-gated Ca2ϩ influx and thereby causes a continuous Ͼ 0.05) was observed in the inhibition of the norepineph- 2ϩ ␮ increase in the [Ca ]i, the inability of isoflurane to cause rine (0.5 M)–induced increase in either force or R340/380 significant increases in R340/380 in the resting state might be between 3% isoflurane and 5% sevoflurane, suggesting that explained by possibly decreased amounts of Ca2ϩ in SR. In the inhibitory action of isoflurane on the norepinephrine a recent study using membrane-permeabilized rabbit pul- response is identical to that of sevoflurane compared at the monary arteries,49 isoflurane caused transient contraction equivalent MAC. However, a significant difference was similar to that observed in this artery. However, on the found in the inhibition of the KCl response between 3% basis of sensitivity to protein kinase C inhibitors, the con- isoflurane and 5% sevoflurane: namely, although no signif- traction in the pulmonary artery appeared to be caused by icant difference was observed in the inhibition of the KCl- protein kinase C activation.49 The difference might be ex- induced increase in force between 3% isoflurane and 5% plained by the species or regional differences. sevoflurane (P Ͼ 0.05), only 5% sevoflurane, not 3% isoflu- On the basis of sensitivity to SKF-96365, a putative rane, significantly inhibited the KCl-induced increase in 29,30 inhibitor of ROCCs, isoflurane was previously pro- R340/380. Thus, the mechanisms behind the depressed con- posed to enhance contractile response to phenylephrine tractile response to KCl appear different between 3% isoflu- by stimulating Ca2ϩ influx through ROCCs in rat aorta.48 rane and 5% sevoflurane. However, our results on SKF-96365 indicate that SKF- In conclusion, isoflurane directly inhibits VSM reactiv- 96365 does not serve as a selective inhibitor of ROCCs, ity by inhibiting both Ca2ϩ mobilization and myofilament consistent with the recent recognition that selective Ca2ϩ sensitivity in systemic resistance arteries. Isoflurane inhibitors of ROCCs or nonselective cation channels are inhibits plasmalemmal Ca2ϩ influx through VGCCs dur- not currently available.50 ing stimulation with either KCl or norepinephrine. How- The observed difference in the effect of isoflurane on ever, the mechanisms behind its inhibition of the volt- the Ca2ϩ–force relation between the fura-2–loaded mem- age-gated Ca2ϩ influx during stimulation with KCl are brane-intact and the ␤-escin–permeabilized conditions presumably at least in part different from those during suggests that its inhibition of the Ca2ϩ activation of stimulation with norepinephrine. During stimulation contractile proteins is presumably mediated by the intact with norepinephrine, isoflurane may prevent activation Ϫ 2ϩ plasma membrane or because of an effect on some in- of the Cl Ca channels and thereby inhibit the Ca influx

Anesthesiology, V 99, No 3, Sep 2003 DIRECT ACTION OF ISOFLURANE ON VASCULAR SMOOTH MUSCLE 677 through VGCCs in a prolonged manner. The presence of 24. Noguera I, Medina P, Segarra G, Martinez MC, Aldasoro M, Vila JM, Lluch S: Potentiation by vasopressin of adrenergic vasoconstriction in the rat isolated the plasma membrane appears essential for isoflurane to mesenteric artery. Br J Pharmacol 1997; 122:431–8 ϩ exert its inhibitory action on the Ca2 -induced activation 25. Akata T, Boyle WA III: Is guanosine-5'-triphosphate involved in calcium- activation of contractile proteins in vascular smooth muscle? Jpn J Pharmacol of contractile proteins. 1997; 75: 1–12 26. Criddle DN, de Moura RS, Greenwood IA, Large WA: Inhibitory action of The authors thank Masae Yamakawa, Research Assistant, Kyushu University, niflumic acid on noradrenaline- and 5-hydroxytryptamine-induced pressor re- Fukuoka, Japan, for her kind assistance in this work. sponses in the isolated mesenteric vascular bed of the rat. Br J Pharmacol 1997; 120:813–8 27. Izumi K, Akata T, Takahashi S: The action of sevoflurane on vascular smooth muscle of isolated mesenteric resistance arteries: I. Role of endothelium. References ANESTHESIOLOGY 2000; 92:1426–40 28. Xiong Z, Sperelakis N: Regulation of L-type calcium channels of vascular 1. Kazama T, Ikeda K: The comparative cardiovascular effects of sevoflurane smooth muscle. J Mol Cell Cardiol 1995; 27:75–91 with halothane and isoflurane. J Anesth (Tokyo) 1988; 2:63–8 29. Hughes AD, Schachter M: Multiple pathways for entry of calcium and Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/99/3/666/336698/0000542-200309000-00023.pdf by guest on 29 September 2021 2. Conzen PF, Vollmar B, Habazettl H, Frink EJ, Peter K, Messmer K: Systemic other divalent cations in a vascular smooth muscle cell line (A7r5). Cell Calcium and regional hemodynamics of isoflurane and sevoflurane in rats. Anesth Analg 1994; 15:317–30 1992; 74:79–88 30. Clementi E, Meldolesi J: Pharmacological and functional properties of 3. Blaise G, Sill JC, Nugent M, Van Dyke AR, Vanhoutte PM: Isoflurane causes ϩ voltage-independent Ca2 channels. Cell Calcium 1996; 19:269–79 endothelium-dependent inhibition of contractile responses of canine coronary 31. Hogg RC, Wang Q, Large WA: Action of niflumic acid on evoked and arteries. ANESTHESIOLOGY 1987; 67:513–7 spontaneous calcium-activated chloride and potassium currents in smooth mus- 4. Sill JC, Eskuri S, Nelson R, Van Dyke RA, Tarara J: The volatile anesthetic isoflurane attenuates Ca2ϩ mobilization in cultured vascular smooth muscle cells. cle cells from rabbit portal vein. Br J Pharmacol 1994; 112:977–84 J Pharmacol Exp Ther 1993; 265:74–80 32. Akata T, Izumi K, Nakashima M: Mechanisms of direct inhibitory action of 5. Tsuchida H, Namba H, Yamakage M, Fujita S, Notsuki E, Namiki A: Effects ketamine on vascular smooth muscle in mesenteric resistance arteries. ANESTHE- of halothane and isoflurane on cytosolic calcium ion concentrations and contrac- SIOLOGY 2001; 95:452–62 tion in vascular smooth muscle of the rat aorta. ANESTHESIOLOGY 1993; 78:531–40 33. Akata T, Boyle WA III: Dual actions of halothane on intracellular calcium 6. Park KW, Dai HB, Lowenstein E, Darvish MA, Sellke FW: Heterogeneous stores of vascular smooth muscle. ANESTHESIOLOGY 1996; 84:580–95 vasomotor responses of rabbit coronary microvessels to isoflurane. ANESTHESIOL- 34. Akata T, Kodama K, Takahashi S: Volatile anesthetic actions on norepi- OGY 1994; 81:1190–7 nephrine-induced contraction of small splanchnic resistance arteries. Can J An- 7. Akata T, Nakashima M, Kodama K, Boyle WA III, Takahashi S: Effects of aesth 1995; 42:1040–50 volatile anesthetics on ACh-induced relaxation in the rabbit mesenteric resistance 35. Drummond JC: MAC for halothane, enflurane, and isoflurane in the New artery. ANESTHESIOLOGY 1995; 82:188–204 Zealand white rabbit: And a test for the validity of MAC determinations. ANESTHE- 8. Akata T, Boyle WA III: Volatile anesthetic actions on contractile proteins in SIOLOGY 1985; 62:336–8 membrane-permeabilized small mesenteric arteries. ANESTHESIOLOGY 1995; 36. Miller FL, Marshall BE: The inhaled anesthetics, Introduction to Anesthesia, 82:700–12 8th edition. Edited by Longnecker DE, Murphy FL. Philadelphia, WB Saunders, 9. Namba H, Tsuchida H: Effect of volatile anesthetics with or without vera- 1992, pp 77–90 pamil on intracellular activity in vascular smooth muscle. ANESTHESIOLOGY 1996; 37. Konishi M, Olson A, Hollingworth S, Baylor SM: Myoplasmic binding of 84:1465–74 fura-2 investigated by steady-state fluorescence and absorbance measurements. 10. Su JY: Mechanisms of action of isoflurane on contraction of rabbit conduit Biophys J 1988; 54:1089–104 artery. ANESTHESIOLOGY 1996; 82:837–42 38. De Lean AP, Munson PJ, Rodbard D: Simultaneous analysis of families of 11. Toda H, Su JY: Mechanisms of isoflurane-increased submaximum Ca2ϩ- sigmoidal curves: Application to bioassay, radioligand assay, and physiological activated force in rabbit skinned femoral arterial strips. ANESTHESIOLOGY 1998; dose-response curves. Am J Physiol 1978; 235:E97–102 89:731–40 39. Kuriyama H, Suzuki H: The effects of acetylcholine on the membrane and 12. Izumi K, Akata T, Takahashi S: Role of endothelium in the action of contractile properties of smooth muscle cells of the rabbit superior mesenteric isoflurane on vascular smooth muscle of isolated mesenteric resistance arteries. artery. Br J Pharmacol 1978; 64:493–501 ANESTHESIOLOGY 2001; 95:990–8 40. Makita Y, Kanmura Y, Itoh T, Suzuki H, Kuriyama H: Effects of nifedipine 13. Yamazaki M, Stekiel TA, Bosnjak ZJ, Kampine JP, Stekiel WJ: Effects of derivatives on smooth muscle cells and neurotransmission in the rabbit mesen- volatile anesthetic agents on in situ vascular smooth muscle transmembrane teric artery. Naunyn Schmiedebergs Arch Pharmacol 1983; 324:302–12 potential in resistance- and capacitance-regulating blood vessels. ANESTHESIOLOGY 41. Mulvany MJ, Nilsson H, Flatman JA: Role of membrane potential in the 1998; 88:1085–95 response of rat small mesenteric arteries to exogenous noradrenaline stimulation. 14. Bevan JA, Laher I, Rowan R: Some implications of the intrasynaptic J Physiol (London) 1982; 332:363–73 norepinephrine concentrations in resistance arteries. Blood Vessels 1987; 24: 42. Nilsson H, Jensen PE, Mulvany MJ: Minor role for direct adrenoceptor- 137–40 2ϩ mediated calcium entry in rat mesenteric small arteries. J Vasc Res 1994; 31: 15. Cauvin C, Weir SW, Buhler FR: Differences in Ca handling along the 314–21 arterial tree: An update including studies in human mesenteric resistance vessels. 43. Greenwood IA, Large WA: Properties of a ClϪ current activated by cell J Cardiovasc Pharmacol 1988; 12:S10–S5 swelling in rabbit portal vein vascular smooth muscle cells. Am J Physiol 1998; 16. Nagao T, Illiano SC, Vanhoutte PM: Heterogenous distribution of endothe- 275:H1524–32 lium-dependent relaxations resistant to nitro-L-arginine in the arterial tree of the 44. Nilsson H: Interactions between membrane-potential and intracellular rat. Am J Physiol 1992; 263:H1090–4 calcium concentration in vascular smooth muscle. Acta Physiol Scand 1998; 17. Akata T, Nakashima M, Izumi K: Comparison of volatile anesthetic actions 164:559–66 on intracellular calcium stores of vascular smooth muscle. ANESTHESIOLOGY 2001; 94:840–50 45. Kitamura K, Yamazaki J: Chloride channels and their functional roles in 18. Akata T, Yoshitake J, Nakashima M, Itoh T: Effects of protamine on smooth muscle tone in the vasculature. Jpn J Pharmacol 2001; 85:351–7 46. Buljubasic N, Rusch NJ, Marijik J, Kampine JP, Bosnjak ZJ: Effects of vascular smooth muscle of rabbit mesenteric artery. ANESTHESIOLOGY 1991; 75: 833–46 halothane and isoflurane on calcium currents in canine coronary arterial cells. 19. Christensen KL, Mulvany MJ: Mesenteric arcade arteries contribute sub- ANESTHESIOLOGY 1992; 76:990–8 stantially to vascular resistance in conscious rats. J Vasc Res 1993; 30:73–9 47. Buljubasic N, Flynn NM, Marijik J, Rusch NJ, Kampine JP, Bosnjak ZJ: ϩ 2ϩ 20. Stekiel TA, Stekiel WJ, Bosnjak ZJ: The peripheral vasculature: Control and Effects of isoflurane on K and Ca conductance in isolated smooth muscle anesthetic actions, Anesthesia: Biologic Foundations, 1st edition. Edited by Yaksh cells of canine cerebral arteries. Anesth Analg 1992; 75:590–6 TL, Lynch C III, Zapol WM, Maze M, Biebuyck JF, Saidman LJ. New York, 48. Hirata S, Enoki T, Kitamura R, Vinh VH, Nakamura K, Mori K: Effects of 2ϩ Lippincott–Raven, 1997, pp 1135–68 isoflurane on receptor-operated Ca channels in rat aortic smooth muscle. Br J 21. Grynkiewicz G, Poenie M, Tsien RY: A new generation of Ca2ϩ indicators Anaesth 1998; 81:578–83 with greatly improved fluorescence properties. J Biol Chem 1985; 260:3440–50 49. Su JY, Vo AC: Role of PKC in isoflurane-induced biphasic contraction in 22. Akata T, Kodama K, Evers A, Takahashi S: Protamine relaxes vascular skinned pulmonary arterial strips. ANESTHESIOLOGY 2002; 96:155–61 ϩ smooth muscle by directly reducing cytosolic calcium concentrations in small 50. Gibson A, McFadzean I, Wallace P, Wayman CP: Capacitative Ca2 entry resistance arteries. J Anesth (Tokyo) 1996; 10:252–9 and the regulation of smooth muscle tone. Trends Pharmacol Sci 1998; 19:266–9 23. Akata T, Izumi K, Nakashima M: The action of sevoflurane on vascular 51. Taheri S, Halsey MJ, Liu J, Eger EI II, Koblin DD, Laster MJ: What solvent smooth muscle of isolated mesenteric resistance arteries: II. Mechanisms of best represents the site of action of inhaled anesthetics in humans, rats, and dogs? endothelium-independent vasorelaxation. ANESTHESIOLOGY 2000; 92:1441–53 Anesth Analg 1991; 72:627–34

Anesthesiology, V 99, No 3, Sep 2003